Screening for tuberculosis (TB) and intensified follow-up of TB cases in people starting antiretroviral therapy and urine-based screening of inpatients with HIV both have the potential to significantly reduce deaths and improve rates of TB treatment in people with HIV, according to results of two large studies presented at the Conference on Retroviruses and Opportunistic Infections (CROI 2018) in Boston.
Despite the introduction of earlier antiretroviral therapy (ART) and wider access to treatment in lower- and middle-income countries, people with HIV in countries with a high burden of TB still have a higher risk of dying of TB after starting antiretroviral treatment than HIV-negative people.
Diagnosing TB in people with HIV who have started ART remains a high priority for reducing deaths and serious illness among people with HIV and for limiting the onward transmission of TB. Screening to exclude active infection is also important so that people without an active infection can be given isoniazid preventive treatment.
Over the past ten years World Health Organisation (WHO) recommendations have sought to improve TB case finding among people with HIV. WHO developed a symptom screen for TB to enable cases of active TB to be identified. WHO has also recommended the integration of the Xpert molecular diagnostic platform into screening protocols, to speed up diagnosis.
Another diagnostic tool with a more limited application, a urine-based lipoarabinomannan (LAM) test, was recommended on a conditional basis for use in people with HIV with advanced HIV disease (a CD4 cell count below 100 cells/mm3) or seriously ill patients at any CD4 cell count. WHO concluded that urine-based LAM was not suitable for use as a general screening tool, either in people with HIV or other groups.
Two studies presented at the 2018 CROI showed that innovations in TB case finding and screening reduced mortality in people on ART. The XPRES study in Botswana found that a package of measures to intensify case finding reduced mortality, while the STAMP trial in South Africa and Malawi confirmed that urine-based LAM screening improved TB diagnosis and treatment, and reduced mortality in people on ART who had been admitted to hospital.
The XPRES study by researchers at the CDC Atlanta, CDC Botswana, Botswana’s Ministry of Health and the London School of Hygiene & Tropical Medicine, examined the phased roll-out of the Xpert MTB/RIF assay as a replacement for smear microscopy in the diagnosis of TB. The study compared three interventions in people starting ART: standard of care plus enhanced care: intensified case finding with additional staff support and intensified tracing for patients who missed clinic appointments; and standard of care plus enhanced care plus use of Xpert MTB/RIF in place of smear microscopy.
The study recruited 14,963 patients at 22 clinics in Botswana. The standard-of-care arm was a retrospective review of 8960 people who had started ART at any of the participating centres in the two years prior to the commencement of roll-out in 2012.
The enhanced care study arm recruited 1786 participants and the enhanced care plus Xpert arm recruited 4215. These arms were non-randomised and recruitment to the enhanced care arm lasted until the availability of the Xpert MTB/RIF assay at a participating centre.
The phased recruitment, dictated by the time needed to make the Xpert diagnostic equipment available to all sites, allowed the investigators to compare the relative contributions of enhanced case finding techniques and a new diagnostic technique to the reduction of TB mortality and morbidity. The primary study outcome was mortality six months after starting ART.
There was no significant difference in participant characteristics between study phases (64% female, median age 35 years) but CD4 count at ART initiation was lower in people starting ART during the standard-of-care phase prior to 2012 (184 cells/mm3 vs 241 cells/mm3 in the enhanced care phase and 246 cells/mm3 in the enhanced care and Xpert phase).
After adjusting for age, sex, pregnancy, CD4 count at ART initiation, ART regimen and haemoglobin level, participants in the enhanced care plus Xpert arm had a 23% reduction in the risk of death six months after starting ART compared to the standard-of-care arm (adjusted hazard ratio 0.77, 95% CI 0.61-0.98, p = 0.031). Enhanced care alone did not result in a significant reduction in six-month mortality compared to the standard of care. But, when 12-month mortality was considered, the investigators found that participants in the enhanced care and the enhanced care plus Xpert arm had a significantly reduced risk of death, with no significant difference between the Xpert arm (AHR 0.76, 95% CI 0.61-0.95, p = 0.015) and the enhanced care arm (AHR = 0.72, 95% CI 0.54-0.97, p = 0.033).
The investigators concluded that active tracing and intensified case finding by healthcare workers should be considered for scale-up. In this study, it was human resources rather than a diagnostic platform that seemed to make the difference when patients were followed for 12 months, suggesting that although diagnostic speed and sensitivity may make a difference in the shorter term, actions by healthcare workers to identify TB and to improve retention in care are more important in reducing the risk of death after starting antiretroviral treatment.
