- Email: [email protected]
Muscle relaxation in mechanically ventilated infants
FETAL AND MEDICINE
NEONATAL Richardr. Behrman,Editor
Muscle relaxation in mechanically ventilated infants We evaluated the effect of muscle paralysis on gas exchange and incidence of pneumothorax in 35 severely ill infants on mechanical ventilation. Pancuronium (0.1 mg/kg) was given repeatedly until spontaneous respirations ceased in infants with inadequate gas exchange with Flo._, > 0.60, or peak inspiratory pressure > 30 cm H.,O, or who were breathing out of phase with the respirator. 0 f 2 7 infants who had an alveolar-arterial oxygen gradient > 300 torr before paralysis, AaDo,_ improved by > 100 torr within one hour of paralysis in only two infants," it worsened in two infants within the same period. By six hours postparalysis, 12 of 2 7 infants had improved, five of whom had had a worsening AaDo._, before administration of pancuronium. Changes in oxygenation were unrelated to changes in arterial carbon dioxide tension in most infants. Peak transpulmonarv pressures after paralysis were lower than during spontaneous breathing, and may explain the low incidence of pneumothorax (3 of 35) during paralysis. Since those who improved could not be distingusihed by birth weight, gestational age, or diagnosis, pancuronium might be worthy o f trial in a mechanically ventilated infant with severe lung disease who is at risk for pneumothorax.
Ann R. Stark, M.D.,* Rebecca Bascom, B.A., and Ivan D. Frantz III, M.D., B o s t o n , M a s s .
SOME INFANTS with severe lung disease may require mechanical ventilation with high inspired oxygen concentrations and high airway pressures to attain adequate gas exchange. These patients are at risk for acute complications, such as pneumothorax, and for chronic lung disease. In addition, although the technique of intermittent mandatory ventilation appears useful during weaning from the respirator, it is not known whether spontaneous respiratory efforts during the acute phase of illness enhance or compromise gas exchange. To evaluate whether arterial oxygenation in mechanically ventilated infants improves following the use of pancuronium, a neuromuscular blocker, we have observed changes in gas exchange before and after paralysis. We also have examined the incidence of pneumothorax in this group of infants, and From the Department of Pediatrics, Harvard Medical School, Children "s Hospital Medical Center, and Boston Hospital for Women. Presented in part at the Society for Pediatric' Research, New York, 1978. Dr. Stark is an E. L. Trudeau Fellow of the American Lung Association and the recipient of a Young Investigator Award from the National Heart. Lung, and Blood Institute. *Reprint address: Joint Program in Neonatologv, 22l Longwood Ave., Boston, MA 02115.
0022-3476/79/300439+05500.50/0 9 1979 The C. V. Mosby Co.
recorded the effect of muscle paralysis on transpulmonary pressure. Abbreviations used AaDo: alveolar-arterial oxygen gradient Flo: fraction of oxygen in inspired gas
SUBJECT AND METHODS During the period September 1, 1976, through October 31, 1977, 35 infants, or about 15% of the infants mechan'ically ventilated in our intensive care units, received pancuronium to control respiration. All infants were on Baby Bird respirators. Pancuronium was administered intravenously in a dose of 0.1 mg/kg, repeated until spontaneous respirations ceased. Indications for paralysis were inadequate gas exchange with FIo~ greater than 0.60, peak inspiratory pressure greater than 30 cm H~O, or "fighting" the respirator, i.e., agitation or breathing out of phase with the respirator. Arterial blood gas values and pH, inspired oxygen concentration, respirator settings, heart rate, spontaneous respiratory rate, and blood pressure were recorded for each infant immediately preceding, within one hour after, and six hours following the administration of pancuronThe Journal of P E D I A T R I C S Vol. 94. No. 3, pp. 439-443
439
440
Stark, Bascom, and Fran t:
The Journal of Pediatrics March t979
I. Response to paralysis in infants with initial AaDo., > 300 torr
Table
Nme after para@s~ (hO
Medical Center) and the Research Advisory Committee (Boston Hospital for Women). RESULTS
No
Better
Worse
change
1
2
2
6
12
2
23 13
ium. All patients had indwelling umbilical or radial artery catheters at the time of the study. The alveolar-arterial oxygen gradient was calculated for each infant prior to and one and six hours following paralysis. A response to paralysis was taken as a change in the AaDo~ greater than 100 torr in either direction. Administration of blood products, bicarbonate, or other medications was also recorded. Of the patients who received pancuronium, 27 had an AaDo, greater than 300 torr. These infants included 15 with hyaline membrane disease, seven with pneumonia, four with persistent fetal circulation, and one with hydrops of unknown etiology. The other eight infants, all with hyaline membrane disease, were less ill and were excluded from the detailed analysis. The 27 infants had birth weights from 910 to 3,840 gm (median, 2,760 gm); 20 of the 27 infants weighed more than 2,000 gm. Mean gestational age was 36 ___ I(SD) weeks. All infants had pulmonary parenchymal changes present on radiographs. All except three patients were given pancuronium during the first 48 hours of life; half were paralyzed during the first day. Infants remained paralyzed up to 166 hours (median, 24 hours). Eight patients had additional measurements of respiration made before and/or after administration of pancuronium. Flow was measured with a pneumotachograph (Fteisch, No. 00), and a Statham PM 15E pressure transducer. The pneumotachograph was placed between the respirator and the endotracheal tube, adding about 1.5 ml of dead space. Tidal volume was determined by electronic integration of the flow signal. Airway pressure was measured at the endotracheal tube using a HewlettPackard No. 270 pressure transducer. Esophageal pressure was measured using a Statham PM6 pressure transducer and a hand-dipped latex balloon (balloon volume 0.2 to 0.3 ml) mounted on a 3.5 French feeding tube. placed through the mouth into the midportion of the esophagus. All measurements were made while the infants were supine and all signals were displayed on a Beckman polygraph. The protocol was approved by the Committee on Clinical Investigation (Children's Hospital
