Implications of Vascular Depression for Successful Cognitive Aging in HIV disease

Introduction: Although older adults with HIV are at high risk for mild neurocognitive disorders, a subset experience successful cognitive aging (SCA). HIV is associated with an increased risk of vascular depression (VasDep), which can affect cognitive and daily functioning. The current study examined whether VasDep impedes SCA among older adults with HIV. Methods: 136 persons with HIV aged 50 years and older were classified as either SCA+ (n=37) or SCA− (n=99) based on a battery of demographically adjusted neurocognitive tests and self-reported cognitive symptoms. Participants were also stratified on the presence of vascular disease (e.g., hypertension) and current depression as determined by the Composite International Diagnostic Interview and the Depression/Dejection scale of the Profile of Mood States. Results: A Cochran-Armitage test revealed a significant additive effect of vascular disease and depression on SCA in this sample of older adults with HIV (z=4.13, p<.0001). Individuals with VasDep had the lowest frequency of SCA+ (0%), which differed significantly from the group with only vascular disease (30%, OR=0.04, CI=0.002,0.68)) and the group with neither vascular disease nor depression (47% OR =0.02, CI=0.33,0.001). Findings were not confounded by demographics, HIV disease severity, or other psychiatric and medical factors (ps>.05). Discussion: These data suggest that presence of VasDep may be a barrier to SCA in older adults with HIV disease. Prospective, longitudinal studies with neuroimaging-based operationalizations of VasDep are needed to further clarify this risk factor’s role in the maintenance of cognitive and brain health in persons with HIV disease.


Introduction
It is currently estimated that more than half of people infected with HIV in the United States are over the age of 50 years (CDC 2022). The prevalence of older persons living with HIV (PLWH) is likely to continue to rise in coming years, as younger adults with HIV who can access and adhere to modern antiretroviral therapies are living longer and older adults represent approximately 10% of new HIV diagnoses (CDC 2022). In general, older PLWH are at greater risk of immune senescence, medical comorbidities (e.g., vascular disease), and polypharmacy (e.g., Langebeek 2019). Moreover, the frequency (e.g., Valcour et al. 2004) and incidence (e.g., Sheppard et al. 2015a) of prodromal (e.g., mild cognitive impairment; Sheppard et al. 2015b; and syndromic neurocognitive disorders tends to be higher in older PLWH. Despite the additive effects of age and HIV on the CNS, there are still considerable differences in cognitive outcomes among older PLWH. Indeed, not all older PLWH will develop cognitive de cits, and many will age quite "successfully" in this regard (Malaspina et al. 2011;Saloner et al. 2019). Malaspina and colleagues (2011) were the rst to describe SCA among PLWH, reporting that approximately one-third of PLWH were without cognitive symptoms in daily life and showed normatively intact performance on a cognitive battery. Subsequent literature suggests that the frequency of SCA is lower in PLWH as compared to seronegative older adults, with an estimated prevalence of between 19% (Moore et al., 2014) and 32% (Malaspina et al., 2011). Not surprisingly, SCA is associated with better everyday functioning, including medication adherence and healthcare self-e cacy The co-occurrence of vascular disease and depression in older adults led Alexopoulos and colleagues (1997) to propose the vascular depression hypothesis (VasDep), which suggests that vascular disease may "predispose, precipitate, or perpetuate" geriatric depression. The VasDep hypothesis speci cally proposes that some aspects of depression result from disrupted mood-related brain networks (i.e., frontal-subcortical circuits) by way of white matter lesions (Krishnan et al. 1997). It is also plausible that depression may increase risk of vascular disease and/or exacerbate pre-existing vascular disease-related conditions that then further amplify depressive symptoms (Kirton et al. 2014;Dotson et al. 2013). Of clinical relevance, individuals with VasDep have greater risk of functional disability and cognitive impairment (e.g., executive dysfunction and slowed processing speed) relative to individuals with non-vascular depression (Aizenstein et al. 2016). It is estimated that 3.4% of adults met criteria for VasDep (González et al. 2012), with the frequency being higher among middle-aged and older persons with a lifetime history of major depressive disorder (MDD) and vascular disease comorbidity.
Additionally, persons who met criteria for VasDep demonstrated increased disease burden relative to nondepression and major depression alone populations.
VasDep may be common and impactful among PLWH. In a cross-sectional study, Beltran-Najera and colleagues (CIDI; World Health Organization 1998). Findings indicated that PLWH had a three-fold increased rate of VasDep (15.7%) relative to seronegative individuals (4.8%). Among PLWH, VasDep was associated with worse historical HIV disease markers (e.g., lower nadir CD4 counts) and older age; furthermore, PLWH with VasDep were over 5 times more likely to be dependent in everyday functioning as compared to PLWH with either vascular disease or depression alone.
As such, it is plausible that VasDep may play a role in SCA among older PLWH. Further support for the contributions of both vascular disease and depression in SCA is evident in studies in healthy adults and PLWH.

