Abstract
Objective. The aim of this study was to describe the prevalence of erectile dysfunction (ED), as well as associated demographic and clinical features, in men with systemic lupus erythematosus (SLE), by means of a systematic, standardized evaluation.
Methods. We performed a transversal study in 8 tertiary care centers in Latin America. We included male patients ≥ 16 years who fulfilled ≥ 4 American College of Rheumatology criteria for SLE and had regular sexual activity, and evaluated them with the International Index of Erectile Function-5 questionnaire. Relevant demographic, clinical, and serological characteristics were recorded. We included 2 control groups: the first was made up of healthy men and the second of men with autoimmune diseases other than SLE (non-SLE group).
Results. We included 590 subjects (174 SLE, 55 non-SLE, and 361 healthy controls). The prevalence of ED in the SLE group was 69%. Mean age in that group was 36.3 ± 1.03 years. Among SLE patients with and without ED, these factors were significantly different: the presence of persistent lymphopenia (p = 0.006), prednisone dose (9.3 ± 1.2 vs 5.3 ± 1.3 mg, p = 0.026), and the Systemic Lupus International Collaborating Clinics damage score (1.25 ± 0.14 vs 0.8 ± 0.16 points, p = 0.042). Independent risk factors for ED in patients with SLE were persistent lymphopenia (OR 2.79, 95% CI 1.37–5.70, p = 0.001) and corticosteroid use in the previous year (OR 2.15, 95% CI 1.37–3.37, p = 0.001).
Conclusion. Regardless of comorbidities, treatment (excluding steroids), and type of disease activity, patients with SLE have a high prevalence of ED, especially considering that most patients are young. Recent corticosteroid use and persistent lymphopenia, which could be related to endothelial dysfunction, are risk factors for this complication in men with SLE.
Systemic lupus erythematosus (SLE) predominantly affects women1, but usually has worse prognosis, higher activity scores, and an increased mortality in men2,3,4, both globally as well as in Latin American patients5. Sexual function, which directly affects quality of life, has different domains that can be affected by disease. In male patients, erectile function is especially relevant, because this disorder is associated with both neuropsychiatric problems (mainly anxiety and depression) and cardiovascular (CV) disease6. Whereas there are different studies regarding sexual function in men with autoimmune diseases [mainly rheumatoid arthritis (RA), spondyloarthritis (SpA), and systemic sclerosis (SSc)]7,8,9,10,11,12,13, as well as studies about reproductive function14,15,16,17,18 and alterations regarding penile anthropometry in SLE19, information about erectile dysfunction (ED) in patients with SLE is quite scant20,21,22.
Males with SLE are mostly young. However, many of them are taking intense immunosuppressive therapies or high-dose steroids, and are at risk for accelerated atherosclerosis2,3,4. Taking these factors into account, as well as the absence of studies performed with a validated instrument that allows a systematic analysis about ED in this group of patients, the main aim of our study was to describe the prevalence, risk factors, and features associated with ED in patients with SLE.
MATERIALS AND METHODS
We performed a transversal study in 8 tertiary care centers throughout Latin America (5 in Mexico, 1 Nicaragua, 1 in El Salvador, and 1 in Colombia) between October 2015 and November 2016. All subjects were included during outpatient visits. Three study groups were formed: men with SLE, men with autoimmune diseases other than SLE (non-SLE group), and healthy controls. In the first group, patients were ≥ 16 years old, fulfilled ≥ 4 American College of Rheumatology classification criteria, and had regular sexual activity in the prior 6 months (with intercourse at least once per week). Patients with other autoimmune diseases were excluded [except for antiphospholipid syndrome (APS), chronic viral infections (human immunodeficiency virus, hepatitis B or C), cancer, and late-onset SLE (diagnosis after the age of 50 yrs]23. We included 2 control groups — 1 with autoimmune diseases other than SLE (non-SLE group, which included RA, SpA, SSc, Sjögren syndrome, systemic vasculitides, adult-onset Still disease, inflammatory myopathies, and primary APS), and the other with healthy controls from each center, who were matched by age with the SLE subjects. Those with diagnosis of a chronic disease or who were using any prescription drug were excluded from this last group. In all groups, patients with incomplete clinical information in their clinical records were excluded.