The response to paralysis of arterial oxygenation is summarized for the infants with initial AaDo=' > 300 torr in Table I. The one-hour interval includes the first blood gas obtained following paralysis. Few changes in inspired oxygen concentration or respirator settings were made during this period, and oxygenation in most infants remained unchanged. By six hours following paralysis, however, 12 infants had improved their oxygenation sufficiently to allow a reduction of inspired oxygen concentration. Five of these infants had had a worsening AaDo~ prior to paralysis. Improvement in oxygenation was unrelated to changes in ventilation in most infants. In the 12 patients who improved their oxygenation, initial values for Pco._, varied widely and individual responses were unpredictable: In four patients, Pco., decreased by 10 torr, in one Pco._, increased, and in the remainder, Pc% remained essentially unchanged. Similar changes were seen in the infants without an improvement in AaDo,. In those infants in whom minute ventilation was measured, changes in Pco., and ventilation were reciprocal as expected. Only three patients developed a pneumothorax while paralyzed. One additional patient had lung rupture after pancuronium was discontinued and four patients had pneumothorax early in their course, prior to paralysis. Few changes in arterial blood pressure were seen one hour after paralysis. Systolic blood pressure decreased by more than 5 torr in five patients and increased in two. Six hours following pancuronium administration systolic blood pressure decreased by more than 5 torr in nine patients and increased in one. Blood pressure fell to less than 45 torr in only two patients; several started with lower values. Blood pressure decreases in three infants may have been secondary to administration of tolazoline. One hour following paralysis, heart rate increases of greater than 10 beats/minute, to a rate of greater than 150 beats/minute, were observed in nine infants; heart rate decreases of greater than 10 beats/minute, to a rate lower than 150 beats/minute, were seen in five infants. At six hours, heart rate had increased by 10 beats/minute in eight patients and decreased in 11. No adverse side effects could be attributed to the drug, even in the infant treated almost seven days, and no surviving patient could not be weaned from the respirator. Excluding those with bronchopulmonary dysplasia, patients required mechanical ventilation for a median of 4.3 days (range 1 to 11). Four infants died. three of whom
Voh~me94 Number3 had initially improved, and three developed bronchopulmonary dysplasia requiring prolonged ventilatory assistance. One of these patients died at 4 months of.age at another hospital. The Figure depicts typical measurements of tidal volume, airway pressure, and esophageal pressure made before and during paralysis in a 1,320 gm infant with hyaline membrane disease. The infant's rapid spontaneous respiratory efforts, at twice the rate of the respirator, generate small volume changes compared to the much larger respirator breaths. However, these efforts are reflected by changes in transpulmonary pressure, the difference between airway pressure and pleural pressure. If a spontaneous inspiration occurs simultaneously with a respirator breath, peak transpulmonary pressure is greater than if no spontaneous effort were made. In this example, before paralysis, peak transpulmonary pressures are about 29 cm H20. After paralysis, however, peak transpulmonary pressures are only about 21 cm H..O. In the patients on whom detailed studies were carried out, average peak transpulmonary pressures decreased from 32.9 -4- 4.6 (SD) cm H.,O before paralysis to 22.9 • 1.8 cm H~O after paralysis (P < 0.025 by paired t test). Compliance of the lung and of the paralyzed chest wall is shown in Table II for the five infants in whom it could be calculated. Values shown are the means of five analyzed breaths. Values for chest wall compliance in the unparalyzed state varied widely (range, 3 to 23 ml/cm H~O), as would be expected because of changing respiratory muscle tone. In each patient, however, chest wall compliance was lower in the unparalyzed state. DISCUSSION In a group of mechanically ventilated infants with severe lung disease, about half had improved arterial oxygenation six hours following paralysis with pancuronium. Five of the 12 infants who improved had had a worsening course prior to administration of pancuronium, suggesting that improvement was a response to paralysis. That the improvement did not occur immediately and was often unassociated with a drop in Pco., was unexpected. However, the time course of response will be better understood by continuous blood gas monitoring, instead of intermittent sampling, as was done in this study. Although we had hoped to define a specific group of patients in whom paralysis would be beneficial, those patients who improved could not be distinguished by birth weight, gestational age, or diagnosis. We expected paralysis to improve oxygenation in infants who were struggling. However, few of our patients were actually resisting the respirator, and most were given pancuronium
