Pulmonary Function Following Severe Acute Respiratory Failure and High Levels of Positive End-Expiratory Pressure

Pulmonary Function Following Severe Acute Respiratory Failure and High Levels of Positive End-Expiratory Pressure* Ma; Michael E. Douglas, MC, USAF;·· and John B. Downs, M.D., F.C.C.P.t

In an ll-mooth period, we treated 561 patients with mechanical ventilation. Fifty-four (10 percent) of these padents bad acute respiratory faDore, requiriDg treatment with positive end-expiratory pressure (pEEP) in escess of 20 mm Hg (1'IUIIe, 20 to 40 DUD Hg). AD patients were allowed to breathe spontaneously between volume-Hmited mechanical breaths deUvered at a rate _cleot to malntain an arterial pH greater than or equal to 7.35. PEEP was applied until calculated pubnOMl')' VeD0U8 admlDure was minimized. Forty-three (80 percent) of

these 54 patients were alive and asymptomatic three months after discharge from the hospital, _d tests of pulmonary function were performed on ten patients within one year after hospitalization. Abnol1lUllities in pulmonary function appeared to be reversible, aad pulmonary function graduaDy approached normal within one year. It appears that neither acute respiratory faDure nor exposure to blgh airway pressures caused slpi&cant permanent pulmonary damage In the ten patients studied.

Because acute respiratory failure has caused death in a significant number of critically ill patients, 1 prior research has been concerned mainly with causation and therapy; however, the use of continuous positive-pressure ventilation (CPPV) may increase the survival rate of patients with acute respiratory failure,2.a and since evaluation of treatment based on survival of patients is relevant only when mortality is excessive, determining the effectiveness of therapy for acute respiratory failure should include consideration of morbidity, as well as mortality.4 Therefore, we tested the pulmonary function of patients with severe acute respiratory failure who were treated with high levels of positive end-expiratory pressure (PEEP).

intrapulmonary shunting of blood (Qsan/Qt) greater than 30 percent of their cardiac output; (3) elevated systolic pulmonary arterial pressure (~ 30 DUD Hg), with normal or low pulmonary arterial wedge pressure (~ 15 DUD Hg); (4) normal or elevated cardiac index (~ 3.0 L/min/sq m); ( 5) chest roentgenographic evidence of interstitial pulmonary edema; and (6) tachypnea and intercostal retraction. Each patient with severe acute respiratory failure was intubated with an orotracheal tube with a low-pressure cuJI and high residual volume. They breathed spontaneously from a continuous How of heated humidi6ed gas supplied by a highflow &ix-oxygen blender (bird Corp.) with a variable flowmeter. A volume-limited ventilator (Mark 9-6, bird Corp.; or J. H. Emerson Co.) provided a mandatory ventilator rate (usually 2 to 4 breaths per minute) to maintain an arterial pH greater than or equal to 7.35, using methods described previously.5 Therapy with PEEP was applied with a threshold resistor exhalation valve (J. H. Emerson Co. ). Endotracheal aspirates were obtained daily for aerobic culture. Pulmonary and radial arterial catheters were inserted percutaneously in all patients. Pulmonary venoUs admixture (Qs/Qt) was calculated at least hourly by standard methods (assuming a respiratory quotient of unity) from the gas tensions and pH measured in simultaneously drawn samples of arterial and mixed venous ( pulmonary arterial) blood.6 ,7 All determinations of gas tensions and pH were obtained using standard electrode techniques (Instrumentation Laboratories). In each patient~ PEEP was applied to a level sufficient to minfmize Qsp/Qt, and the fractional concentration of oxygen in the inspired gas (FI02) was adjusted to maintain Pa02 between 60 and 100 nim Hg. As intrapulmonary derangements responsible for increased Qsp/ Qt improved~ PEEP was gradually decreased; however, Qsp/Qt was not allowed to increase more than 4 percent above the minimum value when PEEP was lowered. Criteria for tracheal extubation were as follows: (1 ) Pa02 greater than or equal to 65 mm Hg with FI02 less than