Participants
The study sample included 140 PLWH aged 50 and older who were enrolled in a study on prospective memory ). Aspects of these data have been published (e.g., Thompson et al. 2022); however, the analyses of VasDep are novel. Individuals were recruited via community-based organizations, advertisements, and local clinics and enrolled through the University of California San Diego's (USCD) HIV Neurobehavioral Research Program. Participants were excluded from the study if there was a positive history of serious psychiatric disorders (e.g., schizophrenia), color blindness, intellectual disability, neurological conditions not due to HIV disease (e.g., non-HIV-related dementia, seizure disorders), head injury with loss of consciousness greater than 30 minutes, current substance dependence as measured by the CIDI version 2.128 (World Health Organization 1998), or positive urine toxicology screen/breathalyzer for illicit drugs on the day of testing. All participants provided written, informed consent prior to completing a comprehensive medical, psychiatric, and neuropsychological assessment that was approved by the institutional review board.  (Patterson et al. 2006). The CIDI is a structured lay interview for MDD. On the POMS, participants indicated how they have been feeling over the past week (e.g., unhappy, sorry for things done) using a ve-point Likert-type scale ranging from zero ("not at all") to four ("extremely"). Possible scores ranged from 0 to 60. POMS scores were transformed into z-scores using age-and sex-based normative data (Nyenhuis et al. 1999), such that higher z-scores re ected worse depressive symptoms. Individuals who obtained a score of 1.5 or higher on the POMS Depression/Dejection subscale or met criteria for current MDD based on the CIDI were classi ed as depressed (Patterson et al. 2006). We used two indicators of depression because the frequency of current MDD diagnoses per the CIDI was fairly low and the multi-modal assessment approach bolsters the rigor of the measurement. Vascular Depression. Participants were strati ed based on vascular factors and depression status, which produced su cient samples of persons: (1) without vascular disease or depression (V-D-) (n = 32), (2) with vascular disease, but without depression (V+D-) (n = 74), and (3) with both vascular disease and depression (V+D+) (n = 30). Three participants did not meet criteria for vascular disease but met criteria for depression (V-D+), however due to such a small sample size this group was excluded from the present study. One participant was missing vascular and depression data and was excluded.

Statistical Analyses
All statistical analyses for the present study were performed in JMP Pro version 16.0 (SAS Institute Inc.) with a critical alpha of .05. Covariates were determined using a data-driven approach (Field-Fote 2019) in which any sociodemographic or clinical variable (see Table 1) that signi cantly related to both the independent and dependent variables, as determined by a series of t-tests or chi-square analyses, were included as a covariate. The Cochran-Armitage test for trends (Armitage 1955) was used to determine if depression and vascular health risks conferred additive detrimental effects on SCA. Follow-up chi-square analyses were conducted to identify signi cant differences in frequencies of VasDep among those who are SCA+ and SCA-.
Finally, a series of logistic regressions were conducted to investigate whether SCA status predicted POMS depression scores and/or number of vascular-related health conditions.

