Postgrad. med. J. (April 1967) 43, 268-279.

Intensive care of the surgical thoracic patientOLOF P. NORLANDER INGRID NORDEIN

Department of Anaesthesiology, Thoracic Clinics, Karolinska Sjukhuset, Karolinska Institute Medical School,Stockholm, Sweden

OUR intensive care unit of eight to twelve beds takesin patients from a unit of eighty surgical thoracicbeds that serves a population of 1.5 million. Approxi-mately 1500 major thoracic surgical procedures,including 150-200 open heart operations, are per-formed each year. This unit functioned from 1958to 1965 in a slightly modified general thoracicsurgical ward but since last year has been housed ina specially designed area that permits isolation,reduces the risk of cross-infection to a minimumand makes working conditions for personnel asattractive as possible (Fig. 1). The unit has a smallrecovery area with three beds for the care of out-patients undergoing diagnostic procedures such asmediastinoscopies and angiocardiograms, and issituated on the same floor as the operating theatres.The main intensive care unit has two single-beddedrooms suitable for the treatment of isolation casesaccording to fever hospital principles, two morefully equipped rooms of two beds each and tworecovery areas, each with two beds. The first fourrooms have facilities for disposal or decontamina-tion of soiled articles and connections for an artificialkidney. Large transparent windows set into thedividing walls of these rooms enable the staff toobserve their colleagues in the adjacent rooms andthis reduces their sense of isolation. The standardequipment available (Fig. 2) includes closed circuittelevision, an intercommunications system and aspecial alarm system that is independent of thegeneral hospital telephone system. As is apparentfrom Fig. 1, fully half the unit is 'ancillary space'housing a blood-gas laboratory with full-timetechnicians, duty officers' and personnel rest room.The entire ward is air-conditioned and is staffed bya permanent complement of fifteen certified nurses,six nurse assistants, fourteen technicians, threeorderlies and some extra technicians (total staff ofabout fifty people) working a shift system. Thehospital authorities run a special course for theintensive care unit staff.

Respiratory problemsLong-standing pulmonary disease that reduces

respiratory reserve, and post-operative complica-tions such as atelectasis, retained bronchial secre-tions and intrathoracic fluid accumulation often

result in post-operative hypoxia or respiratoryacidosis that can be tolerated by a healthy subjectbut not by the high-risk patients that we treat.Early detection of these disorders by clinical meansis usually not possible and we have found it necessaryand worthwhile to measure arterial oxygen tensionsand saturation, arterial carbon dioxide and pH andmeasure respiratory work. Crafoord (1938) sug-gested that increased respiratory work in thesecircumstances may contribute to respiratory andcirculatory failure, and further studies (Bjork &Engstrom, 1955; Damman etal., 1963; Grenvik, 1966)which have contributed to this view are substantiatedby oxygen uptake measurements during spontaneousand controlled ventilation (Engstroim, Herzog &Norlander, 1961), especially in cardiac failure follow-ing open-heart procedures (Figs. 3 and 4) whenoxygen consumption may be reduced from 13% to50% when mechanical ventilation replaces spon-taneous respiration.

Respiratory careWhen lung function is impaired, or cardiac

failure or a greatly increased heart-size ( > 600 cm2/m2) is present, it is routine to follow arterial blood-gases during the anaesthesia and immediately after.If Pao2 decreases during anaesthesia in spite of highconcentrations ofinspired oxygen (Pao2 < 80mmHg),the endotracheal tube is usually kept in place afterthe operation and the patient is transferred to theintensive care unit and connected to an Engstromrespirator. The estimation of pulmonary function,and the degree of right-to-left shunting duringanaesthesia and post-operatively, is facilitated bythe use of the Engstrom machine for the administra-tion of ventilation and volatile anaesthetics as itallows any mixture of oxygen with room air andnitrous oxide. The shunts may be approximatelyestimated by the use of the following formula:

Q PAo2-Pao2Qs + (A-V diff) x 330

PAo2 -Pao2assuming a Pao2 > 150 mmHg (Gordh, Linderholm& Norlander, 1958). Thus in thirteen patients withopen valve repairs the pulmonary shunting increasedfrom 3.8% (range 0.7-9.0%) preoperatively to 12-8%

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

Intensive care of the surgical thoracic patient 269

0

0

0

.oCo0

cd

o0

0ao

ct)

0o

c)o

Cd

0.

at

O

.,

o

cnt

0

*o

-. [-

(W

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

OlofP. Norlander and Ingrid Norddn

FIG. 2. Intensive care treatment after heart surgery. Note the bench with contacts for electricity, outlets for oxygen andcompressed air (Venturi suction), stainless steel rack for catheter receptors and containers. The bench encloses all leadsand is easy to clean on all surfaces. The intensity of light source behind the bed can be regulated and ensures a correctcolour reproduction. On the bench capillary pulse monitor, ECG and arterial pressure monitor. The Engstrom respirator inthe corner and cooling device at the front of the bed.

