Gas exchange measurements in neonates treated with extracorporeal membrane oxygenation

Gas exchange measurements in neonates treated with extracorporeal membrane oxygenation

Gas Exchange Measurements in Neonates Treated With Extracorporeal Membrane Oxygenation By Robert E. Cilley, John R. Wesley, Joseph B. Zwischenberger, ...

541KB Sizes 0 Downloads 25 Views

Gas Exchange Measurements in Neonates Treated With Extracorporeal Membrane Oxygenation By Robert E. Cilley, John R. Wesley, Joseph B. Zwischenberger, and Robert H. Bartlett Ann Arbor, Michigan 9 Closed-circuit spirometry techniques w e r e used to study gas exchange {oxygen and carbon dioxide transport) across the native lung and membrane lung in ten neonates treated with extracorporeal membrane oxygenation (ECMO}. Initially there was negligible oxygen transport across the native lung (0.4 _+ 0.4 m L / k g / m i n ) . An increase in native lung oxygen transport (2.0 _+ 1.5 m L / k g / m i n ) at low ventilator settings and high levels of extracorporeal support heralded recovery of the native lung and predicted weaning from ECMO. Measurement of oxygen consumption and carbon dioxide production permitted the energy expenditure to be calculated in these critically ill neonates. The metabolic rates varied widely in each neonate over time and among neonates. (mean 57 _+ 11 k c a l / k g / d , range 38 to 80 k c a l / k g / d ) . This emphasizes the need to determine energy expenditure in order to guide nutritional support. 9 198B by Grune & Stratton, Inc. INDEX WORDS: Extracorporeal membrane oxygenation; respiratory distress; congenital diaphragmatic hernia; meconium aspiration syndrome.

XTRACORPOREAL membrane oxygenation (ECMO) is prolonged extracorporeal circulation with a modified heart-lung machine using a membrane oxygenator. ECMO provides temporary mechanical support for patients with severe cardiac or respiratory failure. This technique has been used successfully for the treatment of newborn infants with severe respiratory failure. 1 ECMO is considered when infants fail to improve despite maximal mechanical ventilation with 100% oxygen at rapid rates and high airway pressures, which are themselves damaging to the lungs. During ECMO, gas exchange (oxygen and carbon dioxide transport) is achieved in the extracorporeal circuit while the ventilator is set at low nondamaging pressures and oxygen concentrations. Lung function usually returns to normal over a period of days. Recovery is recognized by diuresis and weight loss, improvement of roentgenographic appearance of the lungs, and adequate arterial oxygenation and carbon dioxide removal at progressively lower levels of extracorporeal

E

From the University o f Michigan Medical Center, Mort Children's Hospital, Ann Arbor, MI. Presented in part at The American Pediatric Surgical Association Annual Meeting, May 1, 1985. Address reprint requests to John R. Wesley, MD, University o f Michigan Medical Center, F7516 Mott Children's Hospital, Ann Arbor, M t 48109. 9 1988 by Grune & Stratton, Inc. 0022-3468/88/2304-0002503.00/0 306

circulation. Ultimately, pulmonary recovery has occurred when normal arterial blood gases are achieved using minimal ventilator settings without extracorporeal support. During ECMO a unique opportunity exists to study lung function in infants with severe respiratory failure who are not dependent on native lung function. Since inefficient gas exchange across the patient's native lung is the fundamental problem in respiratory failure, recovery must be characterized by improvement in the efficiency of pulmonary gas exchange. We sought to quantitate gas exchange (ox.ygen transport [VOz] and carbon dioxide transport [VCO2]) across both the neonate's native lung and the artificial membrane lung throughout the course of treatment with ECMO in order to investigate pulmonary recovery. In addition, quantitation of metabolic gas exchange enabled us to calculate energy expenditure. We report gas exchange measurements performed on ten neonates with severe respiratory failure treated with ECMO. MATERIALS AND METHODS Patient Population Newborns are treated with ECMO when they show progressive deterioration or fail to improve while undergoing maximal conventional treatment for respiratory failure. 2 Conventional treatment includes the use of high pressure mechanical ventilation with 100% oxygen, induced alkalosis, inotropic agents, vasodilators, and paralysis with pancuronium. Using ECMO, a survival rate >90% has been achieved in infants older than 34 weeks gestation) '2 Descriptive characteristics of ten patients treated with ECMO who underwent gas exchange measurements are shown in Table 1. ECMO was started between one and eight days of life (mean 3.5 days) and lasted from two to 11 days (mean six days). Venoarterial (VA) ECMO was used in all patients except patient 6 in whom venovenous ECMO was started and then converted to venoarterial perfusion after four days. All patients treated were >34 weeks gestationa! age. The primary diagnoses included meconium aspiration syndrome (MAS), infant respiratory distress syndrome (RDS), congenital diaphragmatic hernia (CDH), and primary persistant fetal circulation (PFC). The pathophysiology of each of these disorders includes pulmonary hypertension. Gas exchange measurements were performed as soon as possible after ECMO was started and daily or twice daily during ECMO. After ECMO, patients were studied prior to extubation if possible. Two patients who required prolonged ventilator support at high ventilator settings (nos. 3 and 9) were not studied after ECMO.

