Objective. The objective of the study was to evaluate the pattern of manifest herpesvirus infections in hospitalized HIV-infected patients and to identify clinical and immunological factors associated with adverse outcomes and mortality.Materials and methods. A prospective-retrospective cohort study was conducted using data from Clinical Infectious Diseases Hospital No. 2 for the period 2018–2023. The total sample for epidemiological analysis included 46,478 patients with HIV infection (46,182 adults and 296 children). For an in-depth clinical and laboratory analysis, 530 adult patients with confirmed HIV infection and moderate-to-severe manifest Human herpesvirus (HHV) infections were selected using a census sampling method (n=530; 322 males and 208 females), including 71 fatal cases. The control group comprised 756 hospitalised patients with HHV infections without HIV infection. Because multiple concurrent herpesvirus infections were common, statistical analysis was performed both at the individual-patient level and according to the absolute number of verified clinical diagnoses. The diagnosis of HIV infection was verified using tests for antibodies to HIV-1, 2/p24 (ARCHITECT HIV Ag/Ab Combo) and a linear immunoblot («MilaBlot-HIV»). HIV-1 RNA was detected using the COBAS AmpliPrep/COBASTaqMan v.2.0 (sensitivity ≥20 copies/ml) and Abbott Real Time (≥40 copies/ml) test systems, followed by log10 transformation of the data. Viral DNA (HSV, VZV, CMV, EBV and HHV-6) was detected by polymerase chain reaction (PCR) using AmpliPrime and AmpliSens reagent kits according to the manufacturer's protocols. Immune status (CD45+/CD4+/CD8+/CD3+) was assessed by flow cytometry (Navios Beckman Coulter). The reference values from the Clinical Immunology Laboratory at the State Scientific Centre of the Institute of Immunology, Federal Medical Biological Agency of Russia, were used as the standard. Statistical analysis was performed using Excel, Statistica 12.0 and Python 3.10 (Pandas, SciPy, Statsmodels). The goodnessof-fit of the resulting logistic regression model was verified using the Hosmer–Lemeshow test (p>0.05).Results and discussion. The proportion of Human herpesvirus (HHV) infections among hospitalized patients with HIV infection was 1.73% (530 out of 30,633). Within this infected cohort, 60.75% were male and 39.25% were female. The overall case fatality rate was 13.40% (n=71), including 37 deceased males and 34 deceased females. The mean age of patients was 42.34±0.36 years (20–76 years); no significant differences were found between genders or in relation to gender and disease outcome (р>0.05). The mean duration of HIV infection was 8.96±0.33 years and was significantly longer among deceased men (Me=11.0 years) than among women (Me=7.0 years; p=0.004), with a borderline tendency toward a longer duration among men overall (Me=8.0 years; p=0.058). ART adherence within the cohort was critically low, standing at 44.7%. Among the survivors, fewer than half of the patients were receiving treatment, with no gender differences (47.4% of men and 48.3% of women; р=0.85). In the group of fatal cases, the proportion of individuals on ART was only 25.35% (р<0.001). The absence of consistent ART is recognised as a leading risk factor for death: the presence of ART reduced the risk of death by 58% (RR=0.42; 95% CI 0.25–0.71), while the absence of regular ART increased the odds of a fatal outcome by 2.70 times (OR=2.70; 95% CI 1.54–4.76; p<0.001). The median HIV RNA level in surviving men was higher than in women, was 5.31 (4.91; 5.68) log10 copies/mL versus 5.06 (4.58; 5.32) log10 copies/mL (p=0.042). The maximum intensity of HIV replication was recorded in the fatal outcome group — 5.97 (5.72; 6.11) log10 copies/mL (р<0.001 compared to survivors) — which was associated with the absence of ART in 74.6% of those who died. The leading cause of hospitalisation was herpes zoster (50.29%), followed by CMV (24.14%) and herpes simplex (HS, 19.19%). Among men, there was a tendency for herpes zoster to predominate (53.04% vs 45.99%; р=0.076), whilst among women, HS monoinfections were more common (21.23% vs 13.71%; р=0.049). Herpes zoster dominated the pattern of monoinfections (67.51%), ahead of HS (16.50%) and CMVI (15.99%). Among the deceased (105 diagnoses in 71 patients), CMVI was the most common (41.90%), ahead of herpes zoster (28.57%) and HS (23.81%). The odds of a fatal outcome in patients with CMV monoinfection were 6.72 times higher than in those with other herpesvirus monoinfections (95% CI 3.84–11.76; p<0.001), with the highest proportion observed among females (42.86%). The immune status of HIV-infected patients differed significantly from that of the HIV-negative control group (р<0.0001): a deficiency in total lymphocytes and CD4+ T-lymphocytes, accompanied by a decrease in the IRI and an increase in the percentage of cytotoxic CD8+ T-lymphocytes. ROC analysis verified the immunoregulatory index (IRI) as a highly accurate predictor of mortality (AUC=0.826, р<0.001) with a critical cut-off value of ≤0.43 (sensitivity 92.3%, specificity 70.6%). The most pronounced changes were observed in fatal cases: a drop in the lymphocyte percentage to a median of 13.5% (7.7; 18.8), a critical deficiency of CD4+ T-lymphocytes (relative count down to 10.0% (7.0; 13.0), absolute — down to 78.0 (50.0; 102.0) cells/mL) and a decrease in the IRI to 0.28 (0.18; 0.36) (р<0.001 compared to survivors), which confirms the IRI as a predictor of poor outcome. To eliminate multicollinearity, the calculated IRI index was excluded from the final multivariate logistic regression model (Logit model). Multivariate logistic regression confirmed that the absolute CD4+ T-lymphocyte count is the strongest independent predictor of fatal outcome: OR=0.70 (95% CI 0.51–0.97; p=0.038).Conclusion. Thus, non-adherence to the required regular ART (25.35%) was significantly associated with an increased risk of mortality among HIV-infected patients with HHV coinfection against the background of severe immunodeficiency.