Diagnostic yield of repeat sampling with immunoassay, real-time PCR, and toxigenic culture for the detection of toxigenic Clostridium difficile in an epidemic and a non-epidemic setting

Abstract

Current international guidelines lack definite conclusions regarding repeat stool sampling for the detection of toxigenic Clostridium difficile. We assessed the value of repeat sampling and compared the diagnostic yield in an epidemic to a non-epidemic setting. Consecutive fecal samples obtained during two time frames were analyzed using direct stool immunoassay toxin testing (enzyme immunoassay [EIA]), direct stool real-time PCR toxin gene testing, and toxigenic culture. Samples collected within 7 days of the initial sample were considered repeat tests. In the epidemic setting 989 patients were analyzed, and in the non-epidemic setting 1,015. In the epidemic setting 204 patients had two or more specimens included for analysis and in the non-epidemic setting 287 patients. In the epidemic setting 136 samples yielded a positive results, either by EIA or toxigenic culture; of these, 108 were positive according to EIA and 123 according to toxigenic culture. In the first test round 98 (90.7 %, 95 % CI 85.3 to 96.2), 114 (92.7 %, 88.1 to 97.3), and 126 (92.6 %, 88.3 to 97.0) positives were detected. Subsequent test rounds yielded 10 (9.3 %, 3.8 to 14.7), 9 (7.3 %, 2.7 to 11.9), and 10 (7.4 %, 3.0 to 11.7) extra positives. In the non-epidemic setting EIA, toxigenic culture and PCR detected 33, 66, and 83 positives. The three tests combined 93 detected positives. In the first test round 30 (90.9 %, 81.1 to 100.7), 63 (95.5 %, 90.4 to 110.5), 76 (91.6 %, 85.6 to 97.5), and 87 (93.5 %, 88.6 to 98.5) positives were detected. Subsequent test rounds yielded 3 (9.1 %, −0.7 to 18.9), 3 (4.5 %, −0.5 to 9.6), 7 (8.4 %, 2.5 to 14.4), and 6 (6.5 %, 1.5 to 11.4) extra positives. In conclusion, repeat testing resulted in 4.5 % to 9.3 % extra positives. No significant difference between the settings studied could be demonstrated. Repeat sampling and multimodality testing may be chosen in an outbreak situation to detect all cases, effectively controlling nosocomial spread.