(range 3 1-20'9%) of systemic output post-opera-tively immediately after the end of surgery (Thung &Norlander, 1966) (Table 1).Due to the favourable gas-flow characteristics of

the Engstrom respirator and its mode of operation(Engstrom, 1963), the synchronization of the patientswith the machine has not been a problem and musclerelaxants have not been used for this purpose.Morphine, pethidine and phenoperidine-chloride in

small doses intravenously are used in combinationwith about 10-20% nitrous oxide in the inspiredgas mixture for patients with an endotracheal tube.Oxygen concentrations of inspired gas are usually50-60% (50% 02+50% room air). With this pro-cedure the patients tolerate an endotracheal tubesurprisingly well. In adults the tube is usually keptfor 24-48 hr. If longer periods of controlled ventila-tion are then deemed necessary a tracheotomy is

270

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

Intensive care of the surgical thoracic patient 271

performed. Plastic, single lumen tracheotomy tubes(Portex) with glued-on rubber cuffs (Eynard) areused and usually changed once a week. The patientsare followed by blood-gas analysis during the criticalor acute phases, but otherwise the respirator treat-ment is mainly managed on the basis of clinical

POrtPAO2= 505 PO l

PAO2= 395

PoO2=88 Poo2=64

_400_ °5~~~-=21% s=0400-Qss.E.' Quiet Tired

E Premed

calm sweatingE Premedicated End of OPc awake300

E Anaesthesia Pred. 0- cons.

251 mt

g 200

0 .~~~~~~07 6-0 72

FIG. 3. Increase in oxygen uptake post-operatively. Observe the increase in cardiacoutput associated with spontaneous respiration at end of operation. Female, 40years-A.S. Starr valve. Cross-hatched columns, respirator; unshaded columns,spontaneous. (Figures on columns are those for cardiac output.)

400 -P0 YPO0I

EEnd

E fO300

EPre-OP ofOPC Pred. 02-cons.

o - - - -

200 OPI 221 MLu00

56 42 36 89 3 135 96 66 110

FIG. 4. Continuous increase in oxygen uptake following re-operation after aortic valvular procedure. AtOP I acute aortic valvular insufficiency occurred which necessitated a ball-valve prothesis at OP II. Male,43 years-A.S. Cross-hatched columns, respirator; unshaded columns, spontaneous.

observations and experience. The ventilatory volumesbeing returned from the patient are followed atregular intervals as well as peak (dynamic) air-waypressures and end-tidal volume pressures (compliancepressure). Those values give us valuable informationabout the conditions of lung and circulation as theyallow estimation of air-way resistance and elasticityconditions of the lung and chest. The respiratoradministers the tidal volume with a gas-flow pattern

which adapts itself to the impedance of the lungs.Even relatively large differences in time constantwithin the lungs do not create uneven distribution ofthe tidal volume (Sabar et al., 1965). The static orno-flow period at the end of each inspiratory periodallows us to draw conclusions from the changes in

compliance, a clinically most important observationwith reference to left-auricular pressure and thedevelopment of pulmonary oedema. Negative pres-sures during the expiratory phase are practicallynever used and no need has been found for frequentdeep sighs from the respirator as based on blood-gasanalysis. The respiratory frequency is usuallyaround 18-20/min and the ventilatory volumesaround 8-14 I/min, depending on the size of the

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

OlofP. Norlander and Ingrid Norden

TABLE 1

Pulmonary shunting during and after open-heart surgery (Thung & Norlander, 1966)

Patient Diagnosis I II III IV V VI VIINo.

