Factors affecting survival in infants weighing 750g or less

Objective To assess which factors independently affect survival in infants weighing 750 g or less. Study design We reviewed the obstetric, neonatal, and placental pathology information of all non-malformed neonates with birth weight of 750 g or less from January 1998 to December 2002. Logistic regression analysis was used to control for the effect of confounding variables. A P < 0.05 was considered significant. Results Fifty nine neonates fulfilled the inclusion criteria; 30 (51%) survived the perinatal period. Surviving neonates were more frequently born after steroid administration ( P = 0.03) and from indicated delivery ( P = 0.01), had greater birth weight ( P = 0.001), gestational age at delivery ( P < 0.001), and 5-min Apgar scores of 7 or more ( P = 0.04) than those who died. There were no significant differences in placental pathology between survivors and neonates who died. Stepwise logistic regression analysis showed that gestational age ( P = 0.01), birth weight ( P = 0.004), female sex ( P = 0.03), 5-min Apgar score (0.026), and steroid administration ( P = 0.04) were independent predictors of survival. Cumulatively these five predictors explained 69% of neonatal survival. Conclusions The predictors of survival among micropremies are the same as those reported for older preterm neonates. The type of preterm delivery (spontaneous versus indicated) and placental pathology do not independently affect survival. Keywords Micropremies Neonatal survival Placental pathology 1 Introduction Although extremely low-birth-weight infants represent <1% of all live births, they account for approximately half of the perinatal deaths. Survivors in this subgroup of preterm infants also have disproportionately high rates of neonatal morbidity. When extremely preterm birth is anticipated, a reliable estimate of neonatal outcome is essential for appropriate counselling of the parents and to guide health care providers who often face difficult management decisions. Estimates of birth weight and gestational age are most commonly used for this purpose. Recently, an additional obstetric factor, which can influence perinatal survival of extremely low-birth-weight infants, has been identified in the willingness of the caregivers to perform a cesarean delivery [1] and to intervene for fetal indications. Perhaps, because of the above reasons, as well as the wide range of gestational age and birth weight reported, the perinatal survival of very low-birth-weight neonates varies widely in the series reported during the past 10 years. Even in the subgroup with birth weight <750 g, survival ranges from 32 to 67% [2] . Severe morbidity is observed in about half of survivors [2] . Our study aims at identifying independent perinatal predictors of neonatal survival at a single institution with the willingness of obstetrician and neonatologist to optimize perinatal survival. 2 Materials and methods Included were all non-malformed, liveborn, singleton babies delivered at San Gerardo Hospital, University of Milano-Bicocca between January 1998 and December 2002 with birth weight of 750 g or less. Maternal records were abstracted for demographic characteristics, antepartum events, intrapartum course, exposure to tocolytic therapy or glucocorticoids, mode of delivery, and type of prematurity (spontaneous versus indicated preterm delivery). Administration of glucocorticoids to enhance fetal maturity was considered for our analysis, only if a complete two-dose course was administered and 48 h elapsed from the initial dose of steroids. Gestational age was calculated on the basis of the last menstrual period or ultrasonographic fetal biometry if the sonographic gestational age differed from the menstrual age by >7 days in the first trimester or >14 days in the second trimester. Small for gestational age was defined as a birth weight <10th centile for sex and gestational age according to Italian standards. During the study period, the overall obstetric and neonatal management of viable babies was aggressive starting at 24 weeks, and it included the use of antenatal corticosteroids and cesarean section for fetal distress. Every liveborn infant was resuscitated when indicated by neonatal condition. Neonatal care included prophylactic continuous positive airways pressure on admission to Neonatal Intensive Care Unit, natural surfactant on intubation, high-frequency oscillatory ventilation, and nitric oxide inhalation therapy when indicated. Prophylactic steroids for prevention of bronchopulmonary dysplasia were never administered. Neonatal records were reviewed for birth weight, gender, Apgar scores, survival to discharge, and occurrence of selected neonatal morbidities. Severe morbidity was defined as occurrence of intraventricular hemorrhage or periventricular leukomalacia, grades III or IV retinopathy of prematurity, necrotizing enterocolitis, or bronchopulmonary displasia, defined as oxygen