- Email: [email protected]
A clinical-pathological study of nonsurvivors of newborn ECMO
A Clinical-Pathological Study of Nonsurvivors of Newborn ECMO By Thomas R. Weber, Susan H. Westfall, Cirilo Sotelo, Carole A. Vogler, and Tom Tracy, Jr St Louis, Missouri 9 Extracorporeal membrane oxygenation (ECMO) is an important means of supporting newborns with respiratory failure. While short- and long-term follow-up of ECMO survivors has been thoroughly addressed, there is no systematic study of nonsurvivors. Nineteen nonsurvivors of newborn ECMO with autopsy results are divided into two groups: group 1 : 1 2 patients who had intracranial lesions as the primary cause of death (hemorrhage 8, encephalomalacia 2, infarct 2); and group 2 : 7 patients with nonintracranial primary causes of death. Patients in group 1 were significantly more acidotic, hypotensive, and smaller in age and birth weight pre-ECMO. Among group 2 patients, two with diaphragmatic hernia died of primary pulmonary disease (diffuse alveolar damage, pulmonary hypoplasia and necrosis, bronchopneumonia). One of 2 patients with persistent fetal circulation (PFC) was treated with massive doses of tolazoline and suffered fatal gastrointestinal hemorrhage and ischemic necrosis of heart, spleen, testes, and adrenals. The other PFC patient had severe pulmonary interstitial fibrosis. Two patients with meconium aspiration and a patient with streptococcal sepsis had diffuse pulmonary damage and multiple organ failure (renal medullary necrosis, and infarcts of adrenal, spleen, liver). In this series, intracranial pathology was the most common cause of death in ECMO patients, related to gestational age, acidosis, hypoxia, and size, but probably unrelated to carotid ligation. Copyright 9 1993 by W.B. Saunders Company INDEX WORDS: Extracorporeal membrane oxygenation (ECMO); respiratory failure, newborn.
T R A C O R P O R E A L membrane oxygenation E X(ECMO) is an important new means of supporting newborns with respiratory failure. Although the majority of newborns requiring mechanical ventilation for respiratory failure are successfully treated, and recover, a number will prove to be unresponsive to conventional therapy and will require ECMO. 1 The complications of ECMO, including death secondary to intracranial hemorrhage, have received mention in previous studiesY However, there have been no comprehensive studies of ECMO nonsurvivors. The present clinical-pathological study of ECMO nonsurvivors was undertaken to determine the prevalent pathological findings and to better define the indications and contraindications for ECMO in the newborn.
achieved just prior to cannulation and continued by continuous drip, to keep the activated clotting time at 200 to 250 seconds. Blood drained from the venous cannula into a small reservoir, from which it was pumped through a membrane oxygenator, then through a heat exchanger for rewarming prior to return to the patient through the arterial cannula. As pulmonary function improved, the flow rate through the ECMO apparatus was decreased, until complete cessation of ECMO could be achieved. Nonsurvival with ECMO therapy was related to several factors. If a baby developed progressive intracerebral hemorrhage, as detected by daily head ultrasound, then ECMO was rapidly weaned or abruptly discontinued, the latter utilized if brain death was imminent. Several patients had recovery of adequate pulmonary function, but died from neurological or multisystem failure after removal from ECMO. Finally, several newborns failed to improve from their primary pulmonary disorder and succumbed after prolonged ECMO runs. This duration of ECMO for these nonsurviving patients ranged from 3 to 16 days (mean, 8 days). Of the 21 nonsurvivors, 19 had an autopsy performed (Table 1). The hospital charts, ECMO flow data sheets including all hemodynamic parameters, and also the age ECMO was initiated, time on ECMO, pre-ECMO oxygenation, level of heparinization, and autopsy findings were reviewed to determine cause of death, prevalent pathological findings, and the relationship of clinical parameters to autopsy findings.