2 AS - 65 15.9 19.5 13.9 10.5 5-39 AI - 4-5 10-8 12-4 8-6 12-6 11.812 AS 0-7 2-2 18-6 14-9 12-6 1513 AS - 5-8 18.5 18-1 25-7 15-5 3-118 AS 9.0 19-2 14-5 - 17-3 18-6 15-720 MI - 2-8 16-1 18-9 33-6 13-2 12-928 MI 4-1 - 14 34 53 12 -52 MI - 5-7 16.7 13-1 25.6 23-6 13-558 AI 3-3 9-7 12-9 10-6 12-5 12.7 -60 AS 4-1 12-6 13.4 15-1 17-5 16-3 13-561 AS - 23-4 15-3 24-2 27.3 21-7 10.664 AI 2-8 11.1 18-6 18-7 17-2 13.8 11-6

AS66 MI 2-5 9.9 23-4 26 36 17.1 20-9

AIxr 3-8 9-5 16-1 18-8 23-0 15-6 12-8

Range 0-7- 2-2- 10.8- 10.6- 8.6- 10-5- 3-19.0 23-4 23-4 34.0 53-0 23-6 20-9

patient. The actual amount of ventilation to be setat the respirator is usually higher than those valuesobtained from calculation of ventilation from thenomogram (Engstrom & Herzog, 1959). This isexplained by an increased metabolism and anincreased dead-space/tidal-volume ratio (Freeman &Nunn, 1963; Hedley-Whyte et al., 1965; Grenvik,1966; Thung & Norlander, 1966). The Engstromrespirator has been used for more than 10 years inour service. All categories of personnel have athorough knowledge of its use and management.This adds to the safety of the treatment and is aguarantee that technical mistakes are not made.

HumidificationAdequate humidification is assured by the use of

an ultrasonic nebulizer in the respiratory circuit.This device, introduced for respirator treatment inour clinic in 1962 has a capacity of humidificationwith aerosol particles of 0.8-1 0 p with a relativehumidity of over 100% at 37° with oxygen atventilatory volumes of over 20 I/min (Herzog,Engstrom & Norlander, 1964). The generator equip-ment is now automatically tuned to a constantultrasonic frequency of 3 megacycles and needs verylittle attention. Distilled water is usually used forhumidification, but favourable results have beenobtained with physiological saline with the additionof bronchodilators such as isoproterenol sulphate.The amount of the drug delivered by the nebulizerdepends on drip rate to the nebulizer head and theventilation of the patient and may be calculatedfrom the humidification nomogram of the ultrasonicdevice. With a drop rate of 6/min and a 0.05%isoproterenol solution, about 0.0075 mg of the drug

is administered to the lungs per breath, which isconsidered sufficient for relief of an asthmaticattack. If continuous administration is required, a0.01% solution may be sufficient (Gay & Long,1949).

Control of circulation and acid-base balanceA number of factors may contribute to a failing

circulation post-operatively and here only some ofthe variables which we try to follow are detailed.All patients with an unsatisfactory circulation, orwhere such is expected to develop, have theircentral venous pressures monitored through acatheter placed in the right auricle or superior venacava. In patients with an open chest operation thismay be done by the surgeons during the operationby the special technique described by Bjork &Grenvik (1967).

Peripheral arterial pressure is monitored from anindwelling radial artery catheter placed percutane-ously. ECG and sometimes peripheral capillary pulseflow are monitored. The treatment of hypotensionand shock is carried out according to surgical andcardiophysiologic principles. Blood-losses are care-fully measured and replaced. Haematocrit andhaemo-globin values are followed as well as chest- and/orabdominal radiographs. Digitalis is commonly usedin patients over 50 years of age after major thoracicsurgery and the dose administered is supervised by aspecial cardiologist.

Urinary output is followed hour by hour in allopen-heart patients and also in those who havepre-operative signs of impaired renal function. Incases of oliguria not due to inadequate fluid-replace-ment, the patients are immediately treated with

272

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

Intensive care of the surgical thoracic patient

hypertonic (40%) glucose solutions and concen-trated human albumin according to the principlesadvocated by renal units. If haemodialysis is neces-sary a Kolff artificial kidney is available on thepremises.

