farid jalali md / may 2020 in normal not have a clinically ... · antihistaminics do not prevent...

Pulmonary Artery

Bronchial Artery

Pulmonary Vein

Azygous Vein

Left Heart

Right HeartBronchial

Vein

Broncho-pulmonary Vein

Broncho-pulmonary Anastomosis

Normal Pulmonary and Bronchial Blood Flow

Alveoli

Alveoli

Alveoli

Farid Jalali MD / May 2020In Normal Lung Circulation, Bronchopulmonary Anastomoses Do Not Have a Clinically Significant Role Unless Either Pulmonary or Bronchial Perfusion is Disturbed

Click For Next Slide

PA

BA

PV

AV

LA

RABV

BPVBPA

In Pulmonary Embolism, Bronchopulmonary Anastomoses (BPA) Allow Diversion of Systemic Oxygenated Perfusion (Black Arrows) from Bronchial to Pulmonary Circulation To Prevent Lung Ischemia

PulmonaryEmbolism

Pulmonary Embolism

BPA: Bronchopulmonary AnastomosisBPV: Bronchopulmonary VeinBA: Bronchial ArteryBV: Bronchial Vein

ALV

ALV

ALV

Deoxygenated Blood

Oxygenated Blood

ShuntPrevents

Lung Ischemia


Farid Jalali MD / May 2020

PA

BA

PV

AV

LA

RABV

BPVBPA

Step 1

COVID19Acute Lung Injury


Alveolar Capillary Near-Occlusive DiseaseAngiotensin II Excess Mediated

VasoconstrictionMicrothrombi ALV

ALV

ALV

SARS-CoV-2 Action on Alveolar Capillary Endothelium Mediated by Ang II Excess Results in a Progressive Alveolar Capillary Occlusive Disease



PA

BA

PV

AV

LA

RABV

BPVBPA

Step 2



ALV

ALV

ALV

With Locally Impaired Pulmonary Circulation, Bronchopulmonary Anastomoses Shunt Oxygenated Blood from Bronchial to Pulmonary Circulation Per Yellow Arrows

Alveolar Capillary Near-Occlusive DiseaseAngiotensin II Excess Mediated

VasoconstrictionMicrothrombi

ShuntPrevents

Lung Ischemia Click For Next Slide


PA

BA

PV

AV

LA

RABV

BPVBPA


Step 3


Minimal PerfusionNormal Ventilation

High V/Q MismatchALV

ALV

ALV

As Lung Endothelial Injury Worsens, Alveolar Capillary Vaso-Occlusive Disease Progresses, Resulting in Development of Dead Space Ventilation

Alveolar Capillary Occlusive DiseaseProgressive Vasoconstriction

Progressive Microthrombi


ShuntPrevents

Lung Ischemia


PA

PV

Step 4


Minimal PerfusionNormal Ventilation

High V/Q MismatchALV

ALV

As Lung Endothelial Injury Worsens, Alveolar Capillary Vaso-Occlusive Disease Progresses, Resulting in Development of Dead Space Ventilation

Alveolar Capillary Occlusive DiseaseProgressive Vasoconstriction

Progressive Microthrombi Click For Next Slide


DorsalAlveolar Hypocapnia

Alveolar Duct Constriction

V RedistributionTo Areas with Less V/Q Mismatch

NADEL, COLEBATCH, AND OLSEN

FIG. 6. A: anatomic relation between vagal efferent nerves and airways. Large dashed lines extending from vagus nerves to terminal airways indicate possible nervous innervation of these areas. Stimulation of vagus nerves constricts all the airways. In- creased time constant (RC) may lead to transient air retention (indicated by solid lines showing overexpanded alveoli). B: anatomic relation between the pulmonary circulation and peripheral airways. R.B. = respiratory bronchiole; A.D. = alveolar duct. Note conspicuous amounts of smooth muscle in peripheral airways relative to total thickness of their walls. Agents in this circulation capable of causing smooth muscle contraction cause these airways to constrict, decreasing pulmonary compliance and expelling air from the lungs (indicated by solid lines showing constricted alveolar duct and “pulling-in” of associated alveoli).

