|
| |||||||||||||||||||||||||||||||||||||||
Background In southern Scandinavia most babies with respiratory distress syndrome are initially treated with nasal continuous positive airway pressure. We performed a multicenter trial to investigate whether the addition of a single dose of porcine surfactant administered during a short intubation before the occurrence of serious deterioration could reduce the subsequent need for mechanical ventilation.
Methods We randomly assigned 35 infants with moderate-to-severe respiratory distress syndrome to surfactant therapy (Curosurf, 200 mg per kilogram of body weight) plus nasal continuous positive airway pressure and 33 infants to nasal continuous positive airway pressure alone. The study was not blinded. The indications for mechanical ventilation were a ratio of arterial to alveolar oxygen tension of less than 0.15, severe apneic attacks, or both.
Results Six hours after randomization, when the median age of the babies was 18 hours, the mean ratio of arterial to alveolar oxygen tension was 0.37 in the surfactant-treated babies, as compared with 0.25 in the controls (P<0.001). The need for subsequent mechanical ventilation was reduced with surfactant therapy (to 43 percent of the surfactant-treated babies, as compared with 85 percent of the controls; P = 0.003). When 17 infants with ratios of arterial-to-alveolar oxygen tension of less than 0.15 at randomization were excluded, the need for mechanical ventilation was still significantly reduced in the surfactant-treated group (to 33 percent [9 of 27 babies], as compared with 83 percent [20 of 24 babies] in the control group; P<0.001). After 28 days, two of the surfactant-treated babies had died, as compared with five of the control babies.
Conclusions In babies with moderate-to-severe respiratory distress syndrome treated with nasal continuous positive airway pressure, a single dose of surfactant reduced the need for subsequent mechanical ventilation.
The combination of surfactant replacement and mechanical ventilation was introduced in 19806. Currently, this combination is used as both prophylactic and rescue therapy for respiratory distress syndrome, with a reduction in mortality7. When surfactant is given prophylactically at birth, some infants who do not need the medication are treated. On the other hand, waiting until respiratory distress syndrome has developed may mean that the treatment is given too late to benefit some babies. Furthermore, mechanical ventilation may have complications. Suction of the trachea may cause hypoxia,8 and suboptimal ventilatory settings may cause leakage of air and depletion of surfactant9,10.
After conducting two pilot studies,11,12 we investigated whether nasal continuous positive airway pressure initiated early (i.e., at the first sign of respiratory distress) in combination with a fast-acting surfactant administered during a short intubation could halt the progression of respiratory distress syndrome, reduce the need for mechanical ventilation, and improve outcomes in newborns.
Methods
Surfactant
Curosurf (Chiesi Farmaceutici, Parma, Italy), the surfactant used, is isolated from minced pig lungs by a combination of washing, centrifugation, extraction with chloroform-methanol, and liquid-gel chromatography13. It contains approximately 99 percent lipids, mainly phospholipids, and 1 percent apoproteins (surfactant proteins B and C)13. The efficacy of this product has been documented in previous trials14,15.
Continuous Positive Airway Pressure
For continuous positive airway pressure, the lightweight system based on Benveniste's pediatric gas-jet valve (Dameca, Copenhagen, Denmark), originally developed for artificial ventilation, was used16,17. This system eliminates the risk that accidental kinking of the tubes in the ventilation system will produce excessive pressure in the airways and the risk that ambient air will be mixed with the inhaled gases18,19. The Benveniste valve generates positive pressures of 4 to 10 cm of water by creating a narrow stream of humidified air mixed with oxygen at a rate of flow of 10 to 16 liters per minute. The mixture is delivered through nasal prongs.
Study Design
Seventy-three infants with moderate-to-severe respiratory distress syndrome treated early with nasal continuous positive airway pressure were randomly assigned either to receive surfactant or to continue receiving nasal continuous positive airway pressure. The study was organized as a multicenter trial involving 11 neonatal intensive care units in Denmark and Sweden. The indication for mechanical ventilation was standardized, and all infants undergoing ventilation could receive surfactant as rescue therapy. The trial was conducted from September 1991 to October 1992.
The primary end point was the percentage of patients requiring mechanical ventilation. On the basis of data from one of the pilot studies,12 we set a goal of reducing this percentage from 80 percent in the controls to 40 percent in the surfactant-treated babies. Assuming a power of 90 percent and a significance level of 5 percent (two-tailed), we estimated that 108 infants would have to be randomized.
The secondary end point was the ratio of arterial to alveolar oxygen tension20 six hours after randomization. Our hypothesis was that the mean ratio in the babies treated with surfactant would be 50 percent higher than the mean ratio in the controls. The protocol called for an interim analysis after half the infants had been randomized. The criterion for stopping the study was a P value of 0.01 for the primary end point.
