Abstract
Objective. No consensus has been reached on sexual dysfunction in men with ankylosing spondylitis (AS). Our study aimed to derive a more precise estimation of the sexual function and its clinical correlations in men with AS.
Methods. A metaanalysis was performed and the related literature were searched in PubMed, Elsevier Science Direct, China National Knowledge Infrastructure, Chinese Biomedical Literature Database, and in reference lists of articles and systematic reviews. Score of the International Index of Erectile Function (IIEF) was used as the outcome measurement, and standardized mean differences (SMD) with 95% CI were calculated.
Results. Eleven studies were included, including 535 men with AS and 430 male controls. Each domain of the IIEF score (erectile function: SMD −0.52, 95% CI −0.68 – −0.37; orgasmic function: −0.72, −1.03 – −0.42; sexual drive: −0.40, −0.62 – −0.18; intercourse satisfaction: −0.86, −1.15 – −0.56; and overall satisfaction: −0.61, −0.91 – −0.32) were lower in men with AS than in controls. In the subgroup analysis, the results did not change except for the sexual drive in the Asians group (−0.15, −0.42–0.13). At metaregression, no study characteristics were significantly associated with effect size of the IIEF score.
Conclusion. Sexual function is impaired in male patients with AS and further studies are necessary to better understand risk factors for sexual dysfunction in this population.
Since the early 1980s, original studies1,2,3,4,5,6,7,8,9,10,11,12,13 and reviews14 have reported sexual dysfunction in men with ankylosing spondylitis (AS). However, the association between AS and male sexual dysfunction (MSD) had been controversial. Sexual dysfunction is currently established on the basis of the International Index of Erectile Function (IIEF)15, which explores 5 domains: erectile function (EF), orgasmic function (OF), sexual drive (SD), intercourse satisfaction (IS), and overall satisfaction (OS). Demographic and clinical characteristics have also been described, such as disease duration, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), degree of morning stiffness3,4,5,8,10,11,12,13, disease activity3,4,5,8,9,10,11,13, and psychological depression3,5,12,13.
One potential reason for those heterogeneous findings is that many studies were of small sample sizes, resulting in the need for large casuistics to confirm the idea of AS as a cause/risk factor for sexual dysfunction7,8,9,10. Other reasons might be confounding factors such as disease duration and disease activity. Metaanalysis is a systematic approach to identifying, appraising, and synthesizing the results of relevant studies to arrive at conclusions about a body of research, which could enhance the statistical power and draw a more reliable conclusion in comparison to a single study16.
To have a full picture of sexual dysfunction in men affected by AS, we performed a systematic review and metaanalysis of studies available in the literature to answer the following questions: (1) is the presence of sexual dysfunction, assessed by the IIEF score [standardized mean difference (SMD)], greater in men with AS than in controls; (2) are there any clinical or demographic factors associated with sexual dysfunction; and (3) is there a sample size that allows predicting a significant difference between men with AS and healthy controls?
MATERIALS AND METHODS
Data sources and searches
We searched the related studies from PubMed, Elsevier Science Direct, China National Knowledge Infrastructure, and the Chinese Biomedical Literature Database (until March 31, 2014) using the terms “ankylosing spondyloarthritis” or “ankylosing” or “spondylitis”, and “sexual function” or “sexual dysfunction”. A manual search was also performed on reference lists from included articles, reviews, editorials, and proceedings of international congresses.
Inclusion and exclusion criteria
Studies considered in this metaanalysis were required to meet the following criteria: (1) the study was focused on the relationship between the MSD and AS, (2) controls were derived from a population within the same geographic area and ethnic background as study patients with AS, and (3) the publication was in English or Chinese. Studies were rejected if they did not meet the inclusion criteria or if they reported duplicated or useless data.
Study selection and data extraction
When results of 1 study were reported in several publications, only the most recent and complete data were considered. The eligibility of trials was assessed by 2 authors independently (DF and LL; Figure 1). Finally, 11 studies with full text met the inclusion criteria3,4,5,6,7,8,9,10,11,12,13 (Table 1). Data concerning studies, patient characteristics, and outcomes were extracted by 2 authors (DF and LL). Discrepancies were resolved by agreement, and a third author was consulted if necessary.