Musclerelaxationin mechanicalventilation
BEFORE PARALYSIS
44 1
PARALYSIS
TIDAL VOLUME
AIRWAY PRESSURE cm 30
*8 1 Sec
_
Figure. Recording of tidal volume, airway pressure, and esophageal pressure before and after paralysis in a mechanically ventilated infant with hyaline membrane disease. Before paralysis, spontaneous breaths appear as negative deflections in esophageal pressure, as well as small deviations in the tidal volume and airway pressure tracings. Table ii. Compliance of the lung and paralyzed chest wall
~
~
Compliance (ml/cmH.,.O)t I Chest Lung wall
Subject 32 34 7 363) 373)
1,320 2,380 3,010 3,300 3.440
28 34 43 39 40
HMD HMD MAS PFC PFC
1.62 --- 0.04 2.23 _+ 0.04 1.10 -2_0.02 2,83 --- 0.10 0.5 _+ 0.08
55 59 30 37 26
-+ 11 4- 21 -+ 5 -+ 12 _+ 1
*HMD = Hyaline membrane disease: MAS = meconium asperation syndrome; PFC = persistentfetal circulation tMean -+ SD. ~:Studiedafter blood gas analysiscompleted. as a "last resort" for severely impaired gas exchange. Most patients did not change their oxygenation immediately after paralysis, and few worsened, while respirator settings and inspired oxygen concentration remained stable, suggesting that spontaneous respiratory efforts while on a respirator contributed little to ventilation in these severely ill patients. This was documented in the patients in whom flow measurements were made (Figure). As lung compliance improves, spontaneous respiratory efforts would be expected to contribute a greater portion of total ventilation. Thus paralysis of a patient with less serious disease or who is recovering might result in deteriorating blood gas values unless compensating respirator adjustments are made.
44 2
Stark, Bascom, and Frantz
No distinguishing factor could be found to predict which infants would respond favorably to paralysis. When infants were grouped according to birth weight, gestational age, or diagnosis, about half of the infants in each category improved. Improvement also appeared unrelated to changes in respirator settings. Changes expected to improve oxygenation were made in only half the infants who improved and in nearly all the others. Blood products, bicarbonate, or tolazoline were often administered to infants who became worse or did not change, but were rarely used in those who improved. Radiographic studies in normal supine adults show different patterns of diaphragm displacement during spontaneous breathing and during controlled ventilation following paralysis. During spontaneous breathing in the supine position, diaphragm movement is greatest in the inferior position, so that ventilation occurs preferentially in dependent areas of the lung. During controlled ventilation, however, most diaphragm displacement occurs in the superior portion, so that nondependent areas are ventilated preferentially, theoretically worsening ventilation-perfusion relations and impairing gas exchange.' Furthermore, mechanical ventilation in anesthetized and paralyzed normal adults significantly changes the distribution of ventilation from that during spontaneous breathing in a way that might be expected to increase the mismatch between ventilation and perfusion. -~ Similar measurements have not been made in infants, but the findings in adults suggest that the use of muscle relaxants to control ventilation in patients with severe respiratory failure might be detrimental. However, we did not observe worsening of arterial oxygenation in our patients. Changes in arterial carbon dioxide tensions following paralysis were variable, and improved oxygenation was unrelated to changes in ventilation. Thus, improvement, when it occurs, may be related to an improvement in ventilation-perfusion relationship resulting in decreased right-to-left shunting. In addition, decreased oxygen consumption during paralysis might contribute to better arterial oxygenation. Despite the severity of their lung disease and the relatively high inspiratory and end-expiratory pressures required, only three of our patients developed a pneumothorax while paralyzed. Although no precisely appropriate comparison group is available, our experience and that described in the literature would indicate a 20 to 30% incidence of pneumothorax in mechanically ventilated infants with hyaline membrane disease. :~-:'The incidence increases when other diagnoses and all types of air leak are considered. ''-~ Although four patients in our series developed pneumothorax prior to paralysis and thus may have been past the period for maximal risk, the lower
The Journal of Pediatrics March 1979