The criteria for entry into the study were as follows: clinical and radiologic findings of respiratory distress syndrome; gestational age of 25 to 35 weeks; age of 2 to 72 hours; a requirement for nasal continuous positive airway pressure
6 cm of water; and a ratio of arterial to alveolar oxygen tension of less than 0.22. Neonates with Apgar scores below 3 five minutes after birth, those whose mothers had had rupture of the membranes for more than 14 days, and those with severe malformations, pneumonia, or pneumothorax were not enrolled. If twins were born, only the first twin to fulfill the criteria was enrolled. Randomization was performed by opening sealed envelopes that were numbered sequentially and kept at each hospital. It was performed in blocks of four so that a similar number of babies would be enrolled in each treatment group at each center.
Prophylaxis with theophylline was optional (5 mg per kilogram of body weight intravenously or orally, followed by 2.5 mg per kilogram twice daily). Transcutaneous partial pressure of oxygen and partial pressure of carbon dioxide were monitored continually (Radiometer, Copenhagen, Denmark, or Hellige, Freiburg, Germany), and these measurements were used to calculate the ratio of arterial to alveolar oxygen tension20.
Infants randomly assigned to treatment with surfactant were treated with morphine (0.1 mg per kilogram intravenously) and atropine (10 µg per kilogram intravenously) before intubation. Treatment with naloxone (10 µg per kilogram intravenously) before extubation was optional. Surfactant (200 mg [2.5 ml] per kilogram) was given in two bolus doses a few minutes apart. After each dose, the infants underwent manual ventilation, usually for two to five minutes. As soon as respiration was judged satisfactory, the infants were extubated. Treatment was considered to have failed in infants who were not extubated within one hour or who were reintubated for ventilator therapy within five days. The study was not blinded.
The indications for mechanical ventilation were a decrease in the oxygen-tension ratio to less than 0.15, severe attacks of apnea, or both. Babies who had oxygen-tension ratios of less than 0.15 at randomization had their transcutaneous gas values monitored for at least 10 minutes before mechanical ventilation was initiated. If the oxygen-tension ratio was improving after the 10-minute period, the decision to administer ventilation was postponed for 20 more minutes -- i.e., until 30 minutes had elapsed after randomization. Babies whose oxygen-tension ratios were still less than 0.15 were then intubated for mechanical ventilation.
Infants treated with mechanical ventilation were permitted to receive surfactant at a dose of 200 mg per kilogram if they required a concentration of inspired oxygen greater than 60 percent12. Additional doses of surfactant could be given 12 and 24 hours later if an infant still required a concentration of inspired oxygen greater than 60 percent. (None of the babies in fact required more than one such rescue dose of surfactant.) Pressure-limited, time-cycled, continuous-flow ventilators were used (Drager, Lubeck, Germany).
Chest films were obtained for all patients at randomization and were assessed retrospectively in a blinded fashion by two of the investigators. The chest films were classified as normal or as indicating mild, moderate, or severe respiratory distress syndrome21. All babies who survived underwent cerebral ultrasonography seven days after birth. Intracerebral hemorrhage was diagnosed and classified according to the method of Papile et al.22 Periventricular leukomalacia was diagnosed by ultrasonography23. Patent ductus arteriosus was diagnosed by clinical signs, by echocardiography, or both.
The trial was approved by the research ethics committee at each institution. Written informed consent was obtained from the parents of the newborns before randomization.
Statistical Analysis
Base-line differences between the groups at randomization were evaluated by the Mann-Whitney test or Fisher's exact test (two-tailed). Logistic-regression analysis24 was used to investigate possible associations between the need for mechanical ventilation and the following variables: gestational age, birth weight, sex, antenatal steroid treatment, birth in the hospital or outside the hospital, cesarean section, perinatal asphyxia (Apgar score,
6 at five minutes), theophylline treatment, age at randomization, transcutaneous partial pressure of carbon dioxide at randomization, chest-film findings at randomization, treatment center, and treatment with surfactant.
We sought to determine the effect of surfactant treatment when the other variables were controlled for. By including interaction terms in the regression model,24 we investigated how the effect of treatment changed as the variables changed -- for example, whether the effect of treatment was the same for boys as for girls. When we used the oxygen-tension ratio at six hours as the dependent variable, ordinary multiple regression analysis was used24. P values of less than 0.05 were considered to indicate statistical significance.
