chapter 25 acute respiratory failure and ventilatory ... · acute respiratory failure and...

Acute Respiratory Failure and Ventilatory Management

667

GERALD HARRINGTON, M.D.*; WILLIAM MARX, D.O.†; AND DANIEL P. STOLTZFUS, M.D.‡

Chapter 25

ACUTE RESPIRATORY FAILURE ANDVENTILATORY MANAGEMENT

*Lieutenant Colonel, Medical Corps, U.S. Army; Department of Surgery, Critical Care Service, Brooke Army Medical Center, San Antonio, Texas 78234-6200†Lieutenant Colonel, Medical Corps, U.S. Army Reserve; formerly, Director, Surgical Critical Care and Staff, General Surgery Service, Fitzsimons Army MedicalCenter, Aurora, Colorado 80045-5001; currently, Assistant Professor of Surgery and Critical Care Medicine, Interim Director, Surgical Critical Care, StateUniversity of New York Health Science Center of Syracuse, 750 East Adams Street, Syracuse, New York 13210

‡Formerly, Major, Medical Corps, U.S. Army; Staff Anesthesiologist, Staff Intensivist, Anesthesia and Operative Services, Walter Reed Army Medical Center,Washington, D. C. 20307-5001; currently, Assistant Professor of Anesthesiology, University of Florida College of Medicine, Gainesville, Florida 32608-1197; Co-director, Surgical Intensive Care Unit, Veterans Affairs Medical Center, P. O. Box 100254, Gainesville, Florida 32610-0254

INTRODUCTION

EPIDEMIOLOGY OF RESPIRATORY FAILURE IN COMBAT CASUALTY CARE

ETIOLOGY OF PULMONARY INSUFFICIENCYMechanical TraumaThermal TraumaChemical WarfareNuclear Warfare

INDIRECT MECHANISMS OF RESPIRATORY INSUFFICIENCYPulmonary AspirationHydrostatic Pulmonary EdemaAdult Respiratory Distress Syndrome

DIAGNOSING RESPIRATORY FAILUREHypoxemiaVentilatory FailureCombined Hypoxemia and Ventilatory Failure in Postoperative Casualties

MANAGING RESPIRATORY FAILURESupportive CareEndotracheal Intubation

PHYSIOLOGICAL EFFECTS OF MECHANICAL VENTILATIONPulmonary EffectsCardiovascular EffectsRenal Effects

COMPLICATIONS FROM MECHANICAL VENTILATORY SUPPORT

ETHICAL DILEMMAS IN USING LIMITED LIFE-SUSTAINING THERAPY

MECHANICAL VENTILATION AND COMBAT CASUALTIESIndications for Mechanical VentilationFunctional Design of Mechanical VentilatorsModes of Positive-Pressure Ventilation

INITIATING MECHANICAL VENTILATIONTidal Volume and Respiratory RateFraction of Inspired OxygenInspiratory Flow RateAdditional SettingsStabilizing the Patient in Acute Respiratory Failure

MONITORING THE EFFECTIVENESS OF VENTILATION AND OXYGENATIONOxygen ConcentrationPulse OximetryGas-Exchange IndicesMonitoring Mechanical Ventilation

WEANING FROM MECHANICAL VENTILATION

VENTILATORS AVAILABLE FOR COMBAT CASUALTY CARE

SUMMARY

Anesthesia and Perioperative Care of the Combat Casualty

668

INTRODUCTION

Respiratory insufficiency in combat casualties isseen in many different guises depending on themechanism and the time of appearance (Exhibit 25-1):

• In some casualties, respiratory insufficiencyis an immediate and direct consequence ofphysical trauma to the lungs or the chestwall. An example might be a soldier who isshot through the chest and sustains a massivetension pneumothorax that will cause deathwithin a few minutes unless immediate careis rendered by a medic or a medical officer.

• In other casualties, respiratory insufficiencymight develop days after direct trauma tothe chest. An example might be a soldierwho sustained a gunshot wound of thelung and subsequently develops pneumo-nia and empyema.

• Unfortunately, some casualties may de-velop respiratory insufficiency that is, inpart, secondary to the therapies given for

an injury in some other part of the body. Anexample might be a soldier who developspulmonary edema following excessive fluidoverload that occurred during the treat-ment of massive blood loss resulting from atraumatic amputation of a leg.

The most common type of respiratory insufficiencythat military anesthesiologists and intensivists (ie,critical care specialists) are likely to see, however,occurs within several days to 1 week following aninjury that did not directly involve the lungs orchest wall: the adult respiratory distress syndrome(ARDS). This condition is now recognized as anexpression of the systemic inflammatory responsesyndrome (SIRS) and is related to sepsis.

This chapter is intended to provide practical helpto the medical officer who is confronted with casu-alties whose conditions are of varying degrees ofseverity, for whom decisions must be made promptlyto avoid further morbidity from cerebral hypoxia or

EXHIBIT 25-1

ETIOLOGY OF RESPIRATORY FAILURE IN COMBAT AND CIVILIAN PATIENTS

Pulmonary aspiration

Foreign-body obstruction of the airway

Soft-tissue obstruction of the airway:

Prolapse of tongue due to unconsciousness

Prolapse due to facial bone fracture

Edema and hematoma formation

Pulmonary injury:

Hemothorax

Pneumothorax

Hemopneumothorax

Lung contusion

Mediastinal emphysema

Rib fractures

Flail chest

Laryngeal fracture (with or without separation)

Tracheobronchial laceration or separation

Acute lung injury secondary to:

Blast overpressure injury

Chemical war gas

Infection

Hypermetabolism

Inhalation of combustion products

Hypoventilation secondary to:

Cervical spinal cord trauma

Cranial trauma

Exposure to nerve agent

Diaphragmatic hernia

Thoracic wall injury

Central nervous system depression from drugsor hypoxemia

Adapted with permission from Grande CM, Stene JK, Bernhard WN. Airway management: Considerations in the traumapatient. Crit Care Clin. 1990; 6(1):43.


669

other visceral organ injury. The chapter empha-sizes (1) the recognition of respiratory insufficiencyand (2) the provision of mechanical ventilation for

casualties with respiratory insufficiency at medicaltreatment facilities from a third-echelon hospital toa medical center in the continental United States.

EPIDEMIOLOGY OF RESPIRATORY FAILURE IN COMBAT CASUALTY CARE

Combat casualties do not commonly present withrespiratory insufficiency. Data from the VietnamWar indicate that probably somewhat less than 5%of those killed in action had injuries primarily tothe respiratory system exclusive of those whoexsanguinated from injuries to the pulmonaryvasculature.1 Respiratory failure at the hospitallevel may be more common. The total number whodie of wounds is about 3% to 4% of hospitalizedcasualties; it is probably reasonable to assume thatall of these casualties—at some time during theirhospital stay—would require respiratory support.Data from the Vietnam War indicate that approxi-mately 0.4% of hospitalized casualties died of whatwould now be called ARDS.2 Today, many morecasualties would probably receive prophylactic res-piratory support to prevent respiratory insuffi-ciency, so it is not unreasonable to infer from thesedata that perhaps 10% of all hospitalized combat

casualties are at risk of developing respiratory in-sufficiency. Furthermore, if nerve agents or en-hanced blast munitions were to be used against U.S.military forces in a future hostile action, a muchhigher percentage of the casualty population willlikely need immediate respiratory support. Forplanning purposes, the Department of Defense’sDeployable Medical Systems (DEPMEDS) treatmentfiles predict that in a major, high-intensity war, 17%of hospitalized casualties will require ventilatorysupport.3

Although the civilian experience depends on thetype of hospital surveyed, probably no more thanseveral percent of hospitalized patients developrespiratory insufficiency. This assumes that about5% of patients who are admitted to civilian hospi-tals are transferred to the intensive care unit (ICU),and that of these, one half require respiratory sup-port for more than 24 hours.4

ETIOLOGY OF PULMONARY INSUFFICIENCY

Mechanical Trauma

The battlefield is replete with potential mecha-nisms for pulmonary injury and the development ofrespiratory failure. Militarily relevant sources ofdirect mechanical trauma include penetrating, blunt,and blast injuries to the lungs and the chest wall. Inaddition, many soldiers with severe penetratinginjuries to the brain are not killed outright andrequire intubation for airway control and mechani-cal ventilation. The injury to the brain results indestruction of the respiratory centers; therefore,ventilatory drive is absent. During the VietnamWar, this was probably the most common indica-tion for using respiratory support.

Penetrating Injury

Penetrating thoracic injuries may, in turn, causesecondary conditions that result in respiratory in-sufficiency. Pneumothorax reduces the ability tocarry out normal gas exchange. The lack of a nega-tive pressure gradient across the lung parenchymaleads to alveolar collapse and a resulting intra-pulmonary shunting of blood. When a sufficientamount of tissue collapses, arterial hypoxemia en-

sues. Tension pneumothorax occurs when air entersthe pleural space during inspiration but is unable toescape during exhalation. The progressive accu-mulation of air produces a shift of the mediastinalstructures toward the unaffected side of the thorax.This event worsens gas exchange by compressing theunaffected lung and reduces venous return to theheart, resulting in shock. Hemothorax and hemo-pneumothorax are found in approximately 50% ofcasualties with penetrating wounds of the lung.5 Thisfact is not surprising, given the close anatomical prox-imity of bronchioles and arterioles within the lungparenchyma. Small fragments that strike the heartmay cause blood to collect within the pericardialspace. The increased pressure within the pericar-dium restricts the ability of the ventricles to acceptblood (tamponade). Cardiac output and blood pres-sure progressively decrease as stroke volume isreduced. If the tamponade is not reversed, cardio-vascular collapse and death result.

Blunt Injury

Blunt chest trauma with rib fractures and poten-tial flail chest (ie, multiple rib fractures that result insegmental chest wall instability) can occur in a


670

variety of battlefield settings. Underlying pulmo-nary contusion may be a far more important causefor respiratory failure than the mechanical dysfunc-tion associated with multiple rib fractures.

Blast Injury

Blast injuries from any type of explosion can becategorized as primary, secondary, and tertiary.Primary blast injury is due to the direct effect of thepressure wave on the body, and is not common.Secondary blast injury occurs when projectiles anddebris from the explosion strike the victim. Finally,tertiary blast injury, which is the major cause of pureblast injury, occurs when the individual is physi-cally displaced by the bulk flow of gases away from

the site of the explosion. The displacement leads tocollision with environmental objects and subse-quent trauma.6 A more complete discussion of themedical consequences of blast injury can be foundin The Medical Consequences of Conventional Warfare:Ballistic, Blast, and Burn Injuries.7

The pressure wave from an explosion movesaway from the epicenter at supersonic speeds,whether through air or water. As primary blastinjury is seen predominantly in the gas-containingorgans, the ears, lungs, and gastrointestinal tractare the most readily affected. The ear is the mostsensitive, with rupture of the tympanic membraneoccurring at pressures above 35 kPa (kilopascals).While injury to the ear causes pain and long-termmorbidity, pulmonary damage causes the greatest

Fig. 25-1. The probabilities of eardrum rupture and death from a pulmonary blast injury. These are shown as functionsof both the peak pressure above atmospheric pressure (measured in either kilopascals [kPa] or pounds per square inch[psi]) and the duration of the overpressure (in milliseconds [ms]). For example, a blast wave with a duration of 1 msthat has an overpressure of 5 psi can be expected to cause rupture of the tympanic membranes in a few people. If theoverpressure reaches 15 psi, one half of the individuals in the exposed population can be expected to have rupturedtympanic membranes. Much greater pressures are required before pulmonary injury occurs. A blast wave with aduration of 1 msec and a peak overpressure of 50 psi can be expected to cause pulmonary injury in a few people.Doubling that pressure is likely to cause a fatal outcome in about one half of the exposed population. The same lethalityis to be expected with a blast wave that lasts 1 s and has a peak overpressure of 12.5 psi. Note that both axes arelogarithmic. Adapted from Bowen T, Bellamy RF, eds. Emergency War Surgery NATO Handbook. 2nd US rev. Washing-ton, DC: US Department of Defense, Government Printing Office; 1988: 76.

5,000

1,000

500

100

10

Pea

k P

ress

ure

(k

Pa)

0.1 1 10 100 1,000Positive Phase Duration (ms)

2

10

100

1,000

50% Eardrum Rupture

Threshold Eardrum Rupture

Threshold Lung Injury

50% Lethality

1% Lethality

Peak

Pres s u

re (ps i)


671

immediate morbidity and mortality (Figure 25-1).Diffuse pulmonary contusions with a marked in-crease in lung water lead to ventilation–perfusionmismatch and arterial hypoxemia. Disruption ofalveoli with the development of alveolar–pulmo-nary venous communications create the potentialfor air emboli, which is the most common cause ofsudden death in such casualties. The gastrointesti-nal tract may be disrupted in portions where gashas collected. The colon tends to suffer the mostgastrointestinal blast injuries, which are manifestedby subserosal and intramural hemorrhage. Gas-trointestinal blast injury may result in delayed rup-ture, days after the traumatic event.8

The clinical presentation of each of these injuriesdepends on the amount of energy transmitted to thevarious tissues in excess of the tissues’ ability todissipate the delivered forces. Significant pulmo-nary and gastrointestinal damage is unlikely tohave occurred in the absence of tympanic mem-brane rupture. Arterial air emboli from alveolar–pulmonary venous communications typicallypresent with signs and symptoms of cerebral dys-function. However, direct trauma to the skull (sec-ondary or tertiary blast injury) is more likely to bethe cause of altered neurological function. Evi-dence of primary blast injury of the lung resemblesthe diffuse pulmonary contusion seen in blunt chesttrauma. Chest pain with tachypnea, use of acces-sory respiratory muscles, and hemoptysis are com-mon findings. The protean manifestations ofbarotrauma (injury resulting from elevated peak air-way pressure) are frequently seen in these patients.Pulmonary contusion, pneumothorax, pneumo-mediastinum, subcutaneous emphysema, and pul-monary interstitial emphysema can be confirmedwith a chest X-ray examination after the patient hasbeen stabilized. Symptoms of blast injury to thelung may be delayed 24 to 48 hours, making differ-entiation from other forms of acute respiratory fail-ure difficult.

The principles of management are detailed inother textbooks. Briefly summarized, the salienttreatment required for improved survival rates ofcasualties with thoracic injuries is as follows:

• Remove the bloody secretions from the air-way.

• Seal open chest wounds.• Evacuate blood and air from the pleural

space.• Treat hypovolemia.• Remove blood from the pericardial sac.• Aggressively treat empyema.

Casualties with blast injuries require additionalcare. This follows the basic principles of traumacare, with the following caveats directed towardpotential complications from the blast:

• Excessive volume resuscitation may com-plicate the pulmonary contusion present inmost of these patients.

• Positive pressure ventilation may increasethe severity of barotrauma and increase thepossibility of air embolism.

• If oxygenation is inadequate with nasalcannula or mask, continuous positive air-way pressure (CPAP) delivered via a tight-fitting mask or endotracheal tube may im-prove oxygenation without worseningbarotrauma.

Thermal Trauma

Although thermal injuries will be encounteredon the battlefield, this mode of injury is uncommonin conventional land warfare, occurring in fewerthan 10% of all casualties (see Chapter 1, CombatTrauma Overview, Figure 1-4). Thermal injuriesare, however, much more common in that subpopu-lation of soldiers who compose the crew of armoredfighting vehicles, and are even more common insailors and air personnel. In fact, fire and associatedsmoke inhalation are the most common causes ofcombat injury on warships.9 Crewed vehicles—whether ships, tanks, or aircraft—are loaded withmunitions and fuel and pose a constant hazard ofexplosions, fire, and thermal injury secondary tobattle damage. Burns to the skin are what weusually think of when thermal injuries are discussed,and in fact, circumferential, full-thickness burns ofthe thoracic wall may sometimes restrict chest-wallmotion to the degree that ventilation is compro-mised. Escharotomy may be required to improvethe restrictive defect. However, the more frequentand more injurious form of thermal injury is dam-age to the respiratory system caused by inhalingsteam and the gaseous and particulate products ofcombustion such as hydrogen chloride and phos-gene from plastic, formaldehyde from wood, andhydrogen cyanide from polyurethane.

Below the level of the vocal cords, thermal injuryfrom heated air is usually prevented by efficientupper-airway cooling. Stridor, hoarseness, and dif-ficulty with phonation all suggest the need to evalu-ate the upper airway. Injury to the hypopharynx maylead to delayed airway obstruction, making promptendotracheal intubation a lifesaving maneuver.


672

These injuries demonstrate a typical acute-lung-injury pattern of increased alveolar-capillary per-meability with noncardiac pulmonary edema.Iatrogenic factors may complicate the lung injury:the massive fluid requirements for resuscitationfrom thermal burns and associated traumatic inju-ries will invariably worsen pulmonary function.Patients may develop respiratory failure 1 to 2 weeksfollowing thermal injury. When delayed respira-tory failure occurs, it is usually caused by infectiouscomplications either within the lung (pneumonia)or at a remote site, giving rise to SIRS, which is themajor cause of death.

Inhalation of the toxic products of combustioncan induce parenchymal lung damage. The effectsmay be immediate or delayed for hours. A com-mon, serious pathophysiological effect of inhalingcombustion products is carbon monoxide intoxica-tion. The marked affinity of carbon monoxide forhemoglobin (> 200-fold higher than that of oxygen)significantly reduces the oxygen-carrying capacityof blood. A carbon monoxide concentration of 0.1%in ambient air leads to a carboxyhemoglobin con-centration of 50%. Carbon monoxide also shifts theoxyhemoglobin dissociation curve to the left, whichfurther reduces oxygen delivery (DO2) to the tis-sues. Patients will complain of headache, nausea,and exertional fatigue with moderate carbon mon-oxide exposure. Damage to the central nervoussystem (CNS) may occur in up to 30% of affectedindividuals unless they receive prompt therapy.The classic “cherry red” coloration is not seen withcarboxy-hemoglobin concentrations lower than 40%.A far more common manifestation is cyanosis.10

Lactic acidosis from reduced DO2 is a common find-ing with severe intoxication.