Location of airway constriction after barium sulfate microembolism. The combination of increase in pulmonary resistance, decrease in pulmonary compliance, and increase in anatomic dead space, which occur after barium sulfate injection, suggests that changes occurred in the peripheral portions of the lungs beyond the anatomic dead space, i.e., the alveoli, alveolar ducts, and respiratory bronchioles (24, 26). The changes could be due to pulmonary edema, alteration of the surface properties of the alveoli, or contraction of smooth muscle in the peripheral airways.

Although pulmonary edema could explain the decreased pulmonary compliance, the animals did not show pulmonary edema postmortem. Primary changes in the surface-active material lining the lungs could also account for the changes. However, intravenous isoproterenol, a potent airway smooth-muscle dilator, prevented or decreased the changes in lung mechanics after barium sulfate microembolism, suggesting that contraction of smooth muscle was responsible for the changes.

Our anatomic studies after barium sulfate microembolism showed predominant constriction of alveolar ducts and less marked constriction of terminal bronchioles; the associated alveoli were pulled in toward the alveolar duct (Fig. 5, A and B). An explanation is that the alveolar mouths contain smooth muscle and are part of the alveolar duct skeleton. Other evidence in-

dicates that right heart injection of various drugs capable of causing smooth muscle contraction (e.g., histamine) results in physiologic and anatomic changes similar to those due to barium sulfate microembolism (I 0).

The existence of smooth muscle in the peripheral airways, including alveolar ducts, has been confirmed by many anatomists (15, 24, 26, 32). The pulmonary arterial branches are distributed to the peripheral airways (24, 26) and the bronchial arterial branches to the large airways (23, 24). Dunn (I 3) described contraction of smooth muscle of the “atria” or bronchiolar termina- tions and the infundibula after injection of starch emboli in goats. He suggested a reflex connection between the impaction of starch granules in the vessels and spasm of the airway muscle. Binger (4) also described irregular narrowing of bronchiolar lumina after starch emboli with areas of partial atelectasis around the emboli. He be- lieved the smooth-muscle contraction was not significant.

Mechanism of airway constriction after barium sulfate microembolism. The lungs of the cat contain histamine (16); injection of 48/80 (a substance which releases histamine from mast cells) into the right side of the heart results in changes in CL and RL similar to those found after barium sulfate microembolism (I 0). After prior injections of 48/80 and the return to the normal state, the changes in CL and RL induced by barium sulfate were significantly reduced. This suggests that the effects depend on the release of histamine. That histamine release occurs after microembolism of the lung is suggested by observations made of cross-circulated dogs: starch embolism of one dog increases the gastric secre- tion of the other dog (I 4, 37). Recent studies indicate that barium sulfate microemboli decrease histamine in the lung tissue (9).

Antihistaminics do not prevent changes in CL and RL in sheep after barium sulfate microembolism (20). How- ever, if barium sulfate releases histamine from a concen- trated source near the peripheral airways, the histamine will reach the nearby muscle cells in high concentration, and intravenously administered antagonists are unlikely to abolish its effects near the site of release.

Histamine has been shown to be released under various circumstances : intravenous injection of cotton-dust extracts releases histamine from the lungs in cats (I) ; these extracts also release histamine from normal human lung tissue in vitro (6). Inhalation of steam appears to release histamine from the lungs in dogs (2).

Kabins et al. (22) have postulated that starch emboli initiate a reflex that stimulates sympathetic efferent nervous activity; pressor amines thus released cause histamine release in the lungs which then results in pulmonary edema. The involvement of sympathetic efferent pathways in the formation of lung edema does not explain the changes in lung mechanics after barium sulfate injection since prior injection of guanethedine did not prevent the change in CL and RL.