incidence during paralysis may represent an improvement. This may be explained by our measurements of transpulmonary pressure, which show that peak pressures are lower following paralysis. In the nonparalyzed state, each spontaneous inspiratory effort is associated with a negative esophageal and pleural pressure. When the spontaneous inspiratory efforts are superimposed on respirator breaths, their effects on transpulmonary pressure are additive. After paralysis, esophageal pressure is affected by the lung volume of the infant, the passive relaxed characteristics of the chest wall, and the volume change produced by each respirator breath. Since the chest walt is very compliant, the esophageal pressure changes are very small. Although the extent to which the alveoli are exposed to the increased transpulmonary pressure is unknown, our data suggest that in mechanically ventilated infants, the pressure required to rupture alveoli is more readily attained during spontaneous breathing efforts. We have included our data on chest wall compliance because, to our knowledge, no similar data for human infants exist in the literature. The studies available on paralyzed infants report total respiratory system compliance? -'' Although our data must be considered preliminary, since values are available from only five babies and lung volume was not measured, it appears that chest wall compliance was greater in the less mature infants. Pancuronium is a competitive blocker to acetylcholine at the neuromuscular junction of skeletal muscle. Although tachycardia can occur following its use, 1~pancuronium, unlike d-tubocurarine, has minimal effects at autonomic ganglia and no histamine-releasing activity, and therefore no appreciable effect on blood pressure. 1:' Although we did not observe significant cardiovascular changes in most patients, there are theoretical reasons why those might have occurred. Since changes in intrathoracic pressure related to spontaneous breathing efforts influence venous return and thus cardiac output, the positive esophageal pressures measured in the paralyzed state may be associated with decreased venous return, reflected in lowered blood pressure in some individuals. In addition, the severe illness in these infants, associated with hypoxemia and acidosis, may contribute to hypotension. Pancuronium has been used safely as a muscle relaxant with neonatal anesthesia TM and in adults. '~' ,5 Its effects are readily reversible with administration of an anticholinesterase and atropine. The major risk to its use in controlled ventilation is accidental disconnection from the respirator. This is of concern even in the nonparalyzed patient with respiratory failure, and appropriate monitoring to avoid this is obviously mandatory. Because of the improvement
Volume 94 Number 3
Muscle relaxation in mechanical ventilation
in oxygenation seen in some infants and the apparent decrease in incidence of pneumothorax, a trial of pancuronium may be indicated in infants with severe respiratory failure during the period of m a x i m u m risk for alveolar rupture. The authors are grateful to the nurses and the house staff of the Neonatal Intensive Care Unit at Children's Hospital Medical Center and the Special Care Nursery at Boston Hospital for Women for their help in identifying patients for our study. We also thank Dr. Mary Ellen Wohl for her assistance with our analysis of respiratory mechanics, and Drs. Mary Ellen Avery and H. William Taeusch, Jr., for their critical review of the manuscript. REFERENCES
1. Froese AB, and Bryan AC: Effects of anesthesia and paralysis on diaphragmatic mechanics in man, Anaesthesiology 41:242, 1974. 2. Rehder K, Sessler AD, and Rodarte JR: Regional intrapulmonary gas distribution in awake and anesthetized-paralyzed man, J Appl Physiol 42:391, 1977. 3. Ogata ES, Gregory GA, Kitterman JA, Phibbs RH, and Tooley WH: Pneumothorax in the respiratory distress syndrome: incidence and effect on vital signs, blood gases, and pH, Pediatrics 58:177, 1976. 4. Boros SJ, and Reynolds JW: Hyaline membrane disease treated with early nasal end-expiratory pressure: one year's experience, Pediatrics 56:218, 1975.
443
5. Madansky D: Personal communication. 6. Berg TJ, Pagtakhan RD, Reed MH, Langston C, and Chernick V: Bronchopulmonary dysplasia and lung rupture in hyaline membrane disease: influence of continuous distending pressure, Pediatrics 55:51, 1975. 7. Fox WW, Berman LS, Downes JJ, and Peckham GJ: The therapeutic application of end-expiratory pressure in the meconium aspiration syndrome, Pediatrics 56:214, 1975. 8. Vidyasagar D, Yeh TF, Harris V, and Pildes RS: Assisted ventilation in infants with meconium aspiration syndrome, Pediatrics 56:208, 1975. 9. Nightingale DA, and Richards CC: Volume-pressure relations of the respiratory system of curarized infants, Anesthesiology 26:710, 1965. 10. Richards CC, and Bachman L: Lung and chest wall compliance of apneic paralyzed infants, J Clin Invest 40:273, 1961. 11. Smythe PM: Studies in neonatal tetanus and on pulmonary compliance of the totally relaxed infant, Br Med J 1:565~ 1963. 12. Kelman GR, and Kennedy BR: Cardiovascular effects of pancuronium in man, Br J Anaesth 43:335, 1971. 13. Roizen MF, and Feeley TW: Pancuronium bromide, Ann Intern Med 88:64, 1978. 14. Bennett El, Ramamurthy S, Dalai FY, and Salem MR: Pancuronium and the neonate, Br J Anaesth 47:75, 1975. 15. Light RW, Bengforr JL, and George RB: The adult respiratory distress syndrome and pancuronium bromide, Anesth Analg 54:219, 1975.
Information for authors
Most of the provisions of the Copyright Act of 1976 became effective on January 1, 1978. Therefore, all manuscripts must be accompanied by the following written statement, signed by one author: "The undersigned author transfers all copyright ownership of the manuscript (title of article) to The C. V. Mosby Company in the event the wor k is published. The undersigned author warrants that the article is original, is not under consideration by another journal, and has not been previously published. I sign for and accept responsibility for releasing this material on behalf of any and all co-authors." Authors will be consulted, when possible, regarding republication of their material.
