Advice
Evidence review
Evidence review
A literature search was conducted which identified 78 references (see search strategy for full details). These references were screened using their titles and abstracts and 26 references were obtained and assessed for relevance.
Three randomised controlled trials (RCTs) identified from the search (Schollhammer et al. 2015, Wolosker et al. 2012 and Costa Jr et al. 2014) were included in this evidence summary. A quasi-randomised controlled trial is also included (Ghaleiha et al. 2012). The included studies were conducted in France, Iran and Brazil. There were no UK-based RCTs identified. A summary of the included studies is shown in table 2 (see evidence tables for full details).
The remaining 22 references were excluded. These are listed in excluded studies with reasons for their exclusion.
Table 2 Summary of included studies
|
Study |
Population |
Intervention and comparison |
Main outcome |
|
RCT |
Adults with hyperhidrosis (generaliseda and localised) and a HDSSb score of 2 or more (n=62) |
Oxybutynin 2.5 mg to 7.5 mg daily versus placebo |
Proportion of patients with an improvement in HDSSb score of 1 or more at week 6 |
|
RCT |
Adults with localised palmar and axillary hyperhidrosis (n=50) |
Oxybutynin 2.5 mg daily to 5 mg twice daily versus placebo |
Patient assessment of hyperhidrosis severity and quality of life at week 6 |
|
RCT |
Women with plantar hyperhidrosis who had previously undergone endoscopic thoracic sympathectomy (n=32) |
Oxybutynin 2.5 mg to 10 mg daily versus placebo |
Transepidermal water loss and quality of life at day 30 |
|
Quasi-RCTc |
Adults with depression and hyperhidrosis secondary to sertraline treatment (n=140) |
Oxybutynin 5 mg daily versus placebo |
Change in HDSSb score at week 2 |
|
Abbreviations: HDSS, Hyperhidrosis Disease Severity Scale; RCT, randomised controlled trial. a Generalised hyperhidrosis was defined as excessive sweating occurring at 2 or more locations (among palmar, plantar, axillary, facial or truncal). b The HDSS is a 4‑point measure of hyperhidrosis severity. A score of 1 or 2 indicates mild or moderate hyperhidrosis. A score of 3 or 4 indicates severe hyperhidrosis. c Treatment groups were determined by participants drawing black or red chips from a ballot box. |
|||
Clinical effectiveness
An overview of the results for clinical effectiveness can be found in results tables.
Patient-reported hyperhidrosis severity
Three of the clinical trials discussed in this evidence summary used a subjective, patient-reported measure of hyperhidrosis severity as a primary or main efficacy outcome measure. Schollhammer et al. (2015) and Ghaleiha et al. (2012) used the validated Hyperhidrosis Disease Severity Scale (HDSS), and Wolosker et al. (2012) used a patient questionnaire that appears to have been developed for this individual study.
Schollhammer et al. (2015) found more people in the oxybutynin group responded to treatment (had an improvement in HDSS score of 1 or more) at 6 weeks than in the placebo group (60% compared with 27%, p<0.01, which is statistically significant). The mean improvement in HDSS score was approximately 1.1 points in the oxybutynin group compared with 0.3 points in the placebo group, from a baseline score of approximately 3.2 in the study population (no statistical analysis reported). It is unclear what level of improvement on the 4‑point HDSS scale is considered clinically significant but, on average, the description changed from 'barely tolerable and frequently interferes with daily activities' to 'tolerable but sometimes interferes with daily activities' in the oxybutynin group.
In Wolosker et al. (2012), at 6 weeks, 74% of people treated with oxybutynin scored their improvement in palmar or axillary hyperhidrosis as 'moderate' or 'great' compared with only 27% of people treated with placebo; the difference between the groups was statistically significant (p<0.001). Statistically significant improvements in plantar hyperhidrosis were also observed, with 92% in the oxybutynin reporting 'moderate' to 'great' improvements, compared with around 13% in the placebo group (p<0.001). However, only 27 patients were included in this analysis and it may have been statistically underpowered to detect any differences between the groups.
In Ghaleiha et al. (2012), at baseline the mean HDSS score was approximately 2.8 (barely tolerable and frequently interferes with my daily activities) across the 2 groups. After 2 weeks, the HDSS score had improved to approximately 1.5 in the oxybutynin group (never noticeable and never interferes with my daily activities) and approximately 2.0 in the placebo group (tolerable but sometimes interferes with my daily activities), and the difference between the groups was statistically significant (p=0.03).