Although there is evidence that histamine is responsible for the changes observed after barium sulfate micro-

Preserved Overall Compliance

Alveolar DuctDilation

VentralAlveolar

HypercapniaProtective Physiologic Response

To Avoid Overdistention of Alveoli with Poor Perfusion To Avoid Further Injurious Capillary Vasoconstriction

PA

BA

PV

AV

LA

RABV

BPVBPA


OccludingCapillary Bed

Step 5


ALV

ALV

ALV

Progressive Occlusion of Alveolar Capillary Bed Results in Backflow of Pulmonary Arteriolar Blood across the Bronchopulmonary Anastomoses along Yellow Arrows

No Evidence of PAH due toFlow Diversion via BPA

High V/Q Mismatch



PA

BA

PV

AV

LA

RABV

BPVBPA


Step 6


Progressive Distention and Back Pressure Buildup in the Direction of Yellow Arrows Causing Hyper-Perfusion in the Intrapulmonary Shunt with Low V/Q Mismatch

ALV

ALV

ALV

Progressive Distention of Pre-Capillary Pulmonary ArterioleUntil It Reaches BPA Will Be Seen

Chest CTA

Bronchial Shunt Drains Across BPV into PV


ShuntPrevents

Lung Ischemia


Increase PA Flow

BA

PV

AV

LA

RABV

BPVIncrease Shunt Flow


Step 7


Factors That Exacerbate Intrapulmonary Shunt PhysiologyIncreased SVR: Valsalva, Systemic VasoconstrictionDecreased PVR: Pulmonary Vasodilators

ALV

ALV

ALV

Both


ShuntPrevents

Lung Ischemia


Reduces Blood Poolingin Dorsal-Predominant Intrapulmonary Shunts

By Gravity

BA

PV

AV

LA

RABV

BPV


Step 8


Prone Position Improves Dorsal-Predominant Intrapulmonary Shunt Physiology

ALV

ALV

ALV

PA

Reduces Shunt FlowClick For

Next Slide

ShuntPrevents

Lung Ischemia


PA

BA

PV

AV

LA

RABV

BPV


Step 9


ALV

ALV

ALV

NO FLOW

Mechanical Ventilation May Accelerate Worsening of High and Low V/Q MismatchDue to Higher Lung Volume (Increased VT, Increased PEEP)

Vascular Resistance ofExtra-Alveolar Vessels

Vascular Resistance ofAlveolar Capillaries

Diminishes AlveolarCapillary Flow

ShuntPreventingIschemia

Occluding

Increases Shunt Flow


ProgressiveVenous

Congestion

1

3

4

1 2 3


Worsens High V/Q Mismatch

Worsens Low V/Q Mismatch

2

4

?

AcceleratedLung Injury

ParenchymalIschemia

Cytokine StormAfter Intubation

PA

BA

PV

AV

LA

RABV

BPV


Step 10


ALV

ALV

ALV

NO FLOW

In Absence of Endothelial Stabilization, Proper Anticoagulation, And Flow Redistribution, Lung Injury Progresses by Worsening High and Low V/Q Mismatch

Diminishing AlveolarCapillary Flow

ShuntPreventingIschemia

Occluding

Increasing Shunt Flow



ProgressiveVenous

Congestion3

42

Worsening High V/Q Mismatch

Worsening Low V/Q Mismatch

ProgressiveLung Injury

1 2 3


4

Diffuse Alveolar Damage

1

ParenchymalIschemia

Endothelial-EpithelialBarrier Breakdown

PA

BA

PV

AV

LA

RABV

BPV


Step 11


ALV

ALV

ALV

NO FLOW

Late Lung Injury is Characterized by Poor Lung ComplianceProgressive Interstitial EdemaProgressive Alveolar Edema and DamageProgressive Bronchial Distortion

BPAClick For

Next Slide


PA

BA

PV

AV

LA

RABV

BPV


Step 12


ALV

ALV

ALV

Significant Reperfusion Injury May Develop As Well with Microthrombi Resolution by Anticoagulation, Thrombolytics, or via Innate Fibrinolysis

BPA

RESTOREDFLOW

Severely Injured Endothelium High Vascular Permeability

AlveolarEdema

InterstitialEdema

End



• Early endothelial stabilization, before hypoxia sets in, is key to prevent SARS-CoV-2 induced, excess Angiotensin II mediated, intense alveolar capillary vasoconstriction as well as the concomitant pro-inflammatory, pro-thrombotic endothelial milieu, all of which form the basis of lung injury in COVID19.