The therapy for respiratory injury caused by ther-mal burn is primarily supportive. Supplementaloxygen and bronchodilators are used to treat hy-poxemia and bronchoconstriction. Mechanical ven-tilation is required for patients with significant acutelung injury. The use of positive end-expiratorypressure (PEEP) increases the functional residualcapacity (FRC, the remaining lung volume afterpassive exhalation) of the lung and allows the frac-tion of inspired oxygen (FIO2) to be reduced tonontoxic levels. The U.S. Army Burn Center, FortSam Houston, Texas, reports that a mode of ventila-tion (high-frequency, percussive) that employs re-peated cycles of very rapid (several hundred perminute) and small volume (less than the dead-spacevolume) ventilations may be more efficacious. Themortality rate found in such patients is markedlyless (16.4%) than in patients with similar thermal

insults who are treated with conventional mechani-cal ventilation (42.7%). The improved survival maybe due to both (a) the low airway pressures found inhigh-frequency percussive ventilation (ie, there isless potential for barotrauma) and (b) improvedremoval of tracheobronchial secretions.11

Prophylactic steroids have not proven beneficialfor inhalation injury and may increase the risk forsubsequent infection. Likewise, antibiotic prophy-laxis is no longer used; it may increase the riskfor resistant bacterial infection, and has not beenshown to reduce morbidity and mortality in clinicaltrials.12

The treatment of carbon monoxide poisoninginvolves supplying supplemental oxygen to reducethe half-life of carboxyhemoglobin. (With 100%oxygen, the half-life is reduced to 40 min, opposedto 250 min in room air.) Hyperbaric oxygen may bebeneficial with severe intoxication; however, thismodality is not likely to be available in the combatzone with the possible exception of aboard largernaval vessels.

Chemical Warfare

Almost all chemical war gases cause respiratoryinsufficiency and manifest a variety of effects thatcould lead to respiratory failure (see Chapter 30,Anesthesia for Casualties of Chemical WarfareAgents, for a discussion of the respiratory effects ofwar gases). The four major categories of war gasesare nerve, blister, choking, and blood agents. De-contamination procedures are not addressed in thischapter; however, because acute respiratory failureis a common cause of death in affected individuals,medical personnel must ensure that the emergentnature of the situation does not lead to hasty actionand inadvertent self-contamination.

Nerve Agents

Nerve agents have the potential to cause vastnumbers of casualties. Soldiers must have suffi-cient personal protective equipment and adequatepharmacological prophylaxis. Military plannersmust emphasize that medical officers may need tocare for mass casualties, all of whom may requiremechanical ventilation.

Nerve agents exert their effects by preventingcholinesterase from hydrolyzing acetylcholine (pre-viously released from the nerve terminal). Thisallows for repetitive stimulation of the postsynapticreceptor, which leads to rapid muscle fatigue. Inha-lational or dermal exposures to nerve agents thus


673

present as a cholinergic syndrome, the severity ofwhich depends on the amount of exposure. Respira-tory effects of nerve agents include broncho-constriction and bronchorrhea along with respiratorymuscle dysfunction. Organophosphate-induced seiz-ures may further complicate the respiratory embar-rassment produced by these agents.

Treatment includes administering both atropine,which blocks the cholinergic receptors, andpralidoxime chloride (2-PAM Cl), which binds toorganophosphate and most carbamate nerve agents,releasing the cholinesterase and allowing normalnerve function to resume. Airway control is ofparamount importance, as copious secretions mayprevent adequate ventilation and oxygenation. Ifrespiratory muscle function is adequate, oral ornasopharyngeal suctioning may be all that is re-quired. However, if respiratory muscle function ispoor, endotracheal intubation and mechanical ven-tilation may be required. Care must be taken toensure that the lungs are not exposed to the exces-sively high pressures that result from the intensebronchoconstriction that follows significant nerveagent exposure. Under these circumstances,barotrauma may occur before bronchospasm hassubsided. Atropine, inhaled beta agonists such asmetaproterenol or terbutaline, and inhalational an-esthetics all reverse bronchospasm.6

Blister Agents

Blister agents (vesicants), of which the mustardsare the best known, are alkylating chemicals thatare thought to act by damaging cellular DNA. Thisleads to cell disruption and death; an intense in-flammatory response causes blisters to form. Inha-lation of vesicants produces severe tracheitis andbronchitis. Symptoms may not develop for 4 to 6hours, requiring medical personnel to observe ca-sualties for delayed effects. Bronchopneumoniafrequently complicates the recovery of these pa-tients. Progressive, irreversible respiratory insuffi-ciency, developing over months to years after inha-lation of mustard, has been described.13 It appearsto be caused by unrelenting fibrosis that leads tostenosis of the entire tracheobronchial tree.

Choking Agents

The war gas that caused the most flagrant ex-ample of respiratory insufficiency in World War Iwas phosgene, a choking agent. Choking agentscause a severe form of respiratory insufficiency andproduce significant irritation of the lower respira-

tory tract. First used in World War I as a toxic agent,phosgene produces distal airway and alveolar in-flammation, which eventually causes pulmonaryedema. As with blister agents, the effects may bedelayed for up to 12 hours; minimal symptom-atology may precede the development of fulminantpulmonary edema.

Treatment is entirely supportive. Supplementaloxygen and positive-pressure ventilation may benecessary for severe cases. The effects are self-limiting, but due to the intense inflammatory re-sponse elicited, the patient’s condition may evolveinto ARDS.

Blood Agents

Blood agents such as hydrogen cyanide combinewith intracellular cytochrome oxidase and halt cel-lular respiration. Exposure to large amounts ofinhaled hydrogen cyanide causes death within min-utes. Exposure to lesser amounts causes vertigo,nausea, vomiting, headache, and tachypnea. Seiz-ures and cardiopulmonary arrest may rapidly en-sue if treatment is delayed.

Treatment involves administering amyl nitrateby inhalation or sodium nitrite by vein. Nitritesoxidize the iron moiety of hemoglobin to producemethemoglobin. Methemoglobin scavenges cya-nide from the circulating blood, which promotes aconcentration gradient for the removal of cyanidefrom cytochrome oxidase. Intravenous sodium thio-sulfate can be administered to provide free sulfur,which will convert cyanide to the far-less-toxic thio-cyanate ion. Mechanical ventilation with supple-mental oxygen will aid in the resuscitation of apneiccasualties; however, expeditious treatment withnitrites is the mainstay of therapy.6

Nuclear Warfare

Thermonuclear weapons produce their devastat-ing effects via the release of massive amounts ofenergy. This energy release produces casualtieswith three types of injuries: blast, thermal, andradiation. Blast and thermal injuries will be treatedin a manner identical to that previously discussed.Neurovascular, gastrointestinal, and hematopoieticsyndromes are associated with nuclear radiationinjuries. Anesthesiologists and intensivists willencounter the neurovascular and gastrointestinalsyndromes only rarely, as these are considered to beuniformly fatal. The hematopoietic syndrome willbe the most common radiation injury requiringmedical care.


674

INDIRECT MECHANISMS OF RESPIRATORY INSUFFICIENCY

Respiratory failure in the combat casualty mayresult from insults other than direct chest trauma.Three of the most frequently encountered etiologiesare pulmonary aspiration, hydrostatic pulmonaryedema, and permeability pulmonary edema.

Pulmonary Aspiration

Medical personnel should assume that every com-bat casualty has a full stomach. The combination ofa traumatic injury and a full stomach puts soldiersat risk for aspirating blood or regurgitated stomachcontents:

The soldier may have eaten just before beingwounded. Even if many hours have passed, painand anxiety greatly delay gastric emptying. Theimportant interval is not the time since the lastmeal, but rather the time from the last meal to thetime of injury. Regurgitation and aspiration of stom-ach contents, particularly of volumes greater than25 mL, with a pH less than 2.5, and in the presenceof solid food, place the casualty at high risk foraspiration pneumonitis.14(p28)

Aspiration of gastric contents may be asympt-omatic or cause several life-threatening syndromes.The underlying common denominator in aspirationis an altered level of consciousness. Many casual-ties will have head injuries that negate the protec-tive airway reflexes, and many will require generalanesthesia or have neurosurgical catastrophes thatpredispose to aspiration.15 However, other predis-posing conditions (ie, neuromuscular or gastrointes-tinal disorders) are not common in this population.

Silent aspiration occurs when a small volume ofgastric contents is introduced into the trachea. Thismay occur somewhat frequently in a civilian medi-cal population—elderly and obtunded—who areundergoing elective surgery, receiving therapeuticnarcotics, whose level of consciousness is depressed,or while a nasogastric tube is being placed. In acombat situation, silent aspiration is probably lesscommon.

When aspiration occurs, apnea or airway ob-struction occur immediately. The acute chemicalinjury damages the tracheal bronchial tree and al-veoli. Pathologic evaluation immediately after as-piration shows bronchiolar epithelial degeneration,pulmonary alveolar edema, and alveolar hemor-rhage. Fibrinous exudate, focal atelectasis, andneutrophilic infiltrates occur later. After 24 hours,focal hemorrhage, atelectasis, empyema, fibrin-filled

alveoli, and pneumonitis are established.16 Bacte-rial pneumonitis can occur within 72 hours, andlung abscesses may occur, usually after days toweeks.

Injury occurs to both Type I and Type II pneumo-cytes. Type I pneumocytes undergo focal necrosisand the basement membrane is exposed. Fluid fillsthe alveolar and perivascular spaces. Type II cellsare initially edematous, then degenerate. Hyalinemembranes form by 48 hours. By 72 hours, alveolarcells regenerate and the inflammatory processabates.

Prevention is the best treatment for aspiration.When prevention fails or is not possible, then mak-ing the correct diagnosis, removing large particu-late matter, and providing appropriate supportivemeasures will help improve morbidity and preventmortality.17

Immediate endotracheal intubation is requiredwhen apnea occurs. Adequate ventilation shouldbe documented by measuring arterial blood gasesor by means of oximetry and capnography, whichare discussed later in this chapter.

Airway protection during intubation withcricothyroid pressure should help prevent aspira-tion. The thumb and middle finger exert pressurelateral to the cricoid cartilage, while the index fin-ger pushes directly down on it. If the technique isproperly performed, this technique can prevent re-gurgitation when intragastric pressures rise to ashigh as 50 cm H2O or more. Normal intragastricpressure is usually less than 16 cm H2O so cricoidpressure should be protective in most instances. Assoon as the patient is intubated, the cuff on theendotracheal tube is inflated before manual pres-sure is released.18

Airway obstruction may be due to laryngospasm,edema, mechanical obstruction by foreign mater-ial, bronchial inflammation, or bronchospasm.Laryngospasm can be broken with either continu-ous positive airway pressure or a rapid-acting de-polarizing muscle relaxant. If the laryngospasmcannot be broken, surgical control of the airway ismandatory. Nebulized racemic epinephrine mayimprove airway edema.15

Suctioning often clears the airway of foreign de-bris, but if residual material remains, bronchoscopywill be required. In many instances, aggressivechest physiotherapy will help clear the small air-ways of particulate matter. This is contraindicatedin patients with head trauma or spinal cord injuries,however.


675

The treatment of aspiration pneumonia shouldbe prophylactic. Elevating the head of the bed toprevent gastroesophageal reflux, neutralizing gas-tric contents, and prophylactic intubation are of theutmost importance. Once aspiration has occurred,adequate oxygenation is mandatory.17

On the battlefield, multiply injured casualtieswill require intubation. Many will also requireurgent or immediate operative intervention.Intubation will prevent further aspiration. Casual-ties with head injuries will require hyperventilationas well.

Bronchial lavage is no longer believed to be indi-cated because it may worsen hypoxia. Positiveairway pressure should be used for diffusepneumonitis; however, its use in localized aspira-tion must be tempered against augmenting ventila-tion–perfusion mismatch. PEEP may be preferen-tially distributed to normally compliant units andnot improve the FRC in the affected lung region. Itmust therefore be used with caution.

Appropriate fluid resuscitation should not betempered by the possibility of lung injury. Fluidoverload should be avoided but aggressive resusci-tation must not. Cardiovascular stability should bemaintained with the proper use of inotropic agentswhen indicated. In more severe cases, a flow-directedpulmonary artery catheter may be beneficial becauseit allows optimization of the hemodynamic status ofthe patient, which should improve survival.

Glucocorticoids have no place in the treatment ofaspiration. Glucocorticoids suppress host defensemechanisms and predispose the patient to infec-tion. Prophylactic antibiotics are not used either, asthey do not prevent pneumonia and tend to selectout resistant organisms, thereby increasing mor-bidity and mortality.17

Hydrostatic Pulmonary Edema

Pulmonary edema is often associated with respi-ratory failure. The type of pulmonary edema mustbe determined before the condition can be properlytreated. These types are hydrostatic pulmonary edema,in which fluid accumulates in the presence of afairly normal blood–air barrier, and permeabilitypulmonary edema, of which ARDS is the classic mani-festation. Hydrostatic pulmonary edema occursbecause fluid accumulates in the presence of a fairlynormal blood–air barrier. In permeability pulmo-nary edema, fluid accumulation occurs because theblood–air barrier is abnormal.19

Hydrostatic pulmonary edema is usually causedby an increase in the intravascular filtration pres-

sure within the pulmonary microvasculature. Theelevated filtration pressure increases the fluid fluxinto the interstitium. If the capacity of the lym-phatic system to reabsorb this fluid is exceeded,then the fluid accumulates in the alveoli. Rarely,fluid may accumulate in two other situations: ifplasma oncotic pressure is reduced or if the capac-ity of the lymphatic system to remove fluid is re-duced. The intravascular pressure must rise above7 to 10 mm Hg for water to flood the interstitialspace.

With the increased flux of fluid into the inter-stitium, the fluid moves down a negative interstitialpressure gradient toward the bronchovascular cuffs.This is where the pulmonary lymphatics begin. Iflymphatic uptake is decreased, fluid accumulatesaround the bronchoalveolar cuff until the intersti-tial pressure is greater than the airway pressure, atwhich time fluid enters the alveolus. This is typi-cally seen when the pulmonary artery occlusionpressure (PAOP) exceeds 20 mm Hg.

In the civilian population, the causes of hydro-static pulmonary edema include cardiac dysfunc-tion, pulmonary venous occlusive disease, neuro-genic pulmonary vasoconstriction, pulmonaryarterial embolization with air, thrombus, or fat;however, these are rarely seen in young, healthysoldiers. In the combat casualty, the most commoncauses of hydrostatic pulmonary edema areiatrogenic. Asthma and foreign bodies that causeairway obstruction, and pulmonary lymphatic ob-struction may also cause hydrostatic pulmonaryedema. Clinically, the classic description is that ofrespiratory distress with pink, frothy sputum andarterial hypoxemia. In trauma patients, this occurswith massive head injury, chest injury, or largeamounts of volume resuscitation.20

Patients will occasionally present less dramati-cally with complaints of mild chest tightness or apressure sensation, a sensation of being unable tocatch the breath, and cough. As the capillary pres-sure continues to increase, and as the fluid fluxbecomes more pronounced, the patient will becomemore symptomatic. The respiratory rate increasesand the ventilatory pattern becomes labored. Thisincreases the work of breathing required to achieveadequate lung expansion. Respiratory failure willoccur because of muscular exhaustion.

Alveolar flooding is accompanied by arterialhypoxemia. Fluid-filled air spaces prevent the pas-sage of gas into the capillary. These air spaces havelow ventilation-to-perfusion ratios (ventilation–perfusion mismatch). As the mismatch increases,the effectiveness of hypoxemic pulmonary vaso-


676

trauma was described in 1932,22 and ARDS wasdefined in 1967.23 For ARDS to occur, the capillaryendothelium of the pulmonary vasculature must beinjured. This results in dyspnea, hypoxemia, de-creased pulmonary compliance, and bilateral pul-monary infiltrates. Pulmonary edema is presentbut not associated with an increased hydrostaticpressure; it is caused by increased permeabilityresulting from diffuse alveolar–capillary endothe-lial injury. ARDS is a common problem affectingpreviously healthy people. The mortality rate ishigh: despite vigorous supportive care, it can begreater than 40%.24

The syndrome is associated with a wide varietyof disease states with multiple underlying etiolo-gies, all demonstrating a single clinical pattern.ARDS can be caused by all types of shock (Exhibit25-2). A 2.2% incidence of ARDS has been reported

constriction decreases and profound arterial hy-poxemia will ensue. Formerly, oxygen was be-lieved to need to diffuse through an abnormallythick layer of water before it reached the red bloodcells, but compared to ventilation–perfusion mis-match, this is probably a minor impediment to theexchange of respiratory gases.

The fluid tends to accumulate first in the depen-dent portions of the lung. Because the pulmonaryblood and air flows to the dependent areas aredisproportionally small, this fluid accumulation ex-aggerates the normal propensity for alveoli in thedependent part of the lungs to be underventilated.The pulmonary blood flow is redistributed to lessedematous, nondependent alveoli. The FRC de-creases with increased critical closing pressure ofthe alveoli. Other unrelated conditions that de-crease FRC (eg, increased intraabdominal pressure,upper abdominal or thoracic incision, and the su-pine position) can compound the hypoxemia.

The treatment of hydrostatic pulmonary edemamust both reverse the changes caused by fluid accu-mulation and eliminate the underlying disorderthat promotes fluid accumulation. The supportivetherapy should minimize arterial hypoxemia anddecrease the work of breathing. Supplemental oxy-gen and positive airway pressure are the mainstaysof treatment. Administration of supplemental oxy-gen through nasal prongs or a face mask is useful.However, endotracheal intubation is often requiredas respiratory failure progresses. If the arterialpartial pressure of oxygen (PaO2) does not increaseafter supplemental oxygen therapy is added, thencontinued hypoxemia is caused by groups of alveoliwith low ventilation–perfusion ratios. Positive air-way pressure is then delivered as PEEP or CPAP.This will increase the PaO2 by increasing the FRC ofthe lungs. (The alveoli are now being ventilatedbecause a higher pressure is being used to forcerespiratory gas into them.) To decrease the work ofbreathing, both positive airway pressure and as-sisted ventilation can be utilized. Positive airwaypressure restores lung volume, increases compli-ance, and increases the cross-sectional area of theairways. It also reduces airway resistance by splint-ing the airway lumen open. Flow rates will de-crease as ventilation becomes less labored, therebyconverting a higher-resistance, turbulent flow intoa lower-resistant, laminar flow pattern.21

Adult Respiratory Distress Syndrome

Lung injury has plagued successful resuscitationof the trauma patient. Lung injury associated with

EXHIBIT 25-2

ETIOLOGY OF SHOCK THAT INDUCESTHE ADULT RESPIRATORY DISTRESSSYNDROME

Burns

Fat emboli

Lung contusion

Head trauma

Near drowning

Septic states

Inhalation of toxic gases:

Smoke

Nitrous oxide

Ammonia

Chlorine

Phosgene

Cadmium

Aspiration of gastric contents

Drug ingestion

Pancreatitis

Multiple transfusions

Disseminated intravascular coagulation

Bowel infarction


677

In casualties with traumatic injuries.25 The syn-drome is more common in casualties with bilateralchest trauma and extensive lung injury than it is incasualties with a localized lung contusion.

Pathophysiology of Alveolar Injury

The normal structural integrity of the alveolus isdisrupted in patients with ARDS. Type I pneu-mocytes cover approximately 95% of the alveolarsurface area while Type II pneumocytes cover theremaining 5%. Because of their greater number,Type I pneumocytes are more likely to be injured.Within 24 hours of injury, Type I pneumocytes aredestroyed, leaving a denuded basement membrane.Type II pneumocytes, which appear to be moreresistant to initial damage, then proliferate andform a continuous layer that covers the previouslydenuded basement membrane. In addition, thedenuded surface may allow aggregates of plasmaproteins, cellular debris, fibrin strands, and rem-nants of surfactant to adhere, which forms the char-acteristic hyaline membrane. Over the next severaldays, the alveolar septum thickens, and not onlyinflammatory cells but also reparative cells prolif-erate. Hyaline membranes organize, and micro-atelectasis occurs. The capillary endothelial dam-age becomes more apparent. Irregularities of theendothelial surface with focal cytoplasmic swellingcan be seen, and fibrosis is apparent in the respira-tory ducts and bronchioles.26

Starling’s equation predicts the fluid flux acrossa semipermeable membrane. It describes the bal-ance between hydrostatic and oncotic pressuresacting in opposition across the endothelial barrier.There are two driving forces (the pressure withinthe capillary, Pc, and the pressure within theinterstitium, Pi) and two retaining forces (the col-loid osmotic pressure within the capillary, πc, andthe interstitium, πi). The symbol σ, which is knownas the reflection coefficient, represents the imper-meability of the membrane to a given substance (ie,if σ = 1, the membrane is completely impermeable;while if σ = 0, the membrane is completely perme-able). K is a constant representing the filtrationcoefficient.

Starling’s equation states:

fluid flux = K[(Pc – Pi) – σ (πc – πi)]

Theoretically, the interstitial hydrostatic pressureis slightly negative due to the elastic properties ofthe lung; therefore, progressively more negative

fluid pressures increase the flux into the alveolarspace. The small amount of fluid filtered into theinterstitial space is readily removed by the exten-sive pulmonary lymphatic system. Hydrostaticpulmonary edema occurs when the factors favoringfluid flux across the capillary membrane exceedthose designed to protect the lungs from develop-ing edema.19

The pulmonary edema that characterizes ARDSis caused by alteration of Starling’s forces. Thisalteration is usually due to an abnormally highpermeability coefficient; the abnormality is madeworse when the capillary hydrostatic pressure in-creases. The latter abnormality may manifest aselevated pulmonary capillary wedge pressure.Decreased colloid oncotic pressure can be caused bysevere hypoproteinemia from overhydration withintravenous crystalloid fluid, liver failure, or thenephrotic syndrome; however, it is rarely respon-sible for lung edema.

Finally, increased interstitial colloid oncotic pres-sure can occur with increased alveolar capillarymembrane permeability (large protein moleculesleak out into the interstitium where their oncoticproperties attract fluid from the vascular bed). Thismay be the primary mechanism responsible for thedevelopment of pulmonary edema in ARDS. Thealtered permeability allows the free movement offluid and protein from the intravascular space, tothe interstitial space, to the intraalveolar space.Edema forms at a rate that exceeds the lymphaticsystem’s ability to remove it.

The alteration of the integrity of the alveolarcapillaries is believed to result from injuries due tothe adverse effects of (a) tumor necrosis factor andleukokinin I, which are released from activatedmacrophages, and (b) oxygen free radicals, whichare released from neutrophils. Intravascular acti-vation of complement leads to stimulation of neu-trophils, sequestration within the capillaries, andsubsequent endothelial damage. Patients with ARDShave increased levels of serum C-5A, the activatedfifth component of the complement cascade.26

Release of C-5A by the alternate pathway causesmacrophages and neutrophils to aggregate. Theneutrophils release oxygen free radicals; prostag-landins; and granules that contain protease, elastase,coagulase, pepsins and lysozymes. Damage to thebasement membrane occurs and increased perme-ability follows. This allows nonhydrostatic pulmo-nary edema to form. Arachidonic acid metabolitesaffect vascular permeability, vascular tone, and air-way reactivity, which causes further vasoconstric-tion and bronchoconstriction.


678

An additional factor resulting in alteration of theintegrity of the alveolar membrane in casualtieswith massive CNS injury is massive sympatheticdischarge, which affects the pulmonary veins. This,in turn, causes profound pulmonary hypertension,which mechanically disrupts the junctions betweenpulmonary capillary endothelial cells.

Pathophysiology Resulting in Abnormal GasExchange

The abnormal gas exchange that occurs in ARDSis due to the increase in extravascular lung water,which tends to close small groups of peripheralalveoli and cause intrapulmonary shunting. Thisprocess could be compounded over time by theprogressive, extensive fibrosis that occurs in mostof these patients. In one study,27 researchers foundthat by using multiple, inert-gas elimination tech-niques in patients with ARDS, some degree of ven-tilation–perfusion mismatch occurred in 40% ofpatients. There was no evidence of a diffusionabnormality as it applies to respiratory gas ex-change.

ARDS also causes changes in lung mechanics.Both the FRC and pulmonary compliance decrease.Loss of surfactant causes alveolar collapse, leadingto further decreases in both FRC and pulmonarycompliance. In addition to surfactant abnormali-ties, flooding of the alveoli decreases alveolar vol-ume and causes further atelectasis. Compliance isfurther reduced by interstitial and alveolar edema,fibrosis, and the abnormal surfactant.

Clinically, during phase I, the patient will bedyspneic and will demonstrate a respiratoryalkalosis with tachycardia and tachypnea (Table 25-1). A latent period, phase II, will occur, lasting from6 to 48 hours after the injury, during which thepatient will appear stable and may appear to beimproving. During phase II, the patient will beginto demonstrate an increase in the work of breathing,with hyperventilation and hypocarbia; the lungsbecome less compliant due to both movement offluid into the alveoli and loss of surfactant. De-creased alveolar volume and widespread atelectasisare the consequences of these changes (Table 25-2).

Physical examination will demonstrate scatteredrales and evidence of the use of the accessory musclesof respiration. The initial chest X-ray examinationwill be normal, but subsequent radiographs willdemonstrate increases in interstitial markings. Withcontinued progression of the disease, the patientwill develop acute respiratory failure (phase III).Lung compliance will be markedly decreased due

to interstitial and alveolar edema, fibrosis, and ab-normal surfactant. Loss of the normal surfactantleads to a decrease in the FRC and thus a furtherdecrease in compliance. Further decreases in theFRC increase the alveolar–arterial oxygen differ-ence and worsen the hypoxia.

The chest radiograph will demonstrate a diffuseinterstitial pattern and, in fact, pneumonia is oftendifficult to rule out. At this time, bronchoalveolarlavage for diagnosis may be indicated to identifythe pneumonic process.

The patient may improve at this time. The physi-cal findings will resolve and the chest radiographwill return to normal. More commonly, however,the disease progresses with severe unresponsivehypoxemia, increased intrapulmonary shunting ofblood, and metabolic and respiratory acidosis. Theradiograph demonstrates honeycombed, patchy,and ill-defined nodular densities.

Treatment

Treatment of the patient with ARDS encompassesmeasures to (a) prevent death from respiratory fail-ure, (b) minimize further lung injury, (c) providegeneral support, and (d) recognize and treat com-plications.

The goal of therapy is to provide adequate oxy-genation at the lowest possible FIO2. High inspiredoxygen concentrations, as well as high airway pres-sures, have the potential to significantly exacerbatethe acute lung injury.

Initially, the lowest FIO2 needed to provide anarterial hemoglobin saturation (SaO2) of 90% ormore is used. Continuous pulse oximetry is used tomonitor therapy. Decreasing the FIO2 to 50% isdesirable, and to achieve it, end-expiratory pres-sure is frequently used. PEEP is begun at 5 cm H2Oand increased incrementally with careful attentionto cardiac output, pulmonary compliance, and peakinspiratory pressure. The medical officer’s clinicaljudgment may be all that is available to assureoptimal oxygenation.

Mechanical ventilation is used to deliver in-creased FIO2 and increased tidal volumes. Tidalvolumes of 10 to 15 mL/kg should be used initially.Careful attention is required to prevent the in-creased peak inspiratory pressure from causingbarotrauma.

The need for high peak inspiratory pressure maybe dictated by decreased compliance of the ARDSlung. However, other causes that may give rise tohigh peak inspiratory pressure are endotrachealtube displacement, pneumothorax, increased pleu-


679

TABLE 25-1

CLINICAL FINDINGS IN ADULT RESPIRATORY DISTRESS SYNDROME

Phase Findings

Phase I: Normal auscultatory examination

Acute Injury Normal chest X-ray examination

Tachycardia

Tachypnea

Respiratory alkalemia on arterial blood gas testing

Phase II: Minor abnormalities on auscultatory examination

Latent Period (6–48 h) Minor abnormalities on chest X-ray examination

Tachycardia and tachypnea persist

Increased work of breathing

Respiratory alkalemia

Increased PAO2 – PaO2

Phase III: High-pitched rales to auscultation

Acute Respiratory Failure Diffuse infiltrates on chest X-ray examination

Marked tachypnea and dyspnea

Decreased lung compliance

Worsening oxygenation and ventilation

Phase IV: Severe, unresponsive hypoxemia

Terminal Abnormalities Increased intrapulmonary shunting

Severe metabolic and respiratory acidemia

PAO2 – PaO2: alveolar-arterial difference in partial pressure of oxygen

TABLE 25-2

PATHOPHYSIOLOGY OF THE ADULT RESPIRATORY DISTRESS SYNDROME

FRC: functional residual capacity; PAOP: pulmonary artery occlusion pressure

Findings Mechanisms

Refractory hypoxemia Increased intrapulmonary shunting

Decreased FRC

Release of inflammatory mediators

Diffuse infiltrates on chest X-ray examination Alveolar edema

Atelectasis

Inflammatory cell infiltrates

Normal PAOP Endothelial damage results in capillary leakage

Decreased pulmonary compliance Edema

Atelectasis

Marked increase in extravascular lung water


680

erate dyspnea on exertion. In the majority, the chestX-ray examination will return to normal, as willtheir lung volumes. Resting arterial blood gaseswill tend to be normal but with exercise, approxi-mately half of these patients will have some de-crease in their PaO2. Surprisingly, the typical ARDSpatient does not die of respiratory insufficiency butof complications of unrelenting sepsis with failureof other organ systems.30

Although not presently applicable to combat situ-ations, the use of extracorporeal membrane oxygen-ation with low-frequency positive pressure ventila-tion may be of some benefit to patients with ARDS.Although several studies of note have shown noimprovement in survival when extracorporeal mem-brane oxygenation was used for oxygenation, oth-ers have used the apparatus with low-frequencypositive pressure ventilation to some success. Lungfunction improved in 73% of patients and 49% sur-vived, compared with the previous mortality rate of91%.31 An intra–vena caval blood oxygen–extracor-poreal carbon dioxide removal device is presentlyundergoing clinical trials. Preliminary data indi-cate that there is no improvement in survival com-pared with conventional therapy.32

ral effusion, increased lung edema, bronchospasm,mucous plugging, and high flow rates. Adjustingthe tidal volume, the level of PEEP, or using para-lytic agents to relax the patient’s respiratory musclesmay be required to decrease barotrauma. The modeof ventilation may be either intermittent, manda-tory ventilation (IMV) or assist control (AC), both ofwhich are discussed later in this chapter. Neithermode has been shown to be superior. The modewith which the operator has the most experienceshould be the mode used.

The use of glucocorticoids in the treatment ofARDS is not resolved. In the past, high-dose gluco-corticoids have been advocated, but a prospectiverandomized trial published in 198728 reported thatno benefit was seen from the use of glucocorticoids.However, a meta-analysis published in 199529 sug-gests that a subset of patients may benefit fromcorticosteroids administered during the fibropro-liferative phase of the disease process.

Even with aggressive, intensive care, the mortal-ity rate for ARDS remains approximately 50%. Ofpatients who survive the disease and its complica-tions, most will become clinically asymptomatic. Asmall percentage, however, will have mild-to-mod-

DIAGNOSING RESPIRATORY FAILURE

DO2 to tissues for metabolic function. Cyanosis isonly detectable when the patient’s blood contains atleast 5 g of deoxygenated hemoglobin per deciliterof blood. Similar findings (ie, signs and symptomsof hypoxemia) are noted when the level of met-hemoglobin reaches 1.5 g/dL, which may be seenfollowing treatment for cyanide toxicity.6 Typi-cally, CNS effects are prominent with significanthypoxemia. Confusion, restlessness, and loss ofjudgment are common manifestations. The lack ofavailable oxygen leads to increased sympathetictone as the body attempts to compensate for inad-equate DO2 to the tissues. Lactic acidosis with acompensatory respiratory alkalosis becomes promi-nent as cells convert to anaerobic metabolism toprovide for energy needs.

The major causes of hypoxemia in the traumapatient are a direct result of abnormalities in venti-lation–perfusion relationships within alveolar cap-illary units. The deoxygenated blood that perfusesthe alveolar capillaries in areas of hypoventilationwithin the lung cannot be normally oxygenated.The oxygenation deficit in this circumstance can becorrected by supplemental oxygen. This will in-crease the relative amount of oxygen within thealveoli in the region of low ventilation.

Because the causes of respiratory failure are sovaried, astute physicians always search for incipi-ent respiratory failure in their patients. The termacute respiratory failure comprises disorders of oxygen-ation or ventilation or both. Its definition, like-wise, has both oxygenation and ventilatory compo-nents:

• hypoxemia: while breathing room air, thepartial pressure of oxygen (PO2) is 50 mmHg or less; and/or

• ventilatory failure: the inability to ventilatesufficiently to maintain arterial blood par-tial pressure of carbon dioxide (PaCO2) lessthan 50 mm Hg.21

Although disorders of oxygenation and ventilationfrequently coexist, consideration of each as a sepa-rate entity allows for a clearer understanding of thepathophysiological processes that lead to respira-tory failure.

Hypoxemia

The signs and symptoms of hypoxemia are theeffects of, and the body’s response to, insufficient


681

Although a number of pathophysiological pro-cesses produce hypoventilation, atelectasis fromdirect trauma to the lung parenchyma is the mostcommon. Overly aggressive volume resuscitationwith increased lung water is also frequently en-countered in the setting of trauma. The secondmechanism for the development of arterial hypox-emia, that of right-to-left intrapulmonary shuntingof blood, can be thought of as the extreme form ofaltered ventilation and perfusion. In this instance,pulmonary arterial blood traverses capillaries inregions of complete alveolar collapse or regionsthat receive no ventilation from airway obstruction.Supplemental oxygen does not correct the oxygen-ation deficit. Correction is accomplished by inflat-ing these collapsed alveoli, and it is necessary toresort to either continuous positive airway pressurevia face mask or endotracheal intubation to im-prove oxygenation.

In combat medical facilities, supplemental oxy-gen is delivered from a tank. Oxygen sieves (concen-trators that can remove and concentrate oxygenfrom ambient air) will be available in future con-flicts to supply supplemental oxygen. In additionto the signs and symptoms of hypoxemia, assess-ment of the patient’s degree of oxygenation re-quires arterial blood-gas measurement and pulseoximetry. The echelon of combat medical care thatwill have blood-gas monitoring capabilities is un-clear; however, transportable pulse oximetry is avail-able and could be utilized at mobile army surgicalhospitals (third echelon), combat support hospitals(third echelon), and possibly even at the clearingstation (second echelon).

Ventilatory Failure

Ventilatory failure may result from a number ofdifferent pathological processes. The two disordersmost commonly associated with respiratory insuf-ficiency leading to ventilatory failure are (1) CNSdepression from trauma or drug effect and (2) com-promise of the respiratory function of the chestwall. Disorders of the chest wall (eg, flail chest)may occasionally lead to ventilatory failure mani-fested by hypercarbia. However, associated abnor-malities of oxygenation tend to represent the morelife-threatening condition. Clinical signs and symp-toms of hypercarbia are similar to those noted withhypoxemia, but with these exceptions: cyanosis isabsent while oxygenation is maintained, andobtundation is more common than agitation. As itdoes with failure of oxygenation, optimum deter-mination of the adequacy of ventilation rests on the

physician’s interpretation of arterial blood-gas de-terminations. Portable capnography provides areasonable estimate of arterial carbon dioxide con-tent and may prove useful in situations in whichblood gas determinations are not available.21

Combined Hypoxemia and Ventilatory Failurein Postoperative Casualties

Postoperative pulmonary complications areamong the commonest causes of morbidity andmortality in this patient population. Of all thepotential respiratory complications, atelectasis iseasily the most common: it may account for upto 90% of postoperative respiratory complica-tions.33 The degree of involvement may range froma small, insignificant group of airways to completecollapse of an entire lung. Other less common butfrequently more life-threatening complicationsinclude pneumonia, bronchospasm, pulmon-ary thromboembolic disease, and respiratory fail-ure. It is possible to define the various abnormali-ties with a combination of radiographic, laboratory,and clinical features. Chest X-ray examinations,blood-gas measurements, and chest ausculta-tion remain the most valuable techniques to definepulmonary abnormalities and determine the needfor more-invasive measures such as bronchos-copy, endotracheal intubation, or pulmonaryarteriography.34

Upper abdominal and thoracic surgery carry thegreatest risk for postoperative pulmonary dysfunc-tion. Splinting to ease pain reduces the patient’sability to cough and breathe deeply, both of whichpredispose to distal airway plugging and collapse.The splinting results in a reduced vital capacity thatleads to hypoventilation, while the airway plug-ging causes atelectasis and hypoxemia. Anotherimportant factor, a direct result of general anesthe-sia, is depression of mucociliary transport. Abnor-mal ciliary action and production of mucus are bothworsened by lengthening anesthesia time. Ciga-rette smoking potentiates this abnormality.

Surgical procedures also predispose to prema-ture closure of the distal airways. Supine position-ing, increased abdominal girth and breathing atreduced lung volumes are nonpulmonary factorsthat contribute to premature airway closure. Inter-stitial edema, airway obstruction from secretionsand bronchoconstriction, and loss of surfactant arepulmonary factors that tend to promote airwayclosure and atelectasis.34

A routine program of respiratory therapy before,during, and after surgery can reduce the number of


682

postoperative respiratory complications and im-prove patient outcome when they do arise. The firstfew hours after the operation are the most criticalfor preventing respiratory complications. Alveolarcollapse, with the resulting loss of FRC, may beprevented or reversed by deep breathing. Maneu-vers that utilize forced exhalation (eg, blow bottles)do not effectively prevent airway collapse and may,in fact, promote this problem if an ineffective inspi-ration precedes the forced exhalation. Deep-breath-ing exercises, with inspiration to total lung capac-ity, are easily the most effective, cost-efficient meansto prevent and reverse atelectasis (and hence, mostpulmonary complications). Incentive spirometersallow patients to participate in their therapy and

free the respiratory therapist to perform other du-ties. Most studies have found that intermittentpositive-pressure breathing is, at best, equal to, andin most cases, inferior to incentive spirometry as aprophylactic measure to prevent postoperative pul-monary complications.33

The likelihood that a patient will develop post-operative respiratory complications is directly re-lated to the type of surgical procedure and theduration of surgery and anesthesia; the patient’ssmoking history, preexisting pulmonary disease,obesity, and overall physical condition; and otherfactors.35 For the most part, soldiers are in excellentphysical condition and do not demonstrate signifi-cant preexisting lung disease.

