Article Text
PDF
Abstract
Introduction People living with HIV (PLHIV) have a higher risk of developing pulmonary hypertension (PH) with subsequent poorer prognosis. This review aimed to determine the (1) survival outcomes and (2) proportion of emergency department (ED) visits and hospitalisations of PLHIV and PH.
Methods We conducted a systematic review and meta-analysis of observational studies reporting survival outcomes for PLHIV and PH. Electronic databases (Medline, EMBASE, PubMed, Web of Science, Global Index Medicus and Cochrane Library), trial registries and conference proceedings were searched until 22 July 2023. We pooled similar measures of effect, assessed apriori subgroups and used meta-regression to determine mortality and associated variables.
Results 5248 studies were identified; 28 studies were included with a total of 5459 PLHIV and PH. The mean survival (95% CI) of PLHIV and PH was 37.4 months (29.9 to 44.8). Participants alive at 1, 2 and 3 years were 85.8% (74.1% to 95.0%), 75.2% (61.9% to 86.7%) and 61.9% (51.8% to 71.6%), respectively. ED visits and hospitalisation rates were 73.3% (32.5% to 99.9%) and 71.2% (42.4% to 94.2%), respectively. More severe disease, measured by echocardiogram, was associated with poorer prognosis (β −0.01, 95% CI −0.02 to 0.00, p=0.009). Survival was higher in high-income countries compared with lower-income countries (β 0.50, 95% CI 0.28 to 0.73, p<0.001) and in Europe compared with the America (β 0.56, 95% CI 0.37 to 0.75, p<0.001).
Conclusion Our study confirms poor prognosis and high healthcare utilisation for PLHIV and PH. Prognosis is associated with country income level, geographic region and PH severity. This highlights the importance of screening in this population.
PROSPERO registration number CRD42023395023.
- Primary Pulmonary Hypertension
- Viral infection
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
People living with HIV (PLHIV) and pulmonary hypertension (PH) have a poor prognosis; however, the specific survival outcomes of healthcare utilisation have never been elucidated.
WHAT THIS STUDY ADDS
The mean survival of PLHIV and PH was 37.4 months, with almost three-quarters presenting to emergency department and being admitted. Prognosis is associated with country income level, geographic region and PH severity.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
This study highlights the poor prognosis and high healthcare utilisation of PLHIV who develop PH, with regional and financial disparities. This emphasises the importance of screening in this population.
Introduction
The prevalence of pulmonary hypertension (PH) globally is estimated to be 1%, with that percentage increasing to 10% in the global population over the age of 65.1 PH is defined as a haemodynamic state with a mean pulmonary arterial pressure (mPAP) >20 mm Hg.2 While a transthoracic echocardiogram (Echo) is used to estimate mPAP and screen for PH, the gold standard for diagnosing PH involves directly measuring the pressure in the pulmonary artery using right heart catheterisation (RHC).3
People living with HIV (PLHIV) have a higher risk of developing PH, especially pulmonary artery hypertension (group 1).4 The pathogenesis of HIV related to PH is unknown, but several proposed mechanisms exist. One such mechanism is that the HIV viral protein Gp120 induces the release of pro-inflammatory cytokines which have proliferative effects on vascular smooth muscle. Another possible explanation is the implication of the negative factor protein encoded by HIV which has been linked to pulmonary vascular remodelling in macaques testing.5
Multiple studies, including a recent systematic review, demonstrate a higher prevalence of PH among PLHIV compared with the general population.4 6 In the antiretroviral therapy (ART) era, PLHIV has improved survival leading to a greater prevalence of non-infectious complications.7 The literature suggests that PLHIV and PH have a poorer prognosis than those living with PH alone, and ART has not been found to significantly reduce the risk of developing PH for PLHIV, although the evidence is controversial.8–10
The outcomes of PH among PLHIV have not been sufficiently described in the literature. This study aims to summarise the evidence on mortality and morbidity among PLHIV who also have PH. This study will inform prognosis, screening risk factors and HIV-related PH management.
Methods
We conducted a systematic review of studies reporting survival and mortality at years 1, 2 and 3, as well as hospitalisations and emergency department (ED) visits in PLHIV and PH. This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.11
The primary outcomes of interest were:
Mean survival from the time of diagnosis or time when first followed by the study until the time of death or end of follow-up.