Results
An interim analysis performed after 54 infants had been randomized showed a statistically significant benefit for those treated with surfactant. The trial was stopped as soon as this analysis was completed. At that time, 73 infants had been randomized (36 treated with surfactant and 37 controls), including 5 who did not meet the entry criteria (2 with gestational ages of 36 weeks, 2 with oxygen-tension ratios of 0.22 and 0.23 at randomization, and 1 with pneumonia at randomization). Therefore, data for 68 patients (35 treated with surfactant and 33 controls) were analyzed. The characteristics of the patients in each group were similar (Table 1). Nasal continuous positive airway pressure was begun at a median age of 19 minutes, and the babies were randomly assigned to the treatment groups at a median age of 12 hours.
|
|
Eight babies with oxygen-tension ratios below 0.15 who were randomly assigned to surfactant therapy responded to treatment with increases in the ratio to more than 0.15. Seven were extubated without having undergone ventilation; the eighth had apnea and required continued ventilation. Five of the seven babies who were extubated required mechanical ventilation 2 to 24 hours after randomization, three because of apnea and two because of oxygen-tension ratios below 0.15. Eight of the control babies with oxygen-tension ratios below 0.15 at randomization were intubated for mechanical ventilation 10 minutes to 7 hours later; the other baby's oxygen-tension ratio increased to more than 0.15, and nasal continuous positive airway pressure was continued. When the 17 infants with oxygen-tension ratios of less than 0.15 at randomization were excluded from the analysis, the need for subsequent mechanical ventilation was still significantly reduced by treatment with surfactant (mechanical ventilation was required by 33 percent of the babies [9 of 27] in the treatment group vs. 83 percent [20 of 24] in the control group; P<0.001).
Total Dose of Surfactant
Three of the 15 infants who were randomly assigned to surfactant treatment and later required mechanical ventilation received an extra dose of surfactant (200 mg per kilogram). Among the control babies who required mechanical ventilation, 19 received one dose of surfactant (200 mg per kilogram). Thus, the 35 babies initially treated with surfactant received a total of 38 doses (35 of which were part of the initial treatment), and the 33 controls received a total of 19 doses. None of the babies undergoing mechanical ventilation required multiple-dose treatment for rescue.
Multiple Regression Analysis
In a logistic-regression model that used the requirement for mechanical ventilation as the dependent variable with no interaction terms, only surfactant treatment was significant in reducing the need for mechanical ventilation (P = 0.003) (Table 2). When interaction terms were included, the effects of treatment with surfactant on the need for mechanical ventilation was greater for boys than for girls (P = 0.008). No other interaction term was statistically significant. There was a trend, however, toward better outcomes in babies with higher birth weights and babies randomized earlier (data not shown). When the oxygen-tension ratio at six hours was used as the dependent variable, treatment with surfactant was associated with a higher mean ratio (P<0.001). In this analysis, the effect of surfactant treatment was also significantly greater for boys than for girls (P = 0.006).
Secondary Outcome Measures
The 15 surfactant-treated and 28 control babies who required mechanical ventilation received it for a median of 2.5 days (the same value for both groups). The median duration of treatment with supplemental oxygen was also similar in both groups.
The outcomes at 28 days of life are summarized in Table 3. No significant differences between the groups were found.
|
Respiratory distress syndrome is a progressive condition with deterioration during the first hours and days of life. Early treatment with continuous positive airway pressure can prevent or temporarily compensate for the increased alveolar-retraction forces that are a consequence of high surface tension caused by deficiency of surfactant9. Our study indicates that the combination of early nasal continuous positive airway pressure and surfactant therapy further retards the progression of this syndrome by improving lung function, as indicated by an increased oxygen-tension ratio, a decreased transcutaneous partial pressure of carbon dioxide, and a reduced need for mechanical ventilation. The babies in our pilot study12 had severe respiratory distress, with an average oxygen-tension ratio of 0.12 when they were treated with surfactant at a mean age of 19 hours. Many of those babies had increasing apnea and required mechanical ventilation despite a good response to surfactant. Therefore, we decided to treat babies at an earlier stage of the disease -- i.e., when they had higher oxygen-tension ratios -- before severe respiratory distress developed.
Few pulmonary and extrapulmonary complications occurred in either the surfactant-treated babies or the controls. This probably reflected both the selection of patients and the treatment strategy. About 6 percent of babies with birth weights below 1500 g who are treated early with nasal continuous positive airway pressure need mechanical ventilation within the first two hours of life4; this represents the approximate proportion of babies with very severe respiratory distress syndrome who were not included in our trial. The frequent use of steroids before birth to mature the lungs may also have reduced the incidence of complications. Nasal continuous positive airway pressure probably mitigated respiratory distress syndrome in both groups of newborns, as in earlier studies in which survival was improved25 and the incidence of bronchopulmonary dysplasia reduced5,25 with the early use of this therapy. The overall incidence of pulmonary complications and mortality was lower in the present study than in the OSIRIS and Curosurf 4 trials26,27; in both these studies, mechanical ventilation and treatment with surfactant were combined in babies with oxygen-tension ratios of less than 0.22. Direct comparison of data from these three trials is not possible, however, because of differences in treatment strategies in the first hours of life.