Bleeding problemsOne of the most common post-operative complica-

tions in surgery, especially in heart-lung surgery, isexcessive bleeding. The bleeding can have a diverseetiology such as (1) inadequate mechanical haemo-stasis, (2) defective vascular haemostasis, (3) fibrino-lysis, (4) other defects in the coagulation mechanism,and (5) in connection with extracorporeal circulation,inadequately neutralized heparin.Many cases of mechanical haemostasis defects

are difficult to diagnose without the analysis of thecoagulation status. In the coagulation defects causingbleeding a complete or at least partial coagulationanalysis is necessary.The most important coagulation assays, in addi-

tion to the counting of platelets, are the coagulationtime and the thrombin generation test. Furthermore,the determination in plasma of the recalcificationtime and the content of heparin, Factor V, pro-thrombin +proconvertin, and fibrinogen as well asthe determination of fibrinolytic activity, is impor-tant.

If normal values are found, a mechanical defectmust first be considered and immediate surgicalintervention is indicated. The heparin assay willshow if a prolonged coagulation time, recalcificationtime, a decreased Factor V and P x P level might bedue to remaining heparin or to a consumption of thecoagulation factors. Heparin is neutralized by prot-amine dichloride, which is preferred to protaminesulphate. After protamine sulphate heparin reboundis said to be found but this is not the case when usingprotamine chloride. Fibrinolysis is treated withkallikrein inhibitor (Trasylol) or epsilonamino-caproic acid (Epsikapron). Fibrinogen deficiency istreated with fibrinogen concentrates prepared ifpossible from small batches of blood in order todecrease the risk of serum-hepatitis as much aspossible.

Decreased content of coagulation factors is treatedwith fresh blood. If a very low platelet content ispresent, platelet concentrates are given.However, if a manifest consumption syndrome

with decrease in coagulation factors and plateletscan be shown to be present, the patient is treatedwith fresh blood and heparin. Trasylol can be usedinstead of heparin to augment the anti-fibrinolyticactivity because Trasylol has been shown to exertanti-coagulant activity.

Acute cardiac failure with pulmonary oedema issometimes seen in thoracic patients. We have madeit a practice to intubate these patients and administer

controlled ventilation with 100% oxygen. If theoedema is not immediately improved by this, aslight positive expiratory pressure is applied on therespirator. This pressure can be regulated accordingto the patient's circulatory condition and usuallyachieves a satisfactory result within a short periodof time. The regular medical measures such asdigitalization, infusion of hypertonic solutions, etc.,are carried out as well.

Isoproterenol sulphate solutions are frequentlyused post-operatively in the cardiac patient, in whoma satisfactory cardiac output is not achieved by theafore-mentioned measures, and dilute solutions areemployed (1-2 mg/1000 ml). Vasoconstrictors areonly used if all other measures to improve thecirculation have failed. However, good results havebeen achieved in isolated cases with diluted adrena-line solutions.

Acid-base balance is repeatedly followed fromarterial samples. Respiratory acidosis is correctedby increased ventilation by the respirator anddeficits in buffer base are compensated with 6%sodium bicarbonate solutions or THAM-buffer.The latter drug is frequently used during and afteropen-heart procedures and is always employed aftercardiac arrest (Nahas, 1963). Changes of more than3 mEq in standard bicarbonate from pre-operativevalues are corrected.

Control of body temperatureAn increase in body temperature may precipitate

circulatory failure by the concomitant increase ofoxygen consumption. This was the only obviousreason for respiratory failure in fifteen patients outof ninety-three who had normal chest radiographsand no previous signs of circulatory failure. Suchfailure was clinically evident when body temperatureincreased above 39° C. Therefore, we try to keepthe temperature around normal with either drugs(salicylates) or by physical means. The patients areplaced on cooling mattresses through which a coldalcohol-water solution is circulated (ThermoriteSystem). Shivering is controlled by the intravenousadministration of opiates (morphine, pethidine) andphenothiazine derivatives of which chlorpromazineis most frequently used, in doses varying from 2-5 to10 mg i.v. each hour, depending on the effect on thecirculation. A body temperature around 35° isaimed at in severe cases, especially where lungfunction is so poor that gas exchange is not sufficientto adequately oxygenate the patient or to eliminatethe carbon dioxide at increased temperatures. Thishas occasionally been seen in pre-operativelycyanotic patients, especially old Fallot's, who developsevere pulmonary changes with congestion andhighly elevated ventilatory pressures on the 3rd to4th post-operative day. Prognosis is usually poor insuch patients.

