A new study to be published in the Nordic journal Acta Obstetricia et Gynecologica Scandinavica compares women’s risk of postpartum infections after vaginal birth or cesarean section, and found significantly increased odds of infection with c-section – even when adjusted for factors such as parity, maternal age, smoking, diabetes, and chronic diseases. The study makes an important addition to existing evidence on the topic, which has been considered inadequate to determine real differences in infection rates by mode of delivery.
The investigators looked at rates of wound, bloodstream and urinary tract infection in the 30 days after birth among 32,468 Danish women during the years 2001-2005. Among the 2.8% of women who developed an infection (most were wound infections), they found increased odds of infection with c-section (OR 4.71, 95% CI 4.08-5.43) as compared with vaginal birth, and a greater risk with emergency as opposed to elective c-section (despite the use of prophylactic antibiotics in emergency cases). They also report that 77% of the UTIs and wound infections were diagnosed after the women had been discharged from the hospital.
The authors conclude: “The risk of postpartum infection seems to be nearly five-fold increased after CS compared with vaginal birth. This may be of concern since the prevalence of CS is increasing.” At 19% for the study period, the Danish rate of c-section was considerably lower than the 31.1% rate in the United States as of 2006.
Amy Romano has covered this topic nicely at Science & Sensibility, describing the previous infection-related findings from an AHRQ systematic review and NIH State-of-the-Science conference report on “maternal request” cesarean, including the limitations of the evidence available for those reports. Notably, the AHRQ report cited one randomized study (the Term Breech Trial) that found no significant difference in infection rates by vaginal birth vs. cesarean delivery, but concluded that the evidence on the topic was weak/limited, and that “These limitations preclude our ability to make conclusive assessments of the maternal infection literature.” Romano also provides an excellent explanation of what a finding of “no significant difference” might actually mean, from “there really is no difference” to issues of study size/design that may obscure real differences.
Despite the lack of adequate previous evidence in the medical literature as explained by Romano and AHRQ, the findings of the new paper echo those reported by women in Childbirth Connection’s New Mothers Speak Out report, in which about 5% of women who had a vaginal birth reported an associated perineal infection, while 19% of women who had c-sections reported an infection associated with that surgery.
On a related note, the new study indirectly highlights one potential benefit of a national health system, as Denmark has – it allows the creation of large registries of health, birth, hospitalization, and other data and access to information on very large numbers of individuals who can be tracked over time for research purposes.
In this case, it allowed researchers to look at a population of more than 32,000 (compared with about 2,000 in the Term Breech study Romano mentions), and, most importantly, to follow the women’s outcomes beyond the initial hospitalization for birth. Such follow-up is generally lacking in U.S. studies because of the difficulties in tracking women post-discharge. Had the Danish study ended when the women left the hospital, it would have failed to include the 77% of all infections that occured after the women had been discharged.