Fatal C. diff infection acquired in community

Fatal C. diff infection acquired in community

Otherwise healthy woman develops sepsis

The death of a healthy woman due to community-associated Clostridium difficile underscores the ominously changing epidemiology of a pathogen rarely noted for causing serious infections beyond the hospital.

"She was in her 30s, an otherwise healthy young woman," recalls Judith O'Donnell, MD, an infectious disease physician at Hahnemann University Hospital in Philadelphia. "She was moderately overweight and had some underlying hypertension, but was a very healthy person. She came in for an elective hysterectomy and developed C. diff on the first postoperative day. We believe that she had the infection when she walked in [because she] manifested disease within 24 hours. We do not believe she acquired the disease nosocomially."

The patient was among the six cases of C. diff infection in healthy women reported in Philadelphia between January and June of 2006. All of the women had taken antibiotics in recent months, although in several cases the antibiotic use was very limited. Investigators say they found no definitive common risk factor and believe that the majority of the women likely contracted their infection outside the hospital setting.

"When we looked back at the records, we hadn't had a single healthy pregnant woman who had complications of C. difficile in the previous year, but in six months we had four," she says. "It may be this population we previously thought to be at low risk is now at increased risk for acquiring C. diff."

Though the laboratory investigation continues investigators suspect that the new epidemic strain of C. diff (ribotype 027) is the infecting agent. "This hypervirulent strain appears to be more capable of causing disease in people who otherwise wouldn't have had problems," says O'Donnell, who reported the cases recently in Toronto at the annual meeting of the Infectious Disease Society of America (ISDA).

Two cases admitted to ICU

Three of the women were pregnant, one woman had given birth three weeks previously and two women — including the fatal case — had recently undergone elective hysterectomies. According to the case reports described at IDSA, the two women who underwent elective hysterectomy received two doses of perioperative antibiotics.¹ Two other women were in their second trimester of pregnancy when admitted with diarrhea after outpatient treatment for bacterial vaginosis. Another woman was admitted three weeks after a cesarean. The final patient was 12 weeks pregnant, admitted eight days after laparoscopic cholecystectomy, for which she received one antibiotic dose. Five of the six cases developed severe diffuse colitis. Four of the six were treated with both intravenous metronidazole and oral vancomycin.

Two cases had to be admitted to the intensive care unit (ICU) for sepsis syndrome and received additional rectal vancomycin. The fatal case required a total colectomy, but continued to have sepsis syndrome and underwent resection of a large portion of ileum. Despite 14 days of therapy, ileostomy fluid still was positive for C. diff toxin and the patient died. The other ICU patient had continuous colonic vancomycin infusion and the sepsis resolved. Two of the four women who did not require ICU care responded to dual antibiotic therapy and were discharged on oral vancomycin. However, both were diagnosed with recurrence of C. diff infection and one of them had to be readmitted to the hospital.

"If this becomes more widespread in the community I think we will probably see more of these severe cases," O'Donnell tells Hospital Infection Control. "We know (ribotype 027) has been in our geographic area. We had it in our hospital, though we think actually the majority of the women in our study were community-associated. We got two of the women's stools sent off for toxin testing now, but we don't know for sure. But we presume [it is the epidemic strain] based upon what has been going on in our region."

However, epidemiologists are hampered to some degree because there is not a commercially available assay to rapidly differentiate between C. diff strains, she notes. "There is no official surveillance as far as C. diff goes," O'Donnell adds. "It is not a disease you report to your local health department, to the state, or the CDC. Here in the hospital, we catch most of the cases because everybody that has diarrhea gets a C. diff test done. We are isolating more aggressively and we have really stepped up our environmental cleaning. We are cleaning C. diff patients' rooms with a one-to-10 dilution of bleach regularly."

The six cases follow a 2005 report of similar C. diff infections in the community, which included reports of infections in peripartum women in Pennsylvania and other states.² "This phenomenon is being seen in other parts of the country," O'Donnell reports. "Talking to folks at CDC and other places, I think everybody is assuming it is because of the [new epidemic] strain."

Nearly one-fifth of cases from community

Indeed, unpublished CDC data also presented at IDSA reveal that almost one-fifth of C. diff infections in a group of six North Carolina hospitals appeared to have community onset.³

Investigators examined medical and laboratory records from January through December 2005 at six North Carolina hospitals in the Raleigh-Durham area. C. diff-associated disease (CDAD) was defined as diarrhea in a patient with a positive C. difficile toxin assay. Community-associated (CA) CDAD was defined as cases with onset in the community or within 72 hours of hospital admission, in a patient with no inpatient health care exposure within the previous two months. A total of 1,137 CDAD cases were identified. Of those, 209 (18%) occurred outside the hospital.

"I don't think people ever thought that one-fifth of cases [of C. diff] would be community-associated," says Clifford McDonald, MD, medical epidemiologist in the division of health care quality promotion at the CDC and Prevention.

The patients were not all necessarily hospitalized, but the review showed their laboratory results in hospital records. "A portion of them were hospitalized, but they were all presenting at hospitals in the sense of coming through the laboratories," he says. "A good number were presenting in ERs and being diagnosed there."

Though the retrospective review did not identify the specific strain in question, the usual suspect is again the epidemic strain emerging in communities and hospitals. "That study suggests that this epidemic strain that is causing all the havoc in hospitals is also out in the community," McDonald says. "We don't know how it is being transmitted."

The limited data on C. diff in the community suggested historical rates may be in the range of eight to 12 per 100,000 population, he added. "These rates are more equivalent to 20 to 30 per 100,000 population," he says. "That means these people are going to come into hospitals and could transmit [to other patients] if they have disease."

References

1. Gupta M, Campbell J, Lata A, et al. Severe Clostridium difficile associated disease (CDAD) in previously healthy women admitted to obstetrical and gynecological (OB/GYN) services at a tertiary care center. Abstract LB-14. Infectious Disease Society of America. Toronto; Oct. 12-15, 2006.
2. Centers for Disease Control and Prevention. Severe Clostridium difficile-associated disease in populations previously at low risk — Four states, 2005. MMWR 2005; 54(47); 1,201-1,205.
3. Kutty P, Benoit S, Woods C, et al. Emerging Clostridium difficile-associated disease in the community and the role of non-antimicrobial risk factors. Abstract LB-28. Infectious Disease Society of America. Toronto; Oct. 12-15, 2006.