Interventions for hyperhidrosis in secondary care: a systematic review and value-of-information analysis

Iontophoresis compared with placebo

Three very small studies (two RCTs, one interrupted time series) compared tap water iontophoresis with a placebo (see Appendix 2, Table 43). 37,40,42

Dahl and Glent-Madsen37 was a half-side comparison RCT including 11 adult patients. Patients had one hand randomised to tap water iontophoresis and the other to a sham treatment. Initial treatment involved sessions of 15 minutes, repeated one to five times a week, and continued until ‘good subjective effect’ was reported. After the initial treatment (duration not reported), six patients continued on maintenance treatment every other week. Efficacy was evaluated with gravimetry, after the initial treatment and after 3 months of maintenance treatment. After the initial treatment, the median difference between treated and placebo hand in sweat reduction was 32%, favouring the treated hand (p < 0.01). The six patients who continued maintenance treatment had an 81% (median) reduction in sweating from baseline at 3 months (p < 0.05). No adverse events were observed.

Stolman42 was a half-side placebo-controlled randomised trial evaluating tap water iontophoresis to the palm delivered for 20 minutes, three times per week, for 3 weeks, in 18 adults. Sweating was evaluated with the iodine starch test. At the end of treatment, 83% patients experienced a ‘marked reduction’ in sweating in the treated hand, whereas no change was noticed in the untreated hand. No serious adverse events were reported.

Karakoc et al. 40 was an interrupted time series study that included 15 patients who received a low-intensity AC (designed as sham treatment) iontophoresis (with tap water) for 15 minutes, eight times over 28 days on both hands, followed by iontophoresis DC (with tap water) 1 week later. Sweating was evaluated by gravimetry up to 1 week after the end of each treatment. There was no reduction in sweating 1 week after the end of the sham treatment, but there was a statistically and clinically significant reduction in sweating 1 week following iontophoresis therapy (mean reduction of 88% in sweating from baseline, from 3.1 ± 0.4 g/hour to 0.4 ± 0.1 g/hour in the right hand and from 3.2 ± 0.3 g/hour to 0.4 ± 0.1 g/hour in the left hand). No data on adverse events were reported.

Iontophoresis compared with no treatment

Two small non-randomised trials compared a ‘dry-type’ iontophoresis device for the hand with no treatment (see Appendix 2, Table 44). 43,44

Na et al. 43 was a non-randomised study evaluating the effectiveness of a ‘dry’ hand-held iontophoresis device. Ten patients were instructed to hold the device with one hand for 30 minutes daily for 1 week, then every other day for another week. The other hand received no treatment. Results were assessed with gravimetry at up to 28 days from treatment initiation. There was a statistically significant difference (p < 0.001) in sweat reduction from baseline favouring the treated hand at the end of treatment (43% vs. 2%). At 2 weeks’ follow-up there was also a statistically significant difference in mean sweat reduction from baseline favouring the intervention (18% vs. 2%), although it is unclear if this result can also be considered clinically significant. The authors stated there were no serious adverse events.

Choi et al. 44 was a non-randomised study evaluating a similar ‘dry’ hand-held device. Twenty-three patients were instructed to hold the device with the left hand for 20 minutes daily for 4 weeks. The right hand received no treatment. Effectiveness was evaluated by gravimetry, Investigator Global Assessment (IGA), Patient Satisfaction Assessment and hydration capacitance. Improvement in IGA and hydration capacitance from 2 to 8 weeks’ follow-up and after 8 weeks was significantly greater in the treated palm than in the untreated hand (p < 0.05), but the clinical significance of this finding is unclear becuase of limited reporting. There was no statistically significant difference between the two palms in gravimetry and Patient Satisfaction Assessment. There were no serious adverse events.

Iontophoresis compared with botulinum toxin

Two studies (one RCT, one non-RCT) compared iontophoresis with BTX injections for palmar hyperhidrosis (see Appendix 2, Table 45). 88,92 It was not stated in the study reports whether tap water or anticholinergic drugs were used in the water bath; however, given the recency of the studies, it is assumed that tap water was used.