NEONATAL Richardr. Behrman,Editor
Muscle relaxation in mechanically ventilated infants We evaluated the effect of muscle paralysis on gas exchange and incidence of pneumothorax in 35 severely ill infants on mechanical ventilation. Pancuronium (0.1 mg/kg) was given repeatedly until spontaneous respirations ceased in infants with inadequate gas exchange with Flo._, > 0.60, or peak inspiratory pressure > 30 cm H.,O, or who were breathing out of phase with the respirator. 0 f 2 7 infants who had an alveolar-arterial oxygen gradient > 300 torr before paralysis, AaDo,_ improved by > 100 torr within one hour of paralysis in only two infants," it worsened in two infants within the same period. By six hours postparalysis, 12 of 2 7 infants had improved, five of whom had had a worsening AaDo._, before administration of pancuronium. Changes in oxygenation were unrelated to changes in arterial carbon dioxide tension in most infants. Peak transpulmonarv pressures after paralysis were lower than during spontaneous breathing, and may explain the low incidence of pneumothorax (3 of 35) during paralysis. Since those who improved could not be distingusihed by birth weight, gestational age, or diagnosis, pancuronium might be worthy o f trial in a mechanically ventilated infant with severe lung disease who is at risk for pneumothorax.
Ann R. Stark, M.D.,* Rebecca Bascom, B.A., and Ivan D. Frantz III, M.D., B o s t o n , M a s s .
SOME INFANTS with severe lung disease may require mechanical ventilation with high inspired oxygen concentrations and high airway pressures to attain adequate gas exchange. These patients are at risk for acute complications, such as pneumothorax, and for chronic lung disease. In addition, although the technique of intermittent mandatory ventilation appears useful during weaning from the respirator, it is not known whether spontaneous respiratory efforts during the acute phase of illness enhance or compromise gas exchange. To evaluate whether arterial oxygenation in mechanically ventilated infants improves following the use of pancuronium, a neuromuscular blocker, we have observed changes in gas exchange before and after paralysis. We also have examined the incidence of pneumothorax in this group of infants, and From the Department of Pediatrics, Harvard Medical School, Children "s Hospital Medical Center, and Boston Hospital for Women. Presented in part at the Society for Pediatric' Research, New York, 1978. Dr. Stark is an E. L. Trudeau Fellow of the American Lung Association and the recipient of a Young Investigator Award from the National Heart. Lung, and Blood Institute. *Reprint address: Joint Program in Neonatologv, 22l Longwood Ave., Boston, MA 02115.
0022-3476/79/300439+05500.50/0 9 1979 The C. V. Mosby Co.
recorded the effect of muscle paralysis on transpulmonary pressure. Abbreviations used AaDo: alveolar-arterial oxygen gradient Flo: fraction of oxygen in inspired gas
SUBJECT AND METHODS During the period September 1, 1976, through October 31, 1977, 35 infants, or about 15% of the infants mechan'ically ventilated in our intensive care units, received pancuronium to control respiration. All infants were on Baby Bird respirators. Pancuronium was administered intravenously in a dose of 0.1 mg/kg, repeated until spontaneous respirations ceased. Indications for paralysis were inadequate gas exchange with FIo~ greater than 0.60, peak inspiratory pressure greater than 30 cm H~O, or "fighting" the respirator, i.e., agitation or breathing out of phase with the respirator. Arterial blood gas values and pH, inspired oxygen concentration, respirator settings, heart rate, spontaneous respiratory rate, and blood pressure were recorded for each infant immediately preceding, within one hour after, and six hours following the administration of pancuronThe Journal of P E D I A T R I C S Vol. 94. No. 3, pp. 439-443
439
440
Stark, Bascom, and Fran t:
The Journal of Pediatrics March t979
I. Response to paralysis in infants with initial AaDo., > 300 torr
Table
Nme after para@s~ (hO
Medical Center) and the Research Advisory Committee (Boston Hospital for Women). RESULTS
No
Better
Worse
change
1
2
2
6
12
2
23 13
ium. All patients had indwelling umbilical or radial artery catheters at the time of the study. The alveolar-arterial oxygen gradient was calculated for each infant prior to and one and six hours following paralysis. A response to paralysis was taken as a change in the AaDo~ greater than 100 torr in either direction. Administration of blood products, bicarbonate, or other medications was also recorded. Of the patients who received pancuronium, 27 had an AaDo, greater than 300 torr. These infants included 15 with hyaline membrane disease, seven with pneumonia, four with persistent fetal circulation, and one with hydrops of unknown etiology. The other eight infants, all with hyaline membrane disease, were less ill and were excluded from the detailed analysis. The 27 infants had birth weights from 910 to 3,840 gm (median, 2,760 gm); 20 of the 27 infants weighed more than 2,000 gm. Mean gestational age was 36 ___ I(SD) weeks. All infants had pulmonary parenchymal changes present on radiographs. All except three patients were given pancuronium during the first 48 hours of life; half were paralyzed during the first day. Infants remained paralyzed up to 166 hours (median, 24 hours). Eight patients had additional measurements of respiration made before and/or after administration of pancuronium. Flow was measured with a pneumotachograph (Fteisch, No. 00), and a Statham PM 15E pressure transducer. The pneumotachograph was placed between the respirator and the endotracheal tube, adding about 1.5 ml of dead space. Tidal volume was determined by electronic integration of the flow signal. Airway pressure was measured at the endotracheal tube using a HewlettPackard No. 270 pressure transducer. Esophageal pressure was measured using a Statham PM6 pressure transducer and a hand-dipped latex balloon (balloon volume 0.2 to 0.3 ml) mounted on a 3.5 French feeding tube. placed through the mouth into the midportion of the esophagus. All measurements were made while the infants were supine and all signals were displayed on a Beckman polygraph. The protocol was approved by the Committee on Clinical Investigation (Children's Hospital
