Sexual dysfunction in men on buprenorphine - naloxone-based substitution therapy.

BACKGROUND: The literature on sexual dysfunction in patients on buprenorphine-naloxone (BNX) substitution is limited.
MATERIALS AND METHODS: This research aimed to study the prevalence and correlates of sexual dysfunction in men on BNX substitution therapy. We recruited consecutive forty men from BNX clinic, who had received BNX for at least 6 months, who were free from any recent illicit drug use (confirmed by urine chromatographic immune assay), and who were either married or had a stable sexual partner. Men with other psychiatric and substance use disorders (except tobacco) were excluded from the study. Data for the control group were obtained from a published study (with similar selection criteria) from our center. We assessed sexual dysfunction with two cross-culturally validated instruments: Arizona Sexual Experience Scale (ASEX) and International Index of Erectile Function.
RESULTS: The sample had a mean age of 31.6 (±8) years; the mean duration of BNX treatment was 9 (±4.2) months and the mean BNX dose was 4.5 (±1.6) mg. ASEX showed the prevalence of sexual dysfunction to be 40%. The IIEF demonstrated intercourse dissatisfaction (95%) and hypoactive sexual desire (92.5%) as almost universal, while 77.5% of the participants reported erectile dysfunction. In comparison to the published data, these figures were significantly more than among the controls. We found no correlation of sexual dysfunction with marital status, age, duration or dose of BNX, duration of illicit opioid use, the severity of opioid dependence, and tobacco dependence.
CONCLUSION: All men on BNX maintenance therapy must be screened for sexual dysfunction. With the rapid scaling up of office-based BNX substitution, assessment and management of sexual dysfunction ought to be incorporated in the training curriculum. Copyright:

INTRODUCTION

Opioid substitution therapy (OST) is the treatment of opioid dependence with highest evidence base; it is effective not only to cut down illicit opioid use, but also to reduce HIV risk behavior, decrease the incidence of hepatitis C, overdose-related deaths, criminal behavior, and for overall improvement in functioning and quality of life.[1234] Medications, commonly used for substitution, are methadone and buprenorphine (with or without combined naloxone).[5] In India, a substantial majority of the OST programs use buprenorphine.[6]

For the effect of OST on sexual functioning, most of the existing research has focused on methadone.[78] The rate of sexual dysfunction with methadone ranges from 14 to 81;[7] commonly encountered problems are decreased libido and orgasmic dysfunctions. A recently published meta-analysis observed an increased risk of erectile dysfunction (ED) among patients on chronic opioid treatment (including opioid analgesics).[8] Sexual dysfunctions due to opioids are thought to be because of their effect on the gonadal–hypothalamic–pituitary axis.[9] This effect is supposedly mediated through the direct agonist action on the mu receptors.[10] Thus, the existing research suggests a causal link between opioid substitution and the chance of experiencing sexual dysfunction, with variable rates across studies. However, none of the studies had done a comprehensive assessment of sexual dysfunction, covering all the domains.

Of the few published studies focusing on sexual dysfunction with buprenorphine-based opioid substitution,[111213] only one is from India.[14] Overall, the rates of sexual dysfunction were lower with buprenorphine, compared to that of methadone.[7] To the best of our knowledge, except the one study from India, no published paper has yet compared the sexual effects of fixed-dose buprenorphine-naloxone combination (BNX) with a control group. This is important to study because, contrary to the usual belief, even sublingually, the bio-availability of naloxone is nearly 25% (as compared to more than 50% buprenorphine), and this systemic exposure to the mu receptor antagonist, naloxone, might influence the sexual function.[1516]

Given the scope of expanding the available literature, we intended to explore first, the rate of sexual dysfunction with BNX-based OST, second to compare the rate with a control group, and, finally, to examine the correlates of sexual dysfunction.

MATERIALS AND METHODS

The study was cross-sectional in design.