Data synthesis and statistical analysis
Difference of the IIEF score in patients with AS and controls were used as study outcomes. Effect size was measured by SMD with 95% CI estimated by a random effects model according to DerSimonian and Laird17. Because all of the 5 domains of the IIEF showed different aspects of sexual function, each domain was analyzed in this metaanalysis separately. To assess the influence of heterogeneity on metaanalyses’ conclusions, subgroup analysis was conducted on the basis of ethnicity (white vs Asian).
Heterogeneity was assessed through Q and I2 statistics for each comparison, and potential sources of heterogeneity were discussed where appropriate18. A p value < 0.10 was considered statistically significant, and I2 values of 25%, 50%, and 75% were defined as low, moderate, and high estimates, respectively.
In forest plots, the vertical line represents no difference in the groups (0 for SMD); squares and horizontal lines represent the point estimates and 95% CI for each comparison; diamonds represent pooled effect size; center represents point estimate and width represents 95% CI. Visual inspection of asymmetry in funnel plots was conducted. Begg rank correlation method and Egger-weighted regression method were also used to statistically assess the publication bias (p ≤ 0.10 indicated the presence of statistically significant publication bias). Finally, a metaregression analysis was performed based on studies included in the metaanalysis with the following factors as independent variables: age, duration of disease, and size of study. The dependent variable was log-transformed SMD. All statistical analyses were performed using STATA 11.0 (Stata-Corp) and Review Manager Version 5.0 (provided by The Cochrane Collaboration, available from www.cc-ims.net/revman).
RESULTS
Studies included in the metaanalysis
A total of 11 published studies, including 535 men with AS and 430 male controls, were included in our study. Of the 11 published studies, 7 were conducted in Turkey, 2 in China, 1 in Korea, and 1 in Tunisia. Among these studies, 6 used the IIEF, 1 used the Brief Male Sexual Function Inventory, 1 used the Glombok-Rust Sexual Satisfaction Scale, 1 used the self-estimated intravaginal ejaculatory latency times, and 2 used self-administrated questionnaires to measure patients’ sexual function. In addition, some studies only reported the frequency of men with sexual dysfunction (FSD), and the others reported mean value of the IIEF score and FSD. The median of mean ages was 36.26 years [interquartile range (IQR) 33.00–38.50], and the median percentage of sexual dysfunction was 33.6% (IQR 15.6–46.1%). The sample size of patients with AS in included studies ranged from 22 to 70. The demographic and clinical characteristics of the patients in each study are summarized in Table 1.
Main results, subgroup, and sensitivity analysis
As shown in Table 2, 6 studies (340 patients with AS, 337 healthy controls) were included in the metaanalysis for EF and AS. Statistical significance was observed (SMD −0.52, 95% CI −0.68 – −0.37) in fixed model (p = 0.39, I2 = 4%; Figure 2). Five, 6, 5, and 6 studies were included in our metaanalysis for OF, SD, IS, and OS, respectively. They were all significantly lower, with an overall SMD of −0.72 for OF (95% CI −1.03 – −0.42), −0.40 for SD (95% CI −0.62 – −0.18), −0.86 for IS (95% CI −1.15 – −0.56), and −0.61 for OS (95% CI −0.91 – −0.32; Figure 3). Because of the significant heterogeneity between studies, they were all determined using the random effect method.
In the subgroup analysis based on ethnicity, subjects were divided into whites and Asians. The results did not change except for SD in the Asian group, where results became insignificant (SMD −0.15, 95% CI −0.42–0.13) in fixed model (p = 0.80, I2 = 0%; Table 2).
Sensitivity analysis was performed in each domain of the IIEF to assess the stability of the metaanalysis. When any single study was deleted, the corresponding pooled SMD were changed slightly (data not shown), with the statistically similar results indicating a good stability of the metaanalysis.
At metaregression, effect size of the IIEF was analyzed as a function of age, disease duration, ESR, CRP, morning stiffness, and disease activity; however, none of these were significantly associated.
Publication bias
Publication bias of the literature was tested by Begg funnel plot and Egger test. A potential publication bias of the metaanalysis on the sexual function was detected by the funnel plot on EF. The graphical funnel plots of the 6 studies appeared to be symmetrical (Figure 4), and we found no evidence of publication bias in the Egger regression test (t = −1.42, p = 0.23, 95% CI −5.38–1.73).