Quality of life
Three of the included studies reported on change in quality of life associated with oxybutynin treatment, although different scales were used to measure this outcome.
Schollhammer et al. (2015) used the validated 30‑point Dermatology Life Quality Index (DLQI) questionnaire to report on changes in quality of life. At baseline the mean DLQI score across the groups was approximately 11 points (out of maximum score of 30), suggesting that the condition was having a very large effect on the patient's lives. After 6 weeks' treatment, the mean improvement in DLQI score was higher in the oxybutynin group (6.9 points) compared with the placebo group (2.3 points), the difference between groups was statistically significant (p<0.01). It is unclear if this improvement is clinically important.
Wolosker et al. (2012) assessed quality of life using patient questionnaires. It is unclear whether those used for measuring change on treatment have been validated, although the authors say the questionnaire assessing the impact of hyperhidrosis on quality of life before treatment has been validated. At baseline, approximately 69% of participants considered their quality of life to be 'very poor', with the remaining participants (31%) reporting 'poor' quality of life. After 6 weeks of treatment with oxybutynin, 35% reported that their quality of life was 'much better', 39% 'a little better' and 26% 'the same'. In contrast, the majority of people in the placebo group (86%) felt that their quality of life remained 'the same' after 6 weeks, with only 14% saying that it was 'a little better' and nobody reporting that it was 'much better'. Nobody in the study reported worsening quality of life. The difference between the groups was statistically significant (p<0.001).
Costa Jr et al. (2014) used a questionnaire for assessing the quality of life associated with hyperhidrosis, which the authors say has been validated. At baseline the adjusted mean quality of life score in the oxybutynin group was 40.4/100 and in the placebo group was 34.8/100; these scores correspond to a rating of 'very good' quality of life. At the end of the treatment period, the mean quality of life score in the oxybutynin group was 17.5 ('excellent', p=0.001 compared to baseline, statistically significant difference from baseline) and in the placebo group was 33.2 (SD 15.3, 'very good', p=0.1 compared to baseline, no statistically significant difference from baseline).
Transepidermal water-loss
Costa Jr et al. (2014) was the only study included in this evidence summary that used an objective measure of sweating: transepidermal water loss. After 30 days' treatment, people in the oxybutynin group had reductions in water loss from the right foot (140.3 g/m2/h to 7.6 g/m2/h), right hand (61.7 g/m2/h to 28.6 g/m2/h), back (38.2 g/m2/h to 10.8 g/m2/h) and abdomen (39.7 g/m2/h to 16.5 g/m2/h); the differences from baseline were all statistically significant (all p<0.01). In contrast, in the placebo group, there were no statistically significant differences in water loss from baseline to day 30 at any of these sites (all p>0.2, see evidence tables for more information).
Other formulations of oxybutynin
No RCTs were identified that investigated the efficacy and safety of modified-release oral oxybutynin or transdermal oxybutynin.
A non-comparative, prospective observational study involving 25 people with hyperhidrosis (mean age 28 years) who received oxybutynin patches twice a week for 10 weeks found that 15/25 (60%) of patients showed an improvement in HDSS score of 1 point or more (Bergón-Sendín et al. 2016). However, without a control arm it is not possible to draw conclusions on the effectiveness of oxybutynin patches for treating hyperhidrosis.
Long-term studies
The RCTs discussed in this evidence summary are all short in duration, although a number of observational studies have looked at the longer-term impact of oxybutynin treatment for hyperhidrosis. Observational studies are more prone to bias and confounding compared with RCTs. Long-term RCTs are needed to confirm the findings of these lower quality studies.
In a retrospective review by Millán-Cayetano et al. (2016), 110 people with hyperhidrosis (mean age 34 years) treated with oxybutynin were evaluated using HDSS at 3 and 12 months. At 3 months, 87/110 (79%) people had partial response to treatment (defined as an improvement in HDSS score of 1 or more), with 69/110 (63%) having an excellent response (defined as an improvement in HDSS score of 2 or more or a HDSS score of 1 at the end of treatment). At 12 months, 101 people remained on treatment, of whom 63/101 (62%) responded to treatment, with 51/101 (50%) having an 'excellent' response.