RESULTS
Based on clinical and autopsy findings, the 19 patients were divided into two groups: group 1:12 patients who had intracranial lesions as the primary cause of death; and group 2:7 patients with nonintracranial primary causes of death, Eight of the 12 patients in group 1 had intracranial hemorrhage: six bilateral and two right-sided only. Two other infants had diffuse encephalomalacia, while the final two had massive infarcts, one rightsided and one left-sided. The primary diagnoses of these patients were persistent fetal circulation (7), sepsis (2), congenital diaphragmatic hernia (2), and meconium aspiration (1). None of these patients had pre-ECMO seizures, while one had pre-ECMO renal failure. The group 1 mean pre-ECMO arterial blood pH was 7.27, mean arterial blood pressure 43 mm Hg,
In a 3-year period, 78 newborns were treated with ECMO. Prior to institution of ECMO, cranial ultrasonography performed on each patient was negative for the presence of intracranial hemorrhage. ECMO was established by cannulating the right internal jugular vein and right common carotid artery using local anesthesia, as previously reported. 5 Anticoagulation with heparin was
From the Division of Pediatric Surgery, Department of Surgery, and the Department of Pathology, Cardinal Glennon Children's Hospital and St Louis University School of Medicine, St Louis, MO. Date accepted: October 8, 1991. Address reprint requests to Thomas R. Weber, MD, Cardinal Glennon Children's Hospital, 1465 S Grand Blvd, St Louis, 340 631O4. Copyright 9 1993 by W.B. Saunders Company 0022-3468/93/2802-0003503.00/0
Journal of Pediatric Surgery, Vo128, No 2 (February), 1993: pp 135-137
135
MATERIALS AND METHODS
WEBER ET AL
138 Table 1. Characteristics of 19 Non-survivors of Newborn ECMO Diagnosis
No.
Gestational Age Range (wk)
Persistent fetal circulation Congenital diaphragmatic hernia Sepsis Meconium aspiration Hydrops
7 4 3 3 2
35-42 34-40 34-40 39-40 34-40
Weight Range (kg) 2.3-3.6 2.2-3.5 1.9-3.6 2.8-3.3 2.0-2.5
had autopsy evidence of small white matter cerebral infarcts, without neurological consequence. Comparison of group I to group II patients showed no significant differences with regard to the age ECMO was initiated, time on ECMO, pre-ECMO oxygenation, and level of heparinization. DISCUSSION
gestational age 37 weeks, and birth weight 2,695 g (Table 2). These patients were treated with ECMO for a mean of 6 days. Associated but nonlethal factors in group 1 patients include cardiac irritability with arrhythmia (2), sepsis (1), elevated liver enzymes (1), and elevated creatinine (l). In contrast, the seven patients in group 2 had nonintracranial lesions as the primary cause of death, based on autopsy results. Three patients died of primary pulmonary disease: severe pulmonary hypoplasia in two patients with congenital diaphragmatic hernia, both of whom also had diffuse alveolar damage and necrosis, and a third patient, with persistent fetal circulation, who had diffuse alveolar damage, interstitial fibrosis, and necrotizing pneumonitis at autopsy. One other patient died of massive hemorrhage and shock. This latter patient with persistent fetal circulation treated with high doses of tolazoline pre-ECMO, had evidence of diffuse upper gastrointestinal and pancreatic hemorrhage, with ischemic infarcts of the heart, spleen, testicles, and adrenal secondary to shock. Three patients in group 2 died from multiple organ failure, with infarcts and hemorrhages in more than one organ system, including lungs, liver, adrenal glands, kidneys, and gastrointestinal tract. One of the seven patients in group 2 had pre-ECMO seizures, while three had pre-ECMO renal failure. The mean group 2 arterial blood pH was 7.39, mean arterial blood pressure 66 mm Hg, gestational age 39 weeks, and birth weight 3,209 g. These parameters were significantly different from those in group 1 (Table 2). Group 2 patients were treated with ECMO an average of 11 days. Two of these patients Table 2. Pre-ECMO Characteristics of Nonsurvivors of Newborn ECMO With an Intracranial (Group 1) or Nonintracranial (Group 2) Cause of Death Group
pH
Arterial Blood Pressure (ram Hg)
Gestational Age (wk)
Birth Weight (g)
1 2
7.27 _+ 0.02 7.39 +_ 0.10"
43 -+ 2.2 66 -+ 4.5t
37.0 --+ 2.0 39.8 --+ 1.6
2,695 --+ 56.2 3,209 -+ 35.1t
NOTE. Data given as mean _+ SEM. *P < ,001 by Student's t test. t P < .05 by Student's t test.