273

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

OlofP. Norlander and Ingrid Norden

Complete A-V block is sometimes seen, either as acomplication following surgery or more commonlyin patients with Adam-Stokes attacks. They aretreated with intracardiac electrodes, either insertedduring surgery or placed intravenously from thejugular vein into the right ventricle. The lattertechnique is now used exclusively in our unit foracute patients and has given excellent immediateand long-term results (Lagergren & Johansson,1963).

Prevention of infectionsThe dangers of cross-infections in intensive care

are well known. Therefore constant attention mustbe paid to hygiene of the unit, its equipment,locality and personnel. Our unit has a hygienic pro-gramme which functions satisfactorily in this respectwithout making the everyday work with the patientstoo difficult. Thus the personnel of the unit wearspecial clothes and are not allowed to leave the unitwithout a complete change except when they aregoing to the nearby operating-room area. Specialfoot-wear is worn and visiting doctors are requiredto change coats before entering the patient's room.Visits from relatives are limited and no regularvisiting hours exist. Non-disposable instrumentsand equipment utilized in the room are left for 1 hrin strong disinfectant solution (phenol-soap) andthen brought to the cleaning area in sealed plasticbags. The isolation room has installations forsteam-cleaning of bedpans and similar utensils. Allfloors are of plastic material and are welded withoutany crevices, so as to facilitate cleaning. Bacterio-logical cultures are done regularly from patient'sequipment and localities.

Cleaning of the respiratorsRespiratory equipment, unless properly cleaned

and sterilized between use on patients, is a mostdangerous source for the spread of infections. Thefollowing procedure is employed for the Engstromrespirator: Before the machine is brought out to thecleaning area, breathing tubes, connection pieces,etc., are placed in a solution of phenol-soap.* Therespirator is cleaned on the outside with a 70%alcohol solution and the engine compartment withalcohol and white spirits. The standard humidifieris emptied and taken apart and the plastic spongediscarded. The humidifier is either steam-auto-claved (the stainless steel type) or boiled for a shortperiod of time (the plastic material type). The patientvalve is taken out and washed in 70% alcohol orexchanged in a case of severe infection. The dosagevalve is dismantled and cleaned with alcohol. Thewater in the water safety-lock is drained and re-

*o-Phenyl-phenol-sodium, 22.5 g; Sapo kalinus venalis, 8;tetracemintetra-sodium 40 %, 4 g; Spiritus denaturatus1 1-5 g Aqua ad 1 000; with 50% water for 2 hr.

placed with fresh distilled water. The breathing bagis exchanged in a case of severe infection. Bacterio-logical cultures are made from the humidifier, dosageand patient valves and at other various sites of themachine and must be found negative before themachine can be used again. In practically every casethe cultures have been negative after this procedure.Presently a new technique is being evaluated for amore rapid and easier sterilization of the respiratorbased on the use of the ultrasonic humidifier. Therespirator is connected in such a way that the gasesmay be ventilated in a closed circuit with nitrogen,and 70% alcohol is nebulized into the circuit by thenebulizer for 1 hr (Petersen et al., 1967). Differentbacteriological tests have demonstrated this to be anefficient method, whereby all the inside parts of themachine in connection with the patient may besterilized. Those tests are now under evaluation andit is our hope that in the future it should be possibleto recommend the method for general use. No air-filters are used on our respirators. The risk of directair contamination is considered low, as most of thepatients are treated in isolation rooms with goodventilation.

Patient materialAbout 100 patients are treated yearly with trache-

ostomy and respirator. During the last year an in-creasing number of shorter treatments for 24-48 hrwith an endotracheal tube have been employed.The number and type of patients treated during theyears 1962-65 is illustrated in Table 2.An illustration as to how the resources of the unit

are employed is given in the following case-report:A 46-year-old woman was operated upon for the

correction of an atrial septal defect and abnormalvenous return with the aid of extracorporeal circu-lation and hypothermia to 32°. The patient exhibitedatrial fibrillation for 6 months prior to operation andthe heart size was 870 ml/m2. A left-to-right shunt offour times the systemic flow was present and thepressure in the pulmonary artery at rest was45 mmHg. The working capacity was low, 200 kpm.The correction was without difficulties and the per-fusion was uneventful. After the operation thepatient was in good general condition, respondedwell and with an adequate spontaneous respiration.She was returned to the intensive care unit and wasleft with spontaneous breathing with 3 litres ofoxygen through a nasal catheter. During the nightthe patient started to bleed and altogether 2500 mlof blood drained from the mediastinum and peri-cardium. Blood pressure and circulation were goodall the time, as the blood was replaced in amountsequal to the loss. Blood pressure fluctuated between80 and 100 mmHg. Central venous pressure had beenhigh since the end of operation-around 20-30 cmH2O-which at the time of operation was