MATERIALS AND METIIODS

We analyzed data on all patients admitted to the surgical intensive care unit at Wilford Hall USAF Medical Center from September 1973 to March 1975. The following conditions were present in those patients considered to. have severe acute respiratory failure: (1) hypoxemia, with an arterial oxygen pressure (PaO 2) less than or equal to 100 DUD Hg while breathing 100 percent oxygen; (2) right-to-left °From the Recovery Room/Surgical Intensive Care Service, Deparbnent of Anesthesiology, Wilford Hall USAF Medical Center, Lackland Air Force Base, Texas· and the Department of AnesthesiolQgy, University of Florida College of Medicine, J. Hillis Miller Health Center, and the Veterans Administration Hospital, Gainesville, F1a. o °Chief, Surgical Intensive Care Service. t Assistant Professor of Anesthesiology and Surgery. Manuscript received March 11; revision accepted June 30. Reprint requests: Department of Anesthesiology, University of Florida College of Medicine, Box ]-254, ]. HiUis Mazm. Health Center, Gainesville, Florida 32610

18 DOUGLAS, DOWNS

CHEST, 71: 1, JANUARY, 1977

Table l-CUnieal Data from Ten PGderau ..lao Surt1iWJd kute Rapirtllory 'cd"'re and UruIenNrd Pulmolllll'1' 'unedon Tali,..

Patient, Age (yr), Sex

Iilitiil··

Following Optimization of PEEPt

A

Diagnosis·

Qsp/Qt·

PaOt, mmHg

Days of Tracheal Intubation

A

I

Qsan/Qt

PaOt, mmHg

PEEP, mmHg

1,39, M

SjP aortobifemoral graft

0.32

60

20

0.08

125

3.5

2,20, M

Motor vehicle accident; multiple fractures

0.33

100

20

0.14

92

6.0

3,19, M

Motor vehicle accident; T-6 spinal cord transection

0.34

100

30

0.10

110

4.0

4,42, F

SIP cholecystectomy; bacterial septicemia

0.39

42

32

0.15

85

16.0

5,22, M

Abdominal gunshot wound; bacterial septicemia

0.34

80

27

0.11

106

6.0

6,67, M

SIP aortobifemoral graft

0.46

50

25

0.12

91

7.0

7,36, M

SIP CASVG

0.40

65

25

0.14

89

3~0

8,18, F

Aspiration pneumonitis

0.40

70

40

0.09

120

3.0

9,58, M

S/pCASVG

0.35

90

33

0.15

93

3.0

Motor vehicle accident; pulmonary contusion; flail chest

0.33

60

32

0.10

105

6.0

0.37±0.01

72±6

28±2

O.12±0.01

102±4

5.8±1.2

10,34, M

Mean±SEt

...

·SjP, Status post; and CASVG, coronary artery saphenous vein graft. ··FIOt,I.0. tMean age, 36 ± 5 years. tFIOt,0.4. or equal to 0.3; (2) Qsp/Qt less than or equal to 0.15; (3) PEEP less than or equal to 3 mm Hg; and (4) ventilator rate less than or equal to one breath per minute with arterial pH greater than or equal to 7.35. The following measurements were obtained to determine appropriate pharmacologic cardiovascular support and fluid administration: (I) cardiac output by thermodilution technique (Edwards Laboratories); ( 2 ) arterial-mixed venous oxygen content difference; ( 3 ) pulmonary arterial wedge pressure; (4) heart rate and arterial blood pressure; and ( 5) hematocrit reading, arterial pH, and urinary output. Appropriate fluids (blood or salt solutions) were administered to maintain these measurements within normal limits. Renal function was monitored by determining free water clearance daily. 8 If there was evidence of renal tubular dysfunction, patients were given increased amounts of fluid and diuretics. 8 All patients received oral or intravenous hyperalimentation within 48 hours of admission to the intensive care unit.

