Advanced HIV Disease at Enrolment in HIV Care: Trends and Associated Factors over a Ten Year Period in Cambodia.

BACKGROUND: Early HIV diagnosis and enrolment in care is needed to achieve early antiretroviral treatment (ART) initiation. Studies on HIV disease stage at enrolment in care from Asian countries are limited. We evaluated trends in and factors associated with late HIV disease presentation over a ten-year period in the largest ART center in Cambodia. METHODS: We conducted a retrospective analysis of program data including all ARV-naïve adults (> 18 years old) enrolling into HIV care from March 2003-December 2013 in a non-governmental hospital in Phnom Penh, Cambodia. We calculated the proportion presenting with advanced stage HIV disease (WHO clinical stage IV or CD4 cell count <100 cells/µL) and the probability of ART initiation by six months after enrolment. Factors associated with late presentation were determined using multivariate logistic regression. RESULTS: From 2003-2013, a total of 5642 HIV-infected patients enrolled in HIV care. The proportion of late presenters decreased from 67% in 2003 to 44% in 2009 and 41% in 2013; a temporary increase to 52% occurred in 2011 coinciding with logistical/budgetary constraints at the national program level. Median CD4 counts increased from 32 cells/µL (IQR 11-127) in 2003 to 239 cells/µL (IQR 63-291) in 2013. Older age and male sex were associated with late presentation across the ten-year period. The probability of ART initiation by six months after enrolment increased from 22.6% in 2003-2006 to 79.9% in 2011-2013. CONCLUSION: Although a gradual improvement was observed over time, a large proportion of patients still enroll late, particularly older or male patients. Interventions to achieve early HIV testing and efficient linkage to care are warranted.

Community-based active tuberculosis case finding in poor urban settlements of Phnom Penh, Cambodia: a feasible and effective strategy.

BACKGROUND: In light of the limitations of the current case finding strategies and the global urgency to improve tuberculosis (TB) case-detection, a renewed interest in active case finding (ACF) has risen. The WHO calls for more evidence on innovative ways of TB screening, especially from low-income countries, to inform global guideline development. We aimed to assess the feasibility of community-based ACF for TB among the urban poor in Cambodia and determine its impact on case detection, treatment uptake and outcome. METHODS: Between 9/2/2012-31/3/2013 the Sihanouk Hospital Center of HOPE conducted a door-to-door survey for TB in deprived communities of Phnom Penh. TB workers and community health volunteers performed symptom screening, collected sputum and facilitated specimen transport to the laboratories. Fluorescence microscopy was introduced at three referral hospitals. The GeneXpert MTB/RIF assay (Xpert) was performed at tertiary level for individuals at increased risk of HIV-associated, drug-resistant or smear-negative TB. Mobile phone/short message system (SMS) was used for same-day issuing of positive results. TB workers contacted diagnosed patients and referred them for care at their local health centre. RESULTS: In 14 months, we screened 315.874 individuals; we identified 12.201 aged ≥ 15 years with symptoms suggestive of TB; 84% provided sputum. We diagnosed 783, including 737 bacteriologically confirmed, TB cases. Xpert testing yielded 41% and 48% additional diagnoses among presumptive HIV-associated and multidrug-resistant TB cases, respectively. The median time from sputum collection to notification (by SMS) of the first positive (microscopy or Xpert) result was 3 days (IQR 2-6). Over 94% commenced TB treatment and 81% successfully completed it. CONCLUSION: Our findings suggest that among the urban poor ACF for TB, using a sensitive symptom screen followed by smear-microscopy and targeted Xpert, contributed to improved case detection of drug-susceptible and drug-resistant TB, shortening the diagnostic delay, and successfully bringing patients into care.

Timing of antiretroviral therapy in Cambodian hospital after diagnosis of tuberculosis: impact of revised WHO guidelines.

OBJECTIVE: To determine if implementation of 2010 World Health Organization (WHO) guidelines on antiretroviral therapy (ART) initiation reduced delay from tuberculosis diagnosis to initiation of ART in a Cambodian urban hospital. METHODS: A retrospective cohort study was conducted in a nongovernmental hospital in Phnom Penh that followed new WHO guidelines in patients with human immunodeficiency virus (HIV) and tuberculosis. All ART-naïve, HIV-positive patients initiated on antituberculosis treatment over the 18 months before and after guideline implementation were included. A competing risk regression model was used. FINDINGS: After implementation of the 2010 WHO guidelines, 190 HIV-positive patients with tuberculosis were identified: 53% males; median age, 38 years; median baseline CD4+ T-lymphocyte (CD4+ cell) count, 43 cells/µL. Before implementation, 262 patients were identified; 56% males; median age, 36 years; median baseline CD4+ cell count, 59 cells/µL. With baseline CD4+ cell counts ≤ 50 cells/µL, median delay to ART declined from 5.8 weeks (interquartile range, IQR: 3.7-9.0) before to 3.0 weeks (IQR: 2.1-4.4) after implementation (P < 0.001); with baseline CD4+ cell counts > 50 cells/µL, delay dropped from 7.0 (IQR: 5.3-11.3) to 3.6 (IQR: 2.9-5.3) weeks (P < 0.001). The probability of ART initiation within 4 and 8 weeks after tuberculosis diagnosis rose from 23% and 65%, respectively, before implementation, to 62% and 90% after implementation. A non-significant increase in 6-month retention and antiretroviral substitution was seen after implementation. CONCLUSION: Implementation of 2010 WHO recommendations in a routine clinical setting shortens delay to ART. Larger studies with longer follow-up are needed to assess impact on patient outcomes.

Incidence and risk factors for tuberculosis in HIV-infected patients while on antiretroviral treatment in Cambodia.