Rajagopal and Mallya 201488 compared the efficacy of iontophoresis with BTX (BTX-A formulation) for the palm in a RCT of 60 patients. Thirty participants were randomised to sessions of iontophoresis (2 × 10 minutes) combined with topical aluminium chloride (20% lotion) three times per week over 4 weeks, and 30 patients received injections of BTX-A [100 units (U) per palm]. Efficacy was assessed 4 weeks after the intervention using HDSS. Patients were also asked to score improvement in their symptoms at follow-up (either mild, good, excellent or no improvement). Participants without a HDSS score improvement from baseline at 4 weeks were crossed over to the other group for another 4 weeks and those with improvement in HDSS were followed up in the same group for 6 months. At 4 weeks, the proportion of reporting an improvement in HDSS of ≥ 2 points was 57% in the BTX-A group, compared with 27% in the iontophoresis group. This difference was clinically and statistically significant (p = 0.037). Similarly, the proportion of patients who achieved a reduction in HDSS of at least 1 point /was higher in the BTX-A group (80%) than in the iontophoresis group (47%). Among non-responders, the rate of improvement was significantly higher in those switching from iontophoresis to BTX-A than in those switching from BTX-A to iontophoresis. The study reported that the effect duration in responders was greater with BTX than with iontophoresis plus topical therapy. There was a statistically significant difference favouring BTX-A in the proportion of patients rating improvement as excellent or good (80% vs. 47%; p = 0.037). There were no severe adverse events. Mild to moderate pain (n = 8) and mild temporary motor weakness (n = 1) were reported in the BTX-A group.

Wachal et al. 92 carried out a non-randomised study of 86 adults with hyperhidrosis of the hand and other upper limb. Participants received one of three interventions: iontophoresis (n = 28), BTX-A (n = 22) or sympathectomy (n = 36). Iontophoresis patients received treatment for 20 minutes every 2–3 days. The total duration of treatment was not reported. Effectiveness and willingness to undergo retreatment were assessed by patients at 1, 6 and 12 months’ follow-up, using two VASs, both ranging from 1 to 10 points (with higher scores indicating greater effectiveness/willingness). The percentage of patients assessing treatment as either good/very good (5–10 points on the VAS) was higher in the BTX-A group than in the iontophoresis group at all follow-up points, but was statistically significant only at 1 month (90.9% vs. 35.7%). Similarly, the percentage of patients who were willing to undergo retreatment was greater in the BTX-A group than in the iontophoresis group, although the difference was statistically significant only at 1 month (81.8% vs. 42.8%). The authors reported that there were no adverse effects associated with iontophoresis and no adverse event data were reported for the BTX-A group.

Iontophoresis compared with iontophoresis with anticholinergics

Two studies (one RCT, one non-RCT) compared iontophoresis alone with iontophoresis combined with anticholinergic therapy (see Appendix 2, Table 46). 38,41

Shimizu et al. 41 carried out a RCT of 52 adult patients randomised to one of three treatments for palmoplantar hyperhidrosis: iontophoresis with AC, iontophoresis (AC) plus oral oxybutynin (4 mg/day) or iontophoresis with DC. Iontophoresis therapy was delivered for 30 minutes once a week to both hands and feet. Efficacy was measured by gravimetry and the results were presented after up to 15 treatment sessions. There was a statistically and clinically significant mean reduction from baseline in all three treatment groups from after the second treatment (p < 0.05), but no statistically significant difference in effectiveness between treatment groups at the end of follow-up (AC group from 0.73 to < 0.25 mg/cm²/minute; AC + oxybutynin from 0.81 to < 0.25 mg/cm²/minute; DC group not reported). No adverse events were reported other than small vesicles on palms in the DC group (n = 5) and dry mouth and eyes in the group receiving oxybutynin (n = 2).

Dolianitis et al. 38 conducted a non-randomised half-side comparison study in which patients acted as their own control, comparing iontophoresis plus a glycopyrrolate solution with standard iontophoresis plus tap water only. Of the 20 patients included, 19 had their palms treated and one had their soles treated. All patients underwent two treatment sessions. In the first treatment, patients were blinded as to which tray contained glycopyrrolate. Patients had one palm/foot treated with iontophoresis plus a 0.05% glycopyrrolate solution and the other side received iontophoresis plus tap water only, for 10 minutes. In a second visit, both palms or soles were treated for 10 minutes with 0.05% glycopyrrolate solution. The interval between both visits was between 2 and 3 weeks for most participants. Patient daily self-assessment was used to evaluate the total number of dry-hand days (total number of days during which hands were totally dry). The intervention resulting in the highest median number of dry-hand days was bilateral glycopyrrolate (11 days, range 0–31 days), followed by unilateral glycopyrrolate (5 days, range 0–17 days) and tap water (3 days, range 0–15 days). The incidence of dry/sore throat (n = 8) was higher after bilateral glycopyrrolate than after unilateral treatment. The authors reported that there were no other adverse events.