We defined survivors as those patients who lived at least three months after discharge from the hospital. The results of pulmonary function tests, arterial blood gas analyses, and chest roentgenograms were obtained as soon as possible after discharge from the hospital and these studies were repeated whenever possible. Pulmonary function testing consisted of studies with a water spirometer and measurements of lung volume and carbon monoxide diffusing capacity ( D L ) • Routine spirometric studies were performed with a IO-L Stead-Wells spirometer. Forced vital capacity (FVC), mean

CHEST, 71: 1, JANUARY, 1977

forced expiratory flow during the middle baH of the FVC (FEF 25-751), and forced expiratory volume at one second ( FEV 1.0) were measured, and percentages of predicted values were reported. 10 Total lung capacity (TLC), funetional residual capacity (FRC), and residual volume (RV) were measured with a body plethsymograph (Collins), 11 were corrected to body temperature and ambient pressure saturated, and were reported as percentages of predicted values. 12 The D L was evaluated by the single-breath method18 adapted for use with gas chromatographic studies,14 and results were expressed as percentages of normal values. IS Results were considered normal if they were between 80 and 120 percent of the predicted values. To facilitate sampling of arterial blood, an indwelling 18-gauge thin-waIled needle was placed into the brachial artery of each patient. Studies were performed while patients breathed room air, sat resting afteI' exercise (we used a 9O-second modification of Masters' twostep prooedure),U and again after they inhaled 100 percent oxygen for 15 minutes. We performed in situ postmortem fixation of lungs within one hour of death on patients for whom autopsy was authorized. Fixation of the lungs was achieved by instilling 2 L of a formaldehyde solution (10 percent buffered formalin) through the endotracheal tube at a pressure equal to the end-expiratory airway pressure. The endotracheal tube was clamped to prevent collapse of the lungs and the thoracic contents were removed en bloc at posbnortem examination. Representative pulmonary tissue from all lobes was prepared for light and electron microscopic examination.

PULMONARY FUNCTION AFTER SEVERE ACm RESPIRATORY FAILURE 19

REsuLTS

of the 43 survivors (Tables I to 4). Patients not tested included five dependent children unable to cooperate for testing, three persons transferred overseas, 12 female patients who were military dependents unable to return for testing, one patient with tracheal stenosis who required a permanent tracheostomy and was unable to perform satisfactory spirometric maneuvers, five who refused pulmonary function testing, and seven who could not be located after military discharge. All surviving patients were asymptomatic of pulmonary dysfunction at the time of discharge from the hospital. All patients' chest x-ray films were abnormal during acute respiratory failure, improved rapidly as Qsp/Qt was minimized with PEEP, and remained improved until extubation or death. Following discharge, chest x-ray films showed no evidence of fibrosis, decreased lung volume, or hyperinflation; and all were interpreted by a radiologist to be within nonnaI limits. No patient complained of intolerance to exercise, physical impairment, increased sputum production, cough, or increased susceptibility to respiratory infections. The Pa02 with the patient breathing 100 percent oxygen was normal in all patients examined. Only patient 4 had both decreased DL and Pa(h following exercise. This patient also had a low FEF2575S and FEV1.0 indicating mild airway obstruction. The FRC was not increased in any patient, in spite of increased RV in patients 6, 8, and 10 (Tables 2 and 4). Pulmonary function was reassessed at least two times in six patients (Nos. 3 to 8 ): Patients 3, 4, and 8 improved after initial testing (Table 4), but we observed no changes in the remaining three patients during the ensuing 12 months. We microscopically examined the pulmonary tissue of six patients who died. Minimal alveolar capillary interstitial widening and edema were present in all patients. There was no evidence Of deposi-