Although exsanguination is the most commoncause of early death among combat casualties, medi-cal officers must not forget that the first step to betaken in the management of any trauma victim is toassure a patent airway.36 Respiratory failure mustbe recognized and treated promptly. Upper airwayobstruction may not be obvious on initial assess-ment in the field, and may recur anytime during themedical-evacuation process. All military medicalpersonnel should be trained in simple maneuvers torelieve upper airway obstruction. Should ahypovolemic trauma casualty require endotrachealintubation and positive-pressure mechanical venti-lation, it is especially important that cardiovascularresuscitation be continued, to mitigate against thepossible adverse effects of the positive pressure.

Supportive Care

Respiratory care begins with the use of supple-mental oxygen. The most common methods toemploy this drug are nasal cannulae and face masks.The FIO2 actually achieved with a nasal cannula isdependent upon the patient’s inspiratory flow rateand respiratory rate. A variety of masks are avail-able for oxygen therapy. Higher percentages of FIO2may be obtained by use of the partial rebreathingmask or the nonrebreathing mask. The latter is thetype most frequently applied to patients followinggeneral anesthesia, and is available in deployablemedical facilities, including field echelons. The useof such a mask, which utilizes three unidirectionalvalves, ensures that carbon dioxide will not berebreathed.

The next most common form of respiratory careused for patients is incentive spirometry. The suc-

cess of this therapy in expanding the patient’s FRCdepends on both achieving adequate inspiratoryvolume and sustaining maximum inspiratory ef-fort. Although preoperative teaching is often donein the setting of elective surgery, this obviously willnot be possible in the combat casualty setting.

Intratracheal suctioning is important in main-taining the patency of the proximal airway and inpreventing secretions from obstructing the alveoli.This is particularly true in the casualty with a pen-etrating chest wound, where the potential exists forbleeding into the airway. The casualty can drownin his own blood. Intratracheal suctioning is notentirely benign: it can be associated with compli-cations such as trauma to the tracheal mucosawith resultant hemorrhage, hypoxia, and the vaso-vagal response leading to bradycardia andhypotension.

Endotracheal Intubation

Although most patients who require endotra-cheal intubation will also require mechanical venti-latory support, there are clinical situations in whichintubation should proceed independent of the crite-ria defining respiratory failure. Intubation is re-quired to provide airway protection and to preventaspiration in a patient who is semicomatose. Thetreatment of upper-airway obstruction followingdirect maxillofacial trauma and the prevention ofcomplete airway obstruction following thermal burninjuries are two other clinical examples. A fulldiscussion of the methods of endotracheal intubationis found in Chapter 3, Airway Management.

Prolonged endotracheal intubation brings with itmany possible complications. The posterior

MANAGING RESPIRATORY FAILURE


683

arytenoid cartilages are the most common sites forlaryngeal damage, which is caused by the anteriorbend imposed on the endotracheal tube at this point.Although the definition of prolonged translaryngealintubation and the timing of tracheostomy are notyet answered in the medical literature, most clini-cians consider 14 to 21 days to be a safe periodbefore performing a tracheostomy.37

The complications of an elective tracheostomyare generally greater than those of translaryngealintubation, although the overall incidence ofcomplications is still low (3%). Early stomal bleed-ing is common; bleeding after 48 hours is more wor-risome because it may indicate a catastrophiccomplication: a fistula between the trachea and theinnominate artery. Tracheal–innominate artery

fistulae occur infrequently but have a high mortal-ity.

An increased incidence of bacteremia andnosocomial pneumonia are other problems associ-ated with tracheostomy. Studies38 have shown thatup to 87% of patients with tracheostomies experi-ence chronic aspiration. For example, trachealstenosis is a common, clinically significant compli-cation. The reported incidence varies, but exceedsthat seen in patients with endotracheal tubes.37

The most common site for tracheal stenosis is thestoma itself. A 75% reduction in the diameter of thetrachea may occur before symptoms are manifested.The patient who is developing tracheal stenosis—who experiences stridor at rest—probably has atracheal diameter less than 5 mm.

PHYSIOLOGICAL EFFECTS OF MECHANICAL VENTILATION

Once an endotracheal tube is used for ventila-tion, the patient’s natural mechanism for the warm-ing and humidification of inspired air are bypassed.Immediately following intubation and the institu-tion of mechanical ventilation, patients may experi-ence physical and psychological discomfort, whichresults in coughing and agitation. Most patients—especially those who are awakening, alert, or mildlyobtunded—attempt to “fight” the ventilator.39

Nearly constant medical supervision is required,and restraints may be necessary to avoid self-extubation.

Pulmonary Effects

Positive pressure applied during inspiration re-sults in an alteration of the normal ventilation–perfusion matching. During spontaneous ventila-tion, most pulmonary blood flow and ventilationoccurs in the dependent parts of the lungs. Thisrelationship is changed during mechanical ventila-tion such that the greatest ventilation now occurs inthe superior aspects of the lung. This is most likelythe result of changes in the diaphragmatic shapeand motion following positive airway pressure.The result is an increase in the physiological deadspace following the institution of mechanicalventilation.

The use of positive-pressure ventilation may alsocreate changes in the lung mechanics. Reductionsin the FRC and pulmonary compliance may resultin an increase in the work of breathing, and in-creased minute ventilation requirement.

The application of expiratory distending pres-sure results in an improvement of gas exchange and

pulmonary compliance secondary to an increase inlung volumes. The mechanism of the increase inFRC is not clearly known but probably is the resultof one or more of the following:

• an increase in the transpulmonary pres-sure;

• “recruitment” of atelectatic alveoli, result-ing in their reexpansion; or simply

• the maintenance of alveolar patency, whichwas achieved during inspiration.

It is unlikely that the application of PEEP can openatelectatic alveoli that were not reexpanded duringpositive-pressure ventilation. Expiratory distend-ing pressure also causes a decrease in the closingvolume below the FRC, which mitigates againstairway collapse during exhalation.

Conversely, an increase in lung volume may bedetrimental if the FRC exceeds the normal volume.Excessive increases in FRC reduce pulmonary com-pliance and increase the pulmonary vascular resis-tance and dead space. Contrary to earlier beliefs,applying expiratory distending pressure actuallyincreases total lung water, although oxygenationusually improves.

Cardiovascular Effects

The effects of expiratory distending pressure canbe divided into (a) the effects on pulmonary vascu-lar resistance and (b) the effect on cardiac output.The pulmonary vascular resistance is lowest whenthe FRC is normal. If the FRC is returned to normalby the recruitment of atelectatic alveoli, and venous


684

The effect of expiratory distending pressure onventricular afterload differs between ventricles. Anincrease in right ventricular afterload is not predict-able and depends upon the amount of increase inthe pulmonary vascular resistance. At higher levelsof PEEP (> 15–20 cm H2O), right ventricular dilationoccurs resulting in a shift of the intraventricular sep-tum and possibly contributing to a decreased leftventricular stroke volume and cardiac output.40 Inthis setting, the PAOP or pulmonary capillary wedgepressure (PCWP) may not reflect the true left ven-tricular end diastolic volume. As a result of an in-creased pleural pressure, the left ventricle may becompressed (ie, assisted). A failing left ventricle per-forms better when the afterload has been reduced.

Renal Effects

High levels of expiratory distending pressurehave been associated with decreased urinary out-put. Animal studies comparing continuous, posi-tive airway pressure to spontaneous breathing (ie,IMV) with PEEP failed to implicate reduced cardiacoutput as the etiologic factor.41 Two possible expla-nations are (1) an increase in venous pressure, re-sulting in a decreased glomerular filtration rate and(2) an alteration of the distribution of intrarenalblood flow.

admixture is decreased, hypoxic pulmonary vaso-constriction will be reduced. In contrast, if thealveolar pressure is increased beyond normal (tho-racic overinflation), pulmonary capillary pressurewill rise, resulting in an increase in the pulmonaryvascular resistance.

The varying effect of expiratory distending pres-sure in different patients depends on thoracic com-pliance, which governs the transmission of pres-sure to the pleural space and great vessels. Whenthe pulmonary compliance is low (eg, in a patientwith ARDS), a minimal amount of expiratory dis-tending pressure is transmitted to the pleural space.Thus, these patients frequently tolerate high levelsof PEEP relatively well. If, however, the thoraciccompliance decreases from abdominal distentionor postoperative atelectasis, much of the expiratorydistending pressure will be transmitted to the pleu-ral space. This results in a potentially significantdecrease in both venous return and cardiac output.The treatment is to increase the peripheral venouspressure, usually by volume administration. Theextent to which this complication occurs may alsobe minimized by permitting spontaneous breathsthrough the use of IMV. Negative-pressure, spon-taneous breaths serve to minimize the effect ofexpiratory distending pressure by augmentingvenous return.

COMPLICATIONS FROM MECHANICAL VENTILATORY SUPPORT

A variety of complications may occur duringmechanical ventilation. Two of the most importantare cardiovascular depression and barotrauma.First, positive-pressure ventilation necessarily in-creases intrathoracic pressure. The increased in-trathoracic pressure reduces venous return to theheart, and, therefore, right ventricular preload. Thishas two effects: (1) the reduction in preload resultsin a decrease in cardiac output; and (2), right-ven-tricular afterload increases, probably as a result ofdirect compression of the pulmonary vasculature.These two detrimental effects act in concert to re-duce cardiac output, and both effects tend to bemagnified by hypovolemia and PEEP.21

The second major complication of positive-pres-sure ventilation is barotrauma. The spectrum ofbarotrauma extends from pulmonary interstitialemphysema through pneumomediastinum, pneu-mothorax, and subcutaneous emphysema.42 Cur-rent theories on barotrauma relate the initial eventas alveolar rupture with dissection of air along thevascular sheath to the mediastinum. Once air en-

ters the mediastinum, it may track along the fascialplanes into the pleural space, subcutaneous tissue,or retroperitoneum.43 It is the development ofpneumothorax, and in particular, of tension pneu-mothorax, that represents the greatest threat to thepatient. Peak airway pressure is believed to be theprimary factor related to the development ofbarotrauma. No barotrauma was found in oneseries of patients whose peak airway pressure couldbe maintained at less than 50 cm H2O.42 Althoughexperts debate the maximal pressure that can betolerated without barotrauma, the practice of mini-mizing peak airway pressure to forestall the devel-opment of barotrauma is standard for managingpatients with respiratory failure.

The finding of pulmonary interstitial emphy-sema on chest X-ray examination should alert theclinician to the real possibility that the patient maysoon develop a pneumothorax. Because pneumo-thorax may present abruptly, the diagnosis shouldbe made on the basis of clinical examination. Thecombination of respiratory distress, hypotension,


685

hyperresonance, diminished breath sounds, andtachycardia should lead the clinician to evacuatethe presumed pneumothorax with a large-boreneedle placed in the second or third intercostalspace in the midclavicular line on the affected side.

Following immediate decompression by needledrainage, a tube thoracostomy should be per-formed, with the tube connected to a Heimlichvalve or a water seal to preclude reaccumulation ofair in the pleural space.

ETHICAL DILEMMAS IN USING LIMITED LIFE-SUSTAINING THERAPY

“Battlefield military medicine has a threefoldmission: to save lives, to alleviate suffering, and toreturn soldiers to duty.”15(p192) The question existsas to whether mechanical ventilation is a usefulcomponent of the medical officer’s armamentariumfor the treatment of combat casualties. In otherwords, is it likely that the casualty who requiresmechanical ventilation will be returned to the battle-field? Probably not, with these two exceptions:when perioperative ventilation is used for residualanesthetic effects, and when mechanical ventilationis used to support casualties of chemical war gases.However, the mission of battlefield medicine is alsoto save lives, and the use of mechanical ventilationto accomplish this goal is without dispute.

When the number of mechanical ventilators islimited, medical officers will be required to decidewhich patients will receive them. This decision

may be based on several considerations. Probablythe most important, and least-biased, factors are (a)the expected reversibility of the inciting injury and(b) the predicted patient outcome. For example, thecasualty who develops ARDS will most likely requireat least 2 to 3 weeks of mechanical ventilatory support.In addition, the mortality rate exceeds 40%.24 If thispatient receives a ventilator, several others with anexpected quick recovery may either sustain neuro-logical injury or death if no respirator is availablefor respiratory support. A second example is thepatient who develops multiple organ failure. If threeor more organ systems have failed for at least 3 daysin the ICU, the mortality exceeds 95% to 97%.44

Another factor that must be considered when alimited number of mechanical ventilators is avail-able is the medical officer’s experience and skill inmanaging critically ill patients nonmechanically.

MECHANICAL VENTILATION AND COMBAT CASUALTIES

The medical officer working in the combat casu-alty environment must (a) understand the types ofpulmonary injuries that may occur and (b) be ableto evaluate the need for mechanical ventilatorysupport. The decision to use a mechanical deviceto assist ventilation is primarily based on thepatient’s inability to sustain sufficient gas exchange.The most common reason for requiring positive-pressure ventilation is refractory hypoxemia. His-torically, this use of positive-pressure ventilationdates back to 1938 for the treatment of pulmonaryedema.45

Military anesthesiologists and critical care spe-cialists must understand the design of mechanicalventilators, be able to choose the proper mode ofventilation, and be able to select the appropriateinitial ventilatory settings. These required settingsinclude tidal volume, rate, peak gas flow, inspira-tory-to-expiratory ratio, and the fractional concen-tration of oxygen. A further challenge to medicalofficers is the lack of sophisticated monitoring forcasualties who are mechanically ventilated. Thus, agood understanding of pulmonary physiology andastute clinical-assessment skills are required to de-

termine if the patient is appropriately ventilatedand oxygenated.

In contrast to the Vietnam War, where the capa-bility for mechanical ventilation was very limited,in future conflicts, state-of-the-art ventilators maybe available to save the lives of casualties whowould otherwise die. During the months followingthe 2 August 1990 invasion of Kuwait by the Iraqiarmy, the Department of Defense purchased manytypes of mechanical ventilators, which were sent tothe Saudi Arabian theatre. An unexpected problemarose, however: although many different kinds ofventilators were available, most of them were unfa-miliar to young physicians new to the military.

Additional problems relating to the provision ofmechanical ventilation in the field were the need foroxygen, oxygen blenders, PEEP valves, and spirom-eters. The principal problem on the battlefield is theavailability of compressed oxygen. During the Viet-nam War, oxygen was provided principally in cyl-inders, with replenishment by liquid oxygen. Infuture conflicts, there are no plans for pipelineoxygen supply in the land hospitals; however, thereshould be an increased availability of liquid oxy-


686

gen. Use of oxygen sieves may become common-place. The logistics of continual provision of filledoxygen tanks is significant.

Indications for Mechanical Ventilation

A combat casualty’s need for mechanical ventila-tory support generally falls under one of three cat-egories: impaired oxygenation, inadequate ventila-tion (eg, the central ventilatory drive or thefunctional integrity of the thoracic cage is impaired),or an excessive amount and/or inefficiency of res-piratory work. A number of clinical problems areassociated with more than one of these deficiencies(Exhibit 25-3).

Several of these indications are self-explanatory;however, a few merit specific comments. Oxygenitself may be toxic to the pulmonary system whenadministered in a high quantity. If a peripheraloxygen saturation above 90% cannot be obtainedwithin a reasonable amount of time (< 12 h) by the useof a face mask–nonrebreathing apparatus (FIO2 > .80),then endotracheal intubation and mechanical venti-lation should be instituted. Improved oxygenationmay be achieved by manipulation of the FIO2, meanairway pressure, or the pattern of ventilation.

Although patients in cardiovascular collapse orshock may not meet the arterial blood-gas criteriafor respiratory failure, supportive care for thesepatients includes using mechanical ventilation toreduce work demands placed on a patient withinadequate cardiac output. Data from animal stud-ies support this approach, as mortality was reducedin the group of animals in shock that were sup-ported with mechanical ventilation.39

Functional Design of Mechanical Ventilators

The type of ventilator is designated by its mecha-nism of cycling (ie, the event that terminates inspi-ration). There are three principal types of cycling:pressure, volume, and time. Both volume and timeallow for a constant level of tidal volume. The tidalvolume varies during use of pressure-cycled venti-lators whenever pulmonary compliance or airwayresistance change. An important component of theventilator’s design is its driving power (ie, whatempowers positive pressure); either compressedgas (usually oxygen) or electrical power may beused. Most modern ventilators, and all that aredesigned for use in ICUs, are electrically powered,have an internal battery for back-up power, and areoften controlled by a microprocessor. For medical

EXHIBIT 25-3

INDICATIONS FOR MECHANICALVENTILATION

Impaired Oxygenation

Direct thoracic trauma with resultingpneumothorax

Hemothorax or pulmonary contusion

Hypovolemia and shock

Pulmonary edema

Progressive atelectasis

Pneumonia or aspiration pneumonitis

Adult respiratory distress syndrome

Impaired Ventilation

Postoperative state with residual anesthesia

Head injury

Chemical warfare injury

Quadriplegia from spinal cord trauma

Diaphragmatic injury

Fractured ribs or flail chest

Excessive Respiratory Work

Multiple trauma

Compensation for severe metabolic acidemia

Severe bronchospasm

Sepsis or multiple organ dysfunction

officers, this advantage is significant: it reduces theamount of compressed oxygen that must be sup-plied to support field hospitals.

Simplistic, mechanical ventilators have only afew components: a gas-controller device, blender,humidifier, PEEP generator, and the circuit tubing.The gas-controller device provides flow either ondemand, based on input from a sensor, or by con-tinuous flow.

The limitations of specific ventilators are dis-cussed later in this chapter.

Pressure Cycled

In pressure-cycled ventilators, inspiration endswhen a preset pressure is reached in the proximalairway, as sensed by a manometer placed within theventilator circuit. The Bennett PR-2 is an example ofthis type of ventilator, and is available in militarywarehouses. It is one of the first kinds of ventilators


687

sent to Saudi Arabia in 1990. However, a pressure-cycled ventilator is used mainly in the patient whois comatose, quadriplegic, or postoperative from gen-eral anesthesia; or who has a large leak around theendotracheal tube and an anesthesiologist or clinicianwho is experienced in reintubation is unavailable.Many limitations discourage the use of the PR-2.

Volume Cycled

In volume-cycled ventilators, the inspiratoryphase ends when a preset volume is delivered to thepatient. This is the most common type of cyclemechanism on modern ventilators and on all ma-chines classified as “ICU ventilators.” Ventilatorsof this type that were sent to the Persian Gulf War tosupport U.S. and coalition forces include the Bear33, Lifecare PLV-100 and PLV-102, Bennett MA-1,and the Puritan-Bennett Companion. A safety fea-ture of an appropriately set peak airway pressureprevents excessive pressure in the airway if eitherpulmonary compliance or airway resistance changesuddenly. When this “pop-off” pressure is reached,the remainder of the tidal volume vents out of thecircuit and remains undelivered to the patient.Volume cycling also has some limitations, includ-ing the need for a good seal around the endotra-cheal tube, and an inspiratory time that is unre-sponsive to the patients’ own respiratory pattern.