Non-comparative studies

The study by Karakoç et al. 39 was a case series involving 112 patients who received iontophoresis for both hands, eight times over 28 days. Sweating was evaluated by gravimetry 20 days after the end of treatment (see Appendix 2, Table 47). A statistically significant reduction in sweat from baseline was reported for 81% of participants and there were no serious adverse events.

Evidence summary for iontophoresis

In summary, 10 studies (four RCTs, five non-RCTs, one case series) of iontophoresis were included. All were rated as being at high or unclear risk of bias.

Three very small studies (two RCTs, one interrupted time series) compared iontophoresis with a placebo37,40,42 and found a positive effect of iontophoresis. Two studies reported a statistically significant difference in sweat reduction (measured by gravimetry) in the short term (approximately 1 to > 3 months from treatment initiation), favouring iontophoresis. 37,40 One RCT reported a marked reduction in hand sweating at 5 days post treatment (iodine starch test) when treatment with iontophoresis was compared with no treatment. 42

Two small non-randomised half-side comparisons with no treatment, in which patients acted as their own control, evaluated a hand-held ‘dry-type’ iontophoresis device;43,44 only one of them found a statistically significant reduction in sweating (gravimetry) and a significant improvement in the treated palm compared with the untreated palm at up to 28 days from treatment initiation.

Two studies (one RCT, one non-RCT) compared iontophoresis alone with iontophoresis combined with anticholinergic therapy. 38,41 One study found no significant difference in sweat reduction (gravimetry) of the hands and feet between iontophoresis alone and iontophoresis combined with oral oxybutynin by the end of the 6-week treatment period. Another study reported that iontophoresis with topical glycopyrrolate resulted in a longer duration of effect (as well as a higher incidence of dry/sore throat) than with iontophoresis alone.

Two studies (one RCT, one non-RCT)88,92 compared iontophoresis with BTX injections for palmar hyperhidrosis. One RCT found a statistically and clinically significant difference in treatment response (HDSS) and patient-reported symptoms between the two interventions favouring BTX at 4 weeks from baseline. One non-RCT92 reported a statistically and clinically significant improvement in patient-reported symptoms and willingness to undergo treatment at 1 month, but the difference was no longer statistically significant at 6 or 12 months. This may reflect a waning of the effect following a single treatment with BTX compared with regular use of iontophoresis, although the evidence is too limited to confirm this hypothesis. There was limited evidence suggesting that effect duration following end of therapy with iontophoresis was shorter than after BTX injection. Patients receiving BTX experienced pain more frequently than with iontophoresis.

One additional case series reported a statistically significant short-term reduction in sweating (gravimetry) for most patients undergoing standard tap water iontophoresis for the hand (20 days’ follow-up). 39

Overall, there is very low-quality but consistent evidence suggesting a beneficial effect of iontophoresis in the treatment of palmoplantar hyperhidrosis compared with placebo or no treatment and that the effect of iontophoresis after treatment discontinuation is short-lived. Compared with iontophoresis alone, the evidence for the effectiveness of combining anticholinergic therapy with iontophoresis is mixed and inconclusive.

There is very low-quality evidence to suggest that iontophoresis is less effective than BTX injections at reducing palmar hyperhidrosis symptoms in the short term and that the effect duration following end of treatment is shorter than with BTX.

Topical glycopyrrolate compared with placebo

Two small RCTs compared glycopyrrolate wipes with placebo, used for hyperhidrosis of the axilla for 4 weeks53 or the face for 10 days (see Appendix 2, Table 48). 51

Mehrotra et al. 53 carried out a three-arm randomised trial that compared 4 weeks of daily treatment glycopyrrolate wipes for axillary hyperhidrosis (4% group or 2% group) with placebo in 38 patients. Efficacy was evaluated with HDSS and gravimetry. At the end of week 4, patients receiving glycopyrrolate experienced a statistically and clinically significant reduction in sweating from baseline (glycopyrrolate 4%, 59% reduction; glycopyrrolate 2%, 48% reduction), and this was significantly greater than the reduction achieved in the placebo-treated group (16%). There was also a significantly greater proportion of responders (i.e. HDSS score reduction of ≥ 2 points) in the glycopyrrolate groups (glycopyrrolate 4%, 50%; glycopyrrolate 2%, 35%) than in the placebo group (9%). Adverse event rates were similar between groups, although it appears that blurred vision, dry mouth and application site discomfort were more frequent in the groups receiving active treatment.