The response to paralysis of arterial oxygenation is summarized for the infants with initial AaDo=' > 300 torr in Table I. The one-hour interval includes the first blood gas obtained following paralysis. Few changes in inspired oxygen concentration or respirator settings were made during this period, and oxygenation in most infants remained unchanged. By six hours following paralysis, however, 12 infants had improved their oxygenation sufficiently to allow a reduction of inspired oxygen concentration. Five of these infants had had a worsening AaDo~ prior to paralysis. Improvement in oxygenation was unrelated to changes in ventilation in most infants. In the 12 patients who improved their oxygenation, initial values for Pco._, varied widely and individual responses were unpredictable: In four patients, Pco., decreased by 10 torr, in one Pco._, increased, and in the remainder, Pc% remained essentially unchanged. Similar changes were seen in the infants without an improvement in AaDo,. In those infants in whom minute ventilation was measured, changes in Pco., and ventilation were reciprocal as expected. Only three patients developed a pneumothorax while paralyzed. One additional patient had lung rupture after pancuronium was discontinued and four patients had pneumothorax early in their course, prior to paralysis. Few changes in arterial blood pressure were seen one hour after paralysis. Systolic blood pressure decreased by more than 5 torr in five patients and increased in two. Six hours following pancuronium administration systolic blood pressure decreased by more than 5 torr in nine patients and increased in one. Blood pressure fell to less than 45 torr in only two patients; several started with lower values. Blood pressure decreases in three infants may have been secondary to administration of tolazoline. One hour following paralysis, heart rate increases of greater than 10 beats/minute, to a rate of greater than 150 beats/minute, were observed in nine infants; heart rate decreases of greater than 10 beats/minute, to a rate lower than 150 beats/minute, were seen in five infants. At six hours, heart rate had increased by 10 beats/minute in eight patients and decreased in 11. No adverse side effects could be attributed to the drug, even in the infant treated almost seven days, and no surviving patient could not be weaned from the respirator. Excluding those with bronchopulmonary dysplasia, patients required mechanical ventilation for a median of 4.3 days (range 1 to 11). Four infants died. three of whom
Voh~me94 Number3 had initially improved, and three developed bronchopulmonary dysplasia requiring prolonged ventilatory assistance. One of these patients died at 4 months of.age at another hospital. The Figure depicts typical measurements of tidal volume, airway pressure, and esophageal pressure made before and during paralysis in a 1,320 gm infant with hyaline membrane disease. The infant's rapid spontaneous respiratory efforts, at twice the rate of the respirator, generate small volume changes compared to the much larger respirator breaths. However, these efforts are reflected by changes in transpulmonary pressure, the difference between airway pressure and pleural pressure. If a spontaneous inspiration occurs simultaneously with a respirator breath, peak transpulmonary pressure is greater than if no spontaneous effort were made. In this example, before paralysis, peak transpulmonary pressures are about 29 cm H20. After paralysis, however, peak transpulmonary pressures are only about 21 cm H..O. In the patients on whom detailed studies were carried out, average peak transpulmonary pressures decreased from 32.9 -4- 4.6 (SD) cm H.,O before paralysis to 22.9 • 1.8 cm H~O after paralysis (P < 0.025 by paired t test). Compliance of the lung and of the paralyzed chest wall is shown in Table II for the five infants in whom it could be calculated. Values shown are the means of five analyzed breaths. Values for chest wall compliance in the unparalyzed state varied widely (range, 3 to 23 ml/cm H~O), as would be expected because of changing respiratory muscle tone. In each patient, however, chest wall compliance was lower in the unparalyzed state. DISCUSSION In a group of mechanically ventilated infants with severe lung disease, about half had improved arterial oxygenation six hours following paralysis with pancuronium. Five of the 12 infants who improved had had a worsening course prior to administration of pancuronium, suggesting that improvement was a response to paralysis. That the improvement did not occur immediately and was often unassociated with a drop in Pco., was unexpected. However, the time course of response will be better understood by continuous blood gas monitoring, instead of intermittent sampling, as was done in this study. Although we had hoped to define a specific group of patients in whom paralysis would be beneficial, those patients who improved could not be distinguished by birth weight, gestational age, or diagnosis. We expected paralysis to improve oxygenation in infants who were struggling. However, few of our patients were actually resisting the respirator, and most were given pancuronium