ECMO for respiratory failure was first successfully performed by Hill et al in an adult in 1972, 6 and extended to newborn respiratory failure several years later by Bartlett et al. 2 If instituted prior to irreversible pulmonary damage, the lungs can regain normal function within a reasonable amount of time on ECMO. Because of risks of heparinization and other complications of ECMO, the procedure is reserved for those who fail conventional therapy. Intracranial hemorrhage, the most common complication o f ECMO in this series as well as others, frequently results in either death or severe neurological impairment. In many cases, this occurs in patients whose lungs recover and otherwise may have survived ECMO. Therefore, cerebral preservation and prevention of intracranial hemorrhage is receiving increasing emphasis in recent series. Other authors have recognized the following risk factors in instituting ECMO: low birth weight, immature gestational age, intracranial hemorrhage, sustained acidosis, cardiac arrest, and hypoxia. 1,3,4,7Patients with these risk factors seem to be most susceptible to the development of intracranial hemorrhage. Previous studies have suggested that carotid artery ligation and systemic heparinization contribute to the development of intracranial hemorrhage. 7-1~ However, the data from the present series, as well as a previous analysis, 11 show that survival and the development of intracranial pathology (hemorrhage, infarction, and encephalomalacia) are associated with acidosis, hypotension, prematurity, and low birth weight, and seem less likely to be due to carotid artery ligation and systemic heparinization. Thus, based on these data and those in the literature, ECMO may be contraindicated in the premature infant with significant hypotension and severe acidosis. Patients who died of nonintracranial pathology invariably had irreversible pulmonary damage or multisystem failure at autopsy. Irreversible lung disease is characterized by pulmonary hypoplasia, diffuse alveolar damage, interstitial fibrosis, necrotizing pneumonitis, and hemorrhagic infarction. 6,7,12 Because some of these pulmonary changes are irreversible, and indeed may be present prior to ECMO,
NONSURVIVORS OF NEWBORN ECMO
137
several authors have suggested performing preECMO lung biopsy. 6,7 We have not used lung biopsy pre-ECMO, feeling that this is not usually possible in newborns requiring ECMO emergently. Nonintracranial causes of death were related to the severity of the primary disease process, and probably were not preventable or reversible. The only exception might be the patient suffering fatal gastrointestinal hemorrhage after treatment with large doses of
tolazoline. This complication has been reported and might be preventable. 13 Although ECMO has proven to be lifesaving for many newborns with unresponsive respiratory failure, constant review of the morbidity and mortality of this technique, such as the present study, will produce more accurate definition of the indications and limits of the procedure.
REFERENCES 1. Loe WA, Graves ED, Oschner JL, et al: Extracorporeal membrane oxygenation for newborn respiratory failure. J Pediatr Surg 20:684-688, 1985 2. Bartlett RH, Andrews AF, Toomasian JM, et al: Extracorporeal circulation in neonatal respiratory failure: Forty-five cases. Surgery 92:425-433, 1982 3. Bartlett RH, Roloff DW, Cornell RG, et al: Extracorporeal circulation in neonatal respiratory failure: A prospective randomized study. Pediatrics 76:479-487, 1985 4. Cilley RE, Zwischenberger JB, Andrews AF, et al: lntracranial hemorrhage during extracorporeal membrane oxygenation in neonates. Pediatrics 78:699-704, 1986 5. Weber TR, Pennington DG, Connors R, et al: Extracorporeal membrane oxygenation for newborn respiratory failure. Ann Thorac Surg 42:529-535, 1986 6. Hill JD, Leval MR, Bramson ML, et al: Acute respiratory insufficiency: Treatment with prolonged extracorporeal oxygenation. J Thorac Cardiovasc Surg 64:551-562, 1972 7. Bartlett RH, Gazzaniga AB, Fong SW, et al: Prolonged
