Table 2.

ChromID MRSA agar culture, LightCycler MRSA Advanced Test, and in-house duplex real-time PCR assay performance in detection of MRSA

Assay	No. of samples with the indicated resulta						Performance (% [95% CI])h
	Positive	Negative	True positiveb	False positive	False negative	True negativec	Sensitivity	Specificity	Positive predictive value	Negative predictive value
Direct ChromID MRSA agar culture	147	1,099	147	0	27	1,072	84.5 (78.3–89.2)	100 (99.7–100.0)	100 (97.8–100.0)	97.5 (96.4–98.3)
LightCycler MRSA Advanced Test	156d	1,090	145e	11f	29	1,061	83.3 (77.0–88.5)	99.0 (98.2–99.5)	92.9 (87.7–96.1)	97.3 (96.2–98.2)
In-house duplex real-time PCR assay	148	1,098g	133	15	41	1,057	76.4 (69.6–82.2)	98.6 (97.7–99.2)	89.9 (83.9–93.9)	96.3 (95.0–97.3)

^aA total of 1,246 samples were used to evaluate the performance of culture and real-time PCR assays for the detection of MRSA nasal colonization.

^bA total of 174 samples were considered true positive for MRSA based on the overall results of cultures and real-time PCR assays (also see Table 1).

^cA total of 1,072 samples were considered true negative for MRSA based on the overall results of cultures and real-time PCR assays (also see Table 1).

^dAmong 156 positive samples from the LightCycler test, 92 showed T_m values of 59°C, whereas 64 showed T_m values of 55°C.

^eAmong 145 true-positive samples indicated by the LightCycler test, 85 showed T_m values of 59°C, whereas 60 showed T_m values of 55°C.

^fThe 11 false-positive samples from the LightCycler test included 7 samples with T_m values of 59°C and 4 samples with T_m values of 55°C.

^gA total of 18 samples with positive amplification for both sau and mecA genes were considered negative in the in-house duplex PCR assay because their ΔCp values were >3 cycles.

^hResults were determined as percentages and 95% confidence intervals (CIs) by an adjusted Wald method.