Hydration during the first days of life and the risk of bronchopulmonary dysplasia in low birth weight infants Linda J, V a n Marter, MD, MPH, Alan Leviton, MD, MS, Elizabeth N. AIIred, MS, M a r c e l l o P a g a n o , PhD, a n d Karl C. K. Kuban, MD From the Divisionof Newborn Medicine, and the Neuroepidemiotogy Unit, Children's Hospital, Boston, and the Department of Biostatistics,Harvard School of Public Health, Boston, Massachusetts.

We conducted a case-control study of antecedents of bronchopulmonary dysplasia (BPD) in 223 infants enrolled in a prospective, randomized clinical trial of phenobarbital prophylaxis for intracranial hemorrhage. The trial took p l a c e at three Boston neonatal intensive care units between June 1981 and April 4984. The 76 babies with BPD had radiographic evidence of the condition and required oxygen therapy for 28 days or more. All 147 control babies survived until d a y 28 of life without meeting either of these criteria for BPD. Compared with control infants, those with BPD received greater quantities of total, crystalloid, and colloid fluids per kilogram per d a y in the first 4 days of life. In addition, infants with BPD generally had a net weight gain in the first 4 days of life in contrast to the normal pattern of weight loss seen in control infants. Finally, the infants with BPD were more likely to be given a clinical diagnosis of patent ductus arteriosus and to have received furosemide on days 3 and 4 of life. From these observations we infer that early postnatal p h e n o m e n a such as excessive fluid therapy may be important in the pathogenesis of BPD. (J PEDIATR1990;446:942-9)

Bronchopulmonary dysplasia most often occurs in infants in the lowest birth weight categories. Although BPD develops almost exclusively in infants recovering from pulmonary disease associated with respiratory distress syndrome, the published trials of surfactant replacement therapy have shown substantial residual rates of BPD in the treated Supported by grants from the Hearst Foundation, National Foundation-March of Dimes, Mead Johnson Nutritional Division, Milton Fund of Harvard University,National Institutes of Neurological and Communicative Disorders and Stroke (grant Nos. NS 20807 and NS 20658), National Heart, Lung, and Blood Institute (grant Nos. HL 40454 and SCOR 2P50HL34616-03), and National Institute of Child Health and Human Development(Mental Retardation Center grant No. HD 06276). Dr. Van Marter's research is also supported by a Charles A. King Trust fellowship from the Medical Foundation. Submitted for publication July 28, 1989; accepted Dec. 12, 1989. Reprint requests: Linda J. Van Marter, MD, Joint Program in Neonatology, 75 Francis St., Boston, MA 02115. 9/23/18939

942

infants, lq3 This suggests that factors other than surfactant deficiency may be important. BPD rates differ among tertiary care centers, even after adjustment for mortality rates, birth weight, race, and sex. 14, 15 These observations suggest that medical care practices in the neonatal care unit may influence the risk of BPD. This hypothesis is also supported BPD PDA RDS

Bronchopulmonarydysplasia Patent ductus arteriosus Respiratorydistress syndrome

by studies demonstrating statistically significant associations between BPD, high fractions of inspired oxygen, 16, 17 and high ventilator pressures. 14, 17-21These findings are in keeping with speculation that barotrauma contributes to the occurrence of BPD. Both past 22 and recent 23 investigators have suggested that judicious use of fluid therapy in the first days of life may improve neonatal outcome. The relevance of this hypothe-