BACKGROUND: Given the lack of detailed studies on tuberculosis (TB) in patients on antiretroviral treatment (ART) in South-East Asia, we aimed to determine the incidence and risk factors for early (after ≤6 months of ART) and late (after >6 months of ART) incident TB in Cambodia. METHODS: We conducted a retrospective analysis of all patients started on ART at a non-governmental hospital in Phnom Penh (March 2003-December 2010). TB diagnosis was performed according to WHO algorithms. Risk factor analysis was performed using multivariate Cox regression modeling. RESULTS: Overall, 2984 patients started ART. The median baseline CD4 count was 89 cells µl(-1) (IQR 25-209), median age 34 years (IQR 29-40). Fifty-three percent of the patients were female. Median follow-up time on ART was 2.4 years. In addition to 932 (31.2%) patients already on TB treatment at ART initiation, 313 (10.5%) developed TB, with an overall incidence rate of 3.9/100 patient-years. Of those developing TB, 179 (6.0%) patients were diagnosed with early TB and 134 (4.5%) with late TB, corresponding with a rate of 13.5 and 2.0 per 100 patient-years respectively. Risk factors for early TB included low body mass index, low baseline CD4 count and low hemoglobin levels. Low on-treatment CD4 counts and hemoglobin levels, being underweight while on ART and prevalent TB were identified as risk factors for late TB. CONCLUSION: The incidence of early TB was high, and predominantly associated with advanced HIV progression markers. Earlier ART initiation and enhanced TB screening prior to and after ART initiation is warranted. Late TB amounts to almost half of the total TB burden, meriting specific preventive and diagnostic approaches.

Evaluation of the 2007 WHO guideline to improve the diagnosis of tuberculosis in ambulatory HIV-positive adults.

BACKGROUND: In 2007 WHO issued a guideline to improve the diagnosis of smear-negative and extrapulmonary tuberculosis (EPTB) in HIV-positive patients. This guideline relies heavily on the acceptance of HIV-testing and availability of chest X-rays. METHODS AND FINDINGS: Cohort study of TB suspects in four tuberculosis (TB) clinics in Phnom Penh, Cambodia. We assessed the operational performance of the guideline, the incremental yield of investigations, and the diagnostic accuracy for smear-negative tuberculosis in HIV-positive patients using culture positivity as reference standard. 1,147 (68.9%) of 1,665 TB suspects presented with unknown HIV status, 1,124 (98.0%) agreed to be tested, 79 (7.0%) were HIV-positive. Compliance with the guideline for chest X-rays and sputum culture requests was 97.1% and 98.3% respectively. Only 35 of 79 HIV-positive patients (44.3%) with a chest X-ray suggestive of TB started TB treatment within 10 days. 105 of 442 HIV-positive TB suspects started TB treatment (56.2% smear-negative pulmonary TB (PTB), 28.6% smear-positive PTB, 15.2% EPTB). The median time to TB treatment initiation was 5 days (IQR: 2-13 days), ranging from 2 days (IQR: 1-11.5 days) for EPTB, over 2.5 days (IQR: 1-4 days) for smear-positive PTB to 9 days (IQR: 3-17 days) for smear-negative PTB. Among the 34 smear-negative TB patients with a confirmed diagnosis, the incremental yield of chest X-ray, clinical suspicion or abdominal ultrasound, and culture was 41.2%, 17.6% and 41.2% respectively. The sensitivity and specificity of the algorithm to diagnose smear-negative TB in HIV-positive TB suspects was 58.8% (95%CI: 42.2%-73.6%) and 79.4% (95%CI: 74.8%-82.4%) respectively. CONCLUSIONS: Pending point-of-care rapid diagnostic tests for TB disease, diagnostic algorithms are needed. The diagnostic accuracy of the 2007 WHO guideline to diagnose smear-negative TB is acceptable. There is, however, reluctance to comply with the guideline in terms of immediate treatment initiation.

Performance of abdominal ultrasound for diagnosis of tuberculosis in HIV-infected persons living in Cambodia.

BACKGROUND: In resource-limited settings, abdominal ultrasound is often used to assist the diagnosis of tuberculosis (TB) in people with HIV (PLHIV), although data on performance characteristics are missing. METHODS: Cross-sectional study of PLHIV in Cambodia receiving a standardized TB diagnostic evaluation, including history, physical examination, chest radiography, microscopy and culture of various specimens, and abdominal ultrasound. Patients with at least one specimen culture positive for Mycobacterium tuberculosis were classified as having TB. RESULTS: TB was diagnosed in 37 (18%) of 212 PLHIV. Abdominal ultrasound was abnormal in 15 of 37 (41%) patients with TB compared with 14 of 175 (8%) without TB (P < 0.01). Predictors of TB disease included multiple enlarged (1.2 cm or greater) abdominal lymph nodes on ultrasound (adjusted odds ratio [OR], 6.4; 95% confidence interval [CI], 1.8-22.4), abnormal chest radiography (OR, 6.8; CI, 2.7-17.0), anorexia (OR, 4.6; CI, 1.8-11.7), and CD4 less than 200 cells/mm (OR, 3.3; CI, 1.2-9.1). Having multiple enlarged abdominal lymph nodes on ultrasound was 97.1% (CI, 93.5%-99.1%) specific for TB with a positive likelihood ratio of 11.4 (CI, 4.3-30.3). CONCLUSIONS: Abdominal ultrasound is a useful diagnostic test for TB disease in PLHIV, increasing the posttest probability of TB when multiple enlarged abdominal lymph nodes are visualized. Its wider use may accelerate access to TB treatment, potentially reducing mortality in PLHIV.