In the study by Hyun et al. ,51 39 patients were treated with 2% topical glycopyrrolate on half of the forehead, while the other half of the forehead was treated with a placebo. Efficacy was evaluated with gravimetry and HDSS. At the end of the 10-day treatment, there was a statistically significant difference of 37% [standard deviation (SD) 11.4%] in reduction of sweat production rate favouring intervention versus placebo (p < 0.025). There was no difference in mean change from baseline in HDSS scores between treatment and placebo, although the risk of within-patient correlation cannot be excluded. One patient reported a transient headache following the intervention and no other adverse events were reported.

Topical glycopyrrolate compared with botulinum toxin type A

One study non-RCT compared topical glycopyrrolate with BTX (see Appendix 2, Table 49). Baker91 allocated patients (n = 40) with axillary hyperhidrosis to one of four groups: glycopyrrolate spray (2% or 1%), BTX-A injections or no treatment. Efficacy was evaluated using HDSS. At 6 weeks following treatment, patients who had received glycopyrrolate 1% had less improvement in HDSS scores than the BTX group; the difference was statistically significant (p < 0.05), although it was not clear if it was also clinically significant. There was no significant difference between glycopyrrolate 2% and BTX-A.

Evidence summary for topical glycopyrrolate

Three studies (two RCTs, one non-RCT) of topical glycopyrrolate were included. Two small, low-quality (rated as having high or unclear risk of bias) RCTs evaluated short-term treatment with glycopyrrolate wipes against placebo, used for hyperhidrosis of the axilla53 or the face. 51 Both studies found that the reduction in sweating (gravimetry) on the treated sides at the end of treatment was significantly greater in the active treatment group than in the placebo group. Compared with placebo, there was evidence of a significant short-term improvement in patient-reported disease severity (HDSS) in patients receiving treatment for axillary hyperhidrosis, but not for facial hyperhidrosis. There was limited and inconclusive evidence from one non-RCT91 regarding the relative effectiveness (HDSS) and safety of BTX injections compared with glycopyrrolate spray for axillary hyperhidrosis.

Overall, there is very low-quality evidence suggesting a short-term benefit of topical glycopyrrolate for axillary and facial hyperhidrosis. No evidence for other treatment sites was found.

Evidence summary for oxybutynin

Three studies, all rated as having an unclear risk of bias, evaluated the effectiveness and safety of oral oxybutynin against placebo for hyperhidrosis of the axilla and palm,60 foot57 and generalised hyperhidrosis. 59

One study reported statistically significant improvement in patient-reported symptoms compared with placebo at the end of treatment, in both the axilla and palm, at 6 weeks. 60 Another study reported a statistically significant difference in sweat reduction (gravimetry) in the foot and the hand at 30 days, although the reduction may be of limited clinical relevance. 57 A third study reported a small difference favouring active treatment compared with placebo in patient-reported disease severity improvement (HDSS) at the end of 6 weeks of treatment, although, again, this reduction was small and may be of limited clinical relevance. 59 All three studies reported a short-term quality-of-life improvement favouring oxybutynin, as well as a significantly higher incidence of dry mouth symptoms in patients receiving active therapy.

Overall, there is low-quality evidence suggesting a short-term small benefit of oxybutynin in hyperhidrosis symptoms and a short-term improvement in quality of life compared with placebo. There is insufficient evidence to indicate whether or not the effectiveness of oxybutynin differs according to target area. Oxybutynin is associated with a high incidence of adverse events, particularly dry mouth.

Evidence summary for methantheline bromide

Two placebo-controlled RCTs of oral methantheline bromide were included. 68,69 Both studies were rated as being at a high or unclear risk of bias. They both reported relatively small reductions in axillary sweating (gravimetry) after 4 weeks of active treatment compared with placebo, but no difference in sweating of the palms. One study also reported small, non-clinically significant, differences in patient-reported disease severity (HDSS) and quality-of-life improvement favouring treatment against placebo. Both studies reported a statistically and clinically significant increase in dry mouth symptoms in patients receiving active treatment compared with those receiving placebo.