Wolosker et al. (2014a) reported on 431 people attending a Brazilian dermatology clinic who received oxybutynin for axillary hyperhidrosis, of whom 181 people received oxybutynin for 6 months or more. After a median treatment duration of 17 months (range 6 to 72 months), 83% (150/181) of people reported a moderate to great improvement in axillary hyperhidrosis (defined as an improvement of 5 to10 points on a 10-point scale).
In a similar study, Wolosker et al. (2014b) reported on 570 people with palmar hyperhidrosis treated with oxybutynin. Of the 246 people treated for more than 6 months (median follow-up 16 months, range 6 to 72 months), 90% (222/246) reported moderate to great improvements in palmar hyperhidrosis (defined as an improvement of 5 to 10 points on a 10‑point scale).
Studies in children
No RCTs of children treated with oxybutynin for hyperhidrosis were identified. In a retrospective observational analysis of patient records by Wolosker et al. (2014c), 45 children (aged 7 to 14 years) were treated with oxybutynin. After 6 weeks' treatment, 22/45 children (49%) reported a large improvement in symptoms (score 8 to 10 on a 10‑point scale), 17/45 (38%) a moderate improvement (score 5 to 7) and 6/45 no or slight improvement (score 0 to 4). No children in the study reported a worsening of symptoms. The investigators also reported on changes in quality of life, with 80% of participants reporting an improvement following treatment.
Safety and tolerability
An overview of the results for safety and tolerability can be found in the results tables.
The SPC for oxybutynin states that the adverse effects of oxybutynin were mainly due to its anticholinergic effects, with dry mouth the most commonly reported. The very common adverse events, occurring in 1 in 10 people or more, are constipation, nausea, dry mouth, dizziness, headache, somnolence, vision blurred and dry skin.
Specialists involved in producing this evidence summary raised concerns that taking oxybutynin long-term may increase the risk of some adverse effects. Antimuscarinic medicines should be used with caution in elderly patients due to the risk of cognitive impairment. Oxybutynin may reduce salivary secretions which could result in dental caries, parodontosis (a periodontal disease) or oral candidiasis (SPC: oxybutynin).
The most common adverse event reported in Schollhammer et al. (2015) was dry mouth, reported by 13/30 people (43%) in the oxybutynin group and 3/28 people (11%) in the placebo group, with a statistically significant difference between the groups (p<0.01). Of the 13 people with dry mouth in the oxybutynin group, 6/13 (46%) evaluated it as being of 'slight intensity', 5/13 (38%) evaluated it as 'mild intensity' and 2/13 (15%) as 'severe intensity'. Four people in the oxybutynin group (13%) also reported blurred vision compared with no people in the placebo group. Other adverse events were reported by 1 or fewer people in each group, and included diarrhoea, headache, dizziness and urinary difficulty.
The only adverse event observed in Wolosker et al. (2012) was dry mouth, which at week 6 was reported by 35% (8/23) of those taking higher doses in the oxybutynin group compared with 9% (2/22) of the placebo group (difference between groups statistically significant, p=0.038). It should be noted that participants in this study were specifically asked about the presence of dry mouth.
In Costa Jr et al. (2014), 16/16 (100%) people in the oxybutynin reported dry mouth, compared with 7/16 (44%) in the placebo group (p=0.001). Other adverse events were constipation and drowsiness, although there were no statistically significant differences between the groups.
In Ghaleiha et al. (2012) participants were asked to report gastrointestinal complications, sedation, dry mouth and urinary complications. In the oxybutynin group, 7/66 people (11%) reported gastrointestinal complications, compared with 6/74 (8%) in the placebo group. Dry mouth and urinary complications were both reported by 3/66 people (4%) in the oxybutynin group compared with 0/74 (0%) in the placebo group. No participants in the study reported sedation.
The studies included in this evidence summary used an oxybutynin dose between 2.5 mg and 10 mg daily. The dose of oxybutynin used for overactive bladder conditions is generally higher, ranging from 7.5 mg to 20 mg daily (SPC: oxybutynin). Since patients in the hyperhidrosis trials received a lower dose that was gradually titrated up, they may have tolerated oxybutynin better than people taking it for bladder conditions.
