A CONTROLLED STUDY OF THE INFLUENCE ON THE NEWBORN OF PROLONGED PREMATURE RUPTURE OF THE AMNIOTIC MEMBRANES AND/OR INFECTION IN THE MOTHER

A controlled study of the influence of prolonged (6 hours or more) premature rupture of amniotic membranes and/or infection in 358 mothers (equally divided between study and control groups) is reported. Bacteriologic studies of the maternal nose and throat swabs were similar in both study and control groups, and the findings did not appear to influence the subsequent course of mother or infant. Maternal blood cultures showed growth in 27% of the study mothers, and in 21% of the control group; the types of organisms appeared similar in both groups and the findings could not be correlated with the infants9 subsequent courses. The cord-blood cultures were positive in 47% of the study group and in 37% of the control group. Clinical sepsis (as defined above) was seen in 31% of the infants delivered of mothers with prolonged premature rupture of amniotic sac, and confirmed bacteriologically in 8%. In the control group of mothers, clinical sepsis was observed in 5%, and proven bacteriologically in 1%. These differences between the study and control mothers are statistically significant. While there were no deaths in the control group, there were seven deaths (4%) in the study group—two stillbirths and five neonatal deaths. Except for one of the stillborns in whom pneumonia was found at autopsy, and one premature infant who died of extreme immaturity, sepsis accounted for the deaths. The umbilical cords were examined histologically in 138 of the infants born after prolonged premature rupture of the membranes, and in 71 of the control group. Inflammatory changes in the blood vessels of the umbilical cords were present in 28% of the infants born of mothers with prematurely ruptured amniotic membranes, and in 6% of the infants in the control group; approximately one-quarter of the former group of infants showed clinical signs of sepsis, while none of the latter became ill. Of the 67 infants whose umbilical cord vessels were normal, 7 (10%) became ill and 2 succumbed to their illness; in addition, there was a stillborn whose umbilical cord vessels were normal, but who showed a pneumonia at postmortem. Of 24 infants showing the combination of positive umbilical cord-blood culture and umbilical cord vasculitis, 5 became ill and 2 of these were subsequently proven to have sepsis; or stated in other terms, only 13% of those with subsequent bacteriologic evidence of sepsis showed this combination. (The remaining cases with proven sepsis had either a positive cordblood culture or umbilical cord vasculitis, except in the single instance of viral (Coxsackie) sepsis where both the cord-blood culture was negative for bacteria and the cord vessels were normal.) Fifteen of the 56 infants with clinical sepsis had histopathologic examination of their umbilical cords; the combination of positive cordblood culture and umbilical vasculitis was observed in 5 (33%), 3 of whom succumbed, accounting for 60% of the deaths (excluding the 2 stillborns). Thus, in those with clinical evidence of sepsis, the additional presence of the combination foretold a grave prognosis. Of course, the numbers are small and rigid conclusions are unwarranted. Additional data are needed and are being accumulated. The incidence of premature birth was approximately six times greater in the mothers with prolonged, premature rupture of membranes than in those mothers whose membranes ruptured at the time of delivery or shortly before. The incidence of premature births in the former group of mothers was approximately two times the over-all incidence of premature births (10-12%) at the Boston City Hospital. In none of the mothers with premature birth was there a bacteriuria at the time of delivery. This finding is in contrast to that reported by Kass et al., who, unfortunately, made no correlation of the bacteriuria found with premature rupture of membranes. The five mothers in the present study who had a significant bacteriuria (>100,000 organisms/ml of urine) at the time of delivery gave birth to full-term infants, and all infants did well clinically. As to the antimicrobial treatment of infants (whether term or premature) in the present series (Table X), all who showed evidence of umbilical cord vasculitis, yet were free from symptoms and not given antimicrobial therapy, did just as well as those who had normal cords but displayed symptoms and were treated. There is no satisfactory evidence to indicate that routine treatment of infants delivered after prolonged premature rupture of membranes is necessary nor desirable. It is our present recommendation that all infants born after prolonged premature rupture of membranes, or of febrile mothers, be labelled suspect and followed closely for the first "danger sign" of possible infection. Once this appears, culture material from nose, throat, blood, urine, and spinal fluid should be obtained, and treatment begun with appropriate antibiotics in proper dosage, always bearing in mind the limited ability of infants, especially premature ones, to conjugate and excrete certain drugs. In infants delivered of febrile mothers it would be wiser, perhaps, to collect all culture material before signs appear; this information might then be available without undue delay should such an infant, who is at higher risk, show signs of illness. Treatment may be altered after results from the laboratory, including sensitivity tests, are available. Under this management the outcome in terms of cure and survival has, in this series, been favorable.