Rapid diagnosis of TB is essential in patients who have been hospitalised so as to ensure rapid treatment initiation. TB in advanced HIV disease is more likely to be smear negative, and more likely to present with atypical symptoms and atypical x-ray evidence, requiring TB culture to make a definitive diagnosis. TB culture requires mycobacteria to be grown in the laboratory from a sample, using specialised laboratory equipment and may take up to four weeks, or longer if drug susceptibility needs to be determined.
Lipoarabinomannan is shed from the cell wall of TB bacteria and is only present in people with active TB. Urine testing for TB-LAM has the potential to speed up TB diagnosis and its use has been shown to reduce the risk of death in hospitalised patients with HIV with CD4 cell counts below 100 cells/mm3. What is not known is whether LAM testing offers additional value in settings where the Xpert MTB/RIF test is available.
The STAMP study by researchers at the London School of Hygiene & Tropical Medicine, Dignitas International, Malawi, Edendale Hospital, South Africa and the Malawi-Liverpool Wellcome Trust Clinical Research Programme, randomised 2,600 consecutive people with HIV admitted to hospital in Edendale, South Africa, and Zomba, Malawi between October 2015 and September 2017. Participants were randomised to the standard-of-care arm (sputum testing by Xpert MTB/RIF) or the intervention arm (sputum testing by Xpert MTB/RIF and urine testing by TB-LAM and Xpert MTB/RIF). After testing, results were returned to the medical team to take action according to routine treatment protocols.
The study population had a median age of 40 years and 57% were female. Eighty-seven per cent of the standard-of-care group and 84% of the intervention group were taking ART and had been doing so for a median of 2.9 years. The median CD4 cell count was 222 cells/mm3 in the standard-of-care group and 231 in the intervention group. Just under 30% had a CD4 cell count below 100 cells/mm3. TB was suspected due to clinical symptoms in 38.5% of the standard-of-care group and 39.1% of the intervention group.
The primary outcome of the study was mortality at 56 days after admission; the secondary outcome was TB events at 56 days.
Overall mortality at 56 days was 21.1% in the standard-of-care arm and 18.3% in the intervention arm, a non-significant reduction in mortality of 2.8% (P = 0.07). A sub-group analysis showed significant reductions in mortality for people with CD4 cell counts below 100 cells/mm3 (-7.1%, 95% CI -13.7%, -0.4%, p = 0.036), for people with baseline haemoglobin below 8 g/dl (-9%, 95% CI -16.6%, -1.3%, p = 0.021) and for patients with suspected TB on admission (-5.7%, 95% CI -11%, -0.5%, p = 0.033).
Patients in the intervention arm with CD4 cell counts below 100 also had a lower risk of death as time elapsed during the 56-day follow-up period, despite staying in hospital for a median of only six days, suggesting that they were less likely to be discharged from hospital with undiagnosed TB.
All patients in the intervention group were significantly more likely to be diagnosed with TB (+7.3%, 95% CI 4.4, 10.2, p < 0.001) and to be treated for TB (+7%, 95% CI 4.1,9.8%, p < 0.001). Baseline CD4 count did not affect the likelihood of being diagnosed. The intervention also had a higher diagnostic yield; whereas almost 40% of hospitalised patients were unable to produce an adequate sputum sample – a frequent problem in advanced HIV disease – 99% of patients were able to provide a urine specimen. The study investigators concluded that the results of the STAMP study support urine-based TB screening for all HIV-positive inpatients, and that the benefit was not restricted to patients with low CD4 cell counts or those already suspected of having TB. “I think there’s enough evidence now for urine-based screening to support a guideline change,” said Ankur Gupta-Wright of the London School of Hygiene and Tropical Medicine at a press briefing after the presentation of the results. He said that improved versions of the TB-LAM test, with a faster turnaround time, would soon become available, making the test suitable for use in a wide variety of settings.
In 2012, at 22 antiretroviral therapy (ART) clinics, Botswana implemented a phased rollout of the Xpert package of interventions, with 3 components: (1) additional nurses and mentoring to support intensified tuberculosis (TB) case finding (ICF) activities, (2) intensified tracing for patients missing clinic appointments, and (3) Xpert MTB/RIF (Xpert) replacing smear microscopy. We evaluated effect of the Xpert package on early (6- and 12-month) ART mortality in the XPRES trial (ClinicalTrials.gov: NCT02538952).
At 22 ART clinics, all adult patients (>12 years old) starting ART were enrolled in three phases: (1) a retrospective standard of care (SOC) phase, (2) a prospective enhanced care (EC) phase, and (3) a prospective EC plus Xpert (EC+X) phase. EC and EC+X phases were enrolled as a stepped-wedge trial. Adults enrolled in the EC phase received SOC plus components 1 (TB ICF) and 2 (intensified tracing) of the Xpert package. Adults enrolled in the EC+X phase received SOC plus all 3 components of the Xpert package. All-cause 6-month ART mortality was the primary outcome. An adjusted analysis, appropriate for study design, controlled for baseline differences in individual-level factors and intra-facility correlation. Trial outcome results are final.