All included subjects filled out the 5-item version of the International Index of Erectile Function (IIEF-5; Supplementary Table 1, available from the authors on request) in the Spanish-validated version24,25. The questionnaire was completed anonymously by each subject, in an isolated room, without any other person present, and was finally placed into a covered box. In the IIEF-5, the lowest score per question is 1 and the highest is 5. A normal erectile function is considered when scores are between 22 and 25, and ED is graded as mild (17–21 points), mild to moderate (12–16 points), moderate (8–11 points), and severe (5–7 points)24. Two last questions were added for patients, asking whether their rheumatologist had queried them about sexual issues during the previous 3 visits, and whether they would like their rheumatologist to inquire about those issues.
For patients in the SLE group, we recorded demographic information, history of disease activity, comorbidities, autoantibody profile, the SLE Disease Activity Index (SLEDAI)26 activity score, the Systemic Lupus International Collaborating Clinics (SLICC) Damage Index27, and immunosuppressive and nonimmunosuppressive medications. Active disease was defined as SLEDAI ≥ 6 points. Lymphopenia was defined as a total lymphocyte count ≤ 1000 cells/m3 and persistent lymphopenia as at least 3 consecutive measurements ≤ 1000 cells/m3. For patients in the non-SLE group, disease activity was established according to the physician’s global assessment. Briefly, we defined disease activity in this group of patients when it was considered after the clinical evaluation by their attending rheumatologist, and immunosuppressive treatment was adjusted accordingly. Regarding healthy controls, age and anthropometric measurements were recorded.
The study was approved by the local Research and Ethics Institutional Committee (Ref 1661). Variables were described in terms of mean and SD or proportions, as convenient. For comparison between groups, the SLE group was compared with each control group; chi-square test was used for categorical variables or Student t test for quantitative variables. Association between variables was assessed by OR with 95% CI. Variables that showed statistical significance in the univariate analysis or those with clinical relevance were included in the multivariate analysis, which was performed through binary logistic regression. A p value < 0.05 was considered statistically significant. Statistical analyses were performed with support of the software SPSS, version 21.
RESULTS
We included 590 subjects (SLE n = 174, non-SLE n = 55, healthy controls n = 361). According to the setting, the patients with SLE were included as follows: Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán (Mexico City, Mexico), 43 patients; Hospital Central Ignacio Morones Prieto (San Luis Potosí, Mexico), 24 patients; Hospital Universitario Dr. José E. González (Monterrey, Mexico), 22 patients; Hospital Metropolitano Vivian Pellas (Managua, Nicaragua), 19 patients; Instituto Salvadoreño del Seguro Social (San Salvador, El Salvador), 18 patients; Instituto Mexicano del Seguro Social (Guadalajara, Mexico), 18 patients; Hospital General de México Dr. Eduardo Liceaga (Mexico City, Mexico), 15 patients; Hospital Universitario de la Samaritana (Bogotá, Colombia), 15 patients. Patients in the non-SLE group had the following diagnoses: RA (34.54%), SpA (20%), primary APS (14.54%), inflammatory myopathies (7.27%), SSc (5.45%), systemic vasculitides (9.09%), juvenile idiopathic arthritis (3.63%), Sjögren syndrome (1.81%), relapsing polychondritis (1.81%), and Still disease (1.81%). Table 1 displays the main clinical and demographic features in the SLE and non-SLE groups. Information regarding healthy controls is shown in Supplementary Table 2 (available from the authors on request).
Demographic, clinical and serological features of SLE and non-SLE patients with and without erectile dysfunction.
