A prospective study of microbial infection in stillbirths and early neonatal death.

No morphologic cause of death was found in 67.8% of 33 perinatal deaths. The mothers had experienced a previous loss in 48% of cases. Inflammation occurred in 65.6% of the cases of perinatal death compared to 4% of a control group (p less than 0.001) and in 73.1% of organism-positive cases of perinatal death compared to 7.1% of organism-positive cases of the control group (p less than 0.001). Incidence of maternal fever or prolonged membrane rupture was not statistically significant. Bacteria were present in 33.3% of the cases of perinatal death (not significant), with more pathogenic strains occurring in this group (p = 0.0028); 75.0% had inflammation compared to 0% of the control group (not significant). Genital mycoplasmas were detected in 78.8% of cases of perinatal death compared to 32.3% of control cases (p less than 0.001). Positive cultures (p = 0.0142) and elevated antibody titers in the fetus or neonate (p = 0.00052) or in the mother (p = 0.0122) occurred significantly more often than in control cases. Inflammation occurred in 78.9% of mycoplasma cases (p = 0.00032); incidences of maternal fever and prolonged membrane rupture were not significantly different. In perinatal death cases 20% had evidence of viruses, and 3.2% had evidence of chlamydia. Evidence of mixed microorganisms occurred in 46% of cases of perinatal death. However, 78.6% (11 of 14) with only one organism had Ureaplasma urealyticum (33.3% overall). Of the Ureaplasma-positive cases, 72.7% had inflammation, 45.5% had fever, and only 18.2% had prolonged membrane rupture compared to 28.6%, 0%, and 16.6%, respectively, in a negative-microorganism group with perinatal deaths. Our observations strongly support the concept that infection is a major cause of perinatal death and that genital mycoplasmas play a significant role.