Method of Neonatal Venoarterial ECMO in Brief Newborns requiring ECMO undergo right internal jugular vein canulation) Venous blood is drained by gravity to a servoregulated roller pump. Blood then passes through a spiral-wound silicone Journal of Pediatric Surgery, Vol 23, No 4 (April), 1988: pp 306-311

GAS EXCHANGE IN NEONATES ON ECMO

307

Table 1. Descriptive Characteristics of Ten Neonates Treated With ECMO on Whom Gas Exchange Measurements W e r e Performed Patient

Birth Weight (kg)

Gestational Age (wk)

Diagnosis

Age at ECMO (h)

Duration of ECMO (h)

Respiratory Outcome

1

2.90

41

PFC

93

79

18 h to extubation

2

4.35

40

MAS

22

120

52 h to extubation

3

2.61

35

RDS

180

264

1 mo ventilator support, mild de-

4

2.52

37

RDS

156

100

75 h to extubation

5

2.87

40

PFC

164

202

Ventilator dependent, late death

6

2.95

35

MAS

88

171

15 h to extubation

7

3.22

38

MAS

36

54

14 h to extubation

8

4.05

40

MAS

39

52

72 h to extubation

9

2,30

38

MAS

37

191

Long-term ventilator support

10

3.16

37

CDH

31

199

8 d to extubation

velopmental delay

Abbreviations: PFC, persistent fetal circulation;

MAS, meconium aspiration syndrome; RDS, respiratory distress syndrome; CDH, congenital

diaphragmatic hernia.

rubber membrane oxygenator (Sci-Med Life Systems, Minneapolis, M N ) and pediatric heat exchanger, and is then returned via the right common carotid artery. During E C M O , ventilator settings are reduced to allow lung recovery (fraction of inspired oxygen [FiO2] = 0.30, inspiratory pressure [[P] = 30 cm HzO, positive end expiratory pressure [PEEP] = 3 cm H20; respiratory rate [RR] = 1S/rain). Extracorporeal flow rates of 60 to 100 m L / k g / m i n are used to provide cardiopulmonary support as determined by adequate arterial oxygenation (PO2 = 55 to 90 m m H g ) , normal mixed venous

oxygen saturation (>75%), and normal tissue perfusion. Patients are anticoagulated with heparin and receive antibiotics while on E C M O . Paralytic drugs are not used.

Nutritional Support The total daily caloric intake was calculated for each neonate from the time of arrival in our neonatal intensive care unit until after the last metabolic study was performed. After E C M O was begun,

Table 2. Native Lung and Membrane Lung Oxygen Transport During and A f t e r ECMO in Ten Neonates Day of Life