Overall, there is low-quality evidence that, compared with placebo, methantheline bromide has a short-term positive effect on axillary hyperhidrosis symptoms and quality of life, although this effect is small and may not be clinically significant. There is no evidence that methantheline bromide improves symptoms of the palm or any other body parts. There is evidence suggesting that methantheline bromide is associated with a high incidence of dry mouth symptoms.

Evidence summary for topical botulinum toxin

Only one very small study (rated as having an unclear risk of bias) evaluated the efficacy of topically applied BTX for axillary hyperhidrosis. 71 There is insufficient evidence to conclude on the effectiveness and safety of topical BTX for primary hyperhidrosis.

Axillary hyperhidrosis

By far the most studied intervention/site combination was BTX for axillary hyperhidrosis: there were many studies versus placebo or no treatment reporting various outcomes.

Botulinum toxin compared with placebo

Of the nine studies that compared subcutaneous injections of BTX with placebo for axillary hyperhidrosis,72,73,75,77,79,80,82,83,85 eight were RCTs and one80 was an open-label continuation study of an included RCT (see Appendix 2, Table 53). 79

Two studies reported using HDSS as a measure of efficacy,77,83 six measured sweating using gravimetry. 75,77,79,80,82,83 Five studies measured quality of life. 73,77,79,80,83 Patient satisfaction was assessed in three placebo-controlled BTX studies, using four different tools. 79,80,83 Three studies reported on duration of effect. 73,77,83 All studies reported data on adverse events.

Out of 10 studies, five were considered sufficiently similar to be pooled in a meta-analysis. 75,77,79,82,83 A range of outcomes were reported as presented in the following sections.

Hyperhidrosis Disease Severity Scale

Two studies estimated treatment response in terms of a reduction in HDSS score of ≥ 2 points. 77,83 These studies were pooled in a meta-analysis. Lowe et al. 77 was a three-armed trial that compared two doses of BTX-A (50 U and 75 U) with placebo, and both active treatment arms were considered sufficiently homogeneous to be pooled. Figure 3 shows that there was a large and statistically significant difference in odds of response at 4 weeks favouring BTX-A [RR 3.30, 95% confidence interval (CI) 2.46 to 4.43]. There was no difference in response rates between the two BTX-A doses in Lowe et al. 77 (50 U and 75 U). There was no evidence of heterogeneity.

FIGURE 3.

Botulinum toxin vs. placebo. Reduction of ≥ 2 points in HDSS at 4 weeks. df, degrees of freedom; IV, inverse variance. In Lowe et al. ,77 the 75-U and 50-U groups were presented separately to explore any dose–response effect and total n/N in the placebo group was 27/108; events and total sample size were divided by two to avoid double counting. This artificially reduces the power of the study in the meta-analysis and does not account for correlation between the two active treatment groups. However, a separate analysis combining the two arms showed no significant difference in results.

Ohshima et al. 83 found that HDSS response rates were maintained in the intervention group at 16 weeks and reported a similarly large difference compared with placebo at this time point (RR 5.96, 95% CI 2.87 to 12.39).

In the Ohshima et al. study,83 the initial treatment phase (which lasted 16–24 weeks) was followed by an open-label second treatment phase; the entire study lasted until 40 weeks after first treatment. Participants initially allocated to BTX received either a re-injection of BTX-A 50 U per axilla or no re-injection. Participants initially allocated to placebo received either one injection of BTX-A 50 U per axilla or no injection. Patients received a BTX-A injection during the second treatment phase if they met ‘re-injection criteria’: mean sweat production beyond 50% of baseline at any time between weeks 16 and 24 [of 78 patients allocated to BTX-A, 34 (44%) were included in the second treatment phase]. Treatment response was defined as ≥ 50% reduction in mean sweat from baseline and a reduction from baseline of ≥ 2 points on the HDSS. Rates of adverse events were higher in the intervention group (54%) than in the placebo group (30%). Duration of effect, defined as the number of days between the initial treatment and the first recording of > 50% of baseline sweat production, was 273 days (95% CI 171 days to a number of days that was not reported) for the BTX group compared with 35 days (95% CI 28 to 56 days) for placebo.