dependence at 28 days. Final infant outcome was ascertained 120 days after birth. Neurologic outcome was assessed every 3 months for the first year, at 18 months, and yearly thereafter using the Milani-Comparetti and Gidoni scoring system [3] . Histopathologic examination of the placenta was performed by observers blinded to the neonatal status. Placental lesions were classified according to standard published protocols [4] . Acute inflammatory lesions and placental vascular lesions were classified as present or absent independently from the severity or extent of the lesions. Statistical analysis was performed with Student's t -test for continuous variables and Fisher's exact test for categorical variables. Stepwise logistic regression analyses was conducted using a forward conditional criterion for the entry of variables to identify the independent perinatal predictors of overall survival. A P < 0.05 or an odds ratio (OR) with 95% confidence interval (CI) not inclusive of unity was considered significant. 3 Results Among the 349 singletons delivered at <34 weeks during the study period, 59 (14%) had birth weight <750 g and constituted the study population. The mean maternal age was 32.2 ± 4.4 years, the rate of nulliparity was 83% (49/59 cases), 25 cases (42.3%) were complicated by preeclampsia. Of the 59 neonates, 30 (51%) survived the neonatal period. Mean gestational age at delivery was 26.5 ± 1.9 weeks and birth weight was 567.6 ± 93.8 g. Table 1 displays the obstetric, and neonatal according to neonatal characteristics, survival at univariate analysis. Survival among micropremies was favourably related to indicated preterm delivery, antenatal administration of steroids, more advanced gestational age, greater birth weight, female gender, and greater Apgar scores at birth. Stratifying the cases in three groups according to birth weight, we observed a significantly higher rate of survival with increasing weight: 5/14 (36%) for neonates < 550 g, 7/21 (33%) for neonates weighting 550–649 g, and 18/24 (75%) for neonates of 650 g or more ( P = 0.01). Interestingly, mode of delivery and umbilical cord pH were not related to neonatal mortality. Placentae were available for histology examination in 93% (28/30) of surviving neonates and in 87% (25/29) of those who died. Only in 9/59 cases (including 6 survivors and 3 neonatal deaths) placental histologic examination failed to show any lesion. A detailed description of observed placental lesions is reported in Table 2 . Placental lesions were not related to neonatal mortality. All potential predictors, regardless of their significance at univariate analysis, were entered into a stepwise logistic regression model using a forward conditional criterion for the entry of variables. The analysis ( Table 3 ) showed that prediction of survival was significantly related to gestational age at delivery, birth weight, female gender, Apgar score at 5 min, and antenatal administration of steroids. Of interest, indicated preterm delivery lost significance at multivariate analysis suggesting that type of prematurity does not independently affect survival at this range of birth weight. The Nagelkerke adaptation of the Cox–Snell R 2 , which measures the cumulative percent of variability in outcome explained by the predictors, showed that 69% of neonatal survival could be explained by these five predictors (with a sensitivity of the model of 87%, specificity of 76%, positive predictive value of 79%, and negative predictive value of 85%). Among the 30 neonates who survived, 8 (27%) did not developed major morbidities, 7 had respiratory distress syndrome and anemia, 1 had only hypoglycemia, anemia, and transitory tachypnea. Major morbidities were observed in 22 infants (73%), including intraventricular hemorrhage in 8 (13%), periventricular leukomalacia in 1 (2%), broncopulmonary dysplasia in 13 (22%), necrotizing enterocolitis in 5 (8%), and retinopathy grade III in 17 (29%) (some infants experienced more than one type of morbidity). No statistical significance was detected for any obstetric, neonatal, and histopathological characteristics variables considered in predicting severe morbidities, with the exception of preeclampsia which was less frequent in cases with major morbidities [9 (41%) in cases with and 7 (88%) in cases without major morbidity, P = 0.045]. Neurodevelopmental delay was present in 6/30 surviving infants at a mean follow-up of 32.5 ± 14.9 months. It was severe in three cases and mild in three. 