274

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

Intensive care of the surgical thoracic patient

TABLE 2

Clinical material treated with respirator as related to type of intervention (1962-65)

Other operations:Heart myasthenia gravis, Non-surgical

Great Oesophagus chest wall tumours, interventions:Operation Open Closed Lungs vessels Stomach mediastinal tumours chest injuries

No. Deaths No. Deaths No. Deaths No. Deaths No. Deaths No. Deaths No. Deaths

1962 24 18 5 1 20 8 11 6 6 5 2 1 13 41963 38 19 9 1 22 8 6 3 8 7 3 0 12 11964 43 15 5 2 16 6 12 5 5 3 6 0 15 41965 44 10 7 3 17 2 11 4 7 5 3 0 14 3

1962-65 149 62 26 7 75 24 40 18 26 20 14 1 54 12

Total number of patients treated 1962-65, 528; survivals, 384; deaths, 144.

thought to be due to a left superior vena cava andrather poor drainage at the point where the venouscatheter had been inserted during the operation.Repeated chest X-rays during the night showed onlya minimal enlargement of the mediastinal space.During the night urinary production decreased andaltogether only about 400 ml of urine had beenpassed since the end of the operation. Serum potas-sium increased during the night and the highestvalue encountered was 6.9 mEq. Resonium, ad-ministered repeatedly, decreased it to 4.9 in theearly morning. A short period, less than 15-30 sec,of asystole occurred at 06.17 hours, but heart actionresumed immediately on external massage. Thecirculation was still clinically satisfactory with anarterial blood pressure around 85-90 mmHg, venouspressure around 25 cmH2O and a pulse frequency of85-90. In order to improve her circulation, a verydilute isoproterenol drip was started which increasedthe pulse rate and blood pressure. An arterial bloodsample at 07.30 hours showed a Pao2 of only67 mmHg in spite of a high flow of nasal oxygen.Hypertonic glucose was administered via the cavacatheter. The diagnosis of the condition lay betweenleft heart failure with oliguria as a complication ofthe pump-procedure and a cardiac tamponade. Thelatter diagnosis did not seem likely considering thechest X-rays and the relatively good circulation whichcould easily be maintained with the isoproterenoldrip. A new arterial sample at 10.05 hours showedthat metabolic acidosis had started to develop. Shewas given 120 mEq of bicarbonate and 24 mEqTHAM-buffer during the next hours, which nor-malized the standard bicarbonate. In spite of a lowarterial oxygen tension she was clinically in satis-factory condition, but it was decided to perform atracheotomy. Two hours later she again had signsof metabolic acidosis, but oxygen tension was103 mmHg. The tracheotomy was now performedunder pethidine-oxygen-succinylcholine anaesthesiawith an endotracheal tube. The blood pressure

decreased at the end of the tracheotomy and theECG, which was monitored the whole time, as wellas the arterial blood pressure, showed an increasingbradycardia. This could be relieved after a fewexternal compressions and increasing the rate ofisoproterenol infusion. The patient was ventilatedwith the Engstrom respirator and the circulationimproved. At 15.30 hours another short period ofcirculatory arrest occurred, again relieved afterexternal massage but with the addition of 200 ,ug ofadrenaline in order to restore the circulation.120 mEq of THAM-buffer was given and the bloodgas analysis 10 min later showed normal values. Thepatient was awake and responded well. However,at 16.05 hours a short arrest occurred again and itwas then decided on to explore to rule out thepossibility of a heart-tamponade. At induction ofanaesthesia, which was made with 2 mg of pheno-peridine chloride and 6 mg of D-tubocurare, thepatient again had a cardiac arrest which waspromptly treated with external massage during thetime the patient was draped for surgery. The chestwas opened and the circulation improved as soon asthe pericardium was opened. Approximately 800 mlof clotted blood were found around the heart, inthe mediastinum and the right pleura. When thishad been removed, the heart resumed good regularbeating with a frequency of 80-90/min. Blood pres-sure 90/60, with a venous pressure of 10 cmH2O.An arterial sample taken at 17.00 hours during re-operation indicated a severe metabolic acidosis,normal Paco2 and an oxygen tension of 127 mmHg.The patient was given 90 mEq of bicarbonate rapidlywithin a period of 10 min, as it was noticed that urineproduction had started again with a good flowduring the operation. The clots were evacuated anda temporary pace-maker electrode was inserted intothe myocardium and connected to an external pace-maker at a frequency of 90/min. Isoproterenol wasdiscontinued. It was noticed that slight continuousbleeding occurred from the right ventricle from the

