BackgroundThere is a need to have uniformed reporting of perinatal mortality for births following assisted reproductive technology (ART) treatment to enable international comparison and benchmarking of ART practice.MethodsThe Australian and New Zealand Assisted Reproduction Database was used in this study. Births of ≥ 20 weeks gestation and/or ≥ 400 grams of birth weight following embryos transfer cycles in Australia and New Zealand during the period 2004 to 2008 were included. Differences in the mortality rates by different perinatal periods from a gestational age cutoff of ≥ 20, ≥ 22, ≥ 24, or ≥ 28 weeks (wks) to a neonatal period cutoff of either < 7 or < 28 days after birth were assessed. Crude and specific (number of embryos transferred and plurality) rates of perinatal mortality were calculated for selected gestational and neonatal periods.ResultsWhen the perinatal period is defined as ≥ 20 wks gestation to < 28 days after birth, the perinatal mortality rate (PMR) was 16.1 per 1000 births (n = 630). A progressive contraction of the gestational age groups resulted in marked reductions in the PMR for deaths at < 28 days (22 wks 11.0; 24 wks 7.7; 28 wks 5.6); and similarly for deaths at < 7 days (20 wks 15.6, 22 wks 10.5; 24 wks 7.3; 28 wks 5.3). In contrast, a contraction of the perinatal period from < 28 to < 7 days after birth only marginally reduced the PMR from 16.2 to 15.6 per 1000 births which was consistent across all gestational ages.The PMR for single embryo transfer (SET) births (≥ 20 weeks gestation to < 7 days post-birth) was significantly lower (12.8 per 1000 SET births) compared to double embryo transfer (DET) births (PMR 18.3 per 1000 DET births; p < 0.001, Fisher’s Exact Test). Similarly, the PMR for SET births (≥ 22 weeks gestation to < 7 days post-birth) was significantly lower (8.8 per 1000 SET births, p < 0.001, Fisher’s Exact Test) when compared to DET births (12.2 per 1000 DET births). The highest PMR (50.5 per 1000 SET births, 95% CI 36.5-64.5) was for twins following SET births (≥ 20 weeks gestation to < 7 days post-birth) compared to twins following DET (23.9 per 1000 DET births, 95% CI 20.8-27.1).ConclusionReporting of perinatal mortality of ART births is an essential component of quality ART practice. This should include measures that monitor the impact on perinatal mortality of multiple embryo transfer. We recommend that reporting of perinatal deaths following ART treatment, should be stratified for three gestation-specific perinatal periods of ≥ 20, ≥ 22 and ≥ 28 completed weeks to < 7 days post-birth; and include plurality specific rates by SET and DET. This would provide a valuable international evidence-base of PMR for use in evaluating ART policy, practice and new research.