Volume 116 Number 6 sis to the study of etiologic mechanisms of pulmonary diseases is supported by data demonstrating the potential for pulmonary edema 24 and patent ductus arteriosus 25' 26 to complicate RDS, and by the well-documented improvement in RDS that occurs in the diuretic phase of this disorder. 27-3~However, the most compelling support for a relationship between fluid therapy and BPD comes from reports linking PDA 3134 and pulmonary edema 35 directly to BPD. METHODS This case-control study of antecedents of BPD in 223 infants was nested in a prospective, randomized, controlled clinical trial of phenobarbital prophylaxis .for neonatal intracranial hemorrhage. 36 The trial prospectively enrolled 280 infants at three neonatal intensive care units affiliated with Harvard Medical School (Brigham and Women's Hospital, Children's Hospital, and Massachusetts General Hospital) between June 1981 and April 1984. Infants were recruited for the phenobarbital study if they required intubation within the first 12 hours of life, their birth weights were 1750 gm or less, and they had no intracranial hemorrhage demonstrable by cranial ultrasonography. During the course of the study of phenobarbital prophylaxis, laboratory and clinical data were prospectively collected, including the clinical diagnoses of such conditions as PDA and BPD. The diagnosis of PDA was based on auscultation of a characteristic murmur, palpation of bounding pulses, and ascertainment of widened pulse pressure. The goals of fluid therapy were generally to maintain adequate blood pressure and urine output and to prevent hyponatremia or hypernatremia. Prior studies of the determinants of BPD have employed diagnostic criteria of either a radiologic37, 38 or a clinical14 nature. The most widely accepted clinical criterion is a requirement for supplemental oxygen for 28 days or more.14, 39 We chose as an additional requirement for case status the presence of stage 2, 3, or 4 radiographic findings of BPD, as described by Northway et al. 37 and Edwards. 38 The 76 infants who met both diagnostic criteria for BPD constituted the case group. Control subjects were the 147 infants who survived until the end of the first postnatal month and did not meet either of the case criteria. Thirty-one infants were excluded because they died before the twenty-eighth postnatal day and therefore were ineligible to fulfill either case or control criteria. An additional 26 infants (15% of the surviving noncase group) were eliminated from analysis because they met only one criterion and thus the diagnosig was unclear. Univariate analyses were performed with the chi-square test, Fisher permutation test, Mantel-Haenszel test for stratified data, and Wilcoxon rank-sum test.

Hydration and bronchopulmonary dysplasia

943

T a b l e I. Descriptive maternal and neonatal variables

Maternal age (yr) Gravidity (No. of pregnancies) Parity (No. of live births) Birth weight (gin)* Gestational age (wk)* Apgar score at 5 rain* Male/female ratio White/black ratio

Infants with BPD (n = 76)

Control subjects (n = 147)

26.9 (0.6) 2.5 (0.2)

26.4 (0.5) 2.2 (0.1)

1.8 (0.1)

1.8 (0.1)

953.0 (31.0) 28.4 (0.2) 4.2 (0.2) 1.11 5.5

1287.0(23.0) 30.9 (0.2) 5.3 (0.2) 0.83 5.8

Data, except ratios, are expressed as mean (standarderror of mean). *p JO

..............................................................

Crystanolds 100

9

/

Fig. 6. Maximum ventilator settings on each of first 4 days of life as function of colloid and crystalloid fluid intakes. Ventilator data are expressed as maximum peak inspiratory pressure (PIP), positive end expiratory pressure (PEEP), rate, and inspired concentration of ox3~gen (F102).

947

948

Van M a r t e r et al.

may retain fluid because of an inherent regulatory abnormality, such as elevated vasopressin levels seen in babies with chronic BPD. 44 In birth weight-stratified analyses, infants with BPD received on average 17% more total fluid and 10% more crystalloid fluid per day than control infants. Howeverl the mean daily colloid intake of infants with BPD was nearly double (193%) that of control subjects. W e also found that the need for higher ventilator settings was more closely linked with colloid than with crystalloid therapy. Despite the disparity in colloid intake between groups, the precise pathophysiologic significance of colloid vs crystalloid therapy is unclear. One possible explanation for a significant link between colloid fluid intake and BPD is that the increased capillary permeability in hyaline membrane disease 45, 46 permits colloid to enter the pulmonary interstitium where it participates in the causal chain leading to BPD. Another explanation is that high colloid intake may serve as a marker for hemodynamic instability or enhanced capillary leak resulting from more severe illness. If so, the association between higher colloid intake and BPD may simply verify that is already well known: infants who are more seriously ill are at higher risk of BPD. Although we found no link between hypotension and BPD, the associations between BPD and both acidemia and administration of pancuronium support the severity-of-illness hypothesis. Arguing against severity of illness as an isolated causative factor is the net weight gain during the first 4 days of life in infants with BPD, compared with weight loss in control subjects. Furthermore, among pancuronium-treated infants, those with greater weight gain were at higher risk of BPD. Weight change is one of the best measures of total body extracellular fluid status in a newborn infant. Normal infants, like our control subject, lose weight in the first 4 days, reflecting loss of the relative excess extracellular fluid they have at birth. Deviations from the normal pattern of weight loss probably reflect persistent miscalculation of free water intake, as well as endogenous factors. Although some infants retain more fluid than others, care givers should modify fluid administration accordingly. Speculation regarding the role of excess extracellular fluid and respiratory disease has a firm historical footing: in 1948 Dr. Clement A. Smith observed that "persistent respiratory distress after birth occurred only i n . . . edematous infants," adding that "we have no means of knowing whether this indicates that the edema involves the lungs as well as other parts of the b o d y " Y Subsequent investigators have affirmed the contribution of pulmonary edema to RDS.24, 35 However, inferences regarding the significance of our results linking fluid therapy and BPD must be made with caution. Although others have reported associations between administration of certain fluids and adverse neonatal outcomes, 22, 42, 47 a conclusion that fluid therapy con-

