Ssible to draw a Kaplan Meier analysis respectively from six GW433908G web months of therapy to clinical and immunological failure. The mean time before developing clinical failure is 54.5 months (SD 0.48, CI 95 ) and forimmunological failure is 58.6 months (SD 0.39, CI 95 ). (Figure 2, 3).Sensitivity, Specificity, and Positive and Negative Predictive ValuesSensitivity of immunological and clinical WHO GDC-0032 criteria to define treatment failure was 23.6 and 18.2 , respectively, compared with virological failure (Table 4). When combining immunological and clinical failure (patient having either one or both) and comparing with virological outcome, sensitivity was 36.4 for predicting virological failure. Specificities for immunological, clinical, and both together for predicting virological failure were 95.4 , 87.0 , and 83.5 , respectively. Positive predictive values (PPV) for immunological and clinical criteria to define virological failure were 24.5 and 8.1 , respectively. When both criteria were analyzed together, PPV was 12.3 . Negative predictive values (NPV) were 95.2 , 94.4 , and 95.4 , respectively, for immunological, clinical, and both criteria.Clinical and Immunological Criteria in HIV/AIDSFigure 3. Kaplan Meier analysis from six months of therapy to immunological failure. doi:10.1371/journal.pone.0049834.gDiscussionIn our study we found a relatively low proportion of virological failure (6.0 ) following 2010 WHO definition (VL .5000 copies/ ml) in patients on ART for more than 12 months as reported in other LMICs, supporting the fact that ART can be provided in resource-poor settings with favorable outcomes [7,11,12]. In this study, the PPV of clinical or immunological monitoring for detecting virological treatment failure was relatively low. Meeet al reported a PPV of CD4 count of 36.8 while Kaiser et al recorded PPV ranging from 9.5 ?8.7 . [11,13,14,15,16,17], which could result in patients with adequate viral suppression being incorrectly identified as failing treatment and being unnecessarily switched to second-line therapy [18]. This would not only reduce treatment options for patients but also potentially increase costs and make follow-up of patients receiving protease inhibitors more difficult. An algorithm for determining treatment failure based on clinical history, hemoglobin level, and CD4 cellTable 4. Sensitivity, specificity, and predictive values of immunological and clinical criteria in identifying virological treatment failure.Test CD4-based immunological failure WHO-defined clinical failure Combination of both CD4-based and WHO-defined clinical 1407003 failureTP 13 10FNFP 42 40 45 113 35TN 829 756Sensitivity, (95 CI) 23.6 (14.4?6.3) 18.2 (10.2?0.3) 36.4 (24.9?9.6)Specificity, (95 CI) 95.4 (93.8?6.6) 87.0 (84.6?9.1) 83.5 (80.9?5.8)Positive predictive value, (95 CI) 24.5 (14.9?7.6) 8.1 (4.5?4.3) 12.3 (8.1?8.2)Negative predictive value, (95 CI) 95.2 (93.5?6.4) 94.4 (92.6?5.8) 95.4 (93.7?6.7)TP = true positive, FN = false negative, FP = false positive, TN = true negative. Results are based in one CD4 and viral load result. doi:10.1371/journal.pone.0049834.tClinical and Immunological Criteria in HIV/AIDScount has recently been proposed, but it has not been validated in routine clinical care [19]. Low sensitivity of clinical and immunological criteria to define treatment failure highlights the need for improved methods to detect treatment failure in the absence of VL testing. In our study, only 8.1 of patients with clinical fail.Ssible to draw a Kaplan Meier analysis respectively from six months of therapy to clinical and immunological failure. The mean time before developing clinical failure is 54.5 months (SD 0.48, CI 95 ) and forimmunological failure is 58.6 months (SD 0.39, CI 95 ). (Figure 2, 3).Sensitivity, Specificity, and Positive and Negative Predictive ValuesSensitivity of immunological and clinical WHO criteria to define treatment failure was 23.6 and 18.2 , respectively, compared with virological failure (Table 4). When combining immunological and clinical failure (patient having either one or both) and comparing with virological outcome, sensitivity was 36.4 for predicting virological failure. Specificities for immunological, clinical, and both together for predicting virological failure were 95.4 , 87.0 , and 83.5 , respectively. Positive predictive values (PPV) for immunological and clinical criteria to define virological failure were 24.5 and 8.1 , respectively. When both criteria were analyzed together, PPV was 12.3 . Negative predictive values (NPV) were 95.2 , 94.4 , and 95.4 , respectively, for immunological, clinical, and both criteria.Clinical and Immunological Criteria in HIV/AIDSFigure 3. Kaplan Meier analysis from six months of therapy to immunological failure. doi:10.1371/journal.pone.0049834.gDiscussionIn our study we found a relatively low proportion of virological failure (6.0 ) following 2010 WHO definition (VL .5000 copies/ ml) in patients on ART for more than 12 months as reported in other LMICs, supporting the fact that ART can be provided in resource-poor settings with favorable outcomes [7,11,12]. In this study, the PPV of clinical or immunological monitoring for detecting virological treatment failure was relatively low. Meeet al reported a PPV of CD4 count of 36.8 while Kaiser et al recorded PPV ranging from 9.5 ?8.7 . [11,13,14,15,16,17], which could result in patients with adequate viral suppression being incorrectly identified as failing treatment and being unnecessarily switched to second-line therapy [18]. This would not only reduce treatment options for patients but also potentially increase costs and make follow-up of patients receiving protease inhibitors more difficult. An algorithm for determining treatment failure based on clinical history, hemoglobin level, and CD4 cellTable 4. Sensitivity, specificity, and predictive values of immunological and clinical criteria in identifying virological treatment failure.Test CD4-based immunological failure WHO-defined clinical failure Combination of both CD4-based and WHO-defined clinical 1407003 failureTP 13 10FNFP 42 40 45 113 35TN 829 756Sensitivity, (95 CI) 23.6 (14.4?6.3) 18.2 (10.2?0.3) 36.4 (24.9?9.6)Specificity, (95 CI) 95.4 (93.8?6.6) 87.0 (84.6?9.1) 83.5 (80.9?5.8)Positive predictive value, (95 CI) 24.5 (14.9?7.6) 8.1 (4.5?4.3) 12.3 (8.1?8.2)Negative predictive value, (95 CI) 95.2 (93.5?6.4) 94.4 (92.6?5.8) 95.4 (93.7?6.7)TP = true positive, FN = false negative, FP = false positive, TN = true negative. Results are based in one CD4 and viral load result. doi:10.1371/journal.pone.0049834.tClinical and Immunological Criteria in HIV/AIDScount has recently been proposed, but it has not been validated in routine clinical care [19]. Low sensitivity of clinical and immunological criteria to define treatment failure highlights the need for improved methods to detect treatment failure in the absence of VL testing. In our study, only 8.1 of patients with clinical fail.