Time Cycled

In time-cycled ventilators, the inspiratory phaseends when the selected time is reached. Ventilatorsof this type include the Servo 900C (Siemens-Elema,Solna, Sweden); the Evita (Dragerwerk, Bonn, Ger-many); and the Univent-750 (Impact Instrumenta-tion, West Caldwell, N.J.). The Univent-750 wasdesigned as a transport ventilator and large quanti-ties were sent to the Persian Gulf during OperationDesert Storm. Similar to the volume-cycled ventila-tor, the time-cycled ventilator delivers a relativelyconstant tidal volume. Tidal volume is determinedby either (a) the inspiratory time and the peakinspiratory flow rate or (b) the minute ventilationand the respiratory rate. For example, if the inspira-tory time is 1 second, and the peak flow rate is 60 L/minute, then the set tidal volume will be 1,000 mL.The Evita, also a time-cycled ventilator, is an ex-tremely flexible and capable ICU ventilator andwas supplied in large quantity to receiving hospi-tals in the U.S. Army Seventh Medical Command, inGermany.

Flow Cycled

Pressure-support ventilation is generally viewedas a mode of ventilation and may be performed oneither volume-cycled or time-cycled machines.Nevertheless, the mechanism for termination ofinspiration (cycling) is unique in that flow to thepatient stops when the patient’s intrinsic inspira-tory flow falls below a certain percentage of theinitial effort.

This topic is discussed more fully below.

Modes of Positive-Pressure Ventilation

There is no uniformly accepted terminology forthe different ventilatory modes. The application ofpositive airway pressure can be divided accordingto the timing during inspiration or exhalation, orduring both phases. When comparing the follow-ing modes of ventilation, one general concept shouldbe remembered: positive intrathoracic pressure maycontribute to decreased venous return and, there-fore, to inadequate cardiac output. The degree towhich this complication becomes clinically signifi-cant depends on the duration and level of positiveairway pressure, whether any spontaneous (nega-tive-pressure) breaths occur, and the patient’s pre-ceding intravascular volume and cardiac status.

Inspiratory Positive Pressure

Controlled Mechanical Ventilation. In the con-trolled mechanical ventilation (CMV) mode, theventilator initiates positive-pressure inspiratorycycles independent of the patient’s spontaneousefforts. This method is indicated when the patientis apneic secondary to heavy sedation, paralysis, orcoma; or when neuromuscular pathology, includ-ing quadriplegia, has occurred. A problem ariseswhen the patient tries to breathe spontaneously: thepatient may compete with the ventilator. The resultwill be a patient who is at best uncomfortable, andat worst subject to pulmonary overinflation andbarotrauma. All inspiratory cycles are positivepressure, thus decreased venous return and de-creased cardiac output are common. If the patientdoes not initiate spontaneous efforts, then the en-tire work of breathing is provided by the ventilator.This allows the patient to rest, but also may lead toatrophy of the diaphragm and intercostal muscles.

Assist-Control Ventilation. Again, all inspira-tory cycles are positive pressure with the AC modeof ventilation. Any breaths taken above the rate set


688

by the physician must be triggered by a spontane-ous effort. Competition between the patient and theventilator is therefore avoided, which reduces thepotential for barotrauma. The primary advantages ofAC ventilation are that (1) respiratory muscles restwhen the peak flow is high enough to decrease thepatient’s work of breathing and (2) the patient deter-mines the PCO2 and the pH. Unfortunately, respira-tory alkalosis is frequently observed in critically illpatients. When AC ventilation is used, the medicalofficer must remember that the set respiratory rate isonly a back-up if the patient’s spontaneous rate fallsbelow that frequency. The actual frequency of fulltidal-volume breaths will be determined by thepatient’s spontaneous rate or the set, mechanicalrate—whichever is the greater number.

Intermittent, Mandatory Ventilation. The IMVmode allows unlimited, negative-pressure, sponta-neous breaths between the delivered mechanicalbreaths. As a result, this mode can be used across aspectrum of minimal to complete ventilatory sup-port, offering flexibility to the physician whosepatient load requires a large number of ventilatorsor who is attempting to wean a patient from theventilator. IMV is indicated primarily during wean-ing. However, when the respiratory rate and tidalvolume are set high enough to provide most of thepatient’s minute ventilation requirement, the patientcan rest, and the work of breathing can be reduced.The use of IMV has several advantages (Exhibit 25-4).

Synchronized, Intermittent, Mandatory Venti-lation. The synchronized, intermittent, mandatoryventilation (SIMV) mode was designed to improvepatient comfort and to prevent the stacking ofbreaths (ie, repeated inhalations without full exha-lation) that may lead to barotrauma. As with IMV,the set rate is the minimum number of positive-

pressure breaths that the patient will receive re-gardless of his intrinsic central respiratory drive.However, in SIMV, the patient’s spontaneous breath-ing and the ventilator are coordinated by means ofa circuit manometer, which detects the onset of aspontaneous breath. Within a “window” duringwhich the ventilator should cycle (eg, every 6 swhen the set rate is 10 breaths per minute), if thepatient makes a spontaneous effort, the mechanicalbreath will be triggered and superimposed at thebeginning of the patient’s respiratory cycle, ratherthan randomly throughout. This timing shouldhelp avoid barotrauma by synchronizing the venti-lator-delivered breath with the inspiratory effort ofthe patient. The only significant disadvantage ofthe IMV or SIMV ventilatory mode is the potentialincrease in work of breathing imposed by the de-mand valve incorporated into the ventilator.

Pressure Support Ventilation. The pressure sup-port (PS) mode of ventilation is triggered by thenegative flow created by a patient’s spontaneousbreath. The initial use of this mode was the inter-mittent, positive-pressure breathing (IPPB) device.The level of pressure support set by the physicianprovides for positive airway pressure for the dura-tion of spontaneous inspiration. Depending on thetype of ventilator employed, the pressure returns tothe end-expiratory pressure level when the inspira-tory flow drops below a preset absolute rate orpercentage of the initial inspiratory flow rate. Thispressure augmentation results in reduced work ofbreathing, which compensates for the increasedresistance from the endotracheal tube, ventilatorcircuit, and mechanical demand valve (which maybe difficult to trigger). The total respiratory rate,tidal volume, and length of inspiration remain un-der the patient’s control; the result is markedlyincreased patient comfort. Pressure support is usu-ally used in conjunction with SIMV, which providesback-up ventilation in the event that patient-initi-ated efforts become inadequate (eg, excessive seda-tion, metabolic encephalopathy).

The disadvantages of pressure-support ventila-tion are that (1) the inspiratory cycle uses positivepressure and (2) the delivered tidal volume varieswhen changes in either airway resistance or pulmo-nary compliance occur.

The initial assist-pressure setting can be esti-mated at approximately one half the peak airwaypressure. An in-line expiratory spirometer shouldbe used to measure spontaneous breaths (5–7 mL/kg exhaled tidal volume is a reasonable goal). Noneof the ventilators that were sent to the Persian GulfWar in 1990 had this capability.

EXHIBIT 25-4

ADVANTAGES OF INTERMITTENT,MANDATORY VENTILATION

Distribution of ventilation is more normal

Physiological dead space is reduced

Detrimental effects on the cardiovascular systemare minimized

Some degree of respiratory muscle tone ismaintained

Respiratory alkalemia is minimized


689

Inverse-Ratio Ventilation. As a patient’s respi-ratory gas exchange becomes progressively inad-equate, increasing the inspiratory-to-expiratory ra-tio to greater than 1 may lead to significantimprovement (by elevating the mean airway pres-sure). The most likely mechanism for this improve-ment is an increase in lung volume resulting fromthe recruitment of atelectatic alveoli through thegeneration of intrinsic end expiratory pressure orso-called “auto-PEEP.” Inverse-ratio ventilationmay be accomplished in the pressure or volumecontrol ventilatory modes. The pressure controlvariety is achieved by maintaining the set pressurelimit for a fixed period of time. The volume controlmode of IRV results when the inspiratory flow rateis reduced to the point whereby inspiratory timeexceeds expiratory time.39 Because this pattern ofventilation is nonphysiological, sedation is usuallyrequired for the patient.

High-Frequency Ventilation. High-frequencyventilation can be provided in three modes: high-frequency, positive-pressure ventilation; high-fre-quency jet ventilation; and high-frequency oscilla-tion. The advantages of high-frequency ventilationare that the mean and peak airway pressures de-crease, resulting in minimized risks of barotraumaand cardiovascular impairment. The Food and DrugAdministration approves the use of high-frequencyventilation in adults for only one condition:bronchopleural fistula. The results of high-fre-quency ventilation in patients with ARDS havebeen very disappointing, but there may be some usefor it in treating inhalation injury.11

Expiratory Positive Pressure

The application of positive pressure during ex-halation, whether it occurs in the spontaneous orthe mechanically ventilated patient, is called expi-ratory distending pressure (EDP). The two princi-pal methods of delivery are (1) CPAP, which usespositive airway pressure throughout the spontane-ous ventilatory cycle, and (2) PEEP, which usespositive pressure during the exhalation fraction ofthe cycle.

Expiratory distending pressure can be appliedvia either a flow resistor or a threshold deviceplaced in the expiratory limb of the ventilatorycircuit. The result is that expiration is retarded,prohibiting full exhalation, which, in effect, main-tains alveolar distension. The flow resistor accom-plishes this by changing the size of the orifice forgas flow, whereas the threshold device allows noexpiratory flow until the pressure exceeds the set

valve pressure. The threshold device is preferredbecause it can accommodate a sudden increase inexpiratory pressure (eg, during coughing) withoutallowing a build-up of pressure in the ventilatorcircuit.

Continuous, Positive Airway Pressure

Large, patient-initiated decreases in airway pres-sure are associated with increased work of breath-ing. CPAP may reduce the work of breathing bygiving an inspiratory assist, which leads to an al-most passive inspiratory cycle. Continuous, posi-tive airway pressure delivered via a face mask mayhave a limited role in the treatment of combat casu-alties because of the frequent occurrence ofobtundation, which increases the risk for pulmo-nary aspiration of gastric contents. With properpatient selection, however, continuous, positive air-way pressure has certain advantages: (1) the main-tenance of the patient’s ability to cough, (2) themaintenance of the patient’s ability to warm andhumidify the inspiratory gases, and (3) the avoid-ance of complications associated with endotrachealintubation. The primary disadvantage is gastricdistension (and the concomitant increased risk ofaspiration), which can be minimized by the place-ment of a nasogastric sump.

Clinical Use of Positive End-Expiratory Pressure

The use of PEEP merits specific attention notonly because this technique is popular but alsobecause the misconceptions regarding its use arenumerous. The only valid reason for using PEEP isto increase lung volume—specifically, the FRC. TheFRC is composed of the residual volume and theexpiratory reserve volume. This lung capacity pro-vides for a reservoir of oxygen-filled alveoli andpermits gas exchange between inspirations. TheFRC is frequently decreased following trauma orsurgery (Table 25-3).

In most patients, the length of the expiratorycycle far exceeds the inspiratory. If alveoli arepermitted to close during the expiratory time, thenintrapulmonary shunting, resulting in hypoxemia,may occur. This premise was used to defend the“super” PEEP that was used during the 1970s, whenPEEP levels greater than 20 cm H2O were titrateduntil the shunt fraction was decreased to the lowestpossible level.46 No prospective study has demon-strated improved patient outcome with the use ofthese high levels of PEEP, leading most clinicians touse PEEP in restricted amounts (eg, < 15 cm H2O).47


690

TABLE 25-3

CAUSES OF DECREASED FUNCTIONAL RESIDUAL CAPACITY

Cause Effect

Residual anesthetic Shifts the CO2 response curve or alters central respiratory control,resulting in shallow tidal volume breathing

Abdominal distension Compresses the diaphragm and the lower pulmonary segments

Splinting of chest-wall Pain prevents deep breathing, which results in collapsed alveolimovements secondary to thoracic pain

Excessive airway secretions Work of breathing increases as the airway is blockedor obstruction

INITIATING MECHANICAL VENTILATION

The initial settings that are selected for mechani-cal support depend on the indication for which thepatient requires ventilatory support. Exhibit 25-5 isa brief guide to setting the ventilator.

Tidal Volume and Respiratory Rate

The recommended initial settings for nonrespir-atory failure–ventilatory support are a tidal volumeof 10 to 15 mL/kg and a respiratory rate of 7 to 10breaths per minute, resulting in a minute ventila-tion of 70 to 150 mL/kg/min. The patient who iscomatose or who remains under the influence ofsedatives will generally be normocarbic at 80 mL/kg/min. Higher rates should be used for the initialsetting when

• hypocapnic therapy is needed to treat in-creased intracranial pressure,

• a severe metabolic acidosis needs immedi-ate compensation,

• a large amount of sodium bicarbonate hasbeen administered,

• a thermal burn has created hypermetabo-lism, or

• pulmonary vascular occlusion has causedan increase in physiological dead space.

The selection of respiratory rate in the AC modeonly determines the minimum amount of ventila-tion that will be ensured if the patient loses hisrespiratory drive. Thus, the selected setting shouldprovide for approximately 70% to 80% of the minuteventilation required for the patient to maintainnormocarbia. To obtain the maximal benefit from

SIMV, once the patient is stable, the respiratory rateshould be set at the minimum amount necessary toprevent respiratory acidosis.

Calculation of the tidal volume should be seton the patient’s actual weight if the weight abovethe estimated lean amount is the result of obesity.Increased weight gain following fluid resusci-tation does not necessitate increasing the tidal vol-ume. Not all the set tidal volume will be deliveredto the patient’s alveoli because of dead space(anatomical and artificial) and volume loss (fromthe compliance of the ventilator circuit tubing).When the patient reaches a peak airway pressureof 40 to 60 cm H2O during ventilation, the aver-age amount of volume lost is 3 to 4 mL/cm H2Opressure.12

Fraction of Inspired Oxygen

Many of the ventilators sent to the Persian Gulf insupport of Operation Desert Storm in 1990 requiredan additional, external oxygen blender to provideoxygen in a relatively safe range (> room air [.21] and< 1.0). All FIO2 settings—even on a sophisticated ICUventilator—should be confirmed by an oxygen ana-lyzer that is placed within the inspiratory limb of thepatient’s breathing circuit. The initial setting for FIO2will depend on whether pulmonary injury has oc-curred. In a postoperative patient who has sustainedonly an extremity or lower-abdominal wound, a slightincrease in the FIO2 to .40 will usually suffice. Ifsignificant pulmonary injury has occurred, then aninitial FIO2 of 1.0 should be selected. All attemptsshould be made to use the lowest concentration ofoxygen that provides at least 90% oxygen saturation


691

Volume-Cycled Ventilators

Apnea Spontaneous Ventilation

Control Mode Assist-Control/Synchronized, Intermittent,Mandatory Ventilation

1. TV: 10–15 mL/kg 1. TV: 10–15 mL/kg

2. Not applicable 2. Sensitivity: –2 to –2.5 cm H2O

3. Respiratory rate: 3. Respiratory rate:

8–12 breaths per min AC: set default rate to 6–8 breaths permin in case of apnea

SIMV: to rest the patient, ensure that80% of breaths are mechanicallyprovided

4. Not applicable 4. PF: 3.5–4.0 • minute ventilation

5. Expiratory time should exceed inspiratory time by 2:1

6. Keep PAP < 30–40 cm H2O

7. Add PEEP in 2- to 4-cm H2O increments if hypoxemiapersists on FIO2 > 50%

8. Set PL at 15–20 cm H2O above PAP during a normalbreath or < 80 cm H2O

Pressure-Cycled Ventilators

1. TV: starting at 20 cm H2O, increase peak pressure until tidal volume is adequate by clinical examinationor arterial blood gas criteria

2. Respiratory rate: provides for a minimum amount if patient becomes apneic. Set to provide 70%–80% ofminute ventilation

3. FIO2: an in-line O2 analyzer is strongly recommended

4. Patient interaction: the spontaneously breathing patient may not tolerate this type of ventilator, necessi-tating either weaning or changing to a different mode of ventilation

Time-Cycled Ventilators

1. TV: set the inspiratory time and PF to provide the calculated tidal volume

2. Respiratory rate: ensures a minimal amount of ventilation

3. As the PF determines the TV, it is necessary to check the inspiratory-to-expiratory time ratio with eachrate change

EXHIBIT 25-5

MODES AND SETTINGS OF POSITIVE PRESSURE VENTILATION

AC: assist-control ventilation; FIO2: fraction of inspired oxygen; PAP: peak airway pressure; PEEP: positive end-expiratorypressure; PF: peak flow; PL: pressure limit; SIMV: synchronized,intermittent, mandatory ventilation; TV: tidal volume

of the arterial hemoglobin. High oxygen concentra-tions are toxic to the lungs and may result in numer-ous complications. These include a deterioration inventilation–perfusion matching, absorption atelec-tasis, decreased mucociliary clearance, and ARDS.48

Decreasing the FIO2, utilizing a pulse oximeter as amonitor, may be safely done in increments of 5% to10% every 5 minutes.

Inspiratory Flow Rate

The peak inspiratory flow rate is particularlyimportant. If it is set too high, ventilation ismaldistributed throughout the lungs. If set too low,the expiratory time may be shortened, creating aproblem for patients with obstructive airway dis-ease. In all cases, the peak inspiratory flow rate


692

must exceed the patient’s intrinsic demand if fur-ther respiratory work by the patient is to be avoided.A setting that is 3.5- to 4-fold greater than thepatient’s minute volume requirement is usuallyadequate. When increasing the peak flow rate, thephysician will generally observe a rise in the peakairway pressure. Occasionally, the peak airwaypressure will decrease, signifying that intrinsic (orauto-) PEEP is occurring. This will resolve withlengthening of the expiratory time.

Additional Settings

The remainder of the mechanical ventilator set-tings are often overlooked by medical officers butwill be set by a respiratory therapist. These includethe inspiratory-to-expiratory ratio, the selection ofan inspiratory waveform (square, ramp, sine), anda variety of alarms. The most important alarm,which is accompanied by a release mechanism, isthe peak airway pressure. Initially, this should beset for approximately 15 to 20 cm H2O higher thanthe peak pressure measured during unlabored, me-chanical breathing. These additional settings areimportant for the intensivist when a patient is par-ticularly difficult to ventilate.

