Article Figures & Data
Figures
- FIG 1
CryptoPS semiquantitative assay. Twenty microliters of a blood specimen is added to the collection well of the cassette, followed by 3 drops of sample diluent, with the result read at 10 min. (A) Negative result (control band positive); (B) :Low-titer positive result (T1 and control bands positive); (C) High-titer positive result (T1, T2, and control bands positive). Patient initials and study identification codes written on the bottoms of the test strips have been concealed to ensure the anonymity of the research participants.
- FIG 2
Discordant CryptoPS and IMMY LFA results. (A) CryptoPS “false-positive” results with secondary testing using the IMMY EIA; 29/30 samples tested negative on the IMMY EIA at an optical density below the positive cutoff, and 1 sample tested positive at an optical density of 0.625. All samples tested negative using the Meridian EIA. (B) CryptoPS “false-negative” results with IMMY LFA dilutional titer and IMMY EIA optical density; 16/16 samples had an IMMY LFA titer of ≤1:160, and 3 tested negative on the IMMY EIA at an optical density below the positive cutoff, all 3 were IMMY LFA positive only in undiluted samples, indicating very low titers.
- FIG 3
Comparison of CryptoPS titer results with dilutional titers obtained using the IMMY LFA. The violin plots show the distribution of dilutional CrAg titers in CryptoPS-negative, T1 (low-titer), and T2 (high-titer) samples, comprising 916 prospectively screened samples and 141 additional archived CrAg-positive plasma samples. Twenty-five CryptoPS-negative samples (16 from the prospective cohort and 9 archived samples) were positive on IMMY testing, all at titers of ≤1:160. Twenty-nine CryptoPS T1-positive samples and 1 CryptoPS T2-positive sample were IMMY LFA negative. Median titers in the T1 and T2 groups with interquartile ranges are given. Note that in the treatment trial from which some of the archived samples were taken, the maximum dilutional titer measured was 1:2,560; hence, some titers at this level may have been higher if further dilutions had been performed.
Tables
- TABLE 1
Baseline characteristics of study participantsa
↵a ART, antiretroviral therapy; CrAg, cryptococcal antigen; IQR, interquartile range; LFA, lateral flow assay.
↵b Except where otherwise indicated, values are given as the median (IQR) or as a percentage (number of individuals with the characteristic/total number).
↵c A total of 916 samples from 870 unique patients were tested; 46 individuals had repeat samples.
↵d Age, sex, and prior cryptococcal meningitis data are reported for the 870 individuals in the CrAg screening cohort at the time of their first CrAg test.
↵e Data were missing for two individuals.
↵f CD4 cell counts, IMMY CrAg LFA results, and ART status at the time of testing (some patients were tested twice during screening period) are shown; of 226 patients not on ART at date of CD4 testing, 17% (39) had defaulted from ART and 83% (187) had no history of ART use.
↵g CrAg results obtained using the IMMY CrAg lateral flow assay.
- TABLE 2
Sensitivity, specificity, and positive, and negative predictive values of the CryptoPS semiquantitative assay versus the conventional IMMY LFAa
↵a Tests were performed on whole-blood samples. PPV, positive predictive value; NPV, negative predictive value; CI, confidence interval; LFA, lateral flow assay.
- TABLE 3
Clinical characteristics and outcomes of patients with false-positive CryptoPS test results on whole-blood specimensa
Supplemental material
- Supplemental file 1 -
Tables S1 and S2
PDF, 46K
- Supplemental file 1 -
