Diagnosis of latent tuberculosis infection is associated with reduced HIV viral load and lower risk for opportunistic infections in people living with HIV

Approximately 28% of the human population have been exposed to Mycobacterium tuberculosis (MTB), with the overwhelming majority of infected individuals not developing disease (latent TB infection (LTBI)). While it is known that uncontrolled HIV infection is a major risk factor for the development of TB, the effect of underlying LTBI on HIV disease progression is less well characterized, in part because longitudinal data are lacking. We sorted all participants of the Swiss HIV Cohort Study (SHCS) with at least 1 documented MTB test into one of the 3 groups: MTB uninfected, LTBI, or active TB. To detect differences in the HIV set point viral load (SPVL), linear regression was used; the frequency of the most common opportunistic infections (OIs) in the SHCS between MTB uninfected patients, patients with LTBI, and patients with active TB were compared using logistic regression and time-to-event analyses. In adjusted models, we corrected for baseline demographic characteristics, i.e., HIV transmission risk group and gender, geographic region, year of HIV diagnosis, and CD4 nadir. A total of 13,943 SHCS patients had at least 1 MTB test documented, of whom 840 (6.0%) had LTBI and 770 (5.5%) developed active TB. Compared to MTB uninfected patients, LTBI was associated with a 0.24 decreased log HIV SPVL in the adjusted model (p < 0.0001). Patients with LTBI had lower odds of having candida stomatitis (adjusted odds ratio (OR) = 0.68, p = 0.0035) and oral hairy leukoplakia (adjusted OR = 0.67, p = 0.033) when compared to MTB uninfected patients. The association of LTBI with a reduced HIV set point virus load and fewer unrelated infections in HIV/TB coinfected patients suggests a more complex interaction between LTBI and HIV than previously assumed.


Number and type of tuberculosis tests
We analyzed information of 13326 tuberculin skin reactivity tests of 10649 patients performed between 1983 and 2019. Most tests were performed in the years 1992 − 1998 (see Figure A). After 2008, predominantly interferon-based screening tests were performed (see B). In total, we analyzed 3978 interferon-based screening tests of 3623 patients, the majority of the tests being quantiferon in-tube followed by liquid quantiferon tests.  1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999  Of the 17 243 tuberculosis test dates, 5750 (33.3%) tuberculosis tests were at date of the SHCS registration, 10 146 (58.8%) tests were within 1 year of the SHCS registration date, 182 (1.1%) more than 1 year before the SHCS registration and 6915 (40.1%) more than one year after SHCS registration.

Years between tuberculosis test and SHCS registriation date
Tuberculosis test year minus SHCS registration year 2 Active tuberculosis Moreover, we looked at the cases of active tuberculosis in the SHCS. These were defined as follows (http: //www.shcs.ch/122-4-cdc-category-c-diagnoses): Pulmonary tuberculosis was defined by either of the following criteria: • by culture (not PCR) • by showing acid-fast bacilli in sputum that is not confirmed by culture or PCR and by responding to specific treatment • suggestive infiltrate on CXR, anamnestic exposure to tuberculosis, or positive PPD test and response to specific treatment Extrapulmonary tuberculosis was defined by either of the following criteria: • by culture (not lung alone) • by showing acid-fast bacilli in stool, blood, body fluid or tissue, or in histology of cervical or hilar lymph nodes, of a species not identified by culture, and there is a concurrent definitive diagnosis (by culture) of pulmonary tuberculosis • responds to standard tuberculosis treatment In total, 367 cases of extrapulmonary and 546 cases of pulmonary tuberculosis were reported in the SHCS (see Figure E).

HIV set point viral load
Of the 13943 patients included in our study population, the HIV set point viral load could be determined for 4516 patients. I.e., for these patients we had at least one ART-naive RNA measurement in the chronic phase of the HIV infection (at least 90 days after the first positive HIV test and before the occurrence of any opportunistic infection). The following table shows the basic characteristics of patients with available HIV set point virus load and those without. Of note, the diagnosis year of patients without SPVL measurement was on median 3 years earlier, 60% had an opportunistic infection and the CD4 nadir was very low with 140.

Opportunistic infections
After excluding weight loss (unspecific disease), pulmonary tuberculosis and cervical dysplasia (as only women are at risk), the 10 most frequent opportunistic diseases diagnosed in our study population are: oral candidiasis, oral hairy leukoplakia, herpes zoster, candidiasis esophageal, pneumocystis pneumonia, HIV-related thrombocytopenia, Kaposi sarcoma, HIV-related encephalopathy, cerebral toxoplasmosis and bacterial pneumonia: In 19708 (62.3%) cases, the diagnosis year of the opportunistic infection was in the same year or after the earliest tuberculosis test of the patient (see Figure F).

Years between tuberculosis test and OI diagnosis year
Tuberculosis test year minus OI diagnosis year In particular, both the tuberculosis test and the diagnosis of the OI were clustered around the registration date of the SHCS, i.e., in most cases shortly after HIV diagnosis (see Figure G).