275

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

276 OlofP. Norlander and Ingrid Norden

site of a needle puncture at the time of operation forpressure measurements. This probably explained theblood loss during the night after the operation, asno other bleeding points could be observed. Thepatient was maintained on the respirator for thenext 8 days, as she exhibited signs of left heartfailure with radiological signs of pulmonary con-gestion and slight peripheral oedema. The urinaryoutput became normal immediately after re-opera-tion and remained around 1200 ml daily. The pace-maker was disconnected after 8 days and the finalrecovery was uneventful.

This report illustrates a common aspect of theproblems of thoracic intensive care. We have apatient with impaired cardiac function after a majorcorrective procedure with post-operative bleedingand oliguria. The problem has many facets, diag-nostic as well as therapeutic. The constant super-vision of the patient with adequate monitoring ofECG, blood pressure and blood-gases, besidesX-rays, etc., made it possible to carry the patientthrough a critical period when the diagnosis was notclear. However, looking retrospectively at the case-history, it is evident that the thoracic 'rule of thumb',that a blood-loss of more than 1500 ml, even if itcan be replaced and circulatory function satisfactorilymaintained, is usually a strong indication for asurgical re-exploration, is a valid one. The respiratortreatment illustrates how the patient could be re-lieved of the respiratory work and guaranteedoptimal blood gas tensions under the prevailingcircumstances. Without the respirator it would havebeen rather difficult to compensate for the severechanges in metabolic acidosis brought about by afailing peripheral circulation, oliguria and decreasedvenous and tissue oxygen tension. It also demon-strates that when tracheotomy is performed in sucha patient the risk of an acute deterioration of cir-culation during the procedure is great and adequatemonitoring must be available. In this case we waitedtoo long before the tracheotomy was made, a delaywhich in this particular case was due to othertechnical reasons. It is interesting to note howrapidly the metabolic acidosis developed after evenshort periods of cardiac arrest, adequately treatedwith external massage. This indicates that bufferslike THAM and bicarbonate should be used in theacute treatment even without previous analysis ofblood gases and acid-base balance.

AcknowledgmentsThe authors wish to express their gratitude to Mrs A.

Linneroth, Department of Hospital Planning, KarolinskaSjukhuset, for her efforts and ideas in planning of theIntensive Care Unit, Thoracic Clinics, and to Dr P. Ohlsson,

Associate Professor, Department of Thoracic Surgery,for the view-points on bleeding and coagulation problemspost-operatively.

ReferencesBJORK, V.0. & ENGSTROM, C.-G. (1955) The treatment of

ventilatory insufficiency after pulmonary resection withtracheotomy and prolonged artificial ventilation. J. thorac.Surg. 30, 356.

BJORK, V.0. and GRENVIK, A. (1967) Percutaneous intra-cardiac catheters for postoperative observations inthoracic surgery. J. Cardiovasc. Surg. (In press).

CRAFOORD, C.L. (1938) On the technique of pneumonectomyin man. Acta chir. scand. Suppl. 54.

DAMMANN, J.F., JR, THUNG, N., CHRISTLIEB, I.I., LITTLE-FIELD, J.B. & MULLER, W.H. (1963) The management ofthe severely ill patient after open-heart surgery. J. thorac.cardiovasc. Surg. 45, 80.

ENGSTR6M, C.-G. (1963) The clinical application of prolongedcontrolled ventilation. With special reference to a methoddeveloped by the author. Acta anaesth. scand. Suppl. XIII.

ENGSTROM, C.-G. & HERZOG, P. (1959) Ventilation nomo-gram for practical use with the Engstrom respirator.Acta chir. scand. 245, 37, Suppl.

ENOSTROM, C.-G., HERZOG, P. & NORLANDER, O.P. (1961)A method for the continuous measurement of oxygenconsumption in the presence of inert gases during control-led ventilation. Acta anaesth. scand. 5, 115.