Study sample

The participants were recruited from the OST clinic of the Drug De-addiction and Treatment Centre under the Department of Psychiatry at the Postgraduate Institute of Medical Education and Research, Chandigarh, situated in North India. Our OST clinic runs once a week and prescribes and dispenses BNX as take-home dosages for 1 week. A majority of the patients/visits are with accompanying family member/s who actively participate in the treatment and medication supervision. The study sample included forty consecutive men who were on BNX-based substitution for at least 6 months, who were aged 21–50 years, and who were either married or had stable relationship for at least 6 months. Excluded from the study were those with other substance dependence (except tobacco), with significant symptoms of acute opioid withdrawal, with other comorbid psychiatric disorders, history of trauma or surgery in the pelvic area, organic brain syndrome, and intellectual disability. Also excluded were those with regular use of medications to treat sexual dysfunctions (such as phosphodiesterase inhibitors) and use of naltrexone or benzodiazepine for more than 2 consecutive weeks.

Control group

To compare the rate of sexual dysfunction with a control group, we used the data from an earlier study published from our center.[17] This study too recruited men between 21 and 50 years, who were either married, or who had a stable heterosexual partner. These individuals were nonblood-related caregivers/visitors of patients attending the hospital, who were not regular users of any psychoactive substance except tobacco (using substances twice or less often per week and <60 ml of standard spirits/alcohol v/v 42.7% on each occasion).

Ethical clearance from our institute's ethics committee for intramural projects and written informed consent from all participants were obtained.

Assessment

The entire assessment was done in a single sitting. The sociodemographic profile of all participants was recorded in a prespecified pro forma and included age and education (completed years), marital status, locality, occupation, religion, type of family, duration of marriage/stable sexual relation, and income. Clinical data collected included the dose and duration of substitution therapy, other side effects, age of onset of opioid use, duration of opioid dependence, presence of other medical comorbidities, and tobacco dependence. The Mini International Neuropsychiatric Interview was administered to confirm the diagnosis of opioid dependence and exclude other drug/alcohol dependence and other psychiatric disorders (major depressive episode, generalized anxiety disorder, panic disorder, and schizophrenia spectrum disorders). Clinical Opiate Withdrawal Scale was used for evaluating withdrawal symptoms excluding participants with score more than 5, suggestive of clinically significant withdrawal.[18] Severity of the Opioid Dependence Questionnaire (SODQ) was applied for assessing the severity of opioid dependence.[19]

Sexual dysfunction among the participants was assessed using two instruments namely International Index of Erectile Function (IIEF) and Arizona Sexual Experience Scale (ASEX). Questionnaire.

The IIEF is a 15-item self-administered instrument with well-established cross-cultural validity for the following domains of sexual function: Erectile Function (6 items), Orgasmic Function (2 items), Sexual Desire (2 items), Intercourse Satisfaction (3 items), and Overall Satisfaction (2 items). The Index correlates positively with the clinical interviews of sexual function[20] and is considered to be the gold standard in evaluating erectile function. It marks the sexual dysfunction as present when the total score of the items is <25 for erectile function; <13 for intercourse satisfaction; and <9 for orgasmic function, sexual desire, and overall satisfaction.

The ASEX is a 5-item self-administered inventory with one question each for the evaluation of five dimensions of sexual functioning in both genders regardless of sexual orientation or relationship with a sexual partner. Consistent with the domains of sexual function described in the Diagnostic and Statistical Manual of Mental Disorders-IV, and currently used inventories of sexual function, it covers drive, arousal, penile erection/vaginal lubrication, ability to reach orgasm, and satisfaction from orgasm. Rated on a 6-point Likert scale, scores for each domain as well as the total scale are taken into account.[21] The clinical sexual dysfunction is defined as a score of >19 for the whole scale, >5 for any domain, or ≥4 for any three domains.

The sexual dysfunction in the control group had also been assessed by the IIEF and the ASEX, in the previous study.[17]

Statistical analysis

We used SPSS software version 14 (SPSS Inc, Chicago, US) for data analysis. Frequencies and percentages were calculated for categorical variables and mean and standard deviations (SDs) were calculated for continuous variables. Student's unpaired t-test was used for comparing the continuous variables between two groups (current study group and control group data from our previously published paper).[17] Chi-square test was used for comparing the frequencies (or percentages) between the two above-mentioned groups.

