Clostridium difficile is a leading cause of infectious diarrhea in hematopoietic stem cell transplant (HSCT) recipients. Asymptomatic colonization of the gastrointestinal tract occurs before development of C. difficile infection (CDI). This prospective study examines the rates, risk factors, and outcomes of colonization with toxigenic and nontoxigenic strains of C. difficile in HSCT patients. This 18-month study was conducted in the HSCT unit at the Karmanos Cancer Center and Wayne State University in Detroit. Stool samples from the patients who consented for the study were taken at admission and weekly until discharge. Anaerobic culture for C. difficile and identification of toxigenic strains by PCR were performed on the stool samples. Demographic information and clinical and laboratory data were collected. Of the 150 patients included in the study, 29% were colonized with C. difficile at admission; 12% with a toxigenic strain and 17% with a nontoxigenic strain. Over a 90-day follow-up, 12 of 44 (26%) patients colonized with any C. difficile strain at admission developed CDI compared with 13 of 106 (12%) of patients not colonized (odds ratio [OR], 2.70; 95% confidence interval [95% CI], 1.11 to 6.48; P = .025). Eleven of 18 (61%) patients colonized with the toxigenic strain and 1 of 26 (4%) of those colonized with nontoxigenic strain developed CDI (OR, 39.30; 95% CI, 4.30 to 359.0; P < .001) at a median of 12 days. On univariate and multivariate analyses, none of the traditional factors associated with high risk for C. difficile colonization or CDI were found to be significant. Recurrent CDI occurred in 28% of cases. Asymptomatic colonization with C. difficile at admission was high in our HSCT population. Colonization with toxigenic C. difficile was predictive of CDI, whereas colonization with a nontoxigenic C. difficile appeared protective. These findings may have implications for infection control strategies and for novel approaches for the prevention and preemptive treatment of CDI in the HSCT patient population.

Staphylococcus aureus meningitis is a challenging disease and little is known about its epidemiology. There are no established management guidelines. We retrospectively reviewed the clinical information, bacteriologic data, and outcomes of all 33 patients with cerebrospinal fluid (CSF) cultures positive for S aureus seen at a single urban teaching hospital from 1999 to 2008. Pulsed-field gel electrophoresis (PFGE) and polymerase chain reaction for staphylococcal cassette chromosome mec (SCCmec), accessory gene regulator (agr) typing, and Panton-Valentine leukocidin (PVL) loci were done on methicillin-resistant S aureus (MRSA) CSF isolates starting in 2005. S aureus caused 12 (36%) cases of postoperative and 21 (64%) cases of hematogenous meningitis. MRSA isolates were found in 6 (50%) cases of postoperative and 10 (48%) cases of hematogenous meningitis. Twelve (75%) of the 16 MRSA infections occurred in the last 5 years of the study. Hematogenous meningitis was associated with older age (p = 0.04), injection drug use (p < 0.01), community-acquired infection (p < 0.01), underlying disease (p = 0.01), staphylococcal infection outside the central nervous system (p = 0.01), altered mental status (p = 0.02), fever (p = 0.01), septic shock (p = 0.03), and bacteremia (p < 0.01). The analysis of the 9 MRSA isolates showed 3 PFGE types: 3 USA100 (33%), 5 USA300 (56%), and 1 USAnot100-1100 (11%). For SCCmec typing, there were 2 (22%) type II and 7 (78%) type IV. All USA300 strains were SCCmec IVa. For agr typing, there were 5 (56%) type I and 4 (44%) type II. Three isolates (33%) were positive for the PVL gene and were USA300 strains. Most patients received nafcillin or vancomycin with or without rifampin or trimethoprim/sulfamethoxazole for a mean period of 17 days (range, 1-42 d). Overall mortality was 36%, and it was associated with community-acquired infection (p = 0.02). Postoperative and hematogenous S aureus meningitis are distinct clinical syndromes. S aureus hematogenous meningitis has devastating clinical consequences and elevated mortality rates, especially if it is acquired in the community. The incidence of MRSA meningitis increased over the last 5 years of the study. Treatment of choice is nafcillin for methicillin-sensitive strains and vancomycin for MRSA strains. The addition of trimethoprim/sulfamethoxazole or rifampin to vancomycin is recommended in severe cases and community-acquired MRSA infections. Linezolid is also a good option due to its good CSF penetration and favorable case reports. The mortality rate is higher in infections acquired in the community.