2

5

6

7

8

9

NatLVO2

0.0

5.6

6.8

9.9

10.4

MemLVO 2

7.2

4.5

3.2

0.9

--

EC flow

59

79

69

21

Off

6.7

7.0

Patient 1

2

3

4

5

3

4

NatLVO

0.4

0.5

0.6

1.0

2.1

MemLVO z

5.3

5.8

6.2

5.9

5.0

EC flow

62

57

85

67

64

Off

7

8

9

10

11

12

13

14

15

16

17

0.3

0.0

0.4

1.0

2.1

2.5

0.6

3.6*

MemLVO 2

5.9

6.3

7.1

6.5

6.2

6.2

6.8

3.5

EC flow

95

117

102

109

102

91

88

47

NatLVO 2

0.0

0.0

1.4

4.3

8.0

4.0

8.2

MemLVO 2

6.4

8.6

8.1

EC flow

83

103

71

24

Off

NatLVO=

0.4

0.0

0.0

0.0

0.0

0.8

1.9

3.7

8.5

MemLVO 2

5.4

7.9

8.9

11.1

11.0

9.6

8.5

6.2

--

63

80

105

115

111

63

Off

NatLVO2

0.8

0.0

98

98

0.0

0.0

3.3

MemLVO 2

7.3

8.1

8.4

9.9

8.3

EC flow

129

112

110

78

71

NatLVO2

0.3

1.4

6.5

30

Off

NatLVO 2

EC flow 6

10

Off 8.9 Off

7.3 Off

6.3

MemLVO 2

4.9

5.2

--

--

EC flow

79

94

Off

Off 6.6

NatLVO2

0.3

2.1

5.5

MemLVO~

4.9

6.1

2.1

--

EC flow

78

61

15

Off

NatLVO2

0.0

0.0

0.0

0.4

0.7

MemLVO 2

5.7

6.4

7.4

7.3

6.6

EC flow

96

70

101

70

79

NatLVO 2

1.5

1.3

2.2

0.4

2.6

0.6

2.1

3.0

MemLVO 2 EC flow

4.0

8.0

5.9

7.2

5.8

8.2

7.4

7.2

60

63

51

51

44

63

47

50

7.3 Off

NatLVO2 = native lung oxygen transport in mL/kg/min. MemLVO 2 ~ membrane lung oxygen transport mL/kg/min. EC f l o w , extracorporeal flow in mL/kg/min. Ventilator settings FiO2 = 0.30, IP = 30 cm H20, RR = 15 except patient 3. *Last study FiO2 = 1.0, IP = 40 cm H20, RR = 40.

308

CILLEY ET AL

parenteral nutrition was used to increase the caloric intake to 80 to 110 Kcal/kg/d. Protein was administered as an amino acid solution beginning at 0.5 g / k g / d and increased to 2.0 g / k g / d . Carbohydrate was given as a 10% to 20% dextrose solution, while fat was given as lipid emulsion at 1 to 4 g / k g / d . Standard guidelines for nutritional support were used in these patients. 4 Nutritional support was not regulated in a prospective fashion.

membrane lung and native lung gas transport was assembled on a compact portable bedside cart. Sequential 15 minute gas exchange measurements were made for one hour. Infants were resting quietly during the studies. All values were reported at standard temperature and pressure dry. Gas exchange was used to calculate energy expenditure by the Weir equation. 6 All values are given as mean +_ 1 SD. Statistical significance was determined by the Student's t-test and analysis of variance.

Method of Gas Exchange Measurement RESULTS

The method of simultaneous measurement of pulmonary and m e m b r a n e lung oxygen and carbon dioxide transport has been presented in detail elsewhere. 5 These methods used adaptations of closed-circuit spirometry techniques. For membrane lung gas exchange, 100% oxygen was supplied to the membrane lung by a roller pump, delivering oxygen at flow rates of 2 to 6 L / m i n , and exhaust gas was returned to the closed circuit. Carbon dioxide in the exhaust gas was measured by infrared capnometry and removed by a soda lime scrubber. Carbon dioxide production was calculated as the product of the fraction of expired CO2 and flow. Oxygen transport across the membrane lung was measured as the volume lost from the closed circuit and replaced volume for volume by 100% oxygen in a volumetric spirometer. Gas exchange across the native lung was measured in a similar fashion, using closed-circuit spirometry. Provision was made to supply heated humidified gas delivered by an infant ventilator with continuous flow at 10 L / m i n . During gas exchange measurements, the ventilator was set at FiO2 = 0.30, IP = 30 cm H20, and R R = 15/min. Equipment for measurement of

Pulmonary Recovery The results of oxygen consumption measurements are shown in Table 2. Sixty-four studies were performed on ten patients. Fifty-two studies were performed during ECMO, measuring both native lung and membrane lung gas exchange, while 12 studies were performed after E C M O was discontinued measuring pulmonary gas exchange alone. At an initial level of extracorporeal flow of 75 _+ 14 m L / k g / m i n , the mean native lung oxygen transport (NatLVO2) was 0.4 + 0.4 m L / k g / m i n , representing 7% of the total oxygen consumption. In three patients there was initially no measurable native lung oxygen transport and in three other patients native lung oxy-

