Acquired immunodeficiency syndrome (AIDS) is an infectious disease caused by human immunodeficiency virus (HIV). According to the World Health Organization, there were approximately 39.9 million (36.1 million–44.6 million) cases of people living with HIV (PLHIV) by the end of 2023, with 1.3 million (1 million-1.7 million) new infections and 630,000 (500,000-820,000) deaths (1). AIDS remains a significant global public health challenge.

With the widespread use of CD4+ T-cell count detection technology, the CD4+ T-cell count has become a crucial indicator for assessing the immune status and clinical stage of HIV-infected individuals (2), initiating antiretroviral therapy (ART) and evaluating the therapeutic effect (3). Numerous studies have investigated the association between CD4+ T-cell counts and mortality risk among PLHIV. However, many of these studies treated the CD4+ T-cell count as a categorical variable, often overlooking its dynamic fluctuations and their evolving impact on mortality risk (4, 5). This dynamic trajectory may exhibit non-linear characteristics, which are often overlooked in linear analyses. The restricted cubic spline (RCS) model effectively integrates spline functions into generalized linear models (e.g., logistic regression and Cox proportional hazards regression) to elucidate the dose–response association between independent and dependent variables (6).

In this study, a 10-year retrospective cohort design was based on the routine treatment database, with a specific focus on PLHIV aged 30, 40, 50, and 60 years to control for age effects. The RCS model was used to analyze the dose–response association between CD4+ T-cell counts and mortality risk in PLHIV.

The Cox proportional hazards regression model indicated that as the CD4+ T-cell count increased, the HR value decreased; however, the HR value did not exhibit a proportional reduction, suggesting a non-linear association between the CD4+ T-cell counts and mortality risk in PLHIV. This finding aligns with those of previous research (9, 10). While regression models are frequently utilized to explore associations between CD4+ T-cell counts and mortality risk, these models often presuppose a linear association (11). If actual data do not conform to this assumption, simply converting the CD4+ T-cell count into a categorical variable may not only overlook or discard the continuous nature of the data but also introduce new biases, thereby potentially affecting the accuracy and reliability of the analysis results. In contrast, the RCS model can treat CD4+ T-cell counts as continuous variables to effectively model their non-linear association with mortality risk through a smooth curve.

Previous studies have demonstrated that age is a significant risk factor for mortality in PLHIV (12, 13). These studies typically categorized PLHIV by age groups, such as Yue Tingting’s study, which divided PLHIV into four groups: under 30 years, 30–39 years, 40–49 years, and 50 years and older (14). Tian Bo used 50 years as the cutoff point, dividing age groups into two: below 50 years and above 50 years (15). Jun-fan Pu categorized age into four groups: 15–29 years, 30–44 years, 45–59 years, and 60 years and older (16). This approach often employs broad intervals of 10 or 15 years, and in some cases, a singular boundary at 50 years, potentially exacerbating disparities in the target analysis variables across different age cohorts (17). Therefore, this study adopted a more refined division strategy, focusing specifically on ages 30, 40, 50, and 60 years to control for potential confounding effects of age on the research results.

The RCS model showed that once the CD4+ T-cell counts reached the respective cutoff values—382 cells/μL for 30 years, 332 cells/μL for 40 years, 334 cells/μL for 50 years, and 215 cells/μL for 60 years—the curve depicting the reduction in mortality risk tended to flatten. Research has indicated that as age increases, the likelihood of opportunistic infections and complications also rises, leading to higher mortality risks (18, 19). Therefore, it is expected that the cutoff value would decrease with advancing age.

Although the cutoff values for each age group vary, they consistently remain around 350 cells/μL. A CD4+ T-cell count of 350 cells/μL is a critical threshold in AIDS-related research. Several studies have used survival curves to clearly illustrate that PLHIV with baseline CD4+ T-cell counts ≥350 cells/μL have higher survival rates throughout the observation period compared to those with counts <350 cells/μL. In addition, the Cox proportional hazards model revealed that the PLHIV with a CD4+ T-cell count <350 cells/μL exhibited a significantly higher mortality risk than those with a count ≥350 cells/μL (16, 20). A study aimed at identifying the determinants of tuberculosis co-infection among PLHIV in Papua found that HIV-infected individuals with a baseline CD4+ T-cell count <350 cells/μL were at a significantly higher risk of opportunistic infections compared to those with a CD4+ T-cell count ≥350 cells/μL (21). Some studies on complications associated with HIV infection have suggested that a CD4+ T-cell count <350 cells/μL can accelerate hepatitis progression (22) and is a risk factor for malignancies (23) and cervical cytological abnormalities (24) in PLHIV. Studies on the impact of HIV-1 genetic diversity on disease progression (25) and the efficacy and safety of a simplified lamivudine plus dolutegravir dual therapy in HIV-1-Infected patients (26) have also identified that a CD4+ T-cell count <350 cells/μL is a risk factor that impedes CD4 + T lymphocyte recovery. Research on pregnancy in women has found that a lower pregnancy incidence (27) and adverse pregnancy outcomes (28) are associated with a CD4+ T-cell count <350 cells/μL. Therefore, considering these findings along with the cutoff value, it is widely recognized that a CD4+ T-cell count <350 cells/μL signifies an elevated risk of mortality and necessitates disease prevention strategies.

In summary, there was a non-linear association between CD4+ T-cell counts and mortality risk in PLHIV. As CD4+ T-cell counts fall below their respective cutoff values, the mortality risk increases significantly. Notably, despite variations in cutoff values across age groups, they consistently remain around 350 cells/μL. Consequently, it is imperative to concentrate on populations whose mortality risk increases rapidly. Comprehensively considering cutoff values—specifically targeting a CD4+ T-cell count <350 cells/μL—is vital for reducing the mortality rate of PLHIV. There are certain limitations to this study. Firstly, it was a retrospective study based on real-world data, which might have resulted in incomplete data collection and insufficient exclusion of confounding factors, such as reasons for attrition and comorbidities. Secondly, only participants aged 30, 40, 50, and 60 years were included, and age was treated as a discrete variable—for example, a person aged 30 years and 11 months was placed in the 30-year age group although they were closer to 40 years old— potentially limiting the generalizability of the results. Thirdly, the single data source might have limited the variability regarding the infection routes. In addition, the study spanned a considerable period but did not account for the progress of the ART protocols over time, which might have introduced bias in mortality risk. Considering these limitations, it is necessary to conduct prospective, multicenter cohort studies to analyze the association between CD4+ T-cell counts and mortality risk of PLHIV, thereby addressing the shortcomings of this study.