4 Comment We have observed that several prenatal variables are independent predictors of survival among micropremies. In our series, the chances of survival double with every additional week of GA, increase about seven-fold for every additional 100 g of birth weight, and are significantly higher for neonates weighting more than 650 g. In addition to these traditionally recognized factors, survival is significantly greater among female infants, who are about seven times as likely to survive as males; it also increases two- to three-fold for every one additional point on the Apgar score at 5 min, and it increases nine-fold with antenatal administration of a complete course of steroids. Only one other study explored the independent predictors of survival among neonates weighing less than 800 g [5] . In agreement with our results, that study found that birth weight, female sex, and antenatal steroids were factors predicting survival. Two factors differed between the two series: we found that gestational age at delivery was an independent predictor, whereas SGA was not; the series of Tyson et al. found that SGA is a predictor of survival and gestational age was not. This discrepancy may reflect differences in populations, as witnessed by differences in SGA (28% versus 54%) and in gestational age (24.8 weeks versus 26.5 weeks) between the two series. The small numbers of neonates in our series precludes any meaningful stratified analysis on major neonatal morbidity. Future series with a larger sample size may sort out potential predictors of neonatal morbidity. Of interest, race was not a predictor in the series of Tyson et al. We cannot comment on the effect of race because our population was comprised almost exclusively of Caucasians. However among less severe preterm neonates (infants weighing <1000 g), Iams and Mercer found that neonatal survival was independently predicted by African–American race [6] . This suggests that the protective effect of race is not manifest in micropremies [5] . The type of prematurity (i.e. spontaneous versus indicated) is similarly not predictive of survival as previously reported [7–10] . This observation confirms that there is no apparent survival advantage for infants born as a result of indicated preterm delivery compared with those born after spontaneous preterm labor or preterm premature rupture of membranes. In line with this finding, we report that placental pathology, which often reflects the type of prematurity [11] , does not have predictive ability for survival. Our results show that more than two-thirds of neonatal death can be explained by the five independent predictors we mentioned above. It should be noted that our study included a population delivered at a single institution by obstetricians, willing to intervene and perform a cesarean section for fetal indications; moreover, all liveborn infants were resuscitated when indicated. It has been shown that survival of extreme micropremies is often a self-fulfilling prophecy, and it increases significantly with operators determined to optimise perinatal survival. In line with this finding, in a multicenter study Tyson at al. reported the odds of adverse neonatal outcome differed significantly among the participating institutions [5] . Our results can, thus be used only by clinicians, with an attitude in neonatal care similar to ours to identify the perinatal risk profile for neonatal survival. References [1] S.F. Bottoms R.H. Paul B.M. Mercer Obstetric determinants of neonatal survival: antenatal predictors of neonatal survival and morbidity in extremely low birth weight infants Am J Obstet Gynecol 180 1999 677 682 [2] M. Hack A.F. Fanaroff Outcomes of children of extremely low birth weight and gestational age in 1990s Semin Neonatol 5 2000 80 106 [3] A. Milani-Comparetti E.A. Gidoni Routine developmental examination in normal and retarded children Dev Med Child Neurol 9 1967 931 936 [4] C. Langston C. Kaplan T. Macpherson Paractice guidelines for examination of the placenta Arch Pathol Lab Med 121 1997 449 476 [5] Tyson JE, Younes N, Verter J, Wright LL for the NICHD MFMU Network. Viability, morbidity, and resource use among newborns 501–800 g birth weight. JAMA 1996;276:1645–51. [6] Iams JD, Mercer BM for the NICHD MFMU Network. What we have learned about antenatal prediction of neonatal morbidity and mortality. Semin Perinatol 2003;27:247–52. [7] D.F. Kimberlin J.C. Hauth J. Owen Indicated versus spontaneous preterm delivery: an evaluation of neonatal morbidity among infants weighing ≤1000 g at birth Am J Obstet Gynecol 180 1999 683 689 [8] C. Bardin P. Zelkowitz A. Papageorgiou Outcome of small for gestational age and appropriate for gestational age infants born before 27 weeks of gestation Pediatrics 100 1997 2 6 [9] J. Owen S.L. Baker J.C. Hauth R.L. Goldenberg R.O. Davis R.L. Copper Is indicated or spontaneous preterm delivery more advantageous for the fetus? Am J Obstet Gynecol 163 1990 868 872 [10] E.J. Wolf A.M. Vintzileos T.S. Rosenkrantz J.F. Rodis C.M. Salafia J.G. Pezzullo Do survival and morbidity of very-low-birth-weight infants vary according to the primary pregnancy complication that results in preterm delivery? Am J Obstet Gynecol 169 1993 1233 1239 [11] C.M. Salafia L.M. Ernst J.C. Pezzullo E.J. Wolf T.S. Rosenkrantz A.M. Vintzileos The very low birth weight infant: maternal complications leading to preterm birth, placental lesions, and intrauterine growth Am J Perinatol 12 1995 106 110