The prevalence of ED in the SLE group was 68.96% versus 23.26% in healthy controls (p = 0.001). There was no statistically significant difference in the ED prevalence when compared with the non-SLE group, but patients with SLE presented with ED at a significantly younger age (36.3 ± 1.03 in SLE vs 46.3 ± 2.2 yrs in non-SLE, p < 0.0001; Table 1). ED in the SLE group was mostly mild to moderate (108/120 patients, 90%). Men with SLE had a mean IIEF-5 score of 17.2 ± 0.33 points. Table 2 shows individual scores per question in patients with SLE, as well as total score and the ED category. Table 1 shows the comparison between patients with and without ED in the SLE and non-SLE groups. Remarkably, prednisone use was more frequent in patients with ED than in those without it (p = 0.026). Regarding comorbidities, the only difference among groups was a higher prevalence of type 2 diabetes mellitus (DM) in non-SLE patients, but there was no difference in patients with or without ED.
Itemized International Index of Erectile Function-5 in patients with SLE.
Table 3 contains relevant information about all patients with SLE. As shown, SLE patients with ED were slightly older than those without ED (36.3 ± 1.03 vs 32.5 ± 1.27 yrs, p = 0.022), without differences in weight, height, body mass index, or time since SLE diagnosis. SLE patients with ED had a lower count of total lymphocytes at the time of the study (p = 0.005), as well as a higher prevalence of persistent lymphopenia (p = 0.006). Regarding immunosuppressive therapy, the use of any of these drugs was equal between groups (p = 0.6). However, patients with ED had a higher frequency of exposure to any dose of corticosteroids in the previous year (p = 0.019), and also a higher prednisone dose at the time of the study (9.31 ± 1.20 vs 5.32 ± 1.29 mg/day, p = 0.02). There were no other significant differences regarding the rest of the immunosuppressive treatment between groups, both at the time of the study and with cumulative doses. There were no relevant differences in comorbidities (p = 0.41) or in nonimmunosuppressive treatments (p = 0.54) between groups. Further, there was no difference regarding disease activity (SLEDAI score 4.89 ± 0.54 vs 3.65 ± 0.52, p = 0.16), but patients with ED had significantly more damage accrual according to the SLICC Damage Index (1.25 ± 0.14 vs 0.80 ± 0.16 points, p = 0.042).
Demographic, clinical, and serological features of SLE patients with and without erectile dysfunction.
After multivariate analysis, risk factors associated with ED in patients with SLE were persistent lymphopenia (OR 2.79, 95% CI 1.37–5.70, p = 0.001) and exposure to any corticosteroid dose in the previous year (OR 2.15, 95% CI 1.37–3.37, p = 0.001).
Finally, we found that most patients who attended the different rheumatology clinics were not questioned about erectile function (86%), at least during the previous 3 visits. Also, most patients (82%) would consider it appropriate to discuss erectile and sexual function in their usual visits (Figure 1).
A. The distribution of responses of patients (n = 174) with SLE to this query: Have you been asked about your erectile function in your last 3 visits to the rheumatologist? B. The percentage of responses from patients with SLE regarding their likelihood of being questioned about their erectile function by their rheumatologists (i.e., would you like your rheumatologist to inquire about your erectile function during your regular visits?). SLE: systemic lupus erythematosus.
DISCUSSION
To our knowledge, this is the first study to evaluate the prevalence of ED in patients with SLE by using an adequate, validated questionnaire24. We found a prevalence close to 70%, whereas in the healthy population the prevalence of ED is between 10 and 22%28. Our findings are similar to those described in patients with DM; however, it is important to note that patients with type 2 diabetes in whom that prevalence has been found were 22 years older than our population, on average29. There are various possible explanations for such a high prevalence of ED in young patients with SLE (mean age 36.3 yrs). Currently, well-known risk factors for ED are obesity (RR 1.9, 95% CI 1.6–2.2), smoking (RR 1.5, 95% CI 1.3–1.7)30, DM (OR 3, 95% CI 1.5–5.8), hypertension (OR 2.05, 95% CI 1.6–2.6), hyperlipidemia (OR 2.2, 95% CI 1.4–3.7), lower urinary tract symptoms (OR 2.2, 95% CI 1.7–2.7), psychological stress (OR 1.6, 95% CI 1.4–1.9), low physical activity (OR 1.3, 95% CI 1.1–1.6), and age31. Whereas the physiopathology of ED involves multiple mechanisms, including the use of certain drugs, history of pelvic surgery or pelvic irradiation, as well as endocrine, neurologic, and psychogenic factors, the main cause associated with ED is local vascular damage32. Interestingly, most of these conventional risk factors were absent in our patients, but the recent use of glucocorticoids was found to be of particular relevance.
