Journal Article Annotations
2018, 2nd Quarter
Annotations by John Grimaldi, MD, and Mary Ann Cohen, MD, FACLP
Type of study: This was a large, multi-site, observational clinical cohort study of patients receiving HIV medical care at six geographically dispersed US academic medical centers from 2005 to 2015.
The finding: This study analysed data from 5,927 participants in the Center for AIDS Research Network of Integrated Clinical Systems (CNICS), comprising eight geographically dispersed US academic medical enters, to determine the effect of depression on risk of missed HIV primary care appointments, treatment failure as measured by a detectable viral load, and all-cause mortality. Using weighted marginal structural models, there was a linear dose-response relationship between days spent in depression and risk of missing appointments and having a detectable viral load. The longer the duration and severity of depression, the more likely participants were to miss follow-up appointments and to have a detectable viral load. The model predicted a threshold rather than a dose-response relationship for mortality so that the presence of any depression doubled the risk of mortality, compared to participants who were never depressed. When time-updated missed appointments were added to the statistical model as an independent variable, its hypothesized mediation of depression as a risk factor for detectable viral load and mortality was attenuated.
Strength and weaknesses: The study’s strengths comprise its large, demographically diverse and geographically dispersed sample population and the use of several outcome measures along the HIV continuum of care. Although it was observational in nature, the study applied advanced causal inference statistical methods to control for alcohol and substance use as well as other psychiatric comorbidities which can also impact HIV outcomes. The measure used to capture chronicity of depression, percentage of days with depression (PDD), is an adaptation of depression-free days which has been used successfully for many years in depression research comparing the benefits of different therapeutic interventions. However, PDD is an estimate that contains measurement error relative to repeated diagnostic evaluations. An additional limitation is that the study included only those individuals who completed at least two self-reported depression scales, (PHQ-9). Thus, the sample may represent a more motivated subgroup of individuals and therefore underestimate the negative impact of depression on HIV outcomes.
Relevance: A significant number of studies have demonstrated that depression has a deleterious effect on the course, treatment and survival of HIV disease across the entire continuum of care. This study extends prior research in several important ways. By measuring the cumulative impact or chronicity of depression rather than identifying its presence or absence, it was able to demonstrate a dose-response relationship to several outcome measures. Thus, the findings suggest the potential benefit of early intervention, and that reducing the severity and length of time spent in a depressive episode may have a meaningful impact on HIV outcomes. These findings also add weight to the growing evidence for the cost-effectiveness of integrated measure-based care that utilizes depression care managers who frequently track and adjust response to treatment for depression. This is especially relevant in light of recent changes in HIV treatment guidelines calling for less frequent laboratory monitoring and office visits.
Type of study: This was a case control study, cross-sectional in design, involving a geographically distinct clinical population of HIV-infected men.
The finding: Using the Heidelberg NSS Scale, this study compared the degree of neurological soft signs (NSS) in three distinct groups of patients with HIV recruited from an ambulatory HIV clinic: 1) asymptomatic neurocognitive impairment (ANI); 2) mild neurocognitive disorder (MND); 3) no neurocognitive impairment. A matched control group of healthy, HIV-negative participants was recruited from the community. Compared to non-neurocognitively impaired patients and HIV-negative controls, patients with MND and ANI demonstrated greater neurological impairment as measured by the Heidelberg NSS Scale. This difference reached statistical significance for all NSS subscales in the MNI subgroup and for the motor coordination subtest in the ANI subgroup.
Strength and weaknesses: The study’s strengths include the use of a healthy control group matched for age and level of education. Patients with HIV-associated dementia, a severe, major psychiatric disorder or active substance use were excluded. The assessment of NSS was blinded to cognitive condition. However, the study’s limitations comprise the inclusion of patients with lifetime depression and substance use. It was cross-sectional in design and involved a small number of subjects, all of whom were male, and who were selected from a clinical convenience sample. Findings therefore may not be generalizable to different geographic settings or demographic groups. Additionally, a significantly greater number of MND patients were not fully virologically suppressed compared to patients in the ANI subgroup.
Relevance: Despite its limitations, this study attempts to fill a significant gap in the search for an effective and efficient means of identifying patients with early HIV-associated neurocognitive impairment. This gap has impeded research into the natural course of HIV-associated neurocognitive disorders (HAND) and the prognostic significance of early indicators of HIV CNS involvement. The study also adds to ongoing research into the link between HAND and other neurodegenerative disorders such as Alzheimer’s disease (AD) and mild cognitive impairment (MCI), both of which may include clinical features of NSS. Elucidating possible common pathophysiological pathways of HAND and AD has been a target of recent research. NSS could also serve as a potential future target for intervention studies whose aim is to alter the outcomes of ANI and MND.
Type of study: This study utilized data from two large, multi-site, observational clinical cohort studies of patients receiving HIV medical care at geographically dispersed US academic medical centers.
The finding: Using data from the Women’s Interagency HIV Study (WIHS) and the Multicenter AIDS Cohort Study (MACS), this study compared cognitive test performance in matched HIV-positive and HIV-negative women and men. Impairment in psychomotor speed, executive functioning and fine motor skills was measured using the Trail Making Tests (TMT) Parts A&B, Grooved Pegboard (GP), Symbol digit Modalities Test (DSMT) and Stroop. Impairment was examined with continuous and categorical demographically-corrected T-scores. No sex differences were observed for HIV-negative participants. However, among HIV-positive individuals, women were significantly more likely to score in the impaired range. For continuous T-scores, women were at significantly greater risk compared to men of scoring in the impaired range on TMTA, TMTB, GP and DSMT. This sex difference was especially marked for categorical T-scores, where women were 2.5 times more likely than men to score in the impaired range on TMTA (p=0.0006) and 5 times more likely to score in the impaired range on GP non-dominant hand (p=0.0001).
Strength and weaknesses: This study has considerable strengths. The WHIS and MACS are the two longest-running, large longitudinal studies of HIV progression in the US. Women are generally underrepresented in US cohort studies of HIV-positive individuals and few studies have directly compared cognitive functioning in HIV-positive women and men. The need to correct this discrepancy is highlighted by the fact that outside the US women account for 50% of HIV cases globally. The study is limited by the absence of comparison between sexes in verbal learning and memory because the WIHS and MACS used different performance measures for these domains. Non-Hispanic blacks were overrepresented and therefore results may not generalize to the broader US population of individuals living with HIV.
Relevance: This study draws attention to the need to consider sex in studying cognitive dysfunction among individuals living with HIV. Findings from studies involving HIV-positive men cannot be generalized to HIV-positive women. Identification of effective interventions to mitigate development and progression of HAND in women cannot rely solely on findings from studies comprised primarily of men. Factors that may play a more prominent role in the evolution of cognitive impairment in women relative to men must first be elucidated. These include biological differences involved in antiretroviral pharmacokinetics as well as hormonal influences on the pathophysiology of HIV CNS involvement. This study extends findings from previous studies demonstrating the adverse effect of psychosocial factors in HIV-positive compared to HIV-negative women: poverty, stress, trauma, mental health and substance use disorders. Future research should examine the interactive effect of cognition and factors known to be elevated in HIV-positive women such as posttraumatic disorder and depression.