Table 3. Oxygen Consumption and Carbon Dioxide Consumption in Ten Neonates Treated W i t h E C M O

Day of Life Patient 1 402 VCO2 Calories 2 402 4CO2 Calories 3 402 4CO2 Calories 4 402 VCO2

1

--

2

-5.7 4.8 25

3

10 6.3 4.8 40

4

5

6

7

8

9

14 6.8 5.2 78

7.2 6.1 42 6.9 6.0 87

10.1 8.5 92 7.1 6.8 93

10.0 9.7 102 6.7* 7.9* 69

10.8 9.3 103 7.0* 6.4* 51 6.2 5.6 38 6.4 5.7 56 5.8 5.2 60 8.4 8.7 104

10.4" 9.0* 97

--

Calories

5 402 VCO2 Calories 6 402 4CO2 Calories 7 402 4CO2 Calories 8 402 4CO2 Calories 9 402 VCO2 Calories 10 402 VCO2 Calories

17

34

41

46

50

29

39 5.2 6.0 39 5.2 5.8 37

46 6.6 7.6 57 8.2 7.9 43 5.7 5.3 31 8.1 6.8 65

35 6.5 6.7 65 7.6 7.7 65 6.4 6.1 55 7.6 6.7 77

33

17

23

5.4 5.1 42

9.2 6.5 50

53 8.1 6.5 63 6.3* 6.0* 68 6.6* 6.5* 68 7.4 7.2 71 8.4 7.7 75

62 8.1 8.0 96

7.7 6.8 87 8.8 7.5 77

80 9.5 8.1 83

All values in mL/kg/min. Calories, total caloric intake in 24 hours (kcal/kg/24 h). *Indicates study was performed after ECMO was discontinued,

55 6.3 6.0 89 8.6 7.1 86 7.9 6.6 95

99

7.3 7.5 92 10.2 8.1 83

10

11

12

13

7.5 8.1 108 9.5 7.8 107 8.9 8.2 79 9.9 9.0 100

7.5 8.0 93 8.3 7.7 106 11.1 9.1 96 11.6 9.3 91

8.3 8.4 85 8.0* 7.0* 98 11.0 10.0 111

8.7 8.3 81

88 10.4 9.4 106

76 8.2* 7.1" 75 10.4 10.1 111

94

68

89

88

71

7.3* 7.5* 72

77

14

15

7.5 6.9 72

9.9 8.4 87 7.3* 7.3* 87

16

17

78

7.1 6.7 86

8.5* 7.2* 91

30

8.9* 8.5* 103

309

G A S E X C H A N G E IN N E O N A T E S O N E C M O

gen transport fell to zero after 24 to 48 hours on ECMO. Later, still at high levels of extracorporeal flow (80 _+ 18 mL/kg/min; difference from initial level not significant), an increase in pulmonary gas exchange was observed. Native lung oxygen consumption increased to 2.0 + 1.5 mL/kg/min representing 22% of the total oxygen consumption, at that time (P < .05 compared with initial NatLVO2 transport; Table 2). In seven patients, ECMO was rapidly weaned, according to the usual criteria for weaning. In these patients, increased gas exchange was observed in the native lung prior to weaning. Patient 3 demonstrated increased pulmonary gas exchange on day 3 of ECMO but failed to wean. Doppler/echocardiographic evaluation revealed a patent ductus arteriosis (PDA) with aorta-to-pulmonary artery shunting. After persistence of the PDA for 24 hours, PDA ligation was performed on day 5 of ECMO. This patient subsequently was weaned from ECMO at high ventilator settings and required prolonged mechanical ventilatory support. Patient 9 demonstrated minimal improvement in pulmonary gas exchange and failed to improve further. ECMO was discontinued and high ventilator settings were required for support. Patient 10 was the only infant with CDH in this study. This 3.2 kg female had an intrauterine diagnosis of left CDH made by ultrasound and underwent repair on her first day of life. She developed progressive pulmonary hypertension and respiratory failure and was put on ECMO at age 31 hours. Although her respiratory failure was as severe as any other patient in the study (pre-ECMO postductal PO2 = 43 mmHg on maximal ventilatory support), she showed high levels of native lung gas exchange (27% of total oxygen consumption) initially. Subsequently, she demonstrated wide swings in the percent of oxygen transport that occurred across her native lung (5% to 36%) while on ECMO.