Stabilizing the Patient in Acute RespiratoryFailure

The approach to initiating mechanical support inthe patient in acute respiratory failure differs fromthe previous discussion. The initial settings recom-

mended above are merely a guideline for initiatingmechanical support. As a result of either hyperme-tabolism or increased dead-space ventilation, a com-bat casualty’s minute ventilation requirement maybe as high as 25 to 30 L/min. Respiratory musclefatigue appears to be a common component of acuterespiratory failure regardless of the etiology.49 Theinitial goal is to ensure a good quality of rest andrecovery. Using a constant tidal volume of 10 to15 mL/kg, the respiratory rate should be increaseduntil the patient is comfortable and rested. Whenthese high levels of mechanical support are re-quired, bedside supervision by an intensivist isnecessary to avoid (a) excessive elevation of airwaypressure and (b) air-trapping, which results from aninadequate expiratory time. Although the AC modemay allow the greatest decrease in the patient’swork of breathing, a very tachypneic or anxiouspatient may cause excessive triggering and stackingof ventilations. Occasionally, the patient may needto be sedated and paralyzed so that hemodynamiccompromise can be minimized and the eliminationof carbon dioxide can be improved.

Patients who do not respond initially to treat-ment for sepsis or shock should be considered forearly intubation and mechanical ventilatory sup-port. Considerable oxygen can be consumed by thediaphragm alone (as much as 30%–40% of the body’stotal amount) in the state of shock or acute respira-tory failure.50

Following a period of stabilization, a progressiveamount of work should be assumed by the patientto prevent atrophy and promote conditioning.51

MONITORING THE EFFECTIVENESS OF VENTILATION AND OXYGENATION

Monitors used in patients with acute respiratoryfailure may be characterized as invasive ornoninvasive, and intermittent or continuous. Amonitor may either follow only the progression ofdisease or give concomitant information regardingthe effect of therapy.

Oxygen Concentration

The FIO2 delivered to the patient is one of themost important and easiest parameters to measure.Once the FIO2 is known, other indices—all of whichserve as markers for the severity of pulmonarydysfunction—can be calculated:

• the difference between the partial pressureof oxygen within the alveoli and that presentwithin the arterial blood (PAO2 – PaO2); nor-mal is 5 to 10 mm Hg;

• the ratio of arterial to alveolar partial pres-sure of oxygen (PaO2/PAO2); normal is 0.75; and

• the ratio of arterial partial pressure of oxy-gen to the fraction of inspired oxygen (PaO2/FIO2); normal is 250 mm Hg.

Pulse Oximetry

The pulse oximeter is a continuous monitor thatdetects the percentage of total hemoglobin satu-rated with oxygen. This is achieved by using aphotoelectric tube to measure the amount of lighttransmitted at two different wavelengths. Only thearterial component of the blood in the tissue isanalyzed. This is done by simultaneously measur-ing pulsatile flow and subtracting the backgroundabsorption, which is the nonpulsatile venous andcapillary blood. Because of a different stereochem-istry of hemoglobin molecules, saturated and un-


693

0 50 100 1500

20

40

60

80

100

Oxygen Pressure (mm Hg)

Oxy

hem

og

lob

in S

atu

rati

on

(%

)

Fig. 25-2. The sigmoid shape of the oxyhemoglobin disso-ciation curve, with its nonlinear relation between thepartial pressure of oxygen dissolved in blood and thepercentage of oxygen saturation of hemoglobin, must beborne in mind when interpreting blood gas data. Theshape and position of the curve can be estimated in termsof these three points, which fix the relation betweenpartial pressure and saturation: (a) the oxygen pressureat which hemoglobin is 50% saturated (P50), (b) the oxy-gen pressure of mixed venous blood, which is normally75% saturated (SmvO2), and (c) the partial pressure atwhich hemoglobin is 90% saturated.

saturated hemoglobin display unique absorptionspectrums. To use this information clinically, thephysician must know the three key points on theoxyhemoglobin dissociation curve (Figure 25-2):

a. the P50, the PO2 at which 50% of the hemo-globin molecules are saturated, which nor-mally occurs at 27 mm Hg;

b. the mixed venous saturation (SmvO2) (75%),corresponding to a PO2 of 35 to 40 mm Hg;and

c. the “knee” of the curve at 90% saturationof hemoglobin molecules, at a PO2 of ap-proximately 60 mm Hg.

This technology has several limitations: pulseoximetry (a) does not compute carboxyhemoglobinor methemoglobin as desaturated hemoglobin, (b)is inaccurate in instances of severe anemia (< 7 g/dL), and (c) does not measure venous oxygen satu-ration because venous blood is only slightly pulsa-tile. The latter situation may occur when largeamounts of PEEP are applied. One important pointis that the patient’s PO2 may decrease from 400 to100 mm Hg, representing a significant change inpulmonary function, yet the pulse oximeter satura-tion display does not change.

The Persian Gulf War was the first conflict inwhich pulse oximeters were widely available for

patient monitoring. Oximeters were present at allanesthetizing locations, within the ICUs, and in theemergency departments.

Gas-Exchange Indices

Partial Pressure of Arterial Oxygen

Factors other than the gas-exchange capabilityand efficiency of the lung may affect the PaO2. Twoclinical situations that exemplify this point are (1)shock, during which the venous PO2 is markedlyreduced, and (2) a state of increased VO2. Thedesaturating effect of venous hypoxemia is exacer-bated in the presence of lung pathology, contribut-ing to intrapulmonary shunting. Hence, when hy-poxemia (low PO2) occurs, it is not a correctassumption that the patient’s clinical problem isone of strictly pulmonary pathology. Conversely, ifthe PO2 is completely normal for a given FIO2, thelungs are assumed to be functioning normally. ThePO2, when measured by an arterial blood-gas deter-mination, is the accepted standard for measuringoxygenation.

Arterial blood gases as a monitor of oxygenationhave the limitation of reflecting the clinical state atonly one moment. Furthermore, the PaO2 measuresonly the oxygen dissolved in the blood, which isonly a small fraction of the total amount of oxygendelivered to the tissues. Most of the arterial oxygencontent is that portion of oxygen bound to hemoglo-bin. However, the PaO2 is the major determinant ofthe position on the oxygen–hemoglobin dissocia-tion curve and thus determines the arterial hemo-globin oxygen saturation.

Alveolar–Arterial Oxygen Tension Difference

The difference between the partial pressure ofoxygen in the alveoli and in the arteries (PAO2 –PaO2) is a specific indicator of pulmonary pathol-ogy, but this index has a significant practical limita-tion: the clinician must know the FIO2. Multipledeterminations of the calculated value for PAO2 –PaO2 are accurate predictors of pulmonary injuryonly when the FIO2 has remained constant. PAO2 –PaO2 is calculated using the following equation:

PAO2 – PaO2 = FIO2 (PB – PH2O) – PCO2/R.Q.

where PB represents barometric pressure (760 mmHg at sea level); PH2O represents partial pressure offully saturated water vapor (47 mm Hg); and R.Q.represents the respiratory quotient, assumed to be0.8 with a normal diet. The normal value for PAO2 –

a

b

c


694

PaO2 is less than 10 mm Hg for a healthy, youngadult.

The Ratio of Arterial to Alveolar Oxygen Tension

The arterial-to-alveolar oxygen tension ratio(PaO2/PAO2) remains more stable with alteration inthe FIO2, and this calculation can be employed topredict the expected PaO2 when the FIO2 is altered.The normal value is 0.75.

The Ratio of Arterial Oxygen to Inspired Oxygen

The arterial-to-inspired oxygen ratio (PaO2/FIO2)is the simplest index to calculate, as it does notrequire use of the alveolar gas equation. The nor-mal value is greater than 250 mm Hg.

Right-to-Left Shunting

Right-to-left intrapulmonary shunting of bloodleads most of the hypoxemia seen with severe respi-ratory failure. The shunt equation applicable tosevere respiratory failure is

Qs/Qt = CcO2 – CaO2/CcO2 – CvO2

where Qs represents the flow to the shunt, Qt repre-sents the total pulmonary blood flow, and CcO2represents the oxygen content of pulmonary cap-illary blood, which equals (1.34 mL O2/g Hb) •(Hb g/dL) • (% SO2). The pulmonary capillaryblood is assumed to be fully saturated (SO2 = 100%).CaO2 represents the arterial oxygen content andCvO2 represents the pulmonary venous oxygen con-tent whose hemoglobin oxygen saturations can bedirectly measured. This equation discounts thesmall contribution of dissolved oxygen in comput-ing the various oxygen contents.

If FIO2 = 1.0, then the following simplified shuntequation may be used:

Qs/Qt = P(A – a)O2/20

Monitoring Mechanical Ventilation

The importance of bedside patient observationcannot be overemphasized. Much information isgained by attentively watching the patient’s spon-taneous ventilatory efforts and use and coordina-tion of the thoracoabdominal muscles of inspiration.In addition to the more-specific pulmonary patholo-gies, tachypnea is an early sign of shock or sepsis.

Airway Pressures

The peak and plateau airway pressures are moni-tored in all patients in respiratory failure who aresupported by mechanical ventilation. From thesemeasurements the dynamic and static compliancemay be calculated as follows:

Cd = change in volume/(PAP – PEEP)

Cs = change in volume/(PLP – PEEP)

where Cd represents dynamic compliance, Cs repre-sents static compliance, change in volume representsventilator tidal volume, PAP represents peak air-way pressure, PEEP represents positive end-expi-ratory pressure, and PLP represents plateau pres-sure. Actual tidal volume should be corrected bysubtraction of the volume lost for circuit expansion(3–4 mL/cm H2O peak airway pressure).

A single, discrete value cannot be given be givenfor the dynamic compliance, which is flow depen-dent. Static compliance has the range 75 to 125 mL/cm H2O. The normal values are position depen-dent, and, for supine measurements, the normalvalue should be reduced approximately 30%.

The measurement of dynamic compliance isaffected by the airway resistance, the complianceof the pulmonary parenchyma, and the compli-ance of the chest wall. The difference in peak andplateau pressures (normal = < 10 cm H2O) indicatesthe difference between these two compliancesand is independent of airway resistance.52 If thisdifference is large, then therapeutic interventionshould be directed toward minimizing or eliminat-ing the contributing factors (Exhibit 25-6).Bronchospasm is a common cause of increased air-way resistance during mechanical ventilation. Stan-

EXHIBIT 25-6

FACTORS CONTRIBUTING TOINCREASED AIRWAY RESISTANCE

Airway secretions

Kinked endotracheal tube

Narrow internal diameter of the endotracheal tube

Condensed water in the ventilator tubing

Bronchospasm


695

dard medical therapy includes inhaled beta ago-nists, inhaled anticholinergic agents, theophylline,and steroids.

If both the peak and plateau airway pressuresare elevated and there is minimal difference be-tween the dynamic and static compliances, then thecompliance of either the pulmonary parenchyma orthe chest wall has been reduced. Most commonly,this is the result of pulmonary injury (Exhibit25-7). All medical officers caring for patients onmechanical ventilation must remember that thepeak airway pressure is not infrequently elevatedas a result of the selection of ventilatory set-tings (peak inspiratory flow, tidal volume, and res-piratory rate), an example of iatrogenic noncompli-ance.

Capnography

Whenever possible, end-tidal carbon dioxide(ETCO2) monitoring (capnography) should be usedin every patient on mechanical ventilation. Theinitial widespread clinical use of capnography wasto confirm the correct placement of an endotrachealtube. If direct visualization and placement of anendotracheal tube between the vocal cords is notachieved, capnography is the only other absoluteconfirmation of correct placement. A malpositionedendotracheal tube is a greater problem in a criticallyill patient in the ICU than in an anesthetized, para-lyzed patient in the operating room. Patients in theICU are obviously intubated for longer periods oftime, are more mobile, more interactive, often moreagitated, and are not continuously observed at thebedside, compared to patients in the operating room.

Self-extubation is a recognized complication in allICUs, with an estimated occurrence of 11% to 13%.53

In addition, because of patient movement and res-piratory therapies, disconnections in the airwaycircuitry (ventilator tubing, humidifier, nebulizer)occur frequently. Capnography provides immedi-ate detection of airway obstruction, extubation, orventilator disconnection. The presence of a regu-larly occurring exhaled carbon dioxide waveformconfirms the continued presence of an intratrachealtube. The use of pulse oximetry is not a replacementfor monitoring these events, as pulmonary oxygenreserve may allow sufficient gas exchange for sev-eral minutes.

An idealized normal capnogram is shown in Fig-ure 25-3. In contrast to pulse oximetry, monitoringof exhaled carbon dioxide can provide more infor-mation than just the ETCO2 (recorded in mm Hg or% CO2). From the shape of the waveform, we caninfer several observations such as the adequacy ofalveolar emptying and the magnitude of pulmo-nary dead space. A well-defined plateau tells usthat alveolar emptying is adequate and pulmonarydead space is minimal; therefore, ETCO2 closelymatches PaCO2.

The difference between PaCO2 and ETCO2 is nor-mally 4 to 7 mm Hg for most monitors currentlyavailable. An increase in this gradient in a patienton mechanical ventilation suggests the further de-velopment of physiological dead space. Exhibit 25-8lists frequent clinical causes of increased dead spaceventilation.

EXHIBIT 25-7

CLINICAL CAUSES OF REDUCEDPULMONARY COMPLIANCE

Pulmonary edema

Pneumonia

Atelectasis

Pulmonary fibrosis

Pneumothorax or hemothorax

Pulmonary aspiration

Chemical pneumonitis

4. Inspiratorydownstroke

2. Expiratoryupstroke

Time

Fig. 25-3. Idealized capnogram waveform, with timeplotted on the x axis and exhaled carbon dioxide concen-tration on the y axis. The numbered segments indicate thefour phases of the ventilatory cycle. Adapted with per-mission from Benumof JL. Anesthesia for Thoracic Surgery.2nd ed. Philadelphia, Pa: WB Saunders; 1995: 244.

Exh

aled

CO

2 C

on

cen

trat

ion

1. Inspiratorybaseline

3. Alveolarplateau


696

Analysis of the carbon dioxide waveform alsoprovides information regarding bronchospasm (theslope of the ascending segment) and the compe-tency of the ventilator expiratory valve (baselinerebreathing of CO2 > 2–4 mm Hg). One study54

indicated that a rise or reappearance of exhaled car-bon dioxide is one of the first indicators of the returnof spontaneous circulation following a cardiopulmo-nary arrest. Additional studies have demonstratedthat ETCO2 monitoring may be a good indicator ofboth the adequacy of pulmonary blood flow andcardiac output, and also a good prognostic indicatorof the success of cardiopulmonary resuscitation.55

Exhaled Volume Monitoring (Spirometry)

It is crucial that exhaled tidal or minute volumebe monitored while mechanical ventilation is used.Several of the ventilators that were sent to the Per-sian Gulf War—including the Univent 750, Bear 33,Lifecare PLV-100 and Lifecare PLV-102—did nothave built-in exhaled volume monitoring. Moni-toring of exhaled volume is particularly importantwhen using either a pressure-cycled or a time-cycledmechanical ventilator. The use of these ventilatorsdoes not ensure the maintenance of a constant minuteventilation and a stable PCO2 level. Even during theuse of volume-cycled ventilation, measurement ofexhaled volume serves as an additional monitorwarning of a disconnection, a malfunctioning in-spiratory or expiratory valve, an endotracheal ortracheostomy cuff leak, or a marked change in thepulmonary compliance (eg, pneumothorax).

EXHIBIT 25-8

FACTORS CONTRIBUTING TOINCREASED DEAD SPACE

Pulmonary

Bronchospasm

Chronic obstructive pulmonary disease

Excessive positive end-expiratory pressure

Circulatory

Hypovolemia

Hypotension

Reduced cardiac output

Pulmonary embolism (thrombus, air, fat)

Clinical Examination of the Patient–Ventilator Unit

The clinical examination of the patient receivingmechanical ventilation (ie, the patient–ventilatorunit) is the most important “monitor” of all. The firstobservation to make is determining whether the pa-tient is ventilating spontaneously. If the patientdemonstrates a lack of ventilatory effort, it usuallyreflects one or more of the criteria discussed in Exhibit25-9. If the patient is making spontaneous ventilatoryefforts, then ensure that the ventilator is not in theCMV mode, as this setting does not allow for venti-lation beyond that designated by the set rate.

It is equally important to evaluate how well themechanical support is meeting the patient’s ventila-tory requirements. If purpose of mechanical venti-lation is to better meet the patient’s ventilation andoxygenation needs, then allowing the patient toperform an excessive amount of the work of breath-ing defeats the purpose of mechanical support. Theamount of minute ventilation required to maintainnormocarbia can provide useful information as tothe amount of dead space ventilation and the over-all metabolic state in the setting of trauma andserious infection.

The level of peak airway pressure is a major riskfactor for the development of barotrauma. For thisreason, should the airway pressures be increased ornoted to be rising, it is prudent to evaluate the patientfor reversible causes of airway pressure elevation.

Lastly, should the patient demonstrate agitationwhile receiving mechanical ventilation, one shouldalways first assume that the problem rests not with thepatient, but rather with the ventilator itself or themanner in which it is being employed. By fullyevaluating the ventilator and assuring that there areno malfunctions or inappropriate ventilatory settings,the clinician can avoid a potential catastrophe.

Oxygen Delivery

The gas exchange capability of the lung, as as-sessed by the PaO2, AaDO2, or other indices, is onlyone of the components that determines DO2 to tis-sues. The two other major components are the levelof hemoglobin and the cardiac output. (The com-ments in this chapter are limited to respiratoryinsufficiency; the reader can find a more detaileddiscussion in Chapter 24, The Syndromes of Sys-temic Inflammatory Response and Multiple OrganDysfunction.) However, a number of studies56–59

give credence to the importance of the calculation ofboth DO2 and VO2 by use of a pulmonary artery


697

EXHIBIT 25-9

CRITERIA FOR OPTIMAL VENTILATOR USE

1. Is the patient ventilating spontaneously?

If NO, then

a. The settings (TV, rate) for mechanical support exceed the patient’s requirement.

b. The patient’s apneic threshold has been altered (eg, narcotics, benzodiazepines).

c. pH compensation is needed for a severe metabolic alkalosis.

d. The patient does not have an intact central ventilatory drive (eg, due to a closed head injury or coma).

e. The patient’s thoracic or diaphragmatic muscular function is inadequate (eg, due to nerve agents,extreme fatigue, malnutrition).

If YES, then

a. Change from control mode to either AC or SIMV, and

b. Evaluate how hard the patient is working while maintaining this spontaneous ventilation (eg, respira-tory rate, diaphoresis, pulse, mental status).

2. What is the minute ventilation requirement?

a. The requirement will vary frequently in the same patient based on the physical demands and thetherapies being performed.

b. The daily trend in mechanical ventilation provides information as to overall patient status in sepsisand trauma.

3. Is the peak airway pressure too high (ie, progressively increasing or > 50 cm H2O)?

If YES, then

a. Evaluate dynamic (TV/PAP – PEEP) and static (TV/plateau – PEEP) compliances.

b. Evaluate whether bronchospasm or secretions are causing increased airway resistance.