As this study was exploratory in nature, a formal a priori sample size calculation was not done. However, we did a post hoc power calculation using the data on the frequency of any sexual dysfunction on ASEX with the softwarehttp://sampsize.sourceforge.net/iface/index.htmlto see if there was adequate power to detect a true difference between the two groups on this outcome variable.

We also compared subgroups with or without sexual dysfunction as per ASEX and IIEF on various sociodemographic (age, marital status, and socioeconomic status) and clinical parameters (age of onset of opioid use, duration of opioid use, dose of buprenorphine, duration of substitution therapy, severity score on opioid dependence, presence of comorbid tobacco dependence, and comorbid medical disorders).

RESULTS

Sociodemographic and clinical profile

The mean age of the participants was 31.6 years (standard deviation [SD] 8.2). Twenty-seven (67.5%) participants were married, whereas all the remaining were single and had a stable sexual partner. Those who had completed school/10 years of formal education were 80%. Ten (25%) participants were unemployed. According to the composite socioeconomic status (covering occupation, education, and income),[22]75% were from the middle category and 25% were from the lower socioeconomic category. In the control group, all the participants were married. Nearly 84% of the control participants had completed school education of 10 years. With respect to the socioeconomic status, only 4% of the control participants belonged to the lower socioeconomic category. Therefore, the BNX and the control group differed statistically significantly in the marital status (P < 0.0001) and socioeconomic status (P = 0.04). However, there was no difference in the educational background.

The mean dose of BNX was 4.5 mg (SD 1.7), and the mean duration of OST was 8.9 months (SD 4.2). Abuse of mixed opioids (synthetic, semi-synthetic, and natural) was the most common (42.5%), followed by heroin (30%) and the natural opioids (opium and poppy husk; 12.5%). The mean SODQ score was 60.5 (SD 8.3). All the participants had an SODQ score of >30, suggestive of severe opioid dependence. As assessed by the treating doctor and corroborated by the urine test for buprenorphine, treatment adherence was almost 95%. Medical comorbidity was recorded in six (15%) participants; four had hepatitis C, and one each diabetes mellitus and chronic pancreatitis. Current tobacco dependence was recorded in 75% of participants. Recent urine analysis for illicit opioids was negative for 37 patients (92.5%), positive for two participants, and equivocal for one. However, the participants denied any history of recent opioid use, which was also corroborated from their family members.

Rates of sexual dysfunction

The common sexual dysfunctions elicited by IIEF were ED (77.5%), ejaculatory dysfunction (67.5%), intercourse dissatisfaction (95%), and decreased sexual desire (92.5%). As per the ASEX, sexual dysfunction was present in 16 (40%) patients, which included sexual desire-related problems in 15 (37.5%) patients, hypo-arousal and ED in 31 (77.5%) patients, orgasmic dissatisfaction in 21 (52.5%) patients, and ejaculatory dysfunction in 18 (45%) patients. For specific sexual dysfunctions (as per respective cutoff scores mentioned in the assessment section), the Cohen's Kappa concordance across the IIEF and ASEX was fair to moderate; 0.23 (P = 0.04) for ED and 0.47 (P = 0.001) for ejaculatory dysfunction.

Comparison of the rates of sexual dysfunction with a control group (data from previously published study)

For comparison, we did not access the individual patient's data; utilizing the published data, we did the statistical calculations through the online calculator (https://www.graphpad.com/quickcalcs). In our study sample, the rate of sexual dysfunction (as per ASEX) was 40%. In our previously published study, from the same setting, the rate of sexual dysfunction among the controls was 8%.[17] Comparison of the groups showed that the difference was statistically significant (Pearson's χ2= 13.166; P = 0.0003).

Post hoc power calculation using the data on the frequency of any sexual dysfunction (40% vs. 8%) showed a power of 95.5%. Hence, there was adequate power to detect a true difference between the groups as regards the prevalence of any sexual dysfunction.