The Journal o f Pediatrics June 1990

tributes to development of BPD is probably premature. Regardless of the causal or indicator role of increased fluid receipt and fluid retention, such experiences and exposures during the first days of the life do appear to predict BPD a month later. Thus continued attention to early postnatal medical care seems appropriate in the search for ways to reduce the occurrence of BPD. We thank Kathleen Finn Sullivan, Kenneth Huff, Elizabeth Brown, and Kalpathy Krishnamoorthy for their work on the original randomized trial and Mary Ellen Avery for support and encouragement throughout this project. REFERENCES

1. Fujiwara T, Chida S, Watanabe Y, et al. Artificial surfactant therapy in hyaline membrane disease. Lancet 1980;1:55-9. 2. Halliday H, Reid M, Meban C, et al. Controlled trial of artificial surfactant to prevent respiratory distress syndrome. Lancet 1984;1:476-8. 3. Enhorning G, Sherman A, Possmayer F, et al. Prevention of neonatal respiratory distress syndrome by tracheal instillation of sur factant: a randomized clinical trial. Pediatrics 1985; 14553. 4. Hallman M, Merritt T, Jarvenpaa AL, et al. Exogenous human surfactant for treatment of severe respiratory distress syndrome: a randomized prospective clinical trial. J PEDIATR 1985;106:963-9. 5. Wilkinson A, Jenkins P, Jeffrey J. Two controlled trials of dry artificial surfactant: early effects of later outcome in babies with surfactant deficiency. Lancet 1985;1:287-91. 6. Merritt T, Hallman M, Bloom B, et al. Prophylactic treatment of very premature infants with human surfactant. N Engl J Med 1986;315:785-90. 7. Gitlin J, Soll R, Parad R, et al. Randomized controlled trial of exogenous surfactant for the treatment of hyaline membrane disease. Pediatrics 1987;79:31-7. 8. Ten Centre Study Group. Ten centre trial of artificial surfacrant (artificial lung expunding compound) in very premature babies. Br Med J 1987;294:991-6. 9. Kendig J, Sinkin R. The effect of surfactant replacement therapy on conditions associated with respiratory distress syndrome: patent ductus arteriosus, intraventricular hemorrhage, and bronchopulmonary dysplasia. Semin Perinatol 1988; 12:255-8. 10. Konishi M, Fujiwara T, Naito T, et al. Surfactant replacement therapy in neonatal respiratory distress syndrome: a multicenter, randomized clinical trial; comparison of high-versus low-dose surfactant TA. Eur J Pediatr 1988;147:20-5. 11. Morley C, Greenough A, Miller N, et al. Randomized trial of artificial surfactant (ALEC) given at birth to babies from 23 to 34 weeks gestation. Early Hum Dev 1988;17:41-54. 12. Morley C. Surfactant therapy for very premature babies. Br Med Bull 1988;44:919-34. 13. Horbar J, Soll R, Sutherland J, et al. A multicenter randomized placebo-controlled trial of snrfactant therapy for respiratory distress syndrome. N Engl J Med 1989;320:959-65. 14. Avery M, Hurd S, Tooley W. Is chronic lung disease in prematurely-born infants preventable? Pediatrics 1987;79:26-30. 15. Horbar J, McAuliffe T, Adler S, et al. Variability in 28-day outcomes for very low birth weight infants: an analysis of 11 neonatal intensive care units. Pediatrics 1988;82:554-9. 16. Bancalari E, Abdenour G, Feller R, Gannon J. Bronchopul-

Volume 116 Number 6

17. 18.

19.

20.

21.

22.

23. 24.

25.

26.

27.

28.

29.

30.

31.