4. Is the patient fighting the ventilator?

If YES, then evaluate for the following:

a. Secretions and the need for suctioning,

b. Bronchospasm, and

c. Stacking of breaths or “auto-PEEP” because of too short an expiratory time or too high a sensitivitysetting.

REMEMBER: the patient is “fighting” for his life. Do not sedate and paralyze the patient until you are confident thatthe ventilator is not malfunctioning and that the settings chosen are appropriate for the patient’s needs.

TV: tidal volume; AC: assist control; SIMV: synchronized, intermittent, mandatory ventilation; PAP: peak airway pressure;PEEP: positive end-expiratory pressure

catheter. Several clinical states including sepsisand ARDS are characterized by what appears to bea pathological oxygen-supply dependency. Sys-temic oxygen need may reach levels as high as 21mL/kg/min.60 (Arterial and mixed-venous oxygencontents and the cardiac index must be measuredsimultaneously to allow these calculations.)

The usefulness of augmenting cardiac output hasbeen difficult to demonstrate in several clinical tri-als. In both the surgical setting and the setting of

myocardial infarction,61 pushing cardiac output withinotropic drugs has not been associated with im-proved outcome. These studies are difficult toperform, since critically ill patients are sometimeshard to match—but meta-analysis and further clini-cal trials will elucidate the true effects of artificallyincreasing oxygen delivery. At this time, the ques-tion remains only partially answered and is thesource of considerable controversy among practi-tioners across the country.62,63


698

offers information pertinent to the patient’s intra-vascular volume status, the presence of conges-tive heart failure, or the possible occurrence of apulmonary embolism. The latter should be sus-pected when a pleura-based, wedge-shaped infil-trate (Hampton’s hump) or regional oligemia(Westermark’s sign) are seen on a chest roentgeno-gram.

Mechanically Ventilated Patients. Patientswho require mechanical ventilation may benefitfrom additional radiographic monitoring for po-tential complications. A daily chest X-ray examina-tion is a good method to use for confirming a safelocation for an endotracheal tube. Although othermonitors (such as capnography) exist to confirmintratracheal placement, a chest X-ray examinationwill reveal the exact location of the cuff and the tipof the endotracheal tube. To prevent eitherintubation of the right main bronchus or accidentalextubation, the optimal distance from the carina tothe tip of the endotracheal tube is 3 to 7 cm. Thecarina is used as a reference point because therelationship of the endotracheal tube tip to theclavicle changes, depending on the angle at whichthe film was exposed. Tracheostomy tubes aremuch more secure, and less frequent radiologicalmonitoring is required.

Mechanical ventilation may be associated withbarotrauma. This is more common when peak air-way pressures exceeding 40 to 50 cm H2O have beenemployed. Barotrauma may take several formsincluding pneumothorax, pneumomediastinum,pneumoperitoneum; or subcutaneous emphysemaof the chest wall, neck, or even the eyes. Patients inrespiratory failure who require high levels of venti-latory support, particularly once evidence ofbarotrauma is present, need frequent, routine chestX-ray examinations, as well as an additional onewhenever a clinical deterioration in either hemody-namic or respiratory status occurs. A smallpneumothorax in a patient being supported withpositive-pressure ventilation may at any time be-come a tension pneumothorax with the accompany-ing hemodynamic compromise.

Hemodynamic Monitoring

Although arterial, central venous, and pulmo-nary artery–catheter monitoring are important inmonitoring the critically ill patient, a full discussionof hemodynamic monitoring is beyond the scope ofthis chapter. Further information may be found inChapter 5, Physiological Monitoring.

Chest Radiography

Daily chest X-ray examinations are an acceptedICU monitor for patients in respiratory failure. Sev-eral studies have validated the importance of dailychest X-ray examinations in critically ill unstable ormechanically ventilated patients: approximately14% to 15% of chest X-ray examinations demon-strate unexpected findings that lead to a change inmanagement.64,65

Most chest X-ray examinations taken of criticallyill patients are made by a portable device at thebedside. These films are taken from an anterior-posterior rather than the standard posteroanteriororientation. This is important for several reasons.The distance from the X-ray beam to the intratho-racic structures is not constant, and objects that arefar from the film (cardiac silhouette) are magnified.Critically ill patients, particularly those in respira-tory failure, are often tachypneic, and obtaining anend-inspiratory chest film is difficult because theycannot hold their breath. It should be rememberedthat a chest X-ray examination cannot properlyevaluate upper-airway obstruction; however, care-ful attention not infrequently reveals tracheal ab-normalities. One example of this is tracheal or leftmain bronchus deviation when a traumatic aorticdissection occurs. The information that can beobtained from a chest X-ray examination dependson the respiratory status of the casualty: whetherthe patient’s ventilations are spontaneous or me-chanically assisted.

Spontaneously Ventilating Patients. Chest radi-ography is used to evaluate the patient for pulmo-nary parenchymal injury. The presence of a radio-logical density may represent collapse of alveoli(eg, atelectasis); or alveolar consolidation from water(pulmonary edema), blood (aspiration or broncho-alveolar hemorrhage); or infected secretions (pneu-monia). These fluids are indistinguishable radio-logically, and clinical examination or otherdiagnostic tests must be conducted to discern theactual cause. The chest X-ray examination may alsodetect pathology in the pleural space (eg, pneumo-thorax, hemothorax, or effusion). Frequently, chestX-ray examinations done in this patient populationare performed in the supine position, and apneumothorax may be missed. One helpful radio-logical sign is the “deep sulcus sign,” a clearcostophrenic angle that often extends well belowthe normal location for the diaphragm.

Additionally, radiological assessment of theheart, great vessels, and pulmonary vasculature


699

WEANING FROM MECHANICAL VENTILATION

EXHIBIT 25-10

CRITERIA FOR WEANING FROMVENTILATORS

Respiratory or Metabolic Demand or Both

Respiratory rate < 25

Minute ventilation < 10 L/min

Dead space ventilation < 30%–40% of totalventilation

Oxygenation

PaO2 > 60 mm Hg on FIO2 < 50%

PEEP < 5 cm H2O

Intrapulmonary shunt < 30%

Adequate hemoglobin concentration (7–10 g/dL)

Respiratory Muscle Strength

Tidal volume > 5 mL/kg

Vital capacity > 10 mL/kg

Negative inspiratory force > –30 cm H2O

Maximal voluntary ventilation > 2-fold theresting minute ventilation

PaO2: partial pressure of arterial oxygen; PEEP: positiveend-expiratory pressure

Weaning the patient from mechanical ventilationrefers to shifting to partial ventilatory support whileobserving the patient’s response to the increasedwork of breathing associated with spontaneous ven-tilation. Weaning is frequently unnecessary in pa-tients who are mechanically supported for a briefperiod of time following surgery or another medi-cal illness and who have not suffered from malnu-trition. These patients should not have developedrespiratory muscle fatigue.

Discontinuation of mechanical ventilation shouldbe considered only if the underlying cause of therespiratory failure (eg, treated pneumonia, resolvedsepsis, arousal from coma) has been reversed.Discontinuation is likely to prove possible only ifthe following preconditions have been met:

• the central respiratory drive is intact, en-suring normocarbia;

• muscular strength is adequate to maintaina clear airway and provide for full sponta-neous breathing;

• the patient is hemodynamically stable (ie,shock is resolved, intravascular volumerestored); and

• carbon dioxide production is not elevated.

Medical officers, however, must consider notonly the casualty’s condition but also the need foraeromedical evacuation to a higher echelon of care.The U.S. Air Force is reluctant to evacuate casualtieson ventilators and, in fact, refused to do so duringthe Persian Gulf War. (See Chapter 27, MilitaryMedical Evacuation, for a more complete discus-sion of this subject.)

Many criteria have been published to evaluate apatient’s ability to be successfully weaned frommechanical ventilatory support (Exhibit 25-10).Some of these criteria are subjective and thus re-quire good clinical acumen as to the physiologicalcapability of the patient. Even objective criteria thatcan be measured and quantified have been ques-tioned, as there are patients who do successfullywean from mechanical support despite their notmeeting the weaning criteria.66,67 In addition, cer-tain of the quantitative measurements are frequentlynot performed properly and thus may give falseinformation. Nevertheless, combining criteria intoa decision tree (Figure 25-4) is usually helpful indeciding if and when to wean. Application of thedecision tree presupposes that the patient is not

sedated to the point where the ventilatory drive isblunted.

Several of these indicators (negative inspiratoryforce, minute ventilation, and maximum voluntaryventilation) were evaluated in one study.68 If theminute ventilation was less than 10 L, then themaximum voluntary ventilation was 2-fold greaterthan the minute volume, and the negative inspira-tory force was greater than –30 cm H2O. All pa-tients were successfully extubated in this study.These parameters are probably of greater use inshort-term ventilator use than they are in prolonged,when the clinical judgment of the physician is asvalid as the measured criteria.

Patients who required prolonged (> 30 d) me-chanical ventilation have been shown to have ahospital mortality identical to patients who requiredonly short-term mechanical ventilation.69 Thus,these patients deserve a well-conceived plan for thegoal of successful extubation and separation frommechanical ventilatory support.


700

Fig. 25-4. Decision tree for weaning from mechanical ventilation.

Is the Disease ProcessThat Led toRespiratory Failure

Resolved or Improved?

Evaluate conventional weaning criteria:1. PaO2 > 60 mm Hg with FIO2 < 50%2. PAO2 – PaO2 < 250 mm Hg3. PaO2/FIO2 > 2004. Vital capacity > 10 mL/kg body weight5. Negative inspiratory effort < –30 cm H2O6. Minute ventilation < 10 L/min7. Respiratory rate < 35/min8. Respiratory rate ÷ tidal volume (in liters) < 100

Weaning from mechanicalsupport is likely to fail

AreAbnormalities

Corrected?

Correct abnormalities:1. Nutrition2. Electrolytes3. Acid–base balance4. Congestive heart failure5. Infection

Patient is likely towean and extubate rapidly

AreAll Criteria

Met?

Weaning from mechanicalsupport is likely to fail

Yes

Yes

Yes

No

No

No


701

The first step in weaning a patient who has beendependent on mechanical ventilatory support is tooptimize the patient’s general status (Exhibit 25-11).This may seem obvious, but it is not fully appreci-ated and is frequently overlooked completely—especially when too much attention is directed to-ward purely respiratory indices. Even if the patientdemonstrates no respiratory distress, it is unwise towean a patient who appears to be in any form ofshock. By definition, the term “shock” means thatthe supply of blood and nutrients (oxygen andglucose) to organs or tissues has not met the de-mand. If such a patient were required to sustain fullspontaneous ventilation, the increase in oxygendemand could worsen the balance between sys-temic oxygen delivery and systemic oxygen need.For example, in sepsis, oxygen utilization by thediaphragm can be very high and, in fact, can consti-tute nearly 30% to 40% of the body’s total VO2.50

Several clinical conditions may lead to an el-evated production of carbon dioxide, necessitatinga higher minute ventilation to maintain normo-carbia. Patients who are febrile, agitated, in pain, orwho have suffered traumatic injuries, thermal burns,or seizures produce increased carbon dioxide. Ex-cessive carbon dioxide production may also ariseiatrogenically as a result of excessive carbohydrateadministration. (For a more complete discussion,see Chapter 23, Metabolic Derangements and Nu-tritional Support.)

It is often impossible to determine the exact causeof a patient’s failure to wean, but recent discussionhas focused on respiratory muscle fatigue and dis-

EXHIBIT 25-11

FACTORS THAT MAY PREVENT DISCONTINUATION OF MECHANICAL VENTILATION

The patient must not be malnourished. Diaphragmatic muscle loss is nearly linearly related to the degree ofmalnutrition.1,2

Significant acid–base and electrolyte abnormalities must be corrected, including metabolic alkalosis thatdemands hypoventilation to normalize the serum pH.

Hypophosphatemia may also contribute to respiratory muscular weakness.

Left-ventricular failure leading to reduced pulmonary compliance may be an occult cause of persistent failureto wean from mechanical support.

Airway secretions may be a significant problem to the patient after extubation, during which time small-airwayobstruction leads to atelectasis and bronchospasm.

Sources: (1) Bell RM, Bynoe RP. Nutrition and respiration. Prob Crit Care. 1987;1(3):413–434. (2) Lewis MI, Belman MJ.Nutrition and the respiratory muscles. Clin Chest Med. 1988;9(2):337–348.

use atrophy as two important causes. It may beclinically impossible to distinguish between thesetwo causes. Respiratory muscle fatigue has beensuspected clinically for some time and electromyo-graphic evidence supports this concern.70 The clini-cian must remain alert to mechanical factors thatexacerbate rather than aid respiratory muscle fa-tigue. Most intensivists know that increased air-way resistance results from using a small endo-tracheal tube (< 7.0 mm internal diameter). Atracheostomy, by increasing the diameter and short-ening the length of the airway, may reduce theresistance. Occasional patients may wean best bysimply proceeding to extubation, totally eliminat-ing the resistance of the artificial airway. (This isbecoming less necessary as advanced mechanicalventilators with multiple ventilatory modes aredeveloped.)

The three principal weaning modalities are (1) aprogressive reduction of mechanical supportthrough the use of IMV, (2) T-piece trials, wherebythe patient is disconnected from the ventilator andattached to a T-piece for progressively longer inter-vals, and (3) pressure-support weaning. The litera-ture does not support the superiority of any onemethod.67

T-piece trials should be initiated in 5-minuteintervals, with the physician or nurse at the bedsideto evaluate any significant changes in the patient’srespiratory rate, pulse, or subjective complaints ofexcessive work. T-piece trials use intermittent peri-ods of vigorous breathing intended to increase thepatient’s respiratory endurance. Rest periods are


702

important to the success of this method. The defini-tion of quality respiratory rest (ie, the respiratorymuscles are resting) has been clouded by recentstudies showing significant amounts of respiratorywork with AC ventilation. In addition, the optimalrest time between periods of T-piece work has notbeen answered in the literature. Our practice hasbeen to allow for a minimum of one hour betweenexercise periods and rest the patient with full ven-tilatory support overnight.

SIMV is another ventilatory strategy commonlyused in weaning. Quite simply, the number ofmechanical breaths per minute provided to the pa-tient are gradually reduced over several days toweeks. As the number of mechanical breaths isreduced, the patient must assume a greater amountof the work of breathing. The potential disadvan-tage of this strategy is the increase in work of breath-ing due to the resistance inherent in the inspiratorydemand valves used in modern mechanical ventila-tors.71 This work may be sufficient to cause theweaning attempt to fail in some patients (eg, thosepatients with severe obstructive lung disease orsignificant respiratory muscle atrophy).

Another relatively new strategy for weaning hasbeen the use of pressure support ventilation, eitherin conjunction with or in place of SIMV. Pressure

support ventilation permits the mechanical flow offresh gas after the patient initiates a breath. Becausethe patient retains control of the total respiratoryrate and the depth and duration of assisted breaths,they are generally more comfortable while receiv-ing mechanical ventilatory support. This modeoffsets the increased work of breathing inherent inthe ventilator circuit due to resistance to gas flow inthe circuit. Most authorities do not recommendreducing the SIMV rate below four breaths perminute unless pressure support ventilation is uti-lized concomitantly.

Several general comments should be made re-garding medical plans for ventilatory weaning:

• Adequate rest periods are essential to pre-vent respiratory muscle fatigue.

• The physician must constantly assess thepatient’s progress during weaning.

• Arterial blood gas measurements are nosubstitute for clinical judgment.

• Hypercarbia is a late finding of inadequaterespiratory muscle strength.

• The patient must be observed for tachypnea,paradoxical respiratory movements, use ofaccessory muscles of inspiration, and dia-phoresis.

VENTILATORS AVAILABLE FOR COMBAT CASUALTY CARE

A large number of mechanical ventilators are inthe inventories of the medical departments of theU.S. armed forces. The following section discussesthe Puritan-Bennett PR-2 (manufactured by theBennett Medical Equipment, Los Angeles, Calif.),which is pressure cycled; the Bear 33 (manufac-tured by Bear Medical Systems, Riverside, Calif.)and the Lifecare PLV-100 and PLV-102 (manufac-tured by Lifecare Corporation, Lafayette, Colo.),which are volume cycled; the Bennett MA-1 (manu-factured by Bennett Medical Equipment, Los Ange-les, Calif.), which is also volume cycled; and theImpact Uni-Vent Model 750 M (manufactured byImpact Instrumentation, West Caldwell, N.J.), whichis time cycled.

These ventilators are available for use with de-ployed medical units.

Puritan-Bennett PR-2

The Puritan-Bennett PR-2 is basically a convertedintermittent, positive-pressure breathing machineand requires a source of compressed gas to func-

tion. The PR-2 is unique among the ventilatorsunder discussion here because it is pressure cycled.The patient’s tidal volume cannot be directly set,but is a result of the pressure limit and the inspira-tory time. The limitations of the PR-2 in providingmechanical ventilatory support to combat casu-alties are as follows:

• If the patient’s compliance decreases or air-way resistance increases (a change in thepatient’s position, secretions in the airway,bronchospasm, or patient agitation), thesame settings of airway pressure will resultin a decreased tidal volume and hypovent-ilation. Thus, we strongly advise that (a)exhaled tidal volume be monitored and (b)an alarm be attached.

• The patient may continue to spontaneouslybreathe; however, competition may occuras there is no mechanism to provide syn-chronization with the patient’s efforts. Thismay result in high airway pressures andbarotrauma.


703

• There is inaccurate control of FIO2 frombreath to breath when oxygen enrichmentis used. The higher the patient’s minuteventilation, the lower the FIO2.

• When the respiratory rate increases and thepressure is kept constant, inspiration mayend prematurely because the constant flowrate results in higher peak airway pres-sures closing the inspiratory valve.

• To provide PEEP, an external valve must beattached to the exhalation limb.

• These ventilators have limited peak flowcapabilities and may not be able to providefor the traumatized casualty whose minuteventilation demands are excessive.

Bear 33 and Lifecare PLV-100 and PLV-102

The Bear 33 and the Lifecare PLV-100 and PLV-102 ventilators are all portable, electrically pow-ered, microprocessor controlled, and volume cycled.They do not require a separate source of compressedgas unless oxygen enrichment is used. All theseventilators can function in AC and SIMV modes,with some limitations. All have an internal batteryand may be connected to an external battery as well.The limitations of these ventilators are as follows:

• “Apnea, breathing-circuit disconnections,leaks, exhalation-valve failures and occlu-sions may remain undetected....”72(p107)

Thus, it is recommended that an exhaledvolume monitor be used with these ventila-tors. There is no built-in capability formonitoring exhaled volume.

• The provision for enriched oxygen supply ispassive and, except in the PLV-102, requiresan externally attached H-valve mechanism.Three problems may be encountered:1. the potential misassembly of the H-

valve, which may lead to obstructionduring patient inspiration;

2. the inaccuracy of the inspired oxygenconcentration, which mandates the useof an oxygen analyzer (the FIO2 will be± 10% if the patient has a stable minuteventilation pattern, and greater if thepatient is unstable);

3. the automatic increase in tidal volume,which occurs with increased oxygenflow (in the Bear 33 and PLV-100), maycause increased airway pressure andbarotrauma.