Comparison of the rates of ED between the BNX group and control group had also shown statistically significantly higher (Pearson's χ2= 34.23; P = 0.0001) occurrence of ED in the BNX group.

Rates of orgasmic dysfunction (Pearson's χ2= 32.065; P = 0.0001), decreased sexual desire (Pearson's χ2= 29.85; P = 0.0001), intercourse dissatisfaction (P = 0.0001), and overall dissatisfaction (Pearson's χ2= 19.56; P = 0.0001) were significantly more in the BNX group. Comparison between the BNX group and the control, with respect to the scores of sexual functions (as per the IIEF), showed significantly lower score (suggestive of dysfunction) in the BNX group, in all domains, namely, erectile function (t = 8.11; P < 0.0001), orgasmic function (t = 7.15; P < 0.0001), sexual desire (t = 9.56; P < 0.0001), intercourse satisfaction (t = 12.9; P < 0.0001), and overall satisfaction (t = 8.1; P < 0.0001). Please see Table 1 for details.

Table 1

Sexual dysfunction: Comparison between the groups on buprenorphine-based opioid substitution and control groups

Sexual dysfunction	OST group (n=40), n (%)	Control group (n=50)#, n (%)	Chi-square/unpaired t-test (P)
As per ASEX, n (%)	16 (40)	4 (8)	13.166 (0.0003)
As per IIEF cut off scores, n (%)
ED (<25)	31 (77.5)	8 (16)	34.229 (<0.0001)
Orgasmic dysfunction (<9)	27 (67.5)	5 (10)	32.065 (<0.0001)
Sexual desire (<9)	37 (92.5)	18 (36)	29.85 (<0.0001)
Intercourse satisfaction (<13)	38 (95)	16 (32)	34.172 (<0.0001)
Overall satisfaction (<9)	33 (82.5)	18 (36)	19.568 (<0.0001)
As per IIEF scores, mean±SD
ED	19.47±5.06	26.3±2.8	8.11 (<0.0001)
Orgasmic dysfunction	7.35±1.92	9.48±0.76	7.15 (<0.0001)
Sexual desire	6.17±1.53	8.76±1.2	9.56 (<0.0001)
Intercourse satisfaction	7.98±2.21	13.04±1.48	12.9 (<0.0001)
Overall satisfaction	6.05±2.29	8.98±1.04	8.1 (<0.0001)

#Data quoted from a published study (Kandappan et al., 2014). OST – Opioid substitution therapy; ASEX – Arizona Sexual Experience Scale Questionnaire; IIEF – International Index of Erectile Function; SD – Standard deviation; ED – Erectile dysfunction

Correlates of sexual dysfunction

Using ASEX criteria when the participants were divided into two groups, with (40%) or without (60%) sexual dysfunction, they showed no significant differences across sociodemographic and clinical parameters.

Using IIEF scores when the participants were divided into two groups, with (n = 31; 77.5%) or without (n = 9; 22.5%) ED, they showed no significant differences across sociodemographic and clinical parameters.

Again using IIEF scores when the participants were divided into two groups, with (n = 27; 67.5%) or without (n = 13; 32.5%) ejaculatory dysfunction, the group with ejaculatory dysfunction had received higher dose of BNX (t = 2.57; P = 0.014), had higher rate of medical co-morbidity (Pearson's χ2= 4.81; P = 0.04), and had higher rate of comorbid tobacco dependence (Pearson's χ2= 4.59; P = 0.03). However, after Bonferroni's correction (P < 0.006) to account for the false-positive association (due to multiple comparisons; total nine comparisons were made), none of the differences remained significant. Table 2 provides the details.