• Once hypoxia sets in, supportive care should include early and aggressive endothelial stabilization interventions, properly dosed anticoagulation to prevent lung microvascular thrombi, HFNC, and awake prone position to redistribute flow away from the forming dorsal-predominant intrapulmonary shunts.

• Alveolar capillary microvascular thrombi are not a pre-requisite for the severe lung injury in COVID19, but are a clear step in the wrong direction if allowed to be formed.

Take Home Points

Take Home PointsFarid Jalali MD / May 2020

• Lung’s natural and physiologic protective response to SARS-CoV-2 induced alveolar capillary vasoconstriction and dead-space ventilation is characterized by alveolar hypocapnic bronchoconstriction at the level of the alveolar ducts to reduce a harmful alveolar expansion in these affected capillaries.

• Naturally, unaffected capillaries and corresponding alveoli will have a higher redistribution of ventilation, will exchange more CO2 into alveolar space, and will therefore have hypercapnic bronchodilation.

• This redistribution keeps the lung compliance preserved in the initial lung injury characterized mainly by dead-space ventilation, forming intrapulmonary shunts, without significant interstitial or alveolar edema.


• Compensatory lower inspiratory volumes characterize the patient’s response, associated with higher respiratory rate, and “shallow rapid breaths” without distress.

• This lower inspiratory volume is needed to prevent expansion of alveoli in the affected vasculopathic areas, as inappropriate expansion compounds the vasoconstriction in these affected alveolar capillaries.

• This will result in a compensatory tendency to develop hypocapnea on blood gas analysis, often concomitant with hypoxia as intrapulmonary shunts also begin to form as lung injury progress.


• Higher lung volumes, and positive pressure ventilation, disturb the fine balance maintained physiologically in the ventilatory redistribution pattern of the COVID19 lung, between high V/Q mismatch areas (poor perfusion, compensatory reduced ventilation to protect against the vasculopathy) and the compensating lower V/Q areas that safely receive higher ventilation in return.

• Therefore, mechanical ventilation may result in worsening of dead-space ventilation by constricting alveolar capillaries in the affected vasculopathic regions, and additionally result in worsening intrapulmonary shunting (next slide) due to reduced resistance in extra-alveolar vessels with higher lung volumes.


• In absence of endothelial stabilization, proper anticoagulation, and flow redistribution, lung Injury progresses to severe form by progressively worsening dead-space ventilation, resulting in intrapulmonary shunt development as described in the the diagrams.

• This advanced stage of lung injury is characterized by progressively diminished flow across the alveolar capillaries, resulting in higher flow across the formed intrapulmonary shunts, eventually culminating into progressive interstitial edema, progressive and diffuse alveolar damage, and alveolar fibrin thrombi deposition.

• Physiologically, this stage resembles “typical ARDS” where alveolar recruitment may be beneficial, but unlikely to reverse the vasculopathic disease process, inevitably resulting in high mortality. Pulmonary vasodilators and systemic vasoconstriction plausibly worsen hypoxia at this stage due to increasing flow across the intrapulmonary shunts.


• Through the action of body’s innate fibrinolytic system, lysis of microthrombi and reversal of flow to an area of injured endothelium may result in cycles of ischemia-reperfusion injury in the lung, mediated early on by monocytes and macrophages, and late by neutrophil activity.

• Reduction in leukocyte trafficking with corticosteroids and other therapeutics can be of value early on in the disease course to mitigate this ischemia-reperfusion injury.

• Late and sudden restoration of flow to a bed of alveolar capillaries that have had a prolonged and deep poor flow, usually in absence of proactive endothelial stabilization and proper anticoagulation, will inevitably result in a severe ischemia-reperfusion injury, significant interstitial and alveolar edema, and sudden demise.

• At this late of a stage in lung injury, ECMO may be the only solution available while pursuing lysis of microthrombi to restore alveolar capillary flow in a controlled fashion, while cardiopulmonary bypass is utilized to reduce risk of hemodynamic demise.

farid jalali md / may 2020 in normal not have a clinically ... · antihistaminics do not prevent...

Documents