extracorporeal cardiopulmonary support in man. J Thorac Cardiovasc Surg 68:918-932, 1974 8. Bartlett RH, Gazzaniga AB, Fong SW, et al: Extracorporeal membrane oxygenator support for cardiopulmonary failure. J Thorac Cardiovasc Surg 73:375-386, 1977 9. Andrews AF, Nixon CA, Cilley RE, et al: One to three-year outcome for 14 neonatal survivors of extracorporeal membrane oxygenation. Pediatrics 78:692-698, 1986 10. Short BL, Pearson GD: Neonatal extracorporeal membrane oxygenation: A review. J Intens Care Med 1:47-54, 1986 11. Weber TR, Connors RH, Tracy TF, et al: Prognostic determinants in extracorporeal membrane oxygenation for respiratory failure in newborns. Ann Thorac Surg 50:720-723, 1990 12. Zapol WM, Snider MT, Hill JD, et al: Extracorporeal membrane oxygenation in severe acute respiratory failure. JAMA 242:2193-2196, 1979 13. Dillard RG: Fatal gastrointestinal hemorrhage in a neonate treated with tolazoline. Clin Pediatr 21:761-762, 1982
T R A C O R P O R E A L membrane oxygenation E X(ECMO) is an important new means of supporting newborns with respiratory failure. Although the majority of newborns requiring mechanical ventilation for respiratory failure are successfully treated, and recover, a number will prove to be unresponsive to conventional therapy and will require ECMO. 1 The complications of ECMO, including death secondary to intracranial hemorrhage, have received mention in previous studiesY However, there have been no comprehensive studies of ECMO nonsurvivors. The present clinical-pathological study of ECMO nonsurvivors was undertaken to determine the prevalent pathological findings and to better define the indications and contraindications for ECMO in the newborn.
achieved just prior to cannulation and continued by continuous drip, to keep the activated clotting time at 200 to 250 seconds. Blood drained from the venous cannula into a small reservoir, from which it was pumped through a membrane oxygenator, then through a heat exchanger for rewarming prior to return to the patient through the arterial cannula. As pulmonary function improved, the flow rate through the ECMO apparatus was decreased, until complete cessation of ECMO could be achieved. Nonsurvival with ECMO therapy was related to several factors. If a baby developed progressive intracerebral hemorrhage, as detected by daily head ultrasound, then ECMO was rapidly weaned or abruptly discontinued, the latter utilized if brain death was imminent. Several patients had recovery of adequate pulmonary function, but died from neurological or multisystem failure after removal from ECMO. Finally, several newborns failed to improve from their primary pulmonary disorder and succumbed after prolonged ECMO runs. This duration of ECMO for these nonsurviving patients ranged from 3 to 16 days (mean, 8 days). Of the 21 nonsurvivors, 19 had an autopsy performed (Table 1). The hospital charts, ECMO flow data sheets including all hemodynamic parameters, and also the age ECMO was initiated, time on ECMO, pre-ECMO oxygenation, level of heparinization, and autopsy findings were reviewed to determine cause of death, prevalent pathological findings, and the relationship of clinical parameters to autopsy findings.
RESULTS
Based on clinical and autopsy findings, the 19 patients were divided into two groups: group 1:12 patients who had intracranial lesions as the primary cause of death; and group 2:7 patients with nonintracranial primary causes of death, Eight of the 12 patients in group 1 had intracranial hemorrhage: six bilateral and two right-sided only. Two other infants had diffuse encephalomalacia, while the final two had massive infarcts, one rightsided and one left-sided. The primary diagnoses of these patients were persistent fetal circulation (7), sepsis (2), congenital diaphragmatic hernia (2), and meconium aspiration (1). None of these patients had pre-ECMO seizures, while one had pre-ECMO renal failure. The group 1 mean pre-ECMO arterial blood pH was 7.27, mean arterial blood pressure 43 mm Hg,
In a 3-year period, 78 newborns were treated with ECMO. Prior to institution of ECMO, cranial ultrasonography performed on each patient was negative for the presence of intracranial hemorrhage. ECMO was established by cannulating the right internal jugular vein and right common carotid artery using local anesthesia, as previously reported. 5 Anticoagulation with heparin was
From the Division of Pediatric Surgery, Department of Surgery, and the Department of Pathology, Cardinal Glennon Children's Hospital and St Louis University School of Medicine, St Louis, MO. Date accepted: October 8, 1991. Address reprint requests to Thomas R. Weber, MD, Cardinal Glennon Children's Hospital, 1465 S Grand Blvd, St Louis, 340 631O4. Copyright 9 1993 by W.B. Saunders Company 0022-3468/93/2802-0003503.00/0
Journal of Pediatric Surgery, Vo128, No 2 (February), 1993: pp 135-137
135
MATERIALS AND METHODS
WEBER ET AL
138 Table 1. Characteristics of 19 Non-survivors of Newborn ECMO Diagnosis
No.