14,963 eligible patients were enrolled; 8,980 in the SOC, 1,768 in the EC, and 4,215 in the EC+X phases. Median age of ART enrollees was 35 years, 64% were female, median weight was 58.4 kg, and median hemoglobin 11.7 g/dL. These characteristics were similar across phases. Pregnancy among females was less common in the SOC than subsequent phases (16% in SOC, 23% in EC, and 32% in EC+X). Median CD4 count at ART initiation was lower in SOC than subsequent phases (184/µL in SOC, 241/µL in EC, and 246/µL in EC+X). In adjusted analysis, compared with the SOC phase, 6-month ART mortality was significantly lower in the EC+X phase, while 12-month ART mortality was significantly lower in both the EC and EC+X phases (Table). When compared with the EC phase, 6- and 12-month mortality rates were not significantly different in the EC+X phase.
In Botswana, interventions to strengthen TB ICF and active tracing were associated with lower early ART mortality and should be considered for scale-up. No additional mortality benefit of replacing smear microscopy with Xpert was observed.
Andrew F Auld, Tefera Agizew, Anikie Mathoma, Rosanna Boyd, Anand Date, Sherri Pals, Christopher Serumola, Unami Mathebula, Heather Alexander, Tedd V Ellerbrock, Goabaone Rankgoane-Pono, Pontsho Pono, Katherine Fielding, Alison Grant, Alyssa Finlay
Tuberculosis (TB) is the major cause of death in people living with HIV (PLHIV), in part due to suboptimal diagnosis. Urine is easily obtained, and urine diagnostics are rapid, complementary to sputum, have good yield, and reflect often-fatal disseminated TB. Urine screening may therefore reduce death and missed TB diagnosis in severely ill PLHIV.
The STAMP trial was a pragmatic, individually randomized controlled trial recruiting consecutive, unselected PLHIV admitted to medical wards in Edendale, South Africa, and Zomba, Malawi. HIV testing was offered if status was unknown. Randomization was stratified by site. Consenting eligible patients (≥18years, not on TB treatment) were allocated to either standard of care (SOC) screening (sputum Xpert MTB/RIF) or intervention arms (SOC plus Determine TB-LAM on neat urine and Xpert MTB/RIF on centrifuged urine) regardless of symptoms. Results were reported as TB screen positive or negative to routine clinicians who managed patients masked to study arm. Mortality (primary outcome) and TB events (secondary) were ascertained at 56 days.
We screened 4788 PLHIV and randomised 2600 (1300/arm) from Oct-2015 to Sep-2017. 26 were excluded, leaving 2574 PLHIV for final analysis, of whom 996 (38.7%) had TB suspected and 1861 (72.3%) were on antiretroviral therapy (ART) at admission. Median CD4 was 227 cells/µL. Baseline characteristics did not differ by arm. 27 (1.0%) were lost to follow-up. By 56 days, 272 (21.1%) and 235 (18.3%) patients had died in SOC and intervention arms respectively (risk difference [RD] -2.8%, 95% confidence interval [CI] -5.8 to 0.3, p=0.07; odds ratio 0.83, 95% CI 0.7 to 1.0). Intervention arm mortality was significantly lower than SOC in pre-specified subgroups: CD4<100 cells/µL (RD -7.1%, 95% CI -13.7 to -0.4); haemoglobin <8g/dl (RD -9.0%, 95% CI -16.6 to -1.3); and TB clinically suspected at admission (RD -5.7%, 95% CI -11.0 to -0.5) (Table). TB diagnosis was significantly more likely in intervention (21.9%) than SOC (14.9%) arm (RD 7.3%, 95% CI 4.4 to 10.2%, p<0.001). Differences in TB diagnosis between arms were not confined to any particular subgroups.
Systematic urine screening of hospitalised PLHIV increased overall TB diagnosis and treatment, and reduced mortality in key subgroups despite high ART coverage. Early mortality differences were minimal outside of these subgroups, although reducing missed TB diagnoses is likely to be of wider value. Trial registration: ISRCTN71603869
Ankur Gupta-Wright, Elizabeth L Corbett, Joep J van Oosterhout, Doug K Wilson, Daniel Grint, Melanie Alufandika-Moyo, Jurgens A Peters, Lingstone Chiume, Stephen D Lawn, Katherine Fielding