Because men with SLE have higher disease activity scores2, they use glucocorticoids more frequently and in higher doses33. We found glucocorticoid use to be a risk factor for ED. Although the use of systemic corticosteroids has not been previously directly related to ED34, up to 70% of patients with Cushing syndrome have sexual dysfunction. It has been suggested that cortisol plays an inhibitory role in male sexual response35. Indeed, hypercortisolism leads to a decreased concentration of luteinizing hormone and testosterone. It also inhibits the pituitary secretion of gonadotrophins and the androgen production in Leydig cells36,37. Also, this effect seems to be reversible, which would explain why recent exposure to glucocorticoids was found to be a risk factor for ED, but the cumulative corticosteroid dose was not. Further, glucocorticoids have been associated with subclinical atherosclerosis in patients with SLE38, and they are considered an additional risk factor for CV disease39. Along with their hormonal actions, their vascular effects could play an additional role in the development of ED in young patients with SLE. ED has been considered as an independent risk factor for major CV events in the general population40. Therefore, it will be relevant to identify SLE patients with ED, not only to offer adequate and timely treatment, but also to assess the presence of other CV risk factors and to begin an appropriate prevention strategy. Further, ED has been found to directly influence quality of life41, which in the case of patients with SLE is already affected by the disease itself42.
While it is widely known that cyclophosphamide may cause oligo/azoospermia, infertility, and alterations in the sex hormone profile in men with SLE15,17, we did not find recent exposure or cumulative doses to be associated with ED.
Further, we also found persistent lymphopenia to be a risk factor for ED in these patients. This could also be related to vascular damage, and specifically to endothelial dysfunction. Low lymphocyte counts have been associated with accelerated atherosclerosis and CV disease43,44. Moreover, our group has described different epidemiological associations between lymphopenia and other conditions related to endothelial dysfunction, such as thrombotic microangiopathy, thrombotic thrombocytopenic purpura, and posterior reversible encephalopathy syndrome, in patients with SLE45,46. Additionally, it has been described that young men with ED have subclinical endothelial dysfunction, even without CV comorbidities47. Therefore, ED could be considered an early clinical sign of increased CV risk, with endothelial dysfunction potentially playing a key role48.
Currently, there is no specific information regarding pharmacological therapies in men with SLE and ED. This should be studied to address their efficacy and safety in this population. However, lifestyle modifications (weight loss, an appropriate diet, and increased physical activity)49 could be recommended for patients with ED, not only to improve erectile function, but also to reduce CV risk. Also, a prompt urology referral could help to individualize each case and the therapeutic options50.
Our work has several limitations. First, it was a transversal study, which did not allow addressing whether there are changes according to disease activity and treatment throughout time. It was performed only in Latin Americans, so these findings may not apply to other populations. Also, there was no neuropsychological assessment at the time of the study, which could have helped to find other factors influencing erectile function. Prospective studies should be performed, with evaluations regarding health-related quality of life and disease perception, as well as functional studies of penile vasculature.
Men with SLE have a strikingly high prevalence of ED, regardless of their young age. Glucocorticoid use and lymphopenia, both of which may cause endothelial dysfunction and lead to vascular damage, are independent novel risk factors for ED in these patients. Men with SLE are rarely asked about sexual function in their regular outpatient visits, even though most of them would agree to such an assessment by their rheumatologist.
- Accepted for publication September 11, 2018.
