Persons living with HIV (PLWH), even when well-treated and without detectable viral replication, experience an increased rate of decline in lung function, as measured by forced expiratory volume in one second (FEV₁) and forced vital capacity (FVC), compared to controls without HIV (1, 2). As a corollary, chronic obstructive pulmonary disease (COPD) is more prevalent in PLWH (3) and, low FEV₁ in PLWH is associated with impaired quality of life and an increased risk of hospitalization (4). Likewise, low FEV₁ in COPD is strongly associated with level of dyspnea, overall symptom burden, risk of hospitalization, and mortality (5–8).

Endothelial dysfunction is highly prevalent in all severities of COPD (9, 10), and it is well established that the pulmonary vascular endothelium is involved in chronic lung disease (11). Furthermore, it is proposed that endothelial dysfunction is not only a consequence of damaged alveolar surfaces, but rather a driving factor behind the decline in lung function and a potential target for treatment (12, 13). Additionally, PLWH exhibits increased prevalence of risk factors for endothelial dysfunction such as smoking, dyslipidemia, and abdominal obesity. Also, HIV infection has been proposed as an independent risk factor possibly through HIV associated immune activity or by direct effect on endothelial cells (14). Recent studies have found, that PLWH have elevated plasma levels of both SDC1 and TM (15, 16).

As the endothelial surface layer is being produced and shed continuously, the most used method to investigate changes in the endothelial surface layer is by measuring shed components (17, 18). Both the membrane bound expression and shedding of surface layer components may be regulated, often in response to inflammation or under conditions of cellular stress (19–21).

Syndecans are transmembrane proteins heavily involved in cellular signaling, with syndecan-1 (SDC1) exclusively found on endothelial cells (17). Elevated levels of soluble SDC1 have been associated with increased endothelial damage and worsened outcomes in COVID-19 (22), heart failure (23), pulmonary embolism (24), and renal disease (25) among others.

Thrombomodulin (TM) is a transmembrane protein involved in the antithrombogenic pathway through its binding of thrombin and activation of Protein C (26). Soluble TM has been associated with coronary heart disease and atherosclerosis (27), endothelial damage in renal disease (27), and acute respiratory distress syndrome (28).

We hypothesize that endothelial dysfunction and injury promotes excess decline in lung function in PLWH, and thus, associates with the presence of elevated shed SDC1 and TM is. Moreover, we hypothesize that elevated SDC1 and TM are associated with increased dyspnea as expressed by the medical research council (MRC) dyspnea score.

We did not find an association between the investigated biomarkers of endothelial injury, TM and SDC1, and decline in lung function or dyspnea. However, in a subgroup of never-smokers (n = 236), we did find a significant association between TM and SDC1 in the crude analysis with only TM remaining significant in the fully adjusted model, although not adjusting for multiple testing.

We hypothesized that endothelial dysfunction and injury might play an important role in driving the decline in lung function. Indeed, our analysis of a never-smoking population suggested that TM and SDC1 could serve as markers of endothelial injury and predict impairment in lung function. However, smoking is a very potent driver of lung function impairment and the main cause of COPD in high income countries (32). Also, smoking is a major contributor to endothelial injury and cardiovascular disease (33, 34), as is also seen in the considerably overlap of COPD and cardiovascular diseases (35). Therefore, it is possible that smoking-related endothelial damage may be not directly related to levels of SDC1 and TM.

The equilibrium between membrane bound and soluble SDC1 and TM is complex and not only influenced by endothelial injury, but also regulated by multiple factors (16, 18). As a result of this complexity, results may be difficult to interpret. A case-cohort study by Salomaa et al. observed decreasing incidence of coronary heart disease with increasing levels of soluble TM. However, the also observed an increase in carotid atherosclerosis with increasing TM (36). Similarly, a recent study by Bundgård et al. reported an association between elevated levels of soluble TM and decreased odds for peripheral arterial disease (31). The explanations may be many, but one offered by the two aforementioned studies was, that increasing levels of soluble TM reflects increasing levels of membrane bound TM which protect the endothelium and ensure an overall anticoagulant state both at and in close proximity to the endothelium and thereby reducing the risk of disease (31, 36). Interestingly, in the present study, a subgroup analysis considering only PLWH with airway limitation (n = 67) found that both elevated TM and SDC1 were associated with protection against decline in FEV₁, however the subgroup was small and the findings non-significant.

PLWH with HIV RNA viral load >100 copies and/or CD4⁺ count ≤300 cells/μl were excluded from the main study, as we hypothesized that uncontrolled HIV may be a confounder both influencing SDC-1 and TM levels and lung function decline thereby distorting the association. However, the sensitivity analysis considering PLWH regardless of HIV RNA viral load and CD4⁺ count did not find any association between markers of endothelial damage and FEV₁ decline, in line with the main analysis findings.

This study was performed on a large, well-characterized population of well-treated PLWH on antiretroviral therapy. Further strengthening the study, spirometry data underwent vigorous validation. Some limitations should be considered: Most notably the follow-up was approximately 2 years which might be too short to demonstrate a significant decline in lung function parameters associated with endothelial injury biomarkers TM and SDC1. Also, TM and SDC1 are only measured once at baseline and therefore temporal alterations in biomarker levels cannot be accounted for. Additionally, TM and SDC1 are not measured locally in the respiratory tract. In future follow-up visits it would be interesting to collect sputum samples, to establish whether biomarkers of endothelial injury in sputum is correlated to lung decline. Last, 150 patients dropped out before follow-up, which may introduce attrition bias to some degree.

To conclude, in this study comprising 698 well-treated PLWH, we did not find an association between elevated biomarkers of endothelial injury, TM and SDC1, and an accelerated decline in lung function nor an increased level of dyspnea.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

This study approved by Committee on Health Research Ethics of the Capital Region of Denmark (approval number H-8-2014-004). This study was conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

CR: Conceptualization, Formal analysis, Methodology, Software, Writing – original draft. AK: Conceptualization, Methodology, Writing – review & editing. NA: Conceptualization, Methodology, Writing – review & editing. RT: Methodology, Writing – review & editing. S-LH: Methodology, Writing – review & editing. PS: Methodology, Writing – review & editing. CU: Methodology, Writing – review & editing. TB: Methodology, Writing – review & editing. SO: Methodology, Writing – review & editing. JJ: Conceptualization, Methodology, Writing – review & editing. SN: Conceptualization, Funding acquisition, Methodology, Supervision, Writing – review & editing.