Evidence strengths and limitations
This evidence summary discusses the results of 3 RCTs that investigated oxybutynin for the treatment of hyperhidrosis, and included a total of 144 participants. The results of a quasi-randomised controlled trial involving 140 participants with drug-induced hyperhidrosis are also considered. There are a number of limitations with the studies.
In all studies, oxybutynin was compared to placebo; there are no RCTs comparing oxybutynin to active treatments for hyperhidrosis. In addition, all studies used immediate release oral oxybutynin; the relative efficacy and safety profile of modified release and transdermal patch formulations of oxybutynin for the treatment of hyperhidrosis is not known.
All studies were short in duration, ranging from 14 to 42 days, including the titration phase. Hyperhidrosis is a chronic condition; data supporting longer-term use of oxybutynin are limited to lower quality, observational studies, which are more subject to bias and confounding than RCTs. The 3 RCTs were small, ranging from 32 to 62 participants. Small studies may not have sufficient power to detect a treatment effect, and may not be large enough to identify less common adverse events. The quasi-randomised trial by Ghaleiha et al. (2012) was larger, involving 140 participants, although this study does not use an appropriate randomisation method, which may have introduced bias.
Although the baseline characteristics recorded were generally similar between the oxybutynin and placebo groups, none of the studies appear to have considered factors which may have affected the degree of hyperhidrosis, such as outdoor or sporting activities and local climate. The studies were conducted in France, Brazil and Iran. None of these studies considered the effect of the local climate on hyperhidrosis. There were no UK-based RCTs.
The RCTs included in this evidence summary did not provide a detailed description of the randomisation method, or how blinding was maintained. It is not clear whether allocation was concealed. In addition to this, the anticholinergic effects of oxybutynin, most notably dry mouth, mean it is possible that participants may have known which treatment they were receiving. Participants in Costa Jr et al. (2014) were told that dry mouth was the most common side effect of anticholinergic treatment, and participants in Ghaleiha et al. (2012) were specially asked to report anticholinergic effects, including dry mouth, sedation and urinary complications. This may have further increased the chance of the participants guessing which treatment they were receiving.
Two studies (Schollhammer et al. 2015 and Ghaleiha et al. 2012) used the validated HDSS scale to assess hyperhidrosis severity. However, being only a 4‑point scale the HDSS is potentially less sensitive than other, larger scales (Schollhammer et al. 2015). The scale used by Wolosker for assessing improvement in symptoms does not appear to have been validated.
Unlike the other 3 studies, Costa Jr et al. (2014) did not use a subjective measure of hyperhidrosis; the study reported on objective improvements using transepidermal water loss. There are no validated objective methods to measure the intensity of hyperhidrosis (Schollhammer et al. 2015). The authors of Schollhammer et al. (2015) justify their use of subjective outcome measures by saying that impairment of quality of life depends not only on the severity of hyperhidrosis, but also on each person's individual adaptation to their condition. Specialists involved in the development of this evidence summary suggested that subjective, person-focused measures of sweating (such as HDSS or DLQI) are more widely used in clinical practice compared with objective measures (such as water loss).
Although Costa Jr et al. (2014) used a validated questionnaire to measure quality of life, the authors acknowledge that the questionnaire was designed to assess palmar hyperhidrosis, with only 2/20 questions asking about plantar hyperhidrosis (the condition affecting people in this trial). The authors suggest that this may be the reason that people in the study reported 'very good' quality of life at baseline.
The DLQI used to assess quality of life in Schollhammer et al. (2015) has been validated. It is unclear however, whether the scale used to measure improvements in quality of life in Wolosker et al. (2012) has also been validated.
Many of the participants in the 3 RCTs had localised hyperhidrosis, for which topical treatments, including antiperspirants, are normally the first-line treatment. Systemic treatments would usually be considered only when topical treatments have not controlled the person's sweating. It is not clear from the studies whether topical treatments had been ineffective in all participants.
The participants in Ghaleiha et al. (2012) were taking sertraline for major depressive disorder and in Costa Jr et al. (2014) had previously undergone endoscopic thoracic sympathectomy. It is unclear whether the results of these studies apply to a more general population with hyperhidrosis. It is also unclear whether the antidepressant treatment used in Ghaleiha et al. (2012) may have improved hyperhidrosis as a result of improving anxiety and depression in the participants, particularly as the placebo group also improved.
An overview of the quality assessment of each included study can be found in evidence tables.