Older studies had reported the incidence of “standard puerperal morbidity,” which was defined by the U.S. Joint Committee on Maternal Welfare as “a temperature of 100.4°F (38.0°C), the temperature to occur in any two of the first 10 days postpartum, exclusive of the first 24 hours, and to be taken by mouth by a standard technique at least four times daily.” Yet the full criteria of the original definition can no longer be applied because of early patient discharge practices. In addition, many infected patients respond to antibiotics so quickly that they do not meet the temperature criteria for standard morbidity.

Low-grade fever (≥100.4°F) or isolated higher temperature elevations occur commonly in the puerperium and often resolve spontaneously, especially after vaginal delivery. Sources of infection in the puerperal period are genital tract infection (endometritis, septic pelvic thrombophlebitis, and abscess), urinary tract infection, mastitis, breast abscess, and complications of anesthesia.

At present, the overall rate of postpartum uterine infection may be estimated at 1% to 2% after vaginal delivery and up to 27% after cesarean delivery, even when prophylactic antibodies are used. Although the absolute risk of death from infection is small among postpartum women, sepsis remains a common cause of maternal death in the United States. In the period 1991 to 1997, infection accounted for 12.6% (242/929) of deaths after live birth in the United States. Infection ranked behind embolism (21.4%), hypertensive disorders (19.4%), and hemorrhage (13.4%) as causes of maternal death. Ranking after infection were cardiomyopathy (9.7%), cerebrovascular accidents (5.3%), and anesthesia complications (1.8%) (see Table 20.1). The pregnancy-related mortality ratio in 1991 to 1997 was 11.5 per 100,000 live-born infants in the United States.

Older studies had indicated that the deaths were preventable in more than two thirds of the cases, with both patient and medical personnel sharing responsibility equally. Patient responsibility consisted primarily of delay in seeking medical care (for example, after prolonged rupture of the membranes [PROM]); physician and hospital responsibility consisted of inadequate evaluation of symptoms, incorrect diagnosis, failure to hospitalize earlier, and inadequate institution of antibiotic therapy.

Epidemic infections occur infrequently on maternity services. Lethal outbreaks of β-hemolytic streptococcal infection occurred as late as 1927, and a number of nonlethal epidemics of streptococcal infection have been reported since 1965. These epidemics were all caused by group A streptococci (usually of one type) and involved from 2 to 20 mothers, 0 to 11 newborn infants, and many of the hospital staff. Although there were no maternal or neonatal deaths, a number of the patients were severely ill. There have been no reported epidemics of puerperal infection due to organisms other than group A streptococci.

The most common cause of puerperal fever is uterine infection; the infection is variously called endometritis, endoparametritis, or simply metritis. Criteria for endomyometritis include fever, uterine tenderness, purulent or foul lochia, peripheral leukocytosis, and exclusion of another infected site. Nonspecific signs and symptoms, such as malaise, abdominal pain, chills, and tachycardia, may be present. However, most patients do not have this complete clinical picture. Many febrile patients with group A or B streptococcal bacteremia have no localizing signs early in the course of the illness. In the vast majority of cases of uterine infection, presenting signs and symptoms develop within the first 5 days after delivery.

When endometritis is suspected, we maintain that an aerobic cervical culture be performed. (There is no available technique to obtain an endometrial specimen free of cervical contamination.) Even though broad-spectrum therapy is used, this culture may identify pathogens that require specific therapy. For example, identification of group A streptococcus should lead to isolation of the patient and should be reported to physicians in the nursery. Isolation of group B streptococcus (GBS) or Neisseria gonorrhoeae should be reported to physicians in the nursery. In addition, there may be isolates that will help direct subsequent antibiotic therapy in case of initial antibiotic failure. Examples include enterococci in initial therapy with a cephalosporin or clindamycin plus gentamicin. Then, from a surveillance viewpoint, it is necessary to determine whether kinds of organisms or their antimicrobial susceptibilities are shifting. Results of these cultures at a given institution also allow for development of specific antibiotic strategies. The commonly used algorithm of not performing a pelvic examination, not performing cultures of the pelvis, and empirically prescribing antibiotics in patients with postpartum infection is unwise. As was recently noted in an editorial, this empiricism “will lead to a generation of specialists in women’s
health who fail to appreciate a logical approach to the evaluation of women with fever and/or infections. It leads not only to empiric diagnosis, but empiric therapy, without the opportunity to document disease. It allows for no future planning in case of emerging resistant organisms.”

About 10% to 20% of patients with endometritis have bacteremia, and blood cultures should be obtained for women with significant puerperal fever. Isolation of an organism from the blood does not imply that this organism, by itself, is responsible for the infection; therefore, antibiotic therapy directed solely against the isolate thus identified might be inadequate.