Mortality, that is, the proportion of people who died at 1, 2 and 3 years.
The secondary outcomes of interest were:
The proportion of PLHIV hospitalised for reasons related to PH during the study observation periods.
The proportion of PLHIV that visited the ED for reasons related to PH during the study observation periods.
Search strategies and data extraction
Electronic databases (Medline, EMBASE, PubMed, Web of Science, Global Index Medicus and the Cochrane Library (CENTRAL)), trial registries (WHO International Clinical Trials Registry Platform and ClinicalTrials.gov) and conference proceedings were searched for peer-reviewed publications since their inception until 22 July 2023. Reference lists from identified articles were manually reviewed for relevant related publications. Abstracts from the American Thoracic Society, European Respiratory Society, American College of Chest Physicians, Conference on Retroviruses and Opportunistic Infections, International Association of Providers of AIDS Care, Canadian Association for HIV Research and International AIDS Society were reviewed from the past year. We contacted experts in the field for unpublished and ongoing studies.
Appropriate keywords and MeSH terms were used in the comprehensive search strategy and included: ‘pulmonary arterial hypertension’, ‘pulmonary hypertension’, ‘PH’, ‘HIV’ and ‘human immune-deficiency virus’. For a full comprehensive search strategy, see online supplemental appendix 1.
Supplemental material
Inclusion and exclusion criteria
We included studies of PH in PLHIV aged 18 years or more. PH must have been diagnosed using RHC or Echo. PH must have been mostly or entirely group 1, pulmonary artery hypertension. We included randomised control trials, prospective, cross-sectional, case–control and case–cohort studies. Publications in all languages were eligible. We included studies that reported survival or other relevant outcomes. We excluded non-human studies, editorials, reviews and commentaries. We excluded studies that had a significant overlap, defined as equal to or more than 10%, of the study population.
Data management
We used a web-based platform to screen and extract data (DistillerSR, Evidence Partners, Ottawa, Canada). Four investigators independently screened the titles and abstracts (MY, CH, MRW, MBS). Full-text articles of potentially relevant studies were scrutinised for eligibility and tagged for inclusion. The investigators discussed any disagreements regarding study inclusion until consensus was achieved. A fifth senior author arbitrated (LM) when consensus was not reached. We contacted the authors of the included studies to clarify missing or unclear information.
Four authors independently extracted data (MY, CH, MRW, MBS), including information regarding study characteristics (authors’ names, study design, length of study, year of publication, country, sample size, median follow-up period), study’s testing for PH diagnosis (Echo (right ventricular systolic pressure (RVSP)/pulmonary arterial pressure), RHC (mPAP, right arterial pressure, pulmonary capillary wedge pressure, pulmonary vascular resistance)), participant demographics (age, sex), number of ED presentations, mean survival with SD and mortality at 1, 2 and 3 years. The study selection process was documented using a PRISMA flow diagram.12 Reasons for excluding articles were recorded and reported in a list of excluded studies.
Risk of bias
Three investigators independently assessed the methodological quality of all studies (CH, MRW, MBS) using the Risk of Bias In Non-randomised Studies of Exposures (ROBINS-E) tool assessing for selection and outcome bias in non-randomised studies.13 This tool was the most appropriate because our interest was in association between PH in PLHIV (exposure) and outcomes (mortality, ED use and hospitalisations) irrespective of the study design. We assessed for publication bias using a funnel plot for outcomes with 10 or more studies.
Patient and public involvement
There were no patients or public involved in this work.
Analysis
We anticipated clinical, methodological and statistical heterogeneity. Only sufficiently similar studies were pooled. Differences in the participant populations, interventions and outcomes measured were evaluated to make judgement on the appropriateness of pooling data.
The primary outcomes were mean survival and mortality at 1, 2 and 3 years. Secondary outcomes were assessed as the proportion of PLHIV hospitalised or presented to the ED for reasons related to PH. We pooled similar measures of effect using a random effects approach. Where possible, medians were converted into means using the approach proposed by Luo et al.14 We used the generic inverse variance approach to incorporate different measures of effect with their standard errors using the meta-command. We pooled proportions using the metaprop command with the Freeman-Tukey arcsine transformation to stabilise the variances. We assessed statistical heterogeneity using the χ2 test for homogeneity with a significance level, alpha=0.10, and the I2 statistic to quantify inconsistency.