Fifty-four patients with severe acute respiratory failure required PEEP greater than or equal to 20 mm Hg to minimize Qsp/Qt. There were 18 female and 36 male patients, with a mean age of 38 ± 8 years (SE). At an FI02 of 1.0 initially, the mean value (-+-SE) for Qsan/Qt was 0.35 -+- 0.06 and for Pa(h was 85 -+- 8 mm Hg. The maximal PEEP to minimize Qsan/Qt was Z7 ± 2 mm Hg (mean ± SE ). Following optimal PEEP (FI02 of 0.4), the mean value (±SE) for Qsp/Qt was 0.14 -+- 0.06 and for Pa(h was 104 -+- 5 mm Hg. The mean number of days of tracheal intubations was 4.6 -+- 2.3 (SE). No patient required an FI02 greater than 0.4 to maintain Pa02 greater than 60 mm Hg at the optimum PEEP level. Forty-three (80 percent) of these 54 patients were alive three months after leaving the hospital. Nine patients died from septicemia and low cardiac output unresponsive to intravenous administration of fluids or infusions of positive inotropic drugs. One patient died after cardiac arrest three weeks after extubation; autopsy revealed avulsion of a ventricular septal patch. One patient suffered a massive intraventricular cerebral hemorrhage, and mechanical ventilatory support was therefore withdrawn. All II patients who died were able to maintain Pa02 greater than 60 mm Hg breathing less than 60 percent oxygen. Hypoxemia was not considered contributory, nor was acute respiratory failure the primary cause in any of these deaths. Cultures of endotracheal aspirate rarely grew bacteria within 24 hours of intubation, but did reveal colonization in all patients intubated for 72 hours or more; however, bacterial pneumonia was not believed to be present or to be the cause of acute respiratory failure in any patient. We performed tests of pulmonary function in ten

Table Z--R. .lt. 01 Spiromearie Studie- and M. . .remenb 01 Luq Yofume

Patient

Time of Final Study after Extubation, mo

1 2 3 4 5 6

7 8 9 10 Mean±SE

0.5 5.0 4.0 12.0 6.0 12.0 11.0 4.0 0.5 0.5 5.6±1.9

Percent of Predicted Normal Values· Smoker

FEV1 .o

FEF25-75%

FVC

FRC

TLC

RV

Yes Yes No Yes No Yes Yes Yes Yes No

95 90 90 84

118 122

71 95 83 83 118

90 92 91 50 85 99 66 86 97 108 86±5

80 90 93 70 95 100 86 88 95 95 89±3

104 88 99 48 80 121 65 125 113 145 99±9

89 64

81 100 77 91 86±6

82

117 147 46 88 110 68 103 l00±9

80

91 76 87 80 86±4

·Values ± 20 percent of predicted were considered normal.

20 DOUGLAS, DOWNS

CHEST, 71: 1, JANUARY, 1977

Table 3---Caroon Monoside DiDlUion (Pereent 01 Predic.ed Normal J'alua) and Arterial Blood Gaa Teuoru (mm H.) at YariOfU Time. FollotfIiq Esaabalion

Patient

DL

1 2 3 4 5 6 7 8 9 10 Mean±SE

74 95 83 65 92 90 84 68 63 94 82±4

At Rest

,

PaOt* 575 605 585 625 615 583 580 590 540 525 582±10

**FIOt == 0.21.

*FIOt == 1.0.

,

A

PaC02 *

PaC02 **

Pa02** 81 82 85 74 107 88 80 86 68 75 83±3

30 35 39 34 38

35 42 32 25 40 35±2

31 36 41 32 37 38 41 34 36 41 37±1

Following Exercise

PaO~**

, PaC02 **

93 94

31 37

A

... t

.. ·t

70 107 101 89 89 72 93 9O±4

30 42 33 41 35 34 33 35±1

tUnable to exercise because of paraplegia.

tion of fibrin, formation of hyaline membranes, or capillary thromboses. In all cases, proliferation of alveolJr type-2 cells was evident, and alveolar microstructure was well preserved.

