Risk factors for mortality in patients with anaerobic bacteremia.

Risk factors for mortality in anaerobic bacteremia have been incompletely defined. The aims of the present study were to determine clinical significance by pathogen for a broad range of obligate anaerobic organisms isolated from blood, and to define the factors independently associated with mortality among those with clinically significant bacteremia. All patients who had anaerobic bacteria isolated from blood over a 19-month period (from 1 September 1998 to 1 April 2000) at two urban teaching hospitals were included in this study. Each case was analyzed for clinical significance by means of a retrospective medical record review using predetermined definitions. Information was collected on a broad range of clinical and microbiological factors, which were evaluated for their association with mortality using a Cox proportional hazards model. Among 166 patients with obligate anaerobic bacteria isolated from blood, 73 (44%) were deemed to have clinically significant bacteremia. Clinical significance ranged from 0% (0/53) for Propionibacterium spp. to 96% (43/45) for Bacteroides spp. The crude mortality rate in patients with clinically significant anaerobic bacteremia was 25% (18/73). Mortality was significantly associated with age, polymicrobial infection, and underlying heart, kidney or liver disease in univariate analysis. Only the presence of liver disease (relative risk, 5.3; 95% confidence interval, 1.7-16.0; P=0.003) and patient age (relative risk, 1.06/y; 95% confidence interval, 1.0-1.1; P=0.005) remained significant in multivariate analysis. Among patients with anaerobic bacteremia, clinical significance varies markedly by pathogen and mortality is independently associated with age and underlying liver disease.

Emergence of ganciclovir-resistant cytomegalovirus disease among recipients of solid-organ transplants.

BACKGROUND: Concerns have been raised about emergence of ganciclovir resistance as a result of the advent of both routine oral ganciclovir prophylaxis and highly potent immunosuppression. We retrospectively assessed the occurrence of ganciclovir-resistant cytomegalovirus disease among transplant recipients who had received oral ganciclovir prophylaxis and highly potent immunosuppression. METHODS: We studied 240 recipients of liver, kidney, or pancreas transplants. Antiviral susceptibility testing of blood cytomegaloviral isolates was done when patients failed to respond to intravenous ganciclovir treatment for symptomatic cytomegalovirus infection. Portions of the UL97 gene associated with ganciclovir resistance were sequenced in cytomegalovirus isolates with phenotypic resistance to ganciclovir. FINDINGS: Ganciclovir-resistant cytomegalovirus disease developed in five (7%) of 67 seronegative recipients of cytomegalovirus-seropositive organs (D+/R-) compared with none of 173 seropositive recipients (p=0.002). Among the 25 (10.4%) patients who developed cytomegalovirus disease within 1 year after transplantation, five had ganciclovir-resistant cytomegalovirus disease. Among D+/R-transplant recipients, ganciclovir-resistant cytomegalovirus disease was more common among the group receiving the most potent immunosuppression--ie, recipients of kidney and pancreas or pancreas alone (four of 19) compared with all other transplant recipients (one of 48, p=0.02). Ganciclovir-resistant cytomegalovirus disease was diagnosed at a median of 10 months after transplantation (range 7-12) after lengthened exposure to ganciclovir, was associated with previously described mutations of the UL97 gene, and led to serious clinical complications. INTERPRETATION: Ganciclovir-resistant cytomegalovirus is an important cause of late morbidity among D+/R- transplant recipients who have had lengthened exposure to ganciclovir and have received highly potent immunosuppression. Strategies to reduce this complication, especially among D+/R- patients, are warranted.