We undertook a number of meta-analyses to estimate more precisely the relationship between neonatal mortality and use of opiates in three groups of women. First, women who continued to use illicit heroin throughout pregnancy; secondly, women stabilized on methadone at the time of conception or shortly after and thirdly, women who use heroin well into pregnancy with late entry into methadone treatment, or who continued to use illicit heroin during pregnancy while receiving methadone.
Findings: The pooled estimates of the relative risks of neonatal mortality for separate heroin and methadone use were both near unity: 1.47 (95% CI 0.88-2.33) and 1.75 (95% CI 0.60-4.59), respectively. The result for heroin may be due to the inclusion in the meta-analysis of a particularly large study, which, unlike the two other smaller studies included found a relative risk near unity. When this study was excluded from the meta-analysis the pooled estimate of the relative risk of neonatal mortality for heroin use was 3.27 (95% CI 0.95-9.60). In contrast to the results for use of methadone only, the pooled relative risk associated with heroin and methadone use was 6.37 (95% CI 2.57-14.68).
Conclusions: The increased relative risk for neonatal mortality associated with women using heroin and methadone during pregnancy, compared to those stabilized on methadone, is probably due to the chaotic and high-risk life-style associated with illicit heroin use and not solely to the use of heroin and methadone per se. It is recommended tht women who use heroin well into pregnancy with late entry into methadone treatment, or who continue to use illicit heroin during pregnancy while receiving methadone, receive special attention over and above that provided to women stabilized on methadone.