Table 2

Clinical and sociodemographic characteristics: Comparison between groups with or without sexual dysfunction

Sexual dysfunction as per ASEX	Mean±SD/n (%)		Chi-square or Fisher’s exact test/unpaired t-test (P)
	Group with sexual dysfunction (n=16)	Group with sexual dysfunction (n=24)
Age	32.8±9.1	30.3±7.5	0.763 (0.45)
Duration of use	120.0±78.1	97.3±67.4	0.97 (0.33)
Duration of OST	9±4.8	8.7±3.8	0.23 (0.82)
Dose of buprenorphine (BNX) (mg)	4±1.6	4.75±1.7	−1.41 (0.16)
SODQ score	59.2±10.1	61.3±7.1	−0.79 (0.43)
SES score	13.6±3.9	12.3±3.1	0.68 (0.49)
Marital status (single)	4.0 (25)	9 (37.5)	2.85 (0.24)
Tobacco dependence (present)	10 (62.5)	20 (83.3)	2.22 (0.13)
Medical co-morbidity (present)	4 (4)	4 (16.7)	0.41 (0.52)
ED as per IIEF	Mean±SD/n (%)		Chi-square or Fisher’s exact test/unpaired t-test (P)
	Group with ED (n=31)	Group without ED (n=9)
Age	31.8±8.0	32.3±8.9	−0.303 (0.76)
Duration of use	98.7±65	103±90	−1.271 (0.21)
Duration of OST	8.9±4.6	8.6±2.9	0.165 (0.86)
Dose of buprenorphine (BNX) (mg)	4.7±1.7	3.5±1.3	1.89 (0.06)
SODQ score	59.6±8.5	63.3±7.8	−1.17 (0.25)
SES score	13.1±3.4	13.0±3.9	0.097 (0.92)
Marital status (single)	11.0 (35.5)	2.0 (22.2)	5.007# (0.08)
Tobacco dependence (present)	24 (77.4)	6.0 (66.7)	0.43# (0.66)
Medical comorbidity (present)	6.0 (19.4)	2.0 (2.22)	0.036# (1.0)
Orgasmic dysfunction as per IIEF	Mean±SD/n (%)		Chi-square or Fisher’s exact test/unpaired t-test (P)
	Group with orgasmic dysfunction (n=27)	Group without orgasmic dysfunction (n=13)
Age	31.7±8.6	31.4±7.4	0.114 (0.9)
Duration of use (month)	102.2±68.4	117.2±80.3	−0.655 (0.51)
Duration of OST (month)	9.4±4.8	7.8±2.7	1.06 (0.29)
Dose of buprenorphine (BNX) (mg)	4.9±1.6	3.5±1.5	2.57 (0.01)
SODQ score	59.4±9.0	62.7±6.5	−1.16 (0.25)
SES score	12.8±3.0	13.7±4.3	−0.47 (0.4)
Marital status (single)	10 (37.0)	3 (23)	3.7 (0.16)
Tobacco dependence (present)	23 (85.2)	7 (53.8)	4.59 (0.03)
Medical comorbidity (present)	8.0 (29.6)	0	4.81# (0.04)

#Fisher’s exact test. OST – Opioid substitution therapy; ASEX – Arizona Sexual Experience Scale Questionnaire; IIEF – International Index of Erectile Function; SODQ – Severity of Opioid Dependence Questionnaire; SES – Socioeconomic status score; BNX – Buprenorphine-naloxone; SD – Standard deviation; ED – Erectile dysfunction

DISCUSSION

Our study is one of the few to look into sexual dysfunction in patients on buprenorphine-based substitution therapy[11121314] and had examined patients on buprenorphine-naloxone fixed-dose combination using two cross-culturally validated instruments (ASEX and IIEF) for a comprehensive evaluation of sexual function. The results of recent urine analysis for both illicit opioids and buprenorphine enhanced the fidelity of our results, i.e., the study group was actually abstinent from illicit opioids and adherent to buprenorphine. The sociodemographic data and the clinical profile of the participants with opioid dependence (on BNX) were similar to the previously published Indian studies.[1723] Although the dose of BNX was lower (compared to the international standards), it was similar to the published literature from India.[24] Therefore, our sample was likely to represent the treatment-seeking population of India. Another point worth mentioning here is all the participants, in our study, had severe opioid dependence (as measured by the SODQ). Comparison with a control group, with data gathered from the same setting, further strengthened the methodology of our study.