Lowe et al. 77 measured the median effect duration in the subgroup of responders, defined as time to return to a HDSS score of 3 or 4 points after treatment. The study reported a median effect duration of 205 days in the 50-U group, 197 days in the 75-U group and 96 days in the placebo group. The difference between each BTX-A group and placebo was statistically significant.

Sweating (gravimetry)

Five RCTs were included in at least one meta-analysis evaluating the effect of BTX on gravimetry against placebo75,77,79,82,83 and one unblinded continuation study of the Naumann et al. 80 RCT was only summarised narratively.

Sweating (gravimetry) response (2–4 weeks’ follow-up)

Figure 4 shows a pooled analysis of four studies75,79,82,83 reporting a reduction of ≥ 50% sweating from baseline at 2–4 weeks’ follow-up and suggests a large and statistically significant difference favouring BTX-A compared with placebo (RR 3.27, 95% CI 1.93 to 5.55).

FIGURE 4.

Botulinum toxin vs. placebo. Reduction in sweating from baseline to 2–4 weeks of ≥ 50%. df, degrees of freedom; IV, inverse variance. Follow-up duration was 2 weeks for Heckmann et al. 75 and 4 weeks for Naumann et al. 79 and Ohshima et al. 83 Median follow-up duration in Odderson82 was 2 weeks (range 1–8 weeks). Data for Odderson82 were extracted from figures.

There was evidence of heterogeneity (I²-statistic = 94.7%), which seems to be due to the Heckmann et al. 75 study. Heckmann et al. 75 used a higher dose than the other studies (200 U vs. 50 U) and had a half-side comparison design. However, Figure 4 shows no evidence of a dose–response effect and suggests that the results in Heckmann et al. 75 were driven by a relatively lower overall response rate in the placebo group (15% in Heckmann et al. 75 vs. 40% overall in the other studies). The reason for the heterogeneity in response rates across placebo groups is unclear. Notwithstanding the heterogeneity the treatment effect favours BTX.

Figure 5 shows a meta-analysis of three studies75,77,82 and suggests a similarly large and statistically significant difference in the odds of patients experiencing a reduction of at least 75% in sweat volumes at 2–4 weeks’ follow-up (RR 6.74, 95% CI 2.84 to 16.03).

FIGURE 5.

Botulinum toxin vs. placebo. Reduction in sweating from baseline to 2–4 weeks of ≥ 75%. df, degrees of freedom; IV, inverse variance. Follow-up duration was 2 weeks for Heckmann et al. 75 and 4 weeks for Lowe et al. 77 Median follow-up duration in Odderson82 was 2 weeks (range 2–8 weeks). Data for Odderson82 were extracted from figures. In Lowe et al. ,77 total n/N in the placebo group was 20/108; events and total sample size were divided by 2 to avoid double-counting.

There was again evidence of heterogeneity (I²-statistic = 92.4%) due to the Heckmann et al. 75 study results, which had a higher RR than the other studies. As with Figure 4, Figure 5 shows no evidence of a dose–response effect and suggests that the results in Heckmann et al. 75 were driven by a relatively lower response rate in the placebo group than in other studies (3% in Heckmann et al. 75 vs. 18% overall in the other studies), rather than a higher response rate in the intervention group (79% in Heckmann et al. 75 vs. 76% overall in the other studies). As before, the reason for the heterogeneity in response rates across placebo groups is unclear.

Sweating (gravimetry) response (16 weeks’ follow-up)

Figure 6 shows a pooled analysis of three studies79,82,83 reporting a reduction of ≥ 50% sweating from baseline at 16 weeks’ follow-up and suggests a large and statistically significant difference favouring BTX-A (RR 2.87, 95% CI 1.94 to 4.26). There was no evidence of heterogeneity.

FIGURE 6.

Botulinum toxin vs. placebo. Reduction in sweating from baseline to 16 weeks of ≥ 50%. df, degrees of freedom; IV, inverse variance. Follow-up duration was 16 weeks for Naumann et al. 79 and Ohshima et al. 83 Median follow-up duration for Odderson82 was 16 weeks (range 10–21 weeks). Data for Odderson82 were extracted from figures.

