The rates of tuberculosis (TB) recurrence and mortality in people with HIV in eastern Europe remain concerningly high and continue to be a clinical and public health challenge, according to a paper published in AIDS. In the first study to report on long-term follow-up of TB/HIV co-infected patients in eastern Europe, one in seven people had TB again and over half died.
Despite a consistent decrease in TB cases across Europe since 2000, with a 25% reduction in 2016-2020 alone, two major challenges remain: a high prevalence of multidrug-resistant TB (MDR-TB) and high rates of TB/HIV co-infection, especially in Russia, Belarus and Ukraine. Co-infections in the region are often associated with injecting drug use and hepatitis C (HCV), which jeopardise health care and outcomes.
A previous study reported that around 30% of TB/HIV co-infected people in eastern Europe died within the year following their TB diagnosis, and as many as 79% of these deaths were TB-related. Also, there was a 40% prevalence of MDR-TB. Nevertheless, studies on TB recurrence and long-term mortality are scarce, with none from eastern Europe.
This brought Dr Christian Kraef from the University of Copenhagen, with colleagues from eastern Europe, to assess TB recurrence, long-term mortality and causes of deaths in TB/HIV co-infected people from the region. For this, the team used the EuroCoord TB-HIV study, a cohort that enrolled people with HIV and active TB in 2011-2013. In 2020-2021, nine of the 18 eastern European clinics that had already contributed to the cohort provided more data from patients who had not died before 2016 – thus extending their follow-up by several years.
The data were collected from patients’ medical records, with a focus on recurrent TB, treatment and drug susceptibility testing for anti-TB medicines, HIV infection (CD4 cell counts, viral loads, antiretroviral therapy), vital status and causes of deaths after 2016.
For the analysis, MDR-TB was defined as resistance of Mycobacterium tuberculosis to at least rifampicin (a standard anti-TB medication). TB recurrence was defined as a new TB diagnosis after TB treatment of at least six months and eight weeks off-treatment.
Clinics in Belarus, Georgia, Latvia, Poland, Romania and the Russian Federation contributed data on 375 patients. Of these, 281 (75%) were male and at the time of their TB diagnosis, 199 (53%) had a history of injecting drug use, 102 (27%) of alcohol abuse and 61 (16%) of imprisonment.
While 90% of participants had been diagnosed with HIV at least three months before their TB diagnosis, only 34% of the cohort had received antiretroviral therapy. Over half (54%) of the patients had low CD4 cell counts (below 200).
MDR-TB was detected in 70 (19%) patients at their first TB diagnosis, while 191 (51%) were not tested for drug susceptibility (indicating that some may well have MDR-TB without anyone knowing about it). HCV antibodies were detected in 203 (54%) patients.
Total follow-up time was 1,713 person years (an average of 4.6 years per person). During this time, TB recurred in 53 (14%) of the 375 participants, leading to an incidence rate of 3.1/100 person years of follow-up (95% confidence interval 2.4-4.0).
In 13 (24%) of these recurrent cases, TB was disseminated: it had spread from the lungs (pulmonary TB) to other parts of the body, with multiple organs affected, and this was quite likely due to poor adherence to treatment and/or follow-up. Disseminated TB can occur within weeks of TB infection or after a period of latency that can last years, and it is more complicated to treat than pulmonary TB.
Among the 33 (62%) recurrent cases for which drug susceptibility results were available, 23 (70%) had MDR-TB, of whom eight (35%) had already tested positive for this at initial TB diagnosis.
Of the 53 participants with recurrent TB, 30 (57%) died. Those who died had a much lower median CD4 cell count, were more likely to have disseminated TB at recurrence and were more likely to have had an episode of TB before enrolment in this study.
Looking at mortality overall, 215 (57% of 375) participants died: 129 (60%) within two years of follow-up, and 86 (40%) afterwards.
Risk factors for mortality in the first two years were a CD4 cell count below 200 (adjusted Hazard Ratio 2.66; Confidence Interval 1.27-5.59), a missing CD4 cell count (aHR 3.27; CI 23-8.60), disseminated TB (aHR 2.43; CI 1.38-4.27) and rifampicin resistance (aHR 3.20; CI 1.18-8.68), while a higher weight was protective (aHR0.97 per kg increase; CI 0.95-0.98).
For mortality beyond two years of follow-up, the study found that participants with disseminated TB (aHR 0.57; CI 0.34-0.95) and those on antiretroviral therapy at the time of TB diagnosis (aHR 0.56; CI 0.31-1.04) were at lower risk of dying subsequently. However, the risk of death beyond two years increased by 65% (aHR 1.65; CI 1.00-2.71) in participants with HCV antibodies at enrolment.
Among the 71 people who died in the first six months following diagnosis, 69% died of TB-related causes, while 11% died of HIV/AIDS. In the 58 who died between six months and two years after diagnosis, 57% died of TB-related causes and 3% of HIV/AIDS. For the 86 people dying more than two years after diagnosis, fewer deaths were due to TB (27%) and more due to HIV/AIDS (15%).
Forty-four deaths (20%) were due to “other causes”, including 34% of the deaths that happened more than two years after diagnosis. Ten participants died from intoxication (overdose), while ten died from bacterial infections and five from violence, including suicide.
The authors underscore the high prevalence of TB recurrence (14%) and death among participants with recurrent TB (over 50%). They consider that the high rate of recurrent TB likely indicates a combination of treatment failure during the initial TB episode – notably reflected by a high rate of disseminated TB – and of TB reinfections. These are due to immunosuppression, probably related to poor adherence to antiretrovirals (when available) and to a social environment with high risk of exposure to TB. At any rate, the findings reflect the vulnerability of a population where injecting drug use, alcohol abuse, imprisonment and substandard access to care are still common.
Interventions to improve these outcomes include access to high quality integrated person-centred care where HIV, TB, HCV care, oral substitution therapy and psychosocial support are all available. Kraef and colleagues also emphasise the need for policy and financing efforts to implement them urgently, in a context of unstable regional security that, they say, “has already led to healthcare disruptions and, almost certainly, a surge of HIV, TB and HCV.”
Kraef C et al. Long-term outcomes after tuberculosis for people with HIV in eastern Europe. AIDS, online ahead of print, July 28, 2023.