The ASEX has higher threshold for the diagnosis of sexual dysfunction; with its restrictive definition, 40% of our participants had current sexual dysfunction. This figure is much higher than the pooled prevalence of 24% (95% confidence interval [CI] 0.12–0.38) for any sexual dysfunction reported in a recent meta-analysis.[7] However, a study from India reported sexual dysfunction among 83% men on buprenorphine.[14] Thus, the observed rate of any sexual dysfunction in our study fell within the range (12%–83%) of the rate reported by previous studies. Use of different instruments for the assessment of sexual dysfunction could account for some variability across the studies.

By IIEF assessment, the most common sexual dysfunctions in our sample were hypoactive sexual desire (92.5%) and intercourse dissatisfaction (95%), followed by ED (77.5%), while orgasmic dysfunction was least common (67.5%). In the meta-analysis which included studies on both methadone- and buprenorphine-based substitution, the most commonly reported dysfunction was hypoactive sexual desire;[7] our study replicates this result. Higher occurrence of ED in our sample is also in line with the existing literature.[8]

In our study, both any and each individual types of sexual dysfunction in patients on BNX were significantly more than the rates among the general population control group. ED has been the most commonly assessed sexual dysfunction. A meta-analysis showed that compared to the controls, the prevalence of ED was significantly more common (relative risk = 1.96; 95% CI: 1.66–2.32) in people who were on chronic opioid therapy.[8] Our results are similar. The possible causes of sexual dysfunction among patients on opioids are still speculative. Major evidence suggests opioid-induced decrease in the testosterone level; potential mechanisms for the same are inhibition of pulsatile release of gonadotropin-releasing hormone, inhibition of secretion of follicle-stimulating hormone and luteinizing hormone, and finally suppression of testosterone production by direct action of opioids on testicular tissue.[9]

We could not find significant association of sexual dysfunction with any of the sociodemographic and clinical factors included in the study. Although the study group was significantly different from the controls in terms of marital and socioeconomic status, we could not find any association between these two sociodemographic variables and sexual dysfunction in the BNX group. In one meta-analysis, younger age (<50 years) and duration of opioid use (>3 years) were found to be significantly associated with ED.[7] In our study, however, all participants were <50 years of age. Moreover, the mean duration of opioid use was 106.4 months (nearly 8 years). Thus, in our study, the relatively young users and long duration of opioid use might have precluded the possibility of finding significant associations. Another possibility could be that the sample size was too small to detect statistical significance. Nevertheless, lack of association with clinical and sociodemographic variables might also indicate the biological nature of the association, as has been described already.

The significant association of BNX substitution and sexual dysfunction has serious implications for clinical practice. Presence of sexual dysfunction could potentially influence treatment adherence (as seen with antidepressants, antihypertensive, and antiretroviral medications) and quality of life.[25262728] In fact, presumed aphrodisiac effect of the illicit opioids might result in relapse.[29] Clinicians who prescribe buprenorphine-based substitution should be aware of this significant association and should routinely assess patients' sexual function.

The results of our study must be considered within the limitations of our study. We did not do related hormonal assays in our patients. In addition, we had no information of sexual dysfunction prior to the initiation of the substitution therapy. Hence, the cross-sectional nature of our study rules out any speculation about the direction of causality between sexual dysfunction and opioid substitution. All our participants being men and majority being young and chronic opioid users, further curbs the generalizability of our study. Finally, although no prior sample size calculation was done, the post hoc power analysis revealed very high probability that the sample size chosen had adequate power (95.5%) to detect a true difference between the two groups on the main outcome variable.

In sum, we believe our study will add important information to the sparse existing evidence of the association between sexual dysfunction and buprenorphine-based substitution. With the rapid rise of office-based BNX substitution, assessment and management of sexual dysfunction ought to be incorporated in the training curriculum. Early detection and treatment of sexual dysfunction should improve the adherence and acceptance of office-based treatment of opioid dependence.

CONCLUSION

In future, prospective multicentric studies with representation from all sex should be able to address some limitations of the present research. Moreover, an inquiry for potential mechanism of sexual dysfunction could also be investigated.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.