Metabolism The total oxygen consumption (native lung and membrane lung oxygen transport) and total carbon dioxide production (native lung and membrane lung carbon dioxide transport) for each gas exchange measurement, as well as the caloric intake for each day of life are shown in Table 3. Caloric intake prior to transfer to our neonatal intensive care unit was not determined. Nine of ten neonates were studied within 18 hours of instituting ECMO (mean 13 _+ 8 hours). Eight of ten neonates were studied after ECMO was stopped. A rise in total oxygen consumption was seen during ECMO followed by a gradual decline (Fig 1). The mean initial oxygen consumption (6.1 m L / k g / min), maximal oxygen consumption (9.2 m L / k g /

12 II

ti

I0 9

VO2

8 7

(cc/kg/min] 6 5 4 3 2

I 0

INITIAL vo2

MAXIMAL vo2 ON ECMO

FINAL VO2

Fig 1. Time course of oxygen consump.tion (~/0 2) in ten neonates during t r e a t m e n t with ECMO. Initial V O 2 measured 1 3 _+ 8 hours after starting ECMO.

min), and final oxygen consumption (7.6 mL/kg/min) differ significantly at the P < .05 confidence level. Energy expenditure ranged from 38 to 80 kcal/kg/d (mean 57 _+ 11 kcal/kg/d). Nutritional support ranged from 10 to 111 kcal/kg/d (mean 75 _+ 25 kcal/kg/d). DISCUSSION

Pulmonary Recovery Gas exchange measurements have not previously been reported in critically ill neonates with respiratory failure who require maximal mechanical ventilation. Infants treated with ECMO provide an important model in which to study lung injury and recovery. The native lung is not needed for gas transport during ECMO, and may be studied without the requirement that it sustain a vital function. Prior to the institution of ECMO, all oxygen transport and carbon dioxide removal occurs across the native lung. One hundred percent oxygen and high ventilator pressure and rate are required for gas exchange sufficient to maintain life. Lung function is best described under these circumstances as inefficient. When infants are placed on ECMO, ventilator pressure, rate, and FiO 2 are reduced. In addition, during venoarterial bypass, 50% to 80% of the venous return to the heart flows through the extracorporeal circuit, and pulmonary blood flow is reduced by that amount. It is not surprising that pulmonary gas exchange was initially negligible due to the combination of decreased ventilator settings and the bypass of pulmonary blood flow. Interestingly, the one patient studied who was on venovenous bypass (in whom pulmonary blood flow was not reduced by ECMO) likewise showed no pulmonary gas exchange initially. Lung recovery was heralded by an increase in pulmonary gas exchange at low ventilator settings while still at high levels of ECMO flow. Subsequently, ECMO flow was decreased, resulting in more pulmonary blood flow and more gas exchange as expected.

310

The critical event that indicated recovery was the change from no or minimal gas exchange (< 10% total) to > 10% of total. The finding that significant pulmonary gas exchange occurred initially in the patient who underwent CDH repair is consistent with the hypothesis that respiratory failure in CDH is primarily a vasospastic disorder and not a result of parenchymal lung injury. The blood that does traverse the pulmonary capillary bed is oxygenated with low ventilator pressures; there was little lung injury in this patient who had been treated with high ventilator pressures for less than one day. We measured gas exchange in all of our patients at one set of ventilatory parameters (FiO2 = 0.30, inspiratory pressure = 30 mH20, rate = 15/min). A more complete characterization of pulmonary injury and recovery would be obtained by quantitating gas exchange during extracorporeal circulation using different ventilatory pressures, modes of ventilation, and inspired oxygen concentrations. Others have used lung compliance, which parallels the course of lung recovery in both animal models and neonates treated with ECMO, as a measure of lung recovery.7'8 The method of gas exchange measurement described in this report is relatively cumbersome and therefore of limited practical value. We are developing simplified methods that indicate the return of pulmonary gas exchange and therefore lung recovery. In the future, this may be of practical clinical value in ECMO therapy. Metabolism Other investigators have measured the metabolic rates of infants using a variety of calorimetric techniques. 9~5 Oxygen consumption values from these studies are given in Table 4. There are few reports on sick infants and only one study was performed on infants undergoing mechanical ventilation. ~4Accurate methods of gas exchange measurement during maximal mechanical ventilation (ie, prior to ECMO therapy) have not been described. Here we have reported the metabolic rates of critically ill neonates with respiratory failure, beginning at the time E C M O therapy was started and continuing throughout their recovery. The use of E C M O permitted the application of closed-circuit spirometry techniques to measure transpulmonary gas exchange and membrane lung gas exchange simultaneously. The summation of the gas exchange across the native lung and membrane lung permitted the calculation of energy expenditure. The metabolic rates varied widely in each neonate over time and among neonates, emphasizing the need to determine individual energy expenditure in order to