FREEMAN, J. & NUNN, J.F. (1963) Ventilation perfusionrelationships after haemorrhage. Clin. Sci. 24, 135.

GAY, L.N. & LONG, J.S. (1949) Clinical evaluation of iso-propylepinephrine in management of bronchial asthma.J. Amer. med. Ass. 139, 452.

GORDH, T., LINDERHOLM, H. & NORLANDER, O.P. (1958)Pulmonary function in relation to anaesthesia and surgeryevaluated by analysis of oxygen tension of arterial blood.Acta anaesth. scand. 2, 15.

GRENVIK, A. (1966) Respiratory, circulatory and metaboliceffects of respirator treatment. A clinical study in post-operative thoracic surgical patients. Acta anaesth. scand.,Suppl. XIX.

HEDLEY-WHYTE, J., CORNING, H., LAVER, M.B., AUSTEN,W.G. & BENDIXEN, H. (1965) Pulmonary ventilation per-fusion relations after heart valve replacement or repair inman. J. clin. Invest. 44, 406.

HERZOG, P., ENGSTROM, C.-G. & NORLANDER, O.P. (1964)Ultra-sonic generation of aerosol for the humidification ofinspired gas during volume-controlled ventilation. Actaanaesth. scand. 8, 75.

LAGERGREN, H. & JOHANSSON, L. (1963) Intracardiac stimu-lation for complete heart block. Acta chir. scand. 125, 562.

NAHAS, G. (1963) The clinical pharmacology of THAM.Clin. Pharmacol. 4, 784.

NORLANDER, O.P., BJORK, V.O., CRAFOORD, CL., FRIBERG,O., HOLMDAHL, M., SVENSSON, A. & WILDMAN, B. (1961)Controlled ventilation in medical practice. Anesthesia,16, 285.

PETERSEN, N.O.A., ENGSTROM, C.-G., HERZOG, P., &NORLANDER, O.P. (1967) Ultra-sonic sterilization ofrespirators with 70% alcohol and nitrogen. Preliminaryreport (To be published).

SABAR, E.F., NORLANDER, O.P., OSBORN, J.J. & GERBODE, F.(1965) Gas distribution studies in experimental unilateralbronchial constriction using an accelerating, volume-controlled respirator. Surgery, 58, 713.

THUNG, N. & NORLANDER, O.P. (1966) Cardio-respiratorychanges during anaesthesia for open-heart surgery. Actaanaesth. scand. 10, 79.

copyright. on M

arch 31, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.43.498.268 on 1 A

pril 1967. Dow

nloaded from

Postgrad. med. J. (July 1967) 43, 512.

Errata

Postgraduate Medical Journal (1967) Vol. 43, No. 498,'Intensive Therapy.'

PLEASE NOTE THE FOLLOWING:

Page 222. The last sentence of the first paragraph should read:'During hypotension a metabolic acidosis was found inall patients but the base deficit showed no correlation withthe concentrations of lactate. There was statisticalcorrelation with AL/P %.'

Page 223. The last sentence of paragraph (b) should read:'Hypotension responded . . . '

Page 254. The legend to Fig. A7 should read:'Pressure - volume loops from the second case whenbreathing spontaneously (A1) and during controlledIPPV (B). A normal loop (A) is shown for comparison.

Page 268. The formula in the second column should read:

Q 1

Qs (A-V diff) x 3301 ---

PAo2 - Pao2

Postgrad. med. J. (July 1967) 43, 512.

Errata

Postgraduate Medical Journal (1967) Vol. 43, No. 498,'Intensive Therapy.'

PLEASE NOTE THE FOLLOWING:

Page 222. The last sentence of the first paragraph should read:'During hypotension a metabolic acidosis was found inall patients but the base deficit showed no correlation withthe concentrations of lactate. There was statisticalcorrelation with AL/P %.'

Page 223. The last sentence of paragraph (b) should read:'Hypotension responded . . . '

Page 254. The legend to Fig. A7 should read:'Pressure - volume loops from the second case whenbreathing spontaneously (A l) and during controlledIPPV (B). A normal loop (A) is shown for comparison.

Page 268. The formula in the second column should read:

Q 1

Qs (A-V diff) x 3301 +

PAo2 - Pao2