Cognitive impairment among older adults with HIV: a systematic review protocol of cohort studies

Transtornos cognitivos em idosos vivendo com HIV: protocolo de revisão sistemática com estudos de coorte

Ibrahim Clós Mahmud

Pontifícia Universidade Católica do Rio Grande do Sul, Brazil

Erick da Rosa Lerner ericklerner2011@gmail.com

Universidade Feevale, Brazil

Yindriana Laguna Rodriguez

Pontifícia Universidade Católica do Rio Grande do Sul, Brazil

Paulo Renato Petersen Behar

Universidade Federal de Ciências da Saúde de Porto Alegre, Brazil

Rodolfo Herberto Schneider

Pontifícia Universidade Católica do Rio Grande do Sul, Brazil

Since its discovery in the 1980s, HIV has affected millions of people around the world. The United Nations reports that approximately 38 million people are living with HIV worldwide. The development of effective antiretroviral therapy has increased the life expectancy of people with HIV.1,2 The progressive spread of HIV among older adults requires constant reassessment and research. However, limited investment in primary and secondary prevention in this age group in low- and middle-income countries, in conjunction with increased life expectancy and an active sex life, are contributing to this public health problem.3 Older adults with HIV are more likely to develop brain pathologies, cognitive impairment, and other neurocognitive disorders earlier than their HIV-negative counterparts, due both to viral damage and the harmful effects of aging. In comparison, among HIV-positive people aged 50 years, life expectancy is lower than among those who become infected at more advanced ages.4,5,6,7

The central nervous system is particularly susceptible to the effects of HIV due to the virus’ high affinity for this tissue type. This is why approximately 50% of people with HIV will suffer from some type of neurocognitive disorder at some point in their lives.7 This has negative consequences not only for patients but for their families and society, since it affects all aspects of memory, attention, and autonomy, increasing the risk of isolation, dependency, and lower treatment adherence.7

Therefore, this study aims to gather information on the incidence of cognitive impairment among older adults with HIV compared to their HIV-negative peers. The expected outcome is to find out higher cognitive impairment in the exposed group. We also will describe the following additional variables: demographic data (country, alcohol use, smoking, advanced age (≥ 80 years), race, immunological stage (mean CD4+) or clinical stage of HIV (according to WHO/CDC criteria), morbidity (other chronic pathologies), antiretroviral therapy and other medications (drug classes and time of use), and dietary factors (sedentary lifestyle, vegetarianism, other diet types).

This protocol is reported according to the preferred reporting items for systematic review and meta-analysis protocols8 and is registered in PROSPERO (CRD42022321914).

The search was performed in August 2022 in the following electronic databases: MEDLINE/PubMed, EMBASE, LILACS, Web of Science, and Scopus.

The reference lists of studies selected for critical appraisal will be screened for other potentially eligible studies. A gray literature search will be performed in Google Scholar.

The MEDLINE/PubMed search strategy, which will be adapted for use in other electronic databases (see Appendix I – supplementary material), will consist of original articles (cohort studies) published between 2012 and 2022, with no language limitations.

Controlled vocabulary will be used whenever possible (ie, MeSH terms for MEDLINE/PubMed, LILACS, WoS, and Scopus and EMTREE terms for EMBASE). The search strategy will combine terms and controlled vocabulary regarding “Cognitive Impairment”, “HIV”, and “Aged.”

The inclusion criteria were determined using the PECO strategy, the most appropriated tool for cohort studies:

• Population: sample aged ≥ 50 years or a mixed population with ≥ 50% of the sample consisting of people aged ≥ 50 years).

• Exposure: this review will consider studies that evaluate patients infected with HIV.

• Comparator/control: This review will consider studies that included older people without HIV infection as an unexposed group.

• Main outcome: studies that report the incidence of cognitive impairment will be included.

Study design: cohort studies, either prospective or retrospective, will be included. Timing and setting: studies lasting ≥ 24 months will be included.

Experimental, in vitro, and animal studies; guidelines and protocols; review studies; letters and editorials; qualitative studies, case reports, and case-control studies will all be eligible for inclusion. Preprints or studies that have not been peer-reviewed will be excluded. After final selection, articles with high risk of bias scores (according the Newcastle- Ottawa Scale) or low quality of evidence scores (C, according the STROBE) will be excluded. Further classification details will be presented in the “methodological quality” and “strength of evidence” sections.