Gestational Age Range (wk)
Persistent fetal circulation Congenital diaphragmatic hernia Sepsis Meconium aspiration Hydrops
7 4 3 3 2
35-42 34-40 34-40 39-40 34-40
Weight Range (kg) 2.3-3.6 2.2-3.5 1.9-3.6 2.8-3.3 2.0-2.5
had autopsy evidence of small white matter cerebral infarcts, without neurological consequence. Comparison of group I to group II patients showed no significant differences with regard to the age ECMO was initiated, time on ECMO, pre-ECMO oxygenation, and level of heparinization. DISCUSSION
gestational age 37 weeks, and birth weight 2,695 g (Table 2). These patients were treated with ECMO for a mean of 6 days. Associated but nonlethal factors in group 1 patients include cardiac irritability with arrhythmia (2), sepsis (1), elevated liver enzymes (1), and elevated creatinine (l). In contrast, the seven patients in group 2 had nonintracranial lesions as the primary cause of death, based on autopsy results. Three patients died of primary pulmonary disease: severe pulmonary hypoplasia in two patients with congenital diaphragmatic hernia, both of whom also had diffuse alveolar damage and necrosis, and a third patient, with persistent fetal circulation, who had diffuse alveolar damage, interstitial fibrosis, and necrotizing pneumonitis at autopsy. One other patient died of massive hemorrhage and shock. This latter patient with persistent fetal circulation treated with high doses of tolazoline pre-ECMO, had evidence of diffuse upper gastrointestinal and pancreatic hemorrhage, with ischemic infarcts of the heart, spleen, testicles, and adrenal secondary to shock. Three patients in group 2 died from multiple organ failure, with infarcts and hemorrhages in more than one organ system, including lungs, liver, adrenal glands, kidneys, and gastrointestinal tract. One of the seven patients in group 2 had pre-ECMO seizures, while three had pre-ECMO renal failure. The mean group 2 arterial blood pH was 7.39, mean arterial blood pressure 66 mm Hg, gestational age 39 weeks, and birth weight 3,209 g. These parameters were significantly different from those in group 1 (Table 2). Group 2 patients were treated with ECMO an average of 11 days. Two of these patients Table 2. Pre-ECMO Characteristics of Nonsurvivors of Newborn ECMO With an Intracranial (Group 1) or Nonintracranial (Group 2) Cause of Death Group
pH
Arterial Blood Pressure (ram Hg)
Gestational Age (wk)
Birth Weight (g)
1 2
7.27 _+ 0.02 7.39 +_ 0.10"
43 -+ 2.2 66 -+ 4.5t
37.0 --+ 2.0 39.8 --+ 1.6
2,695 --+ 56.2 3,209 -+ 35.1t
NOTE. Data given as mean _+ SEM. *P < ,001 by Student's t test. t P < .05 by Student's t test.