DISCUSSION
AS is a chronic inflammatory joint disease that mainly affects the sacroiliac joints and spinal joints, causing pain and stiffness, and progressive fusion of involved joints19. Its common onset age is 20 or 30 years, and it affects men more often than women in a ratio of 2:120. Chronic joint diseases affect millions of people worldwide and have a high burden on society21,22. Because of the axial and peripheral joint involvement and the noticeable inflammation, AS in men may have a severe effect on health status, functionality, and quality of life6,23,24.
Sexuality is an integral part of the personality of human beings. Previous studies have examined sexual problems in patients with AS; the results, however, are controversial because of relatively small sample sizes. In our study, we demonstrated that patients with AS have significantly lower EF, OF, SD, IS, and OS scores on the IIEF than do healthy controls. All findings, with the exception of EF, were of heterogeneity, which was assessed through metaregression analysis. Surprisingly, the heterogeneity was explained by none of the independent variables (i.e., age, disease duration, ESR, CRP, morning stiffness, and disease activity), suggesting that some other factors may need to be considered in future research.
Sexual function is a prominent component of an individual’s quality of life25. All domains of sexual activity may be affected by rheumatic diseases6. It has been reported that patients with AS have significantly lower EF, OF, IS, and OS scores in the IIEF compared to healthy individuals, although no difference was shown in the SD score3. In another study13, EF, OF, and the IIEF total scores were negatively correlated with CRP levels. Bostan, et al26 suggested that the clinical measures of disease activity and functional disability correlated more with CRP than with other laboratory variables. Another study27 pointed out that increases in the CRP levels in patients with AS with axial involvement correlated with the clinical severity of the disease. Patients with higher CRP levels may have more pain and limitation that may influence their quality of sexuality.
The most prevalent concerns of patients with AS are pain (especially backaches), stiffness, and physical restrictions. Because of disease activity and comorbid emotional problems, they may be vulnerable to sexual problems. In our present study, the disease activity and functional conditions were evaluated with the Bath Ankylosing Spondylitis Functional Index (BASFI), the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), and the Bath Ankylosing Spondylitis Metrology Index (BASMI). The effect of the BASFI and BASMI on sexual function has been shown using the linear regression model13. Several previous studies6,24 have also revealed that higher BASFI and BASDAI scores are associated with sexual dysfunction. By limiting the mobility of the axial skeleton, especially in the lower back, AS could lower physical activity during sexual intercourse, and sexual pleasure may be impaired by loss of physical function and by pain. The anticipation that sexual drive or sexual activity will cause back pain could affect psychological status by bringing on depression and anxiety. The sexual health of patients with AS seems to be based on 2 interrelated factors: psychological status and disease activity. Moreover, it is clear that higher disease activity also causes sexual problems and difficulty.
Nonsteroidal antiinflammatory drugs (NSAID) have been demonstrated to increase the risk of erectile dysfunction independent of the indication, and regular use of NSAID was associated with erectile dysfunction28,29. In addition, sexual dysfunction was associated with disease-modifying antirheumatic drugs30. However, 1 study7 reported that anti-tumor necrosis factor-α therapy may reduce sexual dysfunction in male patients with AS, in addition to reducing disease activity. In our study, because of limiting data, we could not evaluate whether there was a relationship between drug usage and sexual problems.
Even though the 11 studies included in this metaanalysis were of high quality, there are some limitations, and thus the result should be interpreted with caution. First, only published studies were included in this metaanalysis, and some relevant studies could not be included in our analysis. Therefore, it is possible that publication bias has been introduced as indicated by the funnel plot, which seemed to be asymmetric. Second, we could not perform further subgroup stratifications analysis because of the limited number of published studies. Finally, studies included in our meta-analysis were of retrospective design, which is subject to some methodological deficiencies.
Our metaanalysis demonstrated that each of 5 domains of IIEF was lower in men with AS than in healthy controls. Sexual dysfunction in patients with AS has not been well studied, with only a few publications about this common but underdiagnosed clinical problem. Future well-designed research with large sample sizes may be required.
Footnotes
-
Supported by grants from the National Natural Science Foundation of China (30771849, 30972530, and 81273169).
- Accepted for publication October 15, 2014.