Data management: Rayyan and Excel will be used for importing and deduplicating records, title/abstract screening, full-text screening, and data extraction. Three reviewers (ICM, ERL and, YLR) will independently screen publications regarding the eligibility criteria, extract the data, and assess the risk of bias and overall quality of evidence in the included studies. When consensus cannot be reached, a fourth author (RHS or PRPB) will be consulted.

Selection process: the initial screening will be performed independently, assessing the titles and abstracts according to the selection criteria and classifying them as eligible, potentially eligible, and ineligible. The results of the search and selection process (total studies found, duplicates, and excluded and included studies) will be recorded, reported in the final review article, and presented in a flow chart.

Data collection process: data will be extracted using a standard form to minimize the risk of error. The extracted data will include specific details about the study characteristics (title, authors, country, year of publication, and aims), the participants (sample size, mean age, proportion of participants ≥ 50 years of age, and sex), methods (study design and follow-up duration), dropouts (all groups), outcomes related to the objective (incidence of cognitive impairment in both groups), additional outcomes (demographic variables, immunological stage [mean CD4+] or clinical stage of HIV [according to WHO/CDC criteria], morbidity, antiretroviral therapy, other medications, and dietary factors) and the results (mean, SD or percentage in both groups and measures of association).

The data extraction form will be tested on 3 articles. If necessary, the authors of included publications will be contacted to request missing or additional data.

We will independently assess the risk of bias of included studies according to:

1. 1. Study design;
2. 2. Inclusion criteria;
3. 3. Data availability;
4. 4. Sample size; and
5. 5. Journal impact.

Studies will be considered high-quality if they have a prospective design, use standardized definitions, and are multicenter studies with large sample sizes.

Three independent, previously trained, and qualified reviewers will assess methodological quality using the Newcastle-Ottawa Scale.9 The score for cohort studies will be calculated according to 3 categories: group selection (0 – 4 points), quality of adjustment for confounding factors (0 – 2 points), and ascertainment of exposure and assessment of outcome (0 – 3 points).

The thresholds for converting the Newcastle-Ottawa scores to Agency for Healthcare Research and Quality categories will be:

• Good quality: 3 or 4 stars in the selection domain AND 1 or 2 stars in the comparability domain AND 2 or 3 stars in the outcome/exposure domain;

• Fair quality: 2 stars in the selection domain AND 1 or 2 stars in the comparability domain AND 2 or 3 stars in the outcome/exposure domain;

• Poor quality: 0 or 1 star in the selection domain OR 0 stars in the comparability domain OR 0 or 1 star in the outcome/exposure domain;

Disagreements between reviewers will be resolved by a fourth reviewer.

The data will be presented in a narrative according to synthesis without meta-analysis10 and meta-analysis of observational studies in epidemiology11 guidelines.

The authors will follow Chapter 12 of the Cochrane Handbook for Systematic Reviews of Interventions,12 using standardized tables to describe the data. The results will be will presented in a forest plot in which articles will be grouped according to risk of bias to focus attention on the most reliable evidence, describing the available association measures (odds ratio, relative risk, or hazard ratio) and 95%CI.

This review will analyze the strength of evidence according to STROBE criteria13,14,15 since it will exclusively include observational studies. Assessment will follow Mendes et al.16 with each of the 22 criteria receiving a score of 0 or 1 (ie, “does not meet” or “meets” the criterion, respectively). As recommended by STROBE and Mataratzis et al.17 the studies will be classified into 3 categories: A (meeting ≥ 80% of the criteria), B (meeting 50% to 80% of the criteria) and C (meeting < 50%) of the criteria.

Publication bias will be evaluated using funnel plots if ≥10 studies investigated the outcome of interest.18 The asymmetry of the funnel plots will be assessed using statistical tests, such as the Egger and Begg test.18 p < 0.05 will be considered statistically significant.

The authors expect to find articles that show a higher incidence of cognitive impairment in older people with HIV than their HIV-negative peers, especially in cases of polypharmacy and among those who have been diagnosed with HIV for the longest time.

The increasing incidence and prevalence of HIV infection among older adults associated with increased life expectancy has led to a new public health problem. Considering the physiological changes involved in the aging process and the systemic changes that result from HIV infection and treatment, research in this area is becoming increasingly important.

If we find associations between health conditions or behavior with cognitive impairment in HIV-positive individuals, dissemination of this evidence could ultimately lead to better quality of life and greater social acceptance for them. The results will be disseminated through peer-reviewed journals, scientific meetings, and direct presentations to the general population.