It is possible that the brief period of intermittent positive-pressure ventilation at the time of treatment with surfactant could have contributed to the improved outcome in the surfactant group. On the basis of previous experience,28 we believe that this is unlikely. In babies who receive early nasal continuous positive airway pressure, any possible positive effects of such ventilation are probably outweighed by the negative effects of handling, intubation, and extubation. For these reasons, we did not consider it ethical to subject the babies in the control group to sham maneuvers that mimicked the administration of surfactant.
The lack of blinding is another potential problem. This factor might have influenced the decision to intubate the babies for mechanical ventilation. Complete blinding, however, would have required using separate dosing teams (medical personnel not otherwise involved in the treatment of the baby) to administer surfactant at all 11 neonatal intensive care units. This was not possible, given budgetary constraints. Furthermore, it would have been difficult to be unaware of the rapid improvement in oxygenation that characteristically occurs very soon after the administration of surfactant12. The typical indications for intubation and mechanical ventilation differed between the groups: an oxygen-tension ratio of less than 0.15 in the controls, and apnea in the babies receiving surfactant.
Substantially more boys than girls responded favorably to early treatment with nasal continuous positive airway pressure and surfactant. In another study, a reduction in the severity of respiratory distress syndrome was reported in boys but not girls after combined treatment with antenatal steroids and prophylactic calf-lung surfactant29. We did not find a similar additive effect of these treatments.
Babies with respiratory distress syndrome who require mechanical ventilation and receive surfactant early in their disease do better than those who are treated late30. Several studies have documented a beneficial effect of prophylactic treatment with surfactant in very-low-birth-weight infants,7 and the administration of surfactant at birth may be superior to treatment when decompensation occurs31. However, prophylactic treatment of all infants under a certain gestational age (e.g., 29 weeks) implies that surfactant would be given to some infants with mature lungs who do not need it. This is an issue of both ethics and economy. Our findings suggest an intermediate approach: the use of surfactant when the diagnosis of respiratory distress syndrome is established, but before serious deterioration has occurred.
Supported by grants from the Danish Medical Research Foundation for the counties of Bornholm, Frederiksborg, Roskilde, Storstroem, and Vestsjaelland; by grants from the Swedish Medical Research Council (projects 3351 and 4732); and by a grant from Oscar II's Jubileumsfond. Curosurf was donated by Chiesi Farmaceutici, Parma, Italy.
We are indebted to the attending nurses and neonatologists in the contributing departments for their support; to Hanne Weinreich, Gitte Enevoldsen, and Christina Ingvarsson for assistance in the preparation of the manuscript; to the Collaborative European Multicenter Study Group for their support; and to Henry Halliday, M.D., David Strayer, M.D., and Ole Christensen, M.D., for constructive advice.
Source Information
From the Department of Pediatrics, Holbaek Central Hospital, Holbaek, Denmark (H.V., P.A.); the Research Unit for Perinatal Pathology, Karolinska Hospital, Stockholm, Sweden (B.R.); the Department of Pediatrics, University Hospital of Copenhagen, Copenhagen, Denmark (G.G., K.L., T.J.); and the Department of Pediatrics, Aalborg Hospital, Aalborg, Denmark (F.E.). The members of the Danish-Swedish Multicenter Study Group are listed in the Appendix.
Address reprint requests to Dr. Verder at the Department of Pediatrics, Holbaek Central Hospital, 4300 Holbaek, Denmark.
References
In addition to the authors, the following persons and institutions participated in the Danish-Swedish Multicenter Study Group.
Departments of pediatrics in Denmark -- Kolding: L. Agertoft and N. Hobolth; Esbjerg: B. Djernes and C. Grytter; Glostrup, University of Copenhagen: J. Hertel; Herning: V. Holm and U.S. Hansen; Odense University Hospital: J. Kamper and K.H. Johansen; Aarhus University Hospital: E. Nathan and A. Lange; Hvidovre, University of Copenhagen: B. Peitersen, L. Pedersen-Bjergaard, and L. Skov. In Sweden -- University of Lund: N. Svenningsen.
Department of Clinical Chemistry, Danderyd Hospital, Stockholm: T. Curstedt; Institute for Medical Statistics, University of Copenhagen: A. Bertelsen; Department of Radiology, Holbaek Central Hospital, Holbaek, Denmark: A. Vestergaard and E. Petersen.
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Case 30-1997: Pulmonary Interstitial Emphysema in Infancy
Fox R. B., Wright A. M., Lefebvre F., Roll C., Hanssler L., Voit T., Wilson J. M.
Extract |
Full Text
N Engl J Med 1998;
338:688-690, Mar 5, 1998.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |