Introduction

Chronic obstructive pulmonary disease (COPD) is highly prevalent and heterogeneous in its clinical pre sentation and evolution¹. It is characterized by chronic and usually progressive and persistent airflow obstruc tion. It is estimated that the prevalence worldwide is around 10%, 14% in Latin America², and 8.9% In Colombia³. In the last decades, the prevalence of COPD has grown exponentially⁴.

Erectile dysfunction (ED) is a common disease among adult men defined as the inability to achieve and maintain an erection enough to allow satisfactory sex ual performance⁵. It is a benign disorder that affects physical and psychosocial health affecting the quality of life of those who suffer from it and their families⁶. The prevalence in the general population of this dis ease varies in the different age groups, being more frequent in men > 60. The DENSA study, conducted to estimate the prevalence of ED in Latin American coun tries (Colombia, Ecuador, and Venezuela), found a prevalence of ED in Colombia of 52.8%⁷ A global multivariate analysis estimates that the risk of ED increases by at least 10% for each year of age⁸. It is estimated that by 2025, 322 million men will suffer from this disease^9,10.

In COPD, a prevalence study carried out in 1982 found that approximately six out of 20 COPD patients were suffering from erectile dysfunction¹¹. However, in more recent studies conducted in cities located at sea level, it was found that between 72% and 78% of patients with COPD suffer from some degree of ED^12-15. Although a recent meta-analysis did not find a higher prevalence of ED in COPD patients when compared to patients without COPD, it documented an increased risk of moderate or severe ED in COPD patients¹⁶.

Dyspnea, cough, muscle weakness, and decreased physical activity associated with low testosterone levels have been postulated as the leading causes of decreased sexual activity in patients suffering from COPD^13,15,17.

International Index Erectile Function (IIEF) was written in English and validated in 12 countries and ten lan guages, including Spanish. It was validated in Peru^18,19, a country with cultural and epidemiological characteris tics like other Latin American countries. While evaluat ing these patients, physicians ignore the association with some comorbidities, such as ED^12,19. Only one quality of life questionnaire for patients with asthma and/ or COPD includes questions about sexuality Quality-of-life for Respiratory Illness Questionnaire (QOL-RIQ)²⁰. A study showed that 87% of patients with COPD do not discuss their sexual problems with their doctor, and 78% do not share these problems with their partner²¹.

The studies that evaluated the prevalence of ED in the COPD population were conducted at sea level, and the extrapolation of these criteria to high altitudes is questionable. Based on sea-level studies, it is impos sible to ensure that the prevalence is the same, which is why the findings cannot be extrapolated^12,22 With these data, we could be underestimating the preva lence of ED negatively affecting the quality of life of patients with COPD, who are doomed to settle for low sexual satisfaction.

We aimed to assess if there is an association between ED and airflow obstruction in patients with COPD living at high altitudes and its prevalence and association with other medications or comorbidities. This study was conducted in a reference institution for pulmonary dis eases in Bogota, Colombia, which is 2640 m above sea level, likely representative of populations living at high altitudes (≥ 2500 and < 3500 m).

Materials and methods

This study was conducted as an observational cross-sectional survey in 150 male patients between 40 and 80 years of age with a confirmed diagnosis of COPD defined as the presence of a risk factor: smoking with ≥ 10 pack-year (PY) history and/or exposure to wood smoke ≥ 10 years in a closed room; and persistent obstructive airflow limitation (forced expiratory volume in the first second [FEV1/forced vital capacity [FVC] < 70% after bronchodilator). The exclusion criteria were clinical instability during the past 8 weeks (defined by the need for hospitalizations or consultation for acute changes in symptoms or current medication), patients who refused to sign the informed consent, and patients who did not have or are not interested in having sexual activity. An Institutional Review Board approved the study proto col, and all subjects provided written informed consent.

Results

We included 169 men with a diagnosis of COPD con firmed by spirometry, with an average age of 69.9 years. The general characteristics are described in table. Most patients had moderate to severe airflow obstruc tion, and only 10.7% had a mild obstruction. We found decreased arterial oxygen pressure (PaO₂), oxygen sat uration, and diffusing capacity for carbon monoxide (DLCO) percentage (Table). Regarding erectile func tion, the prevalence of ED was found in 78.8% of patients, with an average score of 15 in the IIEF (Table 3); however, in 27.2% of patients, erectile function was not evaluable (Table 3).

It was verified that the cross-cultural adaptation of the version in Spanish made in Peru is valid, following existing recommendations for scales that have already been translated and validated in the same language but in a different country than the subjects interviewed.

Regarding the transcultural adaptation and verifica tion of the psychometric characteristics of the IIEF-15 scale, the interdisciplinary committee of experts evalu ated each item of the scale in both the original version created by the author Raymond Rosen and the scale validated in Spanish in Peru by Zegarra et al., as well as the relevance of the application of the scale in the evaluation of the disease, the semantic, idiomatic and conceptual equivalence of the items, the applicability in the different regions of the country, and the daily routine of the situations raised in the questionnaire for the Colombian population, which determined that the scale validated in Peru is understandable and culturally equivalent to the Colombian population; however, minor changes in the syntactic order and the addition of synonyms in parentheses to increase understanding were made. Regarding the feasibility of the survey, 99% of the patients filled out the questionnaire entirely in the test and the re-test, ≤ 1% of the answers were invalid or marked twice, and the average completion time was 6 (IQR: 5-10) min. Most of the patients had a high school or lower educational level. Finally, when assessing the psychometric characteristics of the IIEF-15 scale, Cronbach's a was 0.959, and the absolute intraclass correlation coefficient was 0.936 (95% CI: 0.873; 0.968) (p < 0.001).

A logistic regression model was constructed sequentially with the presence of ED (yes/no) as the dependent variable and using DAGS; in the bivariate analysis, the variables associated with ED with p < 0.25 were: age, number of PY smoked, years of exposition to wood smoke, dyspnea, Beck's depres sion score, FEV1/FVC, meters in the 6-min walk test, marital status, heart failure, use of beta-blocker, angiotensin converting enzyme inhibitors, and statins (Table 4); such variables were included in the multivariate model. The most parsimonious multivariate model included beta-blocker and statin use; the vari able FEV1 was also included in the analysis because it was part of the main objective. However, no statis tical association was found in the bivariate analysis (p = 0.603) or the multivariate model. In the final model, the use of a beta-blocker had an odds ratio of 7 (p = 0.016), statins 0.35 (p = 0.032), and FEV1 0.004 (p = 0.730) (Table 5). Finally, the model's goodness of fit was demonstrated by the Hosmer-Lemeshow test (p = 0.199).

Discussion

The results of this study showed that the prevalence of ED patients with COPD living at a high altitude is high (78.8%). The findings align with the previous stud ies in cities located at sea level, where they found a prevalence between 72% and 78%^12-14. According to our study, there is no association between FEV1 and the presence or severity of ED in patients with COPD. However, such an association cannot be completely rule-out, considering the limitations of our study design. Our results also support that PaO₂ is not associated with the degree of ED.

The high prevalence of ED in patients with COPD compared with the general population, supported by the findings of our study and by the literature, leads us to believe that even if airflow obstruction seems not to be a contributing factor to ED, other factors present in COPD, like the inflammatory cascade perhaps related to the history of heavy smoking, could be contributing to the pathophysiology of ED²³, and should be explored in future studies.

It should also be considered that this high prevalence of ED may be because the diagnosis of COPD is usu ally made after several years of the onset of symptoms when the patient reaches an advanced age. Therefore, in our cohort, as in similar studies, most patients were > 60 years, and it has been widely described that age is strongly associated with ED²⁴, probably due to the progressive decrease in age of sex hormones, as well as an androgenic deficiency that causes adipocyte pro liferation in the corpora cavernosa and causes tissue damage in the smooth muscle, which interferes with the veno-occlusive mechanism of the erection, and initiates the cycle of ischemia, tissue damage, and fibrosis²⁵.

Finally, we can emphasize that regardless of the severity of COPD, the prevalence of ED is very high; therefore, COPD patients should be screened for this condition. In our population, we found a strong associ ation between the use of beta-blockers and the pres ence of ED, previously described in patients with cardiovascular disease^26-28, especially if they consume this group of medications²⁸. Although the association of ED with beta-blockers had not been previously described in patients with COPD, it is biologically plausible, espe cially in patients with additional risk factors for ED. The use of nebivolol, a third-generation beta-blocker, does not seem to negatively influence the sexual function of patients who consume it^26,29, and may have favorable effects in users of traditional beta-blockers when switch ing to this drug²⁷. In this group of patients, statins could be a protective factor for ED. Therefore, we recommend that evaluating erectile function in patients with COPD should be a routine procedure in the usual consultation. This screening should be done with the instrument that we have validated in this study in the Colombian popu lation since it is easy and rapid completion (6 min) to make comprehensive assessments and timely interven tions, such as addressing urology specialists, which impact the quality of life of patients with COPD.

Conclusion

Erectile dysfunction is an under-questioned topic by doctors caring for COPD patients, and patients rarely comment on sexual problems with their doctors. Furthermore, there is only one COPD quality of life ques tionnaire, including items about sexuality (QOL-RIQ) ⁽²⁰.

Although the severity of COPD is not associated with ED, the prevalence of ED in COPD is higher than in the general population. We recommend screening for ED in patients with COPD using the IIEF questionnaire. Beta-blockers have a strong association with ED in COPD patients that had not been previously described in COPD.