CILLEY ET AL

Table 4. Metabolic Rates of Neonates and Infants as Reported by Various Investigators 402

Age at Measurement

4.3 5.5

13-40 h 41-64 h

9.0 6.1

91-110 h 111-140 h

4.8 6.6 6.7

3h 1d 3d

7.0

7d

4.6

3h

5.3 6.2 6.2

ld 3d 7d

Senterre and Karlberg ~2 (SGA newborns) (term newborns)

4.8 5.4

Birth

Brooke ~3 (premature infants)

8.0

2 wk

Levison et al9 (newborns with RDS)

Hill and Rahimtulla 1~ (term newborns)

Scopes and Ahmed ~ (term newborns)

Birth

Richardson et al ~4 (newborns with RDS)

Dechert et al is

8.3

1d

6.5

2d

8-9

Variable

Oxygen consumption given in mL/kg/min. Abbreviations: SGA, small for gestational

age; RDS, respiratory

distress syndrome.

guide nutritional support and prevent over or under feeding. Since the caloric intake necessary to promote growth may exceed the resting energy expenditure by up to 45%, exact caloric needs can only be estimated from indirect calorimetry. 16 The increase in oxygen consumption that occurred during treatment with E C M O occurred concurrently with an increase in the caloric intake. The question of whether there is a causal relationship between increased caloric intake and rising oxygen consumption cannot be answered from these data. The increase in oxygen consumption while on E C M O could also be a result of extracorporeal circulation itself. The release of catacholamines and activation of WBCs that occur with extracorporeal circulation could cause an immediate rise in oxygen consumption] 7 It is unlikely, however, that the gradual rise, over days, and sustained elevation of oxygen consumption is due to extracorporeal circulation alone. In one canine study where oxygen consumption was determined before, during, and after extracorporeal circulation, the oxygen consumption during extracorporeal circulation was never greater than oxygen consumption before extracorporeal circulation. ~8 In normal newborn mammals there is a rise in

GAS EXCHANGE IN NEONATES ON ECMO

311

oxygen c o n s u m p t i o n during the first several days after birth. 9t1'19 Proposed reasons for this rise include (1) a rise in general level of sympathetic activity, (2) a rise in the arterial PO2, (3) increased tone in skeletal muscles, (4) greater activity of the gastrointestinal tract, (5) the specific d y n a m i c action of foodstuffs, (6) the work of breathing, and (7) the need for temperature regulation. Prior to birth, every fetus lives u n d e r conditions of relatively low oxygen delivery. H u m a n umbilical vein hemoglobin s a t u r a t i o n is 48% to 64%; while umbilical arterial hemoglobin s a t u r a t i o n ranges from 16% to 31%. 20 This condition of relatively low oxygen delivery n o r m a l l y changes at birth as arterial blood becomes nearly s a t u r a t e d and oxygen delivery therefore increases (provided that cardiac output a n d hemoglobin level r e m a i n constant). It is possible that the rise in oxygen consumption seen after birth is the result of an increase in oxygen delivery with an accomp a n y i n g increase in oxygen-utilizing metabolic pathways. N e o n a t e s with severe respiratory failure m a y never achieve n o r m a l levels of oxygen delivery. Their arterial blood is not saturated with oxygen a n d their cardiac output m a y be secondarily depressed. W h e n E C M O is begun, the neonate is i m m e d i a t e l y c h a n g e d

to a state of n o r m a l or l u x u r i a n t oxygen delivery. The g r a d u a l increase in oxygen c o n s u m p t i o n observed during E C M O m a y represent a similar a d a p t a t i o n to increased oxygen delivery that n o r m a l l y occurs shortly after birth. Since the E C M O patients' t e m p e r a t u r e s were controlled by a heat exchanger in the extracorporeal circuit, it is unlikely that oxygen c o n s u m p t i o n changes were due to t h e r m o r e g u l a t o r y effects. I n f a n t s were on standardized p a r e n t e r a l nutrition, m a k i n g the d y n a m i c action of food a n unlikely cause of oxygen c o n s u m p t i o n changes. The rise in oxygen c o n s u m p t i o n was seen in neonates who were n o r m a l l y active as well as those who had little spontaneous m o v e m e n t d u r i n g E C M O . Residual paralysis m a y have c o n t r i b u t e d to the decreased metabolic rate found on the first meas u r e m e n t after E C M O was begun. T h e relationship between oxygen c o n s u m p t i o n and oxygen delivery in newborns remains largely u n k n o w n . ACKNOWLEDGMENT

