Further evidence that infection is an infrequent cause of first trimester spontaneous abortion.

A previous cohort study found no clinical evidence that infection occurred more often in subjects experiencing pregnancy loss compared with those experiencing successful pregnancy [Simpson et al. (1996) Hum. Reprod., 11, 668-672]. Given these surprising findings, we conducted a similar analysis on another cohort also followed prospectively. Using couples practising natural family planning for conception or contraception, information on clinical evidence of infection was gathered beginning with week 5 of gestation. Information on fever and signs of overt infection was specifically sought by interview and physical examination. Frequencies of urinary, vaginal and other infections in subjects experiencing pregnancy loss were 11.1, 9.5 and 8.7% respectively, not significantly different from rates in subjects having liveborns (10.1, 10.2 and 10.3% respectively). Thus, no association between clinical infection and early pregnancy loss (< or = 16 weeks) was observed. Cohort studies utilizing biologically based assays are awaited because extant data do not provide evidence that clinically evident infections play major roles in first trimester pregnancy losses.