• “Substantial increases in work of breathingvaried with each ventilator so that the workof breathing approached that of a patientwith high airway resistance: Bear 33, 41%;Lifecare PLV-100 and PLV-102, 88%....”72(p107)

While supported on these ventilators, thepatient may appear dyspneic, and weaningby IMV is likely to be unsuccessful unlessan external H-valve attachment is as-sembled.

• If a microprocessor failure occurs, the Bear33 will not return to the previously setparameters, but will revert to “apnea set-tings” of a tidal volume of 500 mL and arespiratory rate of 16 breaths per minute.

• Lead–acid batteries are used in all theseventilators. The life of this type of batterywill be shortened if it is allowed to remaindischarged. If fully functioning andcharged, the internal battery life is expectedto be 1 to 3 hours.

• To provide PEEP therapy, an externalvalve needs to be attached on the exhala-tion limb.

Bennett MA-1

The Bennett MA-1 has been used for many yearsin ICUs. This ventilator is electronically poweredand volume cycled (termination of inspiration). Itslimitations are as follows:

• There is no built-in monitoring of exhaledvolume.

• An externally attached H-valve apparatusmust be used to allow IMV to be used.Hence, IMV is not synchronized and me-chanical breaths may “stack” on top of spon-taneous breaths.

• To provide PEEP, an external valve must beattached to the exhalation limb.

Impact Uni-Vent Model 750 M

The Impact Uni-Vent Model 750 is the newestventilator to be accepted for military use. TheImpact is a portable, electrically controlled, time-cycled, pressure-limited mechanical ventilator. It ismicroprocessor-controlled and is capable of con-trol, AC, or SIMV ventilatory modes. It has aninternal battery with a fully charged time capabilityof 9 hours. This mechanical ventilator does requirea source of compressed gas to provide patient flow,


704

except during spontaneous breaths in the SIMVmode. The limitations are as follows:

• There is no built-in capability for exhaledvolume monitoring.

• To institute PEEP therapy, an external PEEPvalve needs to be attached and several con-trol settings must be adjusted to allow moni-toring of the amount of PEEP within thepatient’s circuit.

• The work of breathing is not well studiedfor this ventilator. An optional demandvalve (connected to a source of compressedgas) is available, which provides an in-spiratory flow of 60 L/min and is reportedto provide a small pressure assist (support)of 2 to 3 cm H2O.

• If the optional demand valve is not used,

room air will be entrained through theantiasphyxiation port during the patient’sspontaneous breaths. This will decreasethe fractional concentration of oxygen pro-vided to the patient.

• There is no measurement of the size ofspontaneous breaths taken during SIMV.

• The digital display of peak inspiratory pres-sure is hard to quantitate in the most com-mon range of pressures (10–50 cm H2O).

Many of the mechanical ventilators projected foruse in the combat casualty setting are different fromthose used in most peacetime hospitals. For thisreason, it would be prudent for medical personnelwho may be called on to care for patients with theseventilators to familiarize themselves with their op-erating capabilities.

SUMMARY

Respiratory failure of a magnitude that requiresmechanical ventilation can be expected to occur in10% to 15% of hospitalized casualties. The mostlikely causes are ARDS, severe brain injury, andsevere torso injury during the postoperative phase.The diagnosis of respiratory failure depends on theobservation of hypoxemia (PaO2 < 50 mm Hg whilebreathing room air) and ventilatory failure (PaCO2 >50 mm Hg). Although these findings indicate thepresence of severe pathophysiology, respiratory sup-port with supplemental oxygen, incentive spirom-etry, and intratracheal suctioning may decrease theneed for intubation and mechanical ventilation.

Mechanical ventilation has several beneficial ef-fects, including eliminating the work of breathingfor the casualty, maximizing the concentration ofinspired oxygen, and ventilating collapsed andpoorly ventilated alveoli. The latter changes causea decrease in the amount of intrapulmonary shunt-ing and an increase in the functional residual capac-ity of the lung. However, high airway pressureduring expiration—especially in the presence ofhypovolemia—may lead to decreased venous re-turn to the heart, a fall in cardiac output, and adecrease in oxygen delivery. Barotrauma leading totension pneumothorax is a potentially fatal compli-cation of mechanical ventilation but is unlikely tooccur if peak airway pressure is maintained below50 cm H2O.

Military anesthesiologists and intensivists mustunderstand the design of deployable mechanicalventilators, be able to choose the proper mode ofventilation, and be able to select the appropriate

initial ventilatory settings. These required settingsinclude tidal volume, rate, peak gas flow, inspira-tory-to-expiratory ratio, and the fractional concen-tration of oxygen. In addition, anesthesiologistsand intensivists must be familiar with the logisticsof oxygen supply.

Mechanical ventilators use three types of cy-cling: pressure, volume, and time. Both volume-and time-cycled ventilators allow for a constantlevel of tidal volume, which makes them more ef-fective than pressure-cycled ventilators wheneverpulmonary compliance or airway resistance is el-evated. The initial settings that are selected formechanical support depend on the indication forwhich the patient requires ventilatory support. Plau-sible initial settings are tidal volume, 10 to 15 mL/kg; respiratory rate, 7 to 10 breaths per minute; andinspired oxygen concentration, 40%. Both the res-piratory rate and the oxygen concentration mayhave to be increased, and flow rates and end expira-tory pressures adjusted, if pulmonary injury ispresent. A variety of parameters may be measuredto judge the effectiveness of ventilation and oxy-genation (eg, systemic DO2 and the magnitude ofthe arterial-venous shunt), but useful informationis also obtained from the subjective appearance ofthe casualty. In addition, in deployable hospitals,from the practical standpoint, an assessment as tothe effectiveness of ventilation will depend on mea-surement of PaO2 and PaCO2 measured with a bloodgas analyzer, SaO2 measured with a pulse oximeter,a chest radiogram, and the airway flow and pres-sure measurements from the controls and sensors of


705

the mechanical ventilator. The goal is to maximizeoxygenation while minimizing airway pressuresand inspired oxygen concentration.

Discontinuation of mechanical ventilation shouldbe considered only if the underlying cause of therespiratory failure has been reversed. Multiplecriteria have been advanced to guide the weaningprocess once the decision has been made to discon-tinue mechanical support. Among them are (a) PaO2

greater than 60 mm Hg when FIO2 is less than 0.5, (b)the ability to produce a spontaneous vital capacitygreater than 10 mL/kg, and (c) the ability to gener-ate a negative inspiratory effort greater than 30 cmH2O. None of these criteria are absolute. Prema-ture discontinuation of mechanical ventilation andextubation to make possible rapid evacuation—especially when it is by air—is to be avoided incombat casualties.

REFERENCES

1. Bellamy RF. Colonel, Medical Corps, US Army. Personal communication, July 1994. The single most usefuldatabase describing combat wounds and the circumstances of wounding is the Wound Data and MunitionsEffectiveness Team (WDMET) study prepared by the US Army Materiel Command during the Vietnam War.These data are stored at the National Naval Medical Center, Bethesda, Maryland. Access is controlled by theUniformed Services University of the Health Sciences, Bethesda, Maryland 20814-4799; telephone (301) 295-6262. Three summary volumes contain extensive abstracts of the statistical data and can be obtained fromDefense Documentation Center, Cameron Station, Alexandria, Virginia 22304-6145; telephone (703) 545-6700and (703) 274-7633.

2. Arnold K, Cutting RT. Causes of death in United States military personnel hospitalized in Vietnam. Milit Med.1978;143:161–164.

3. Wright S. US Army Medical Department Center and School, Fort Sam Houston, Tex. Personal communication,29 July 1993.

4. Whatmore D. Lieutenant Colonel, Medical Corps, US Army. Personal communication, August 1994.

5. McNamara J, Messersmith J. Thoracic injuries in combat casualties in Vietnam. Ann Thor Surg. 1970;10(4):389–401.

6. Bowen TE, Bellamy RF, eds. Emergency War Surgery NATO Handbook. 2nd rev US ed. Washington, DC:Department of Defense, Government Printing Office; 1988.

7. Bellamy RF, Zajtchuk R, eds. Medical Consequences of Conventional Warfare. In: Zajtchuk R, Jenkins DP, BellamyRF, eds. Textbook of Military Medicine. Washington, DC: US Department of the Army, Office of The SurgeonGeneral, and Borden Institute; 1991.

8. Huller T, Bazini Y. Blast injuries of the chest and abdomen. Arch Surg. 1970;100:24–30.

9. Blood CG. Analyses of battle casualties by weapon type aboard US Navy warships. Milit Med. 1992;157(3):124–130.

10. Fein A, Leff A, Hopewell P. Pathophysiology and management of the complications resulting from fire and theinhaled products of combustion. Crit Care Med. 1980;8(2):94–98.

11. Rue LW, Cioffi WG, Mason AD, McManus WF, Pruitt BA. Improved survival of burned patients with inhalationinjury. Arch Surg. 1993;128:772–780.

12. Levine B, Petroff P, Slade C, Pruitt B. Prospective trials of dexamethasone and aerosolized gentamicin in thetreatment of inhalation injury in the burned patient. J Trauma. 1978;18(2):188–193.

13. Freitag L, Firusian N, Stamatis G, Greschuchria D. The role of bronchoscopy in pulmonary complications dueto mustard gas inhalation. Chest. 1991;100:1436–1441.

14. Bellamy RF, ed. Combat Casualty Care Guidelines: Operation Desert Storm. Washington, DC: US Department of theArmy, Office of The Surgeon General, and Borden Institute; 1991: 28.


706

15. Dolev E. Wartime trauma: Lessons and perspectives. Internat Anes Clin. 1987;25(1):191–202.

16. Teabeaut J. Aspiration of gastric contents: An experimental study. Am J Pathol. 1951;28:51.

17. Patterson A. Pulmonary Aspiration Syndromes in Respiratory Failure. Chicago, Ill: Yearbook Medical Publishers; 1986.

18. Fanning G. The efficacy of cricoid pressure in preventing regurgitation of gastric contents. Anesthesia.1970;32(6):553–555.

19. Allen S, Drake R, Williams J, Laine G, Gabel J. Recent advances in pulmonary edema. Crit Care Med.1987;15(10):963–970.

20. Grande CM, Stene JK, Bernhard WN, Barton CR. Trauma anesthesia and critical care: the concept and rationalefor a new subspecialty. Crit Care Clin. 1990;6(1):1–11.

21. Tobin MJ, Yang K. Weaning from mechanical ventilation. Crit Care Clin. 1990;6(3):725–747.

22. Moon VH, Kennedy PJ. Pathology of shock. Arch Path. 1932;14:360–371.

23. Ashbaugh D, Bigelow D, Petty T, Levine B. Acute respiratory distress in adults. Lancet. 1967;2(7511):319–323.

24. Milberg JA, Davis DR, Steinberg KP, Hudson LD. Improved survival of patients with acute respiratory distresssyndrome (ARDS): 1983–1993. JAMA. 1995;273(4):306–309.

25. Olcott C, Barber R, Blaisdell F. Diagnosis and treatment of respiratory failure after civilian trauma. Am J Surg.1971;122:260–268.

26. Hammerschmidt D, Weaver L. Association of complement activation in elevated plasma C5A with ARDS:Pathophysiological relevance and possible prognostic value. Lancet. 1980;1:947–949.

27. Sutter PM, Schlobohm RM. Determination of functional residual capacity during mechanical ventilation.Anesthesiology. 1974;41:605–607.

28. Bernard GR, Luce JM, Sprung CL. High dose corticosteriods in patients with the Adult Respiratory DistressSyndrome. N Engl J Med. 1987;317(25):1565–1570.

29. Kollef MH, Schuster DP. The acute respiratory distress syndrome. N Engl J Med. 1995;332(1):27–35.

30. Montgomery AB, Stager MA, Carrico J. Cause of mortality in patients with the adult respiratory distresssyndrome. Am Rev Respir Dis. 1985;132(5):485–489.

31. Zapol WM, Snider MT, Hill JD. Extracorporeal membrane oxygenation in severe acute pulmonary failure: Arandomized prospective study. JAMA. 1979;242(20):2193–2196.

32. Abrams JH, Gilmour IJ, Kriett JM, et al. Low frequency positive pressure ventilation with extracorporeal carbondioxide removal. Crit Care Med. 1990;18(2):218–220.

33. Van de Water J, Watring W, Linton L. Prevention of postoperative pulmonary complications. Surg GynecolObstet. 1978;135:129.

34. Latimer RG, Dickman M, Day WC, Gunn ML, Schmidt CD. Ventilatory patterns and pulmonary complicationsafter upper abdominal surgery determined by preoperative and postoperative computerized spirometry andblood gas analysis. Am J Surg. 1971;122(5):622–632.

35. Ward RJ, Danzier F, Bonica JJ, Allen GD, Bowes J. An evaluation of postoperative respiratory maneuvers. SurgGynecol Obstet. 1966;123:51–54.


707

36. Committee on Trauma, American College of Surgeons. Advanced Trauma Life Support Program for Physicians:Instructor Manual. 5th ed. Chicago, Ill: American College of Surgeons; 1993.

37. Berlauk JF. Prolonged endotracheal intubation vs tracheostomy. Crit Care Med. 1986;14(8):742–745.

38. Nash M. Swallowing problems in the tracheotomized patient. Otolaryngol Clin North Am. 1988;21(4):701–709.

39. Marini J, Wheeler A. Mechanical ventilation. In: Critical Care Medicine—The Essentials. Baltimore, Md: Williams& Wilkins; 1989: 65.

40. Biondi JW, Shulman DS, Soufer R, et al. The effect of incremental positive end-expiratory pressure on rightventricular hemodynamics and ejection fraction. Anesth Analg. 1988;67(2):144–151.

41. Shelhamer JH, Natanson C, Parrillo JE. Positive end-expiratory pressure and adults. JAMA. 1984;251(20):2692–2695.

42. Petersen GW, Baier H. Incidence of pulmonary barotrauma in a medical ICU. Crit Care Med. 1983;11(2):67–69.

43. Haake R, Schlictig R, Ulstad D, Henschen R. Barotrauma, pathophysiology, risk factors and prevention. Chest.1987;91(4):608–613.

44. Knaus WA, Wagner DP. Multiple systems organ failure: Epidemiology and prognosis. Crit Care Clin. 1989;5(2):221–232.

45. Barach A, Martin J, Eckman M. Positive pressure respiration and its application to treatment of acute pulmonaryedema. Ann Int Med. 1938;12:754.

46. Gallagher TJ, Civetta JM, Kirby RR. Terminology update: Optimal PEEP. Crit Care Med. 1978;6(5):323–326.

47. Petty TL. The use, abuse, and mystique of positive end-expiratory pressure. Am Rev Respir Dis. 1988;138(2):475–478.

48. Lodato R. Oxygen toxicity. Crit Care Clin. 1990;6(3):749–765.

49. Tobin M, Lodato R. Respiratory monitoring in the ICU. Pulmonary and Critical Care Update. 1988;3(22):1.

50. Roussos C, Macklem P. The respiratory muscles. N Engl J Med. 1982;307(13):786–797.

51. MacIntyre NR. Respiratory function during pressure support ventilator. Chest. 1986;89(5):677–683.

52. Hubmayr RD, Gay PC, Tayyab MM. Respiratory system mechanics in ventilated patients: Techniques andindications. Mayo Clin Proc. 1987;62(5):358–368.

53. Coppolo DP, May JJ. Self-extubations: A 12-month experience. Chest. 1990;99(5):1319–1320.

54. Garnett AR, Ornato JP, Gonzales ER, Johnson EB. End-tidal carbon dioxide monitoring during cardiopulmonaryresuscitation. JAMA. 1987;257(4):512–515.

55. Sanders AB, Kern KB, Otto CW, Milander MM, Eury GA. End-tidal carbon dioxide monitoring duringcardiopulmonary resuscitation: A prognostic indication for survival. JAMA. 1989;2629100:1347–1351.

56. Shoemaker WC, Appel PL, Kram HB, Waxman K, Lee TS. Prospective trial of supranormal values of survivorsas therapeutic goals in high-risk surgical patients. Chest. 1988;94(6):1176–1186.

57. Wolf Y, Cotev S, Perel A, Manny J. Dependence of oxygen consumption on cardiac output in sepsis. Crit CareMed. 1987;15(3):198–203.


708

58. Cain SM. Supply dependence of oxygen uptake in ARDS: Myth or reality? Am J Med Sci. 1984;288(3):119–124.

59. Gutierrez G, Pohil RJ. Oxygen consumption is linearly related to O2 supply in critically ill patients. J Crit Care.1986;1(1):45.

60. Mohsenifar Z, Goldback P, Tashkin D, Campisi D. Relationship between O2 delivery and O2 consumption in theAdult Respiratory Distress Syndrome. Chest. 1983;84(3):267–271.

61. Yu M, Takanishi D, Myers SA, et al. Frequency of mortality and myocardial infarction during maximizingoxygen delivery: A prospective, randomized trial. Crit Care Med. 1995;23(6):1025–1032.

62. Knox JB. Oxygen consumption–oxygen delivery dependency in adult respiratory distress syndrome. NewHorizons. 1993;1(3):381–387.

63. Demling RH. Adult respiratory distress syndrome: Current concepts. New Horizons. 1993;1(3):388–401.

64. Brunel W, Coleman D, Schwartz D, Peper E, Cohen N. Assessment of routine chest roentgenograms and thephysical examination to confirm endotracheal tube position. Chest. 1989;96(5):1043–1045.

65. Strain DJ, Kinasluity GT, Vereen LE, George RB. Value of routine daily chest X-rays in the medical intensive careunit. Crit Care Med. 1985;13(7):534–536.

66. Sporn PH, Morganroth ML. Discontinuation of mechanical ventilation. Clin Chest Med. 1988;9(1):113–126.

67. Morganroth M, Grum C. Weaning from mechanical ventilation. J Intern Care Med. 1988;3(2):109.

68. Sahn SA, Lakshninarayan S. Bedside criteria for the discontinuation of mechanical ventilation. Chest.1973;63(6):1002–1005.

69. Morganroth ML, Morganroth JL, Nett LM, Perry TL. Criteria for weaning for prolonged mechanical ventilation.Arch Intern Med. 1984;144(5):1012–1016.

70. Cohen C, Zagelbaum G, Gross D, Roussos C, Macklem P. Clinical manifestations of inspiratory muscle fatigue.Am J Med. 1982;73(3):308–316.

71. Hughes C, Poparich J. Overcoming common problems with mechanical ventilation valve systems. J Crit Illness.1988;1:13.

72. ECRI. Portable volume ventilators. Health Devices. 1988;17(4):107–131. Special issue.

chapter 25 acute respiratory failure and ventilatory ... · acute respiratory failure and...

Documents