When data were available, subgroup analysis was conducted by the severity of the disease, country income level and WHO geographic regions. Disease severity was stratified into three subgroups, with mild, moderate and severe disease being defined as (cohort’s mean) RVSP of ≤65 mm Hg, 60–80 mm Hg and ≥80 mm Hg as measured by Echo and (cohort’s mean) mPAP of ≤40 mm Hg, 40–50 mm Hg and >51 mm Hg as measured by RHC. The countries where the studies were conducted were stratified into low-income, low-middle-income, high-middle-income and high-income countries according to the World Bank classification15 and by the WHO geographic regions (Africa, America, Southeast Asia, Europe, Eastern Mediterranean and Western Pacific).16
Univariable and multivariable meta-regression was used to determine which variables are associated with mortality (using the metareg command). In the multivariable models, we adjusted for age and sex. We reported effect sizes and beta-coefficients with their respective 95% CI. Variables with less than 10 included studies reporting data were excluded from meta-regression. All analyses were conducted in Stata V.17 (StataCorp 2021). We hypothesised that people with less severe diseases, people in higher income countries and people in more recent studies would have lower mortality and longer mean survival.
Results
Our search retrieved 5248 studies, of which 4781 were excluded during title and abstract screening. During the full-text assessment of the remaining 467 studies, 438 were excluded for not reporting survival or secondary outcomes (n=387), not including PLHIV (n=215) or not being eligible study types (n=312) as outlined in figure 1. Thus, 28 studies were included, of which 14 (50%) were prospective cohorts, 5 (18%) were retrospective cohorts and 9 (32%) were case series (citations in online supplemental material). There were no randomised controlled trials. The median year of publication was 2009 (IQR, 2001–2015). The median study duration was 62 months (IQR, 36–116).
Supplemental material
Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram. ED, emergency department; PH, pulmonary hypertension.
Among the 28 included studies, there were a total of 5459 PLHIV with a median number of 93 participants (IQR, 4–89), a median age of 40 years (IQR, 34–43) and a median of 11% males (IQR, 2–45%) included in each study. 24 studies were conducted in high-income countries, three in upper-middle-income countries and one in a low-income country. The characteristics of the included studies are reported in table 1.
Characteristics of the included studies
Primary outcomes
The pooled estimate for mean survival time of PLHIV and PH was 37.4 months (95% CI 29.9 to 44.8). The pooled proportions of participants alive (95% CI) at years 1, 2 and 3 were 85.8% (74.1% to 95.0%), 75.2% (61.9% to 86.7%) and 61.9% (51.8% to 71.6%), respectively. The results of pooled estimates are summarised in tables 2 and 3.
Study and participants’ characteristics of the included studies
The pooled mean and proportions of pulmonary hypertension outcomes in people living with HIV
Subgroup analysis—disease severity
At each time interval, as disease severity progressed, as measured by Echo and RHC, the proportion of participants alive decreased. In the RHC subgroups, people with mild disease had a mean survival (95% CI) of 93.3 months (78.2 to 108.2) and people with moderate disease had a mean survival of 19.5 months (11.0 to 28.0). Likewise, for the pooled proportion of participants alive, effect sizes (95% CI) at 1 year in the mild, moderate and severe subgroups were 97.8% (85.1% to 100%), 89.0% (78.5% to 96.8%) and 79.4% (57.5% to 96%), respectively. In the Echo subgroups, people with mild disease had a pooled estimated mean survival (95% CI) of 41.8 months (33.2 to 50.5) and those with moderate disease had a pooled estimated mean survival of 25.5 months (11.4 to 39.7). Again, for the pooled proportion of participants alive, effect sizes (95% CI) at 1 year in the mild, moderate and severe subgroups were 83.7% (43.1% to 100%), 80.6% (39.7% to 100.0%) and 40.0% (11.8% to 76.9%), respectively. Similarly, in meta-regression, more severe disease, as measured by Echo, was associated with poorer prognosis (β −0.01, 95% CI −0.02 to 0.00, p=0.009). The results of meta-regression are summarised in tables 4 and 5.