months after extubation; however, this patient was 67 years old and had smoked one to two packs of cigarettes per day for 52 years. Three patients had increased RV, but in 'no patient was the FRC elevated, indicating poor cooperation of patients, since the increase in RV was associated with airway obstruction in only one patient. Patients 4 and 7 had a significant decrease in RV, FRC, and TLC, indicating pulmonary restriction; however, both patients had normal FVC. D L was decreased in four patients, all cigarette smokers, but was accompanied by a decrease in Pa02 after exercise in only one patient (No.4). The Pa02 was less than 80 mm Hg with an FIo2 of 0.21 in only three patients and was within normal limits following inhalation of 100 percent oxygen. These values indicate the possible presence of ventilation-to-perfusion ~Jn-., equality with minimal intrapulmonary right-to-Ieft shunting of blood or abnormality of oxygen diffu" sion. 19 Patients 3, 4, and 8 were studied on several occasions (Table 4). Initially, patient 3 had a reduction in FVC and D L , but a normal D L adjusted for alveolar volume. The FVC may have been reduced secondary to a T-6 spinal cord transection and the D L may have been low because of reduced lung volume, since D L corrected for alveolar volume was

DISCUSSION

Although the cause of acute respiratory failure is not clearly understood, the pathologic processes are well defined. Patients dying of acute respiratory failure usually have progressive interstitial, perivascular, and alveolar edema believed to be sec'ondary to diffuse destruction of alveolar-capillary membranes. The absence of alveolar type-1 cells, the deposition and organization of fibrin, and, eventually, fibrosis indicate loss of structural integrity.16.17 Conventional therapy has been used in an attempt to"reverse these derangements,7.18 but incomplete reversal may cause restriction of lung volume, decreased diffusing capacity for oxygen, and ventilation-to-perfusion abnormalities. 4 Our patients had minimal laboratory observations indicating pulmonary dysfunction. Mean time for final examination of all patients was 5.6 months. Many tests performed earlier "revealed minimal abnormality, indicating rapid recovery from severely deranged respiratory function. One patient (No.6) had significant airway obstruction 12

Table 4--Serial Deeerminmioru of Pulmf'nary Funedon FoUOM1iq Acute R~piratory Failure·

Patient 3 Measurement FVC FEV1 •O FEF25-75% FRC

TLC RV

DL

A

Initial**

Mter 3 mo"

63 92 84 95 86 113 68

83 90

r

Patient 4

,

,

Initial**

After 12 mo

40 100 130 58 55 43 39

83 84 117 50 70

82

91 92 99 93

Patient 8 r

,

A

Initial**

After 4 mo

55 100 90 102 81

76 100 110 86 88 125 68

88

125

65

50

*Table values repre$ent percent of predicted normal values. **At discharge from hospital.

CHEST, 71: 1, JANUARY, 1977

PULMONARY FUNCTION AFTER SEVERE ACUTE RESPIRATORY FAILURE 21

normal. Normal pulmonary function returned within three months. Initially, patient 4 had a reduction in FVC, all static lung volumes, and D L • Within one year, these measurements improved, but not to normal values. This patient was the first subject treated with high levels of PEEP, and minimization of Qsp/Qt was not maintained, nor was reduction in PEEP limited by a 2-percent increase in Qsp/Qt, as they were in all other subjects. 2o In addition, this patient received supplemental oxygen therapy, which no other patient required, for two weeks after extubation. The use less aggressive therapy may be responsible for the persistent abnormalities of pulmonary function. Patient 8 had an initial reduction in FVC and D L , which was present four months later. The abnormal D L may have been a result of reduced lung volume or cigarette smoking, or both. Abnormalities that were present in each of these three subjects within two weeks of tracheal extubation improved with time. Three patients (Nos. 1, 9, and 10) examined within two weeks of extubation had almost normal pulmonary function. Few reports have been published concerning recovery of pulmonary function following acute respiratory failure. Lakshminarayan et al21 reported a 43-percent survival rate in a group of 55 patients with acute respiratory failure who received conventional therapy. Ten survivors had pulmonary function testing an average of 21 months after acute ~espiratory failure. Four had obstructive defects, with FEF25-75~ ranging from 11 to 50 percent of predicted values. Two showed a restrictive pattern., with FVC being 65 percent of predicted values. The D L was abnormal in most patients, although arterial blood gas levels usually were normal. Additional case reports4.22 revealed early restrictive and gas diffusion defects after acute respiratory failure. The defects improved with time in one case;4 exCept for survival rate, these findings are all similar to ours. Acute respiratory failure is manifested by a progressively increasing closure of small airways and loss of lung volume, which produces right-to-left intrapulmonary shunting of blood and a p~gres­ sively decreasing pulmonary compliance. 2 The' pulmonary compliance of patients with acute respiratory failure may be ree;Juced so severely that spontaneous respiration cannot provide sufficient alveolar ventilation for elimination of carbon dioxide. Controlled mechanical ventilation, when used to prevent hypoventilation, will increase intrapleural pressure, decrease venous return, and increase resistance to blood Howat the pulmonary capillary level. 7 Cardiac output and, therefore, oxygen delivery may be compromised by CPPV.2 Thus, ap-