ECMO for respiratory failure was first successfully performed by Hill et al in an adult in 1972, 6 and extended to newborn respiratory failure several years later by Bartlett et al. 2 If instituted prior to irreversible pulmonary damage, the lungs can regain normal function within a reasonable amount of time on ECMO. Because of risks of heparinization and other complications of ECMO, the procedure is reserved for those who fail conventional therapy. Intracranial hemorrhage, the most common complication o f ECMO in this series as well as others, frequently results in either death or severe neurological impairment. In many cases, this occurs in patients whose lungs recover and otherwise may have survived ECMO. Therefore, cerebral preservation and prevention of intracranial hemorrhage is receiving increasing emphasis in recent series. Other authors have recognized the following risk factors in instituting ECMO: low birth weight, immature gestational age, intracranial hemorrhage, sustained acidosis, cardiac arrest, and hypoxia. 1,3,4,7Patients with these risk factors seem to be most susceptible to the development of intracranial hemorrhage. Previous studies have suggested that carotid artery ligation and systemic heparinization contribute to the development of intracranial hemorrhage. 7-1~ However, the data from the present series, as well as a previous analysis, 11 show that survival and the development of intracranial pathology (hemorrhage, infarction, and encephalomalacia) are associated with acidosis, hypotension, prematurity, and low birth weight, and seem less likely to be due to carotid artery ligation and systemic heparinization. Thus, based on these data and those in the literature, ECMO may be contraindicated in the premature infant with significant hypotension and severe acidosis. Patients who died of nonintracranial pathology invariably had irreversible pulmonary damage or multisystem failure at autopsy. Irreversible lung disease is characterized by pulmonary hypoplasia, diffuse alveolar damage, interstitial fibrosis, necrotizing pneumonitis, and hemorrhagic infarction. 6,7,12 Because some of these pulmonary changes are irreversible, and indeed may be present prior to ECMO,
NONSURVIVORS OF NEWBORN ECMO
137
several authors have suggested performing preECMO lung biopsy. 6,7 We have not used lung biopsy pre-ECMO, feeling that this is not usually possible in newborns requiring ECMO emergently. Nonintracranial causes of death were related to the severity of the primary disease process, and probably were not preventable or reversible. The only exception might be the patient suffering fatal gastrointestinal hemorrhage after treatment with large doses of
tolazoline. This complication has been reported and might be preventable. 13 Although ECMO has proven to be lifesaving for many newborns with unresponsive respiratory failure, constant review of the morbidity and mortality of this technique, such as the present study, will produce more accurate definition of the indications and limits of the procedure.
REFERENCES 1. Loe WA, Graves ED, Oschner JL, et al: Extracorporeal membrane oxygenation for newborn respiratory failure. J Pediatr Surg 20:684-688, 1985 2. Bartlett RH, Andrews AF, Toomasian JM, et al: Extracorporeal circulation in neonatal respiratory failure: Forty-five cases. Surgery 92:425-433, 1982 3. Bartlett RH, Roloff DW, Cornell RG, et al: Extracorporeal circulation in neonatal respiratory failure: A prospective randomized study. Pediatrics 76:479-487, 1985 4. Cilley RE, Zwischenberger JB, Andrews AF, et al: lntracranial hemorrhage during extracorporeal membrane oxygenation in neonates. Pediatrics 78:699-704, 1986 5. Weber TR, Pennington DG, Connors R, et al: Extracorporeal membrane oxygenation for newborn respiratory failure. Ann Thorac Surg 42:529-535, 1986 6. Hill JD, Leval MR, Bramson ML, et al: Acute respiratory insufficiency: Treatment with prolonged extracorporeal oxygenation. J Thorac Cardiovasc Surg 64:551-562, 1972 7. Bartlett RH, Gazzaniga AB, Fong SW, et al: Prolonged
extracorporeal cardiopulmonary support in man. J Thorac Cardiovasc Surg 68:918-932, 1974 8. Bartlett RH, Gazzaniga AB, Fong SW, et al: Extracorporeal membrane oxygenator support for cardiopulmonary failure. J Thorac Cardiovasc Surg 73:375-386, 1977 9. Andrews AF, Nixon CA, Cilley RE, et al: One to three-year outcome for 14 neonatal survivors of extracorporeal membrane oxygenation. Pediatrics 78:692-698, 1986 10. Short BL, Pearson GD: Neonatal extracorporeal membrane oxygenation: A review. J Intens Care Med 1:47-54, 1986 11. Weber TR, Connors RH, Tracy TF, et al: Prognostic determinants in extracorporeal membrane oxygenation for respiratory failure in newborns. Ann Thorac Surg 50:720-723, 1990 12. Zapol WM, Snider MT, Hill JD, et al: Extracorporeal membrane oxygenation in severe acute respiratory failure. JAMA 242:2193-2196, 1979 13. Dillard RG: Fatal gastrointestinal hemorrhage in a neonate treated with tolazoline. Clin Pediatr 21:761-762, 1982