The pooled proportions of pulmonary hypertension outcomes by subgroups of populations, severity and studies
The results of meta-regression of pulmonary hypertension outcomes in PLHIV
Subgroup analysis—country income level
Studies completed in upper-middle-income countries had survival (95% CI) of 95.2% (81.7% to 100%), 87.5% (60.6% to 100%) and 80.5% (54.3% to 97.9%) at 1, 2 and 3 years, respectively. In studies conducted in high-income countries, the survival (95% CI) was 88.0% (76.8% to 96.5%), 76.4% (64.2% to 86.9%) and 62.2% (56.6% to 68.7%) at 1, 2 and 3 years, respectively. Studies conducted in lower-middle-income countries had the lowest survival (95% CI) of 35.1% (27.5% to 43.5%), 35.1% (27.5% to 43.5%) and 23.9% (17.5% to 31.8%) for years 1, 2 and 3, respectively. In our multivariate meta-regression models, studies conducted in higher income countries compared with lower income countries had a statistically significantly higher proportion alive at year 1 (β 0.50, 95% CI 0.28 to 0.73, p<0.001), year 2 (β 0.42, 95% CI 0.16 to 0.67, p=0.004) and year 3 (β 0.36, 95% CI 0.16 to 0.57, p=0.002).
Subgroup analysis—year of publication
In our meta-regression models, the year of publication was not associated with the proportion alive at 1 year (β 0.001, p=0.892), 2 years (β 0.001, p=0.827) or 3 years (β 0.002, p=0.785).
Subgroup analysis—WHO geographic regions
Studies conducted in Europe had the highest proportion of participants alive (95% CI) at year 1 with 91.6% (83.7% to 97.5%) compared with the America with 79.5% (56.8% to 96.4%) and Africa with 35.1% (27.5% to 43.5%). In univariate analysis, it was found that in comparison to studies conducted in America, studies conducted in Europe had a statistically significant increase in proportion alive at 1 (β 0.575, 95% CI 0.37 to 0.78, p=0.001), 2 (β 0.467, 95% CI 0.01 to 0.72, p=0.046) and 3 years (β 0.438, 95% CI 0.10 to 0.78, p=0.026). In multivariate analysis, the relationship was significant at 1 year (β 0.561, 95% CI 0.37 to 0.75, p<0.001) but was not significant at 2 years (β 0.481, p=0.106) or 3 years (β 0.444, p=0.139).
Secondary outcomes
The proportions of PLHIV who visited the ED or were hospitalised for reasons associated with PH during the studies observation periods (95% CI) were 73.3% (32.5% to 99.9%) and 71.2% (42.4% to 94.2%), respectively. In our multivariable meta-regression models, country income level (β 0.272, p=0.583) and year of publication (β −0.037, p=0.080) were not associated with hospitalisation.
Risk of bias
Overall, the risk of bias was high in 19 studies (68%), moderate in 6 (21%) and low in 333 (11%). The two most common concerns were the risk of bias due to confounding and the risk of bias arising from measurement of the exposure. The risk of bias domains is reported in figure 2, and the risk of bias is summarised in figure 3.
Risk of bias domains for included studies.
Risk of bias summary for included studies.
Discussion
In this systematic review and meta-analysis, we have shown that PLHIV who also have PH have poor survival outcomes with high rates of ED visits and hospitalisations. This population’s survival outcomes appear to be associated with country income level, geographic region and PH severity. To the best of our knowledge, this is the most up-to-date and comprehensive systematic review of PH outcomes in PLHIV, highlighting the importance of screening for PH in PLHIV.
When comparing our study’s mortality rates with those of modern cohorts of PH, we find that we obtain comparable data to high-risk patients. Using the Pulmonary Hypertension Association Registry in the USA for those enrolled between September 2015 and September 2020, the mortality rates at 1, 2 and 3 years for high-risk patients were 12–19%, 22–38% and 28–55%, respectively.17 Thus, our study’s mortality rate at 1, 2 and 3 years of 14.2%, 24.8% and 38.1%, respectively, falls into the range presented in the literature of those with high-risk PH.