ot

22 DOUGLAS, DOWNS

plication of levels of PEEP necessary to normalize lung volume and minimize Qsp/ Qt may not be possible. 7 In addition, controlled ventilation will cause maldistribution of the alveolar ventilation-to-perfusion ratio,23-25 a pathologic condition that should be avoided. Allowing spontaneous ventilatory efforts to persist and adding ventilatory support as needed results in lower intrapleural pressure and improves thoracic venous blood return, cardiac output, and oxygen delivery.28 Therefore, higher levels of PEEP and, presumably, more complete reversal of the closure of small airways can be used safely without compromising cardiovascular function. 27.28 Examination of the pulmonary microstructure in those patients who died did not demonstrate the pathologic picture reported by other investigators. IS Rather, the integrity of alveolar capillaries was maintained, with minimal evidence of interstitial edema or hemorrhage. Early reparative processes were evident with hyperplasia of alveolar type-2 cells. 29 Only minor pathologic findings suggest that early aggressive therapy with PEEP was not harmful, and previously reported clinical data have confinned its clinical efficacy. 20.27 Aggressive therapy with levels of PEEP necessary to minimize Qspl Qt, and intensive cardiopulmonary monitoring permitted a survival rate of 80 percent (43/54) in our patients with severe acute respiratory failure. In spite of high sustained airway pressures, such therapy allowed early return to almost normal pulmonary function. It was impossible to determine to what extent treatment influenced findings, but in all ten cases studied, residual abnormalities were not of clinical significance. ACKNOWLEDGMENTS: We thank the nurses and technicians of the Surgical Intensive Unit, Wilford Hall USAF Medical Center for extreme dedication to patient care; Lt.Col G. Harkleroad for performing tests of pUlmonary function; Ms. Dianna Kosman and Ms. Pauline Snider for editorial assistance; and J. H. Modell, M.D., for continuous encouragement. .

Care

REFERENCES

1 Fairley HB, Schlobohm RM, Singer MM, et al: The appropriateness of intensive respiratory care. Crit Care Med 1:115, 1973 2 Sugennan HJ, Rogers RM, Miller LD: Positive endexpiratory pressure (PEEP): Indications and physiologic considerations. Chest 62 (suppl) :86S-94S, 1972 3 Kumar A, Falke KJ, Geffin B, et a1: Continuous positivepressure ventilation in acute respiratory failure. N Eng! J Med 283:1430-1436, 1970 4 Downs JB, Olsen GN: Pulmonary function following adult respiratory. distress syndrome. Chest 65: 92-93, 1974 5 Downs JB, Perkins HM, Modell J8: Intermittent mandatory ventilation: An evaluation. Arch Surg 109:519523, 1974