Musclerelaxationin mechanicalventilation
BEFORE PARALYSIS
44 1
PARALYSIS
TIDAL VOLUME
AIRWAY PRESSURE cm 30
*8 1 Sec
_
Figure. Recording of tidal volume, airway pressure, and esophageal pressure before and after paralysis in a mechanically ventilated infant with hyaline membrane disease. Before paralysis, spontaneous breaths appear as negative deflections in esophageal pressure, as well as small deviations in the tidal volume and airway pressure tracings. Table ii. Compliance of the lung and paralyzed chest wall
~
~
Compliance (ml/cmH.,.O)t I Chest Lung wall
Subject 32 34 7 363) 373)
1,320 2,380 3,010 3,300 3.440
28 34 43 39 40
HMD HMD MAS PFC PFC
1.62 --- 0.04 2.23 _+ 0.04 1.10 -2_0.02 2,83 --- 0.10 0.5 _+ 0.08
55 59 30 37 26
-+ 11 4- 21 -+ 5 -+ 12 _+ 1
*HMD = Hyaline membrane disease: MAS = meconium asperation syndrome; PFC = persistentfetal circulation tMean -+ SD. ~:Studiedafter blood gas analysiscompleted. as a "last resort" for severely impaired gas exchange. Most patients did not change their oxygenation immediately after paralysis, and few worsened, while respirator settings and inspired oxygen concentration remained stable, suggesting that spontaneous respiratory efforts while on a respirator contributed little to ventilation in these severely ill patients. This was documented in the patients in whom flow measurements were made (Figure). As lung compliance improves, spontaneous respiratory efforts would be expected to contribute a greater portion of total ventilation. Thus paralysis of a patient with less serious disease or who is recovering might result in deteriorating blood gas values unless compensating respirator adjustments are made.
44 2
Stark, Bascom, and Frantz
No distinguishing factor could be found to predict which infants would respond favorably to paralysis. When infants were grouped according to birth weight, gestational age, or diagnosis, about half of the infants in each category improved. Improvement also appeared unrelated to changes in respirator settings. Changes expected to improve oxygenation were made in only half the infants who improved and in nearly all the others. Blood products, bicarbonate, or tolazoline were often administered to infants who became worse or did not change, but were rarely used in those who improved. Radiographic studies in normal supine adults show different patterns of diaphragm displacement during spontaneous breathing and during controlled ventilation following paralysis. During spontaneous breathing in the supine position, diaphragm movement is greatest in the inferior position, so that ventilation occurs preferentially in dependent areas of the lung. During controlled ventilation, however, most diaphragm displacement occurs in the superior portion, so that nondependent areas are ventilated preferentially, theoretically worsening ventilation-perfusion relations and impairing gas exchange.' Furthermore, mechanical ventilation in anesthetized and paralyzed normal adults significantly changes the distribution of ventilation from that during spontaneous breathing in a way that might be expected to increase the mismatch between ventilation and perfusion. -~ Similar measurements have not been made in infants, but the findings in adults suggest that the use of muscle relaxants to control ventilation in patients with severe respiratory failure might be detrimental. However, we did not observe worsening of arterial oxygenation in our patients. Changes in arterial carbon dioxide tensions following paralysis were variable, and improved oxygenation was unrelated to changes in ventilation. Thus, improvement, when it occurs, may be related to an improvement in ventilation-perfusion relationship resulting in decreased right-to-left shunting. In addition, decreased oxygen consumption during paralysis might contribute to better arterial oxygenation. Despite the severity of their lung disease and the relatively high inspiratory and end-expiratory pressures required, only three of our patients developed a pneumothorax while paralyzed. Although no precisely appropriate comparison group is available, our experience and that described in the literature would indicate a 20 to 30% incidence of pneumothorax in mechanically ventilated infants with hyaline membrane disease. :~-:'The incidence increases when other diagnoses and all types of air leak are considered. ''-~ Although four patients in our series developed pneumothorax prior to paralysis and thus may have been past the period for maximal risk, the lower
The Journal of Pediatrics March 1979