Our study showed that as PH disease progressed in PLHIV, survival outcomes worsened throughout the follow-up period. Severe PH is known to increase afterload and right ventricle failure and impair gas exchange, triggering systemic vasoconstriction that leads to clinical deterioration and poorer survival outcomes for PLHIV with higher disease severity.18–22 PLHIV with more severe PH are more likely to experience a higher burden of comorbidities and exhibit lower response to treatment, contributing to lower survival rates.23 These findings underscore the prognostic implications of disease severity and reinforce the importance of the early detection and accurate assessment of disease severity in PLHIV with PH to improve survival outcomes for this vulnerable population.24
PLHIV with PH in high-income and upper-middle-income countries had the highest proportion of participants alive at all follow-up points, surpassing those in lower-middle-income countries. It is well-known that countries with lower income levels suffer from the heavy burden of HIV prevalence, exacerbated by limited availability and access to healthcare resources.25 In these resource-limited settings, healthcare providers face barriers to diagnosing PH by RHC or echocardiography, which likely leads to the underdiagnosis and suboptimal treatment of PH for PLHIV.24 In contrast, high-income countries often have greater access to advanced healthcare resources, allowing for enhanced diagnostic and treatment capabilities.26 27 Furthermore, countries with lower income may not have engaged in detailed phenotyping of PH that may have affected comparisons between income groups.
Our study also found that studies conducted in Europe showed lower mortality than those in the America. This could be attributed to a multitude of factors related to healthcare infrastructure, treatment availability and patient demographics.26 28–31 Cultural and socioeconomic factors may also play a role in these disparities and impact the overall healthcare-seeking behaviour of individuals in different regions.31–34
We also found that there was no association between the year of publication of studies and the PH survival outcomes in PLHIV. Although the introduction of ART and its progressive and extensive use have led to improved overall survival outcomes for PLHIV, these have not translated to PH in PLHIV.17 35 Studies that have assessed the incidence and severity of PH in PLHIV before and during the ART era have not shown a definitive change.17 36 37 On the contrary, these studies continue to comment on the improved prognosis of PLHIV and the subsequent importance of screening for increasing non-infectious complications including PH.35–37
The literature suggests that in PLHIV, women experience a greater risk of developing cardiovascular disease than men.38 39 When looking specifically at PLHIV and PH, studies from Europe and Africa have demonstrated an independent association between the female sex and PH as well as a larger proportion of female participants.40 41 Consistent with the literature, our study demonstrated a larger proportion of female participants.
This study has several strengths. This is the first study to comprehensively examine PH survival outcomes in PLHIV. We completed an exhaustive search of the literature, including hand search, abstract screen and articles not in English. We measured survival as continuous and binary. We used meta-regression to quantify associations.
The limitations of this study include heterogeneity and lack of studies from certain regions. Heterogeneity in study design and participant populations may have arisen from heterogeneity in testing (invasive vs non-invasive), treatments, care settings (rural, community or academic sites), social risk factors as well as evolving clinical practice over the chronologic range of included studies. Additionally, the included studies had a high risk of bias due to confounding factors and potential errors in the measurement of exposure. The limited availability of studies from lower-middle-income and low-income countries as well as those from Southeast Asia, Eastern Mediterranean and West Pacific regions highlights the need for further research in these regions to address healthcare disparities and improve outcomes for PLHIV with PH. Given the heterogeneity, bias and limited availability of studies, the findings of the meta-analysis should be interpreted with caution. Future studies should address these gaps by including comprehensive data for better meta-analysis and prognostication. To minimise survival bias associated with outliers, we had planned to report median survival. Unfortunately, due to the nature of the data collected and challenges in pooling medians, we opted to report mean survival despite this limitation instead.
Conclusion
This study highlights the poor prognosis and high healthcare utilisation of PLHIV who also have PH, with regional and financial disparities. This highlights the importance of screening for PH in PLHIV. This study identified country income level, geographic region and PH severity as features that are associated with poorer prognosis and may assist in targeting screening and treatment efforts. This study also highlights the heterogeneity, bias and limited availability of studies pointing to the need for future studies in this field.
Supplemental material
Supplemental material
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
X @Mark__You
Contributors LM was responsible for the overall content of the manuscript as the guarantor. LM conceptualised the research question, contributed to data extraction, data analysis, statistical analysis, manuscript draft and critical revision and final approval of the manuscript. MY conceptualised the research question, contributed to data acquisition, data extraction, data analysis, statistical analysis, manuscript draft and critical revision and final approval of the manuscript. MBS, CH, and MRW contributed to data acquisition, data extraction, data interpretation, manuscript draft and critical revision and final approval of the manuscript. NR contributed to statistical analysis, manuscript draft and critical revision and final approval of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.