Special thanks to Ronald E. Dechert for assistance with statistical analysis, John M. Toomasian for assistance with laboratory development in this project, and Elizabeth E. Noble and Cynthia L. Folsom for clerical assistance.

REFERENCES

1. Bartlett RH, Gazzaniga A, Toomasian J, et al: Extracorporeal membrane oxygenation (ECMO) in neonatal respiratory failure: 100 cases. Surgery 204:236-245, 1986 2. Bartlett RH, Roloff DW, Cornell RG, et al: Extracorporeal circulation in neonatal respiratory failure: A prospectiverandomized study. Pediatrics 76:479-487, 1985 3. German JC, Worcester C, Gazzaniga AB, et al: Technical aspects on the management of the meconium aspiration syndrome with extracorporeal circulation. J Pediatr Surg 15:378-383, 1980 4. Cox MA, Thrift MC: Nutrition, in Cloherty JP, Stark AR (eds): Manual of Neonatal Care. Boston, Little, Brown, 1980, pp 315-348 5. Cilley RE, Wesley JR, Zwischenberger JB, et al: Method of pulmonary and membrane lung gas exchange measurement during extracorporeal membrane oxygenation. Trans ASAIO 32:525-529, 1986 6. Wier JB deV: New methods for calculating metabolic rate with special reference to protein metabolism. J Physiol 109:1-9, 1949 7. Solca M, KolobowT, Huang H, et al: Management of antenatal preterm fetal lung in the prevention of respiratory distress syndrome in lambs. Biol Neonate 44:93-101, 1983 8. Lotze A, Short BL, Taylor GA: The use of lung compliance as a parameter for improvement in lung function in newborns with respiratory failure requiring ECMO. Crit Care Med 15:226-229, 1987 9. Levison H, Delivoria-Papadopoulos M, Swyer PR: Oxygen consumption in newly born infants with the respiratory distress syndrome. Biol Neonate 7:255-269, 1964 10. Hill JR, Rahimtulla KA: Heat balance and the metabolic rate of newborn babies in relation to environmental temperatures;

and the effect of age and weight on basal metabolic rate. J Physiol 180:239-265, 1965 11. Scopes JW, Ahmed I: Minimal rates of oxygen consumption in sick and premature newborn infants. Arch Dis Child 41:407-416, 1966 12. Senterre J, Karlberg P: Respiratory quotient and metabolic rate in normal full-term and small-for-date newborn infants. Acta Pediatr Scand 59:653-658, 1976 13. Brooke OG: Energy requirements and utilization of the low birthweight infant. Acta Pediatr Scand 296:67-70, 1982 (suppl) 14. Richardson P, Bose CL, Bucciarelli RL, et al: Oxygen consumption in infants with respiratory distress syndrome. Biol Neonate 46:53-56, 1984 15. Dechert R, Wesley J, Schafer L, et al: Comparison of oxygen consumption,carbon dioxide production and resting energy expenditure in premature and full-term infants. J Pediatr Surg 20:792-798, 1985 16. LaMond S, Wesley JR, Dechert R, et al: Correlation of cumulative caloric balance to weight gain in sick infants. J Parent Enteral Nutr 9:120, 1985 (abstr) 17. Bartlett RH, Gazzaniga AB: Extracorporeal circulation for cardiopulmonary failure, in Current Problems in Surgery. Chicago, Year Book Medical, 1978, pp 36-40 18. Anderson MN, Senning A: Studies in oxygen consumption during extracorporeal circulation with a pump-oxygenator. Ann Surg 148:59-65, 1958 19. Dawes GS: Oxygen consumption and temperature regulation in the newborn, in Fetal and Neonatal Physiology. Chicago, Year Book Medical, 1968, pp 191-197 20. Harned HS: Respiration and the respiratory system, in Stave U (ed): Perinatal Physiology.New York, Plenum, 1978, pp 54-57