incidence during paralysis may represent an improvement. This may be explained by our measurements of transpulmonary pressure, which show that peak pressures are lower following paralysis. In the nonparalyzed state, each spontaneous inspiratory effort is associated with a negative esophageal and pleural pressure. When the spontaneous inspiratory efforts are superimposed on respirator breaths, their effects on transpulmonary pressure are additive. After paralysis, esophageal pressure is affected by the lung volume of the infant, the passive relaxed characteristics of the chest wall, and the volume change produced by each respirator breath. Since the chest walt is very compliant, the esophageal pressure changes are very small. Although the extent to which the alveoli are exposed to the increased transpulmonary pressure is unknown, our data suggest that in mechanically ventilated infants, the pressure required to rupture alveoli is more readily attained during spontaneous breathing efforts. We have included our data on chest wall compliance because, to our knowledge, no similar data for human infants exist in the literature. The studies available on paralyzed infants report total respiratory system compliance? -'' Although our data must be considered preliminary, since values are available from only five babies and lung volume was not measured, it appears that chest wall compliance was greater in the less mature infants. Pancuronium is a competitive blocker to acetylcholine at the neuromuscular junction of skeletal muscle. Although tachycardia can occur following its use, 1~pancuronium, unlike d-tubocurarine, has minimal effects at autonomic ganglia and no histamine-releasing activity, and therefore no appreciable effect on blood pressure. 1:' Although we did not observe significant cardiovascular changes in most patients, there are theoretical reasons why those might have occurred. Since changes in intrathoracic pressure related to spontaneous breathing efforts influence venous return and thus cardiac output, the positive esophageal pressures measured in the paralyzed state may be associated with decreased venous return, reflected in lowered blood pressure in some individuals. In addition, the severe illness in these infants, associated with hypoxemia and acidosis, may contribute to hypotension. Pancuronium has been used safely as a muscle relaxant with neonatal anesthesia TM and in adults. '~' ,5 Its effects are readily reversible with administration of an anticholinesterase and atropine. The major risk to its use in controlled ventilation is accidental disconnection from the respirator. This is of concern even in the nonparalyzed patient with respiratory failure, and appropriate monitoring to avoid this is obviously mandatory. Because of the improvement
Volume 94 Number 3
Muscle relaxation in mechanical ventilation
in oxygenation seen in some infants and the apparent decrease in incidence of pneumothorax, a trial of pancuronium may be indicated in infants with severe respiratory failure during the period of m a x i m u m risk for alveolar rupture. The authors are grateful to the nurses and the house staff of the Neonatal Intensive Care Unit at Children's Hospital Medical Center and the Special Care Nursery at Boston Hospital for Women for their help in identifying patients for our study. We also thank Dr. Mary Ellen Wohl for her assistance with our analysis of respiratory mechanics, and Drs. Mary Ellen Avery and H. William Taeusch, Jr., for their critical review of the manuscript. REFERENCES
1. Froese AB, and Bryan AC: Effects of anesthesia and paralysis on diaphragmatic mechanics in man, Anaesthesiology 41:242, 1974. 2. Rehder K, Sessler AD, and Rodarte JR: Regional intrapulmonary gas distribution in awake and anesthetized-paralyzed man, J Appl Physiol 42:391, 1977. 3. Ogata ES, Gregory GA, Kitterman JA, Phibbs RH, and Tooley WH: Pneumothorax in the respiratory distress syndrome: incidence and effect on vital signs, blood gases, and pH, Pediatrics 58:177, 1976. 4. Boros SJ, and Reynolds JW: Hyaline membrane disease treated with early nasal end-expiratory pressure: one year's experience, Pediatrics 56:218, 1975.
443
5. Madansky D: Personal communication. 6. Berg TJ, Pagtakhan RD, Reed MH, Langston C, and Chernick V: Bronchopulmonary dysplasia and lung rupture in hyaline membrane disease: influence of continuous distending pressure, Pediatrics 55:51, 1975. 7. Fox WW, Berman LS, Downes JJ, and Peckham GJ: The therapeutic application of end-expiratory pressure in the meconium aspiration syndrome, Pediatrics 56:214, 1975. 8. Vidyasagar D, Yeh TF, Harris V, and Pildes RS: Assisted ventilation in infants with meconium aspiration syndrome, Pediatrics 56:208, 1975. 9. Nightingale DA, and Richards CC: Volume-pressure relations of the respiratory system of curarized infants, Anesthesiology 26:710, 1965. 10. Richards CC, and Bachman L: Lung and chest wall compliance of apneic paralyzed infants, J Clin Invest 40:273, 1961. 11. Smythe PM: Studies in neonatal tetanus and on pulmonary compliance of the totally relaxed infant, Br Med J 1:565~ 1963. 12. Kelman GR, and Kennedy BR: Cardiovascular effects of pancuronium in man, Br J Anaesth 43:335, 1971. 13. Roizen MF, and Feeley TW: Pancuronium bromide, Ann Intern Med 88:64, 1978. 14. Bennett El, Ramamurthy S, Dalai FY, and Salem MR: Pancuronium and the neonate, Br J Anaesth 47:75, 1975. 15. Light RW, Bengforr JL, and George RB: The adult respiratory distress syndrome and pancuronium bromide, Anesth Analg 54:219, 1975.
Information for authors
Most of the provisions of the Copyright Act of 1976 became effective on January 1, 1978. Therefore, all manuscripts must be accompanied by the following written statement, signed by one author: "The undersigned author transfers all copyright ownership of the manuscript (title of article) to The C. V. Mosby Company in the event the wor k is published. The undersigned author warrants that the article is original, is not under consideration by another journal, and has not been previously published. I sign for and accept responsibility for releasing this material on behalf of any and all co-authors." Authors will be consulted, when possible, regarding republication of their material.