monary dysplasia: clinical presentation. J PEDIATR 1979; 95:819-23. Rhodes P, Hall R, Leonidas J. Chronic pulmonary disease in neonates with assisted ventilation. Pediatrics 1975;55:788-96. Berg T, Pagtakhan R, Reed M, et al. Bronchopulmonary dysplasia and lung rupture in hyaline membrane disease: influence of continuous distending pressure. Pediatrics 1975;55:51-4. Stocker J, Madewall J. Persistent interstitial pulmonary emphysema: another complication of the respiratory distres s syndrome. Pediatrics 1977;59:847-5% Moylan F, Walker A, Kramer S, et al. Alveolar rupture as an independent predictor of bronchopulmonary dysplasia. Crit Care Med 1978;6:10-3. Stern L. The role of respirators in the etiology and pathogenesis of bronchopulmonary dysplasia. J PEDIATR 1979;95: 867-9. Smith C, Yudkin S, Young W, et al. Adjustment of electrolytes and water following premature birth (with special reference to edema). Pediatrics 1949;3:34-48, Arant B. Fluid therapy in the neonate- concepts in transition. J P~)~A'r~ 1982;101:387-9. Bland R. Edema formation in the lungs and its relationship to neonatal respiratory distress. Acta Paediatr Scand [Suppl] 1983;103:92-9. Stevenson J. Fluid administration in the association of patent ductus arteriosus complicating respiratory distress syndrome. J PEDIATR 1977;90:257-61. Pongiglione G, Marasioni M, Silvestri G, et al. Early treatment of patent ductus arteriosus in premature infants with severe respiratory distress syndrome. Pediatr Cardiol 1988;91-4. Langman C, Engle W, Baumgart S, et al. The diuretic phase of respiratory distress syndrome and its relationship to oxygenation. J PEDIATR 1981;98:462-6. Heal D, Belik J, Spitzer A, et al. Changes in pulmonary function during the diuretic phase of respiratory distress syndrome. J PEDIATR 1982;101:103-7. Green T, Thompson T, Johnson D, et al. Diuresis and pulmonary function in premature infants with respiratory distress syndrome, J PEDIATR 1983;103:618-23. Engle W, Arnat B, Wiriyathian S, et al. Diuresis and respiratory distress syndrome: physiologic mechanisms and therapeutic implications. J PEDIATR 1983; 102:912- 7. Gay J, Daily W, Meyer B, et al. Ligation of the patent ductus arteriosus in premature infants: report of 45 cases. J Pediatr Surg 1973;8:677-83.

Hydration and bronchopulrnonary dysplasia

949

32. Brown E. Increased risk of bronchopulmonary dysplasia in infants with patent ductus arteriosus. J PEDIATR 1979;95:865-6. 33. Naulty C, Horn S, Conry J, Avery G. Improved lung compliance after tigation of patent ductus arteriosus in hyaline membrane disease. J PEI)IATR 1978;93:682-4. 34. Kamtorn V, Wei T, Bautista A, et al. Early vs. late preterm patent ductus arteriosus ligation. Pediatr Res 1988;23:434A. 35. Brown E, Stark A, Sosenko I, et al, Bronchopulmonary dysplasia: possible relationship to pulmonary edema. J PEDIATR 1978;92:982-4. 36. Kuban K, Leviton A, Krishnamoorthy K, et al. Neonatal intracranial hemorrhage and phenobarbital. Pediatrics 1986;77: 443-50. 37. Northway W, Rosan R, Porter D. Pulmonary disease following respiratory therapy of hyaline-membrane disease. N Engl J Med 1967;276:357-68. 38. Edwards D. Radiographic aspects of bronchopalmonary dysplasia. J PEDIATR 1979;95:823-9. 39. Edwards D, Dyer W, Northway W. Twelve years' experience with bronchopulmonary dysplasia. Pediatrics 1977;59:839-45. 40. Simmons C Jr, Jose J. Fluid and electrolyte management of the newborn. In: Cloherty J and Stark A, eds. Manual of neonatal care, 2nd ed. Boston: Little, Brown, 1985;359-68. 41. Lorenz J, Kleinman L, Kotagal U, et al. Water balance in very low birth-weight infants: relationship to water and sodium intake and effect on outcome. J PEDIATR 1982;101:423-32. 42. Bell E, Warburton D, Stonestreet B, Oh W. High-volume fluid intake predisposes premature infants to necrotising enterocolitis. Lancet 1979;2(8133):90. 43. Jacob J, Gluck L, DiSessa T, et al. The contribution of PDA in the neonate with severe RDS. J PEDIATR 1980;96:79-87. 44. Hazinski T, Blalock W, Engelhardt B. Control of water balance in infants with bronetiopulmonary dysplasia: role of endogenous vasopressin. Pediatr Res 1988;23:86-8, 45. deSa D. Pulmonary fluid content in infants with respiratory distress. J Pathol 1969;97:469-78. 46. Jeffries A, Coates G, O'Brodovich H. Pulmonary epithelial permeability in hyaline membrane disease. N Engl J Med 1984;311:1075-80. 47. MeGrady G, Rettig P, Jistre G, et al. An outbreak of necrotizing enterocolitis--assoeiation with transfusions of packed red blood cells. Am J Epidemiol 1986;126:1165-72.