CHEST, 71: 1, JANUARY, 1977

6 Pulmonary terms and symbols: A report of the ACCPATS Joint Committee on Pulmonary Nomenclature. Chest 67:583-593, 1975 7 Downs JB, Klein EF, Modell JH, et al: The ~ffeet'''Of incremental PEEP on Pa02 in patients with respiratory failure. Anesth Analg (Cleve) 52:210-215, 1973 8 Baek SM, Brown RS, Shoemaker WC: Early prediction of acute renal failure and recovery: 1. Sequential measurement of free water clearance. Ann Surg 177:253258, 1973 9 Baek SM, Brown RS, Shoemaker WC: Early prediction of acute renal failure and recovery: 2. Renal function response to furosemide. Ann Surg 178:605-608, 1973 10 Morris JF, Koski A, Johnson LC: Spirometric standards for healthy, nonsmoking adults. Am Rev Respir Dis 103:57-67, 1971 11 Boren HG, Kory RC, Syner JC: The Veterans Administration-Army cooperative study of pulmonary function: 2. The lung volume and its subdivisions in normal men. Am J Med 41:96-114, 1966 12 Bates DV, Macklem PT, Christie RV: Respiratory Functions in Disease: An Introduction to the Integrated S~dy of the Lung (2nd 00). Philadelphia, WB Saunders Co, 1971 13 Weg JG, Krumholz RA, Hackleroad LE: Unilateral hyperlucent lung. Ann Intern Moo 62:675-684, 1965 14 Cotes JE: Lung Function: Assessment and Application of Medicine (2nd ed). London, Blackwell Scientific Publications, 1968 J5 Gould KG Jr, Cooper KH, Harkleroad LE: Pulmonary function and work capacity in the absence of physiologic' dorsal kyphosis of the spine. Dis Chest 55:405410, 1969 16 Blaisdel FW, Lim RC, Stallone RJ: The mechanism of pulmonary damage following traumatic shock. Surg Gynecol Obstet 130: 15-22, 1970 17 Nash G, Blennerhassett JB, Pontoppidan H: Pulmonary lesions associated with oxygen therapy and artificial ventilation. N Engl J Moo 276:368-374, 1967

CHEST, 71: 1, JANUARY, 1977

18 Ashbaugh DG, Petty TL: Positive end-expiratory pressure: Physiology, indications and contraindications. J Thorac Cardiovasc Surg 65:165-170, 1973 19 Rahn H, Farhi LE: Ventilation, perfusion and gas exchange: VA/Q concept. In Fenn WO, Rahn H (eds): Handbook of Physiology and Respiration: Respiration (sec 3, vol 2). Washington, DC, American Physiological Society, 1964, pp 735-766 20 Carter GL, Downs JB, Dannemiller FJ: "Hyper" endexpiratory pressure in the treabnent of adult respiratory insufficiency: A case report. Anesth Analg ( Cleve) 54:31-34, 1975 21 Lakshminarayan S, Petty TL, Stanford RE: Recovery after adult respiratory distress syndrome. Abstracts of American Thoracic Society Meeting, 1975, pp 31-32 22 Llamas R: Adult respiratory distress syndrome: Report of survival after two episodes. Chest 65:468-469, 1974 23 Downs JB, Mitchell LA, Dannemiller F], et al: Modification of airway closure and pulmonary gas exchange following cardiac surgery. Crit Care Med 3:41, 1975 24 Modell. JH: Ventilation/perfusion changes during mechanical ventilation. Dis Chest 55:447-451, 1969 25 Froese AB, Bryan AC: Effect of anesthesia and paralysis on diaphragmatic mechanics in man. Anesthesiology 41:242-255, 1974 26 Kirby RR, Perry JC, Calderwood HW, et al: Cardiorespiratory effects of high positive end-expiratory pressure. Anesthesiology 43:533-539, 1975 27 Kirby RR, Downs JB, Civetta JM, et al: High level positive end expiratory pressure (PEEP) in acute respiratory insufficiency. Chest 67: 156-163, 1975 28 Cournand A, Motley HL, Werko L, et al: Physiological studies of the effects of intermittent positive pressure breathing on cardiac output in man. Am J Physiol 152: 162-174, 1948 29 Bachofen M, Weibel ER: Basic pattern of tissue repair in human lungs following unspecific injury. Chest 65 (suppl) :14S-19S, 1974

PULMONARY FUNCTION AFTER' SEVERE ACUTE RESPIRATORY FAILURE 23