Results
No signi cant differences were identi ed among participants in the V-D-, V + D-, or V + D + groups on any sociodemographic or clinical variables listed in Table 1 (see Table 2; ps > .05); therefore, no covariates were necessary in the primary analyses. A Cochran-Armitage test for trends revealed a signi cant additive trend of VasDep group and SCA+/-status (z = 4.13, p < .0001). Speci cally, individuals with vascular disease and depression (V + D+) had the lowest rates of SCA+ (Fig. 1) and negative outcomes. Absence of depression has been previously associated with self-rated successful aging in PLWH aged 56 to 65 (Rooney et al. 2019); however, participants in that study self-reported successful aging on a one-item measure based on their own conceptualization. Our outcome of SCA captures optimal aging in a more comprehensive manner than self-report of subjective successful aging or examination of single cognitive tests or domains, as the operational de nition encompassed intact performance across cognitive domains as well as scores < 1.5 SD above the mean on self-reported executive dysfunction, memory slips in everyday life, and mood.
In addition to the independent relationships of depression and vascular disease with SCA, we found that individuals with VasDep, based on comorbid depression and vascular disease, were least likely to be in the SCA + group. This was true in comparison to individuals without depression or vascular risk as well as those with vascular disease but no depression. The latter suggests that depression confers a risk to cognitive health in PLWH beyond vascular disease alone, mirroring our previous nding that VasDep in PLWH increases the risk for functional disability to a greater degree than depression or vascular disease alone (Beltran-Najera et al. The interrelationships between vascular disease, depression, in ammation, and cognitive functioning suggest SCA may be promoted in PLWH by prevention and treatment efforts that reduce in ammation and vascular risk. Physical activity and diet are particularly important. Higher levels of physical activity are associated with bene ts to brain health, general health, and mental health across the lifespan, including in older adults and in PLWH (Erlandson et al. 2021;Fazeli et al. 2023;Ibeneme et al. 2022). Physical activity also reduces chronic in ammation (Monteiro-Junior et al. 2018), reduces white matter lesions in the brain, and leads to other direct bene ts to the brain such as increasing regional volumes and promoting the production of brain-derived neurotrophic factors (Umegaki et al. 2021). There is evidence that PLWH have lower levels of physical activity, and that physical inactivity is associated with higher levels of depressive symptoms (Qin et al. 2022). In the same study, depressive symptoms and physical activity mediated the association between HIV status and cognitive functioning. Thus, promoting physical activity may promote SCA in VasDep by reducing risk factors for vascular disease and depression and by directly bene ting brain health.
Similarly, diet and nutrition may support vascular health and reduce in ammation. A diet rich in antioxidants has been recommended for cognitive health in PLWH due to the role of antioxidants in reducing systemic neuroin ammation (Vance et al. 2022). Heart-healthy diets such as Mediterranean-style diets are associated with better vascular health and brain health in older adults (Roman et al. 2019) and could be promoted in PLWH as well. In support of the potential bene ts of heart-healthy diets in PLWH, a randomized parallel-group pilot trial found that compared to general dietary advice to reduce saturated fat intake, PLWH who received support to adopt a Mediterranean-style diet combined with cholesterol-lowering foods for six months had lower LDLcholesterol and systolic blood pressure after the intervention (Stradling et al. 2021). More research is needed in this limited area to determine the long-term bene ts of different dietary patterns on the mechanisms that contribute to vascular disease and depression in PLWH.
PLWH are one of several groups of individuals who face barriers to SCA, including physiological effects of chronic diseases and treatments, a higher burden of chronic stress, and obstacles such as stigma that result in healthcare disparities. These factors can be compounded in older individuals with HIV who not only have a lifetime of accumulated risk factors but also face additional barriers resulting from ageism, as well as biological risks related to aging. The lack of data related to these important considerations is a limitation of the current study, as is the predominance of male, White, highly educated participants in our sample. Longitudinal studies in demographically diverse samples will further clarify the impact of VasDep on SCA in diverse PLWH and the potential for different behavioral and somatic interventions to improve cognitive outcomes. There is clearly no simple solution that can guarantee SCA in older PLWH; however, our nding that vascular disease and depression independently and in combination impact SCA suggest vascular and depression risk reduction are key targets for treatment. Results also suggest a particular need for clinicians to assess for vascular disease or depression in PLWH who present with one or the other condition, since the presence of both is associated with a lower likelihood of SCA.   Figure 1