Figure 6 suggests that the effect observed in patients undergoing BTX-A treatment at 2–4 weeks (see Figure 4) is largely maintained at 16 weeks, albeit slightly reduced: the absolute rate of BTX-A patients with ≥ 50% reduction in sweating at 2–4 weeks was 94% compared with 83% at 16 weeks.

Sweating (gravimetry): mean percentage sweat reduction (2–4 weeks)

Figure 7 shows a meta-analysis of four studies77,79,82,83 and suggests a large and statistically significant difference in mean percentage sweat reduction from baseline favouring BTX-A. Compared with placebo, patients receiving BTX-A had an approximately 57% greater reduction in sweating at 2–4 weeks’ follow-up (MD –56.83%, 95% CI –64.61% to –49.04%). There was no evidence of heterogeneity.

FIGURE 7.

Botulinum toxin vs. placebo. Mean per cent reduction in sweating from baseline to 2–4 weeks. df, degrees of freedom; IV, inverse variance. Follow-up duration was 4 weeks for Lowe et al.,77 Naumann et al. 79 and Ohshima et al. 83 Median follow-up duration in Odderson82 was 2 weeks (range 1–8 weeks). Data for Odderson82 were extracted and calculated from figures.

Sweating (gravimetry): mean percentage sweat reduction (16 weeks)

Figure 8 shows a pooled analysis of three studies,79,82,83 which suggests a slightly greater difference in mean percentage sweat reduction at 16 weeks’ follow-up. Compared with placebo, patients receiving BTX-A had an approximately 67% greater reduction in sweating (MD –66.93%, 95% CI –82.76% to –51.10%). There was no evidence of heterogeneity.

FIGURE 8.

Botulinum toxin vs. placebo. Mean per cent reduction in sweating from baseline to 16 weeks. df, degrees of freedom; IV, inverse variance. Follow-up duration was 16 weeks for Naumann et al. 79 and Ohshima et al. 83 Median follow-up duration for Odderson82 was 16 weeks (range 10–20 weeks). Data for Odderson82 were extracted and calculated from figures.

Naumann et al. 80 was a 16-month open-label continuation study of the Naumann et al. 79 RCT and was not included in the meta-analyses. After 4 months of the initial trial, in which patients were randomised to either BTX-A 50 U or placebo, participants could receive up to three further treatments with open-label BTX-A over 12 months. Of the 207 participants enrolled during the entire 16-month period, 80 received one treatment, 93 had two treatments, 30 received three treatments and four only received placebo during the initial phase of the trial. Outcomes assessed included treatment response (gravimetry), patient satisfaction and quality of life (bespoke questionnaires) and adverse events. The proportion of responders, defined as those achieving ≥ 50% reduction in spontaneous sweating, was higher in the BTX groups than in the placebo group both 4 weeks after treatments (BTX group: first treatment, 96.1%; second treatment, 91.1%; third treatment, 83.3%; placebo group: 34.7%) and 16 weeks after treatments (BTX group; first treatment, 85.7%; second treatment, 87.3%; third treatment, 80.5%; placebo group: 20.6%).

Sweating (sweat area)

Two studies evaluated sweat outcomes using non-gravimetric methods. Schnider et al. 85 measured sweat production by digitising and analysing ninhydrin-stained sheets using standardised image analysis software. Balzani et al. 72 used the Minor’s iodine starch test. Both studies reported clinically relevant improvements in sweating at follow-up in the BTX group compared with the placebo group.

Quality of life

Of the five studies that measured quality-of-life outcomes, two used DLQI,77,83 one used SF-1279 and two used a bespoke questionnaire. 73,80

Two studies were combined in a meta-analysis. Figure 9 shows that there was a statistically significant difference in mean reduction in DLQI score from baseline to 4 weeks’ follow-up favouring BTX-A over placebo (MD –4.80, 95% CI –5.67 to –3.94). There was no evidence of significant heterogeneity. Ohshima et al. 83 reported further results at up to 16 weeks following initial treatment, which suggested that the effect was maintained over this period.

FIGURE 9.

Botulinum toxin vs. placebo. Mean reduction in DLQI score from baseline to 4 weeks. df, degrees of freedom; IV, inverse variance. In Lowe et al. ,77 the total sample size of the placebo group (n = 108) was divided by 2 to avoid double-counting.

Naumann et al. 79 reported a statistically significant difference in mean change in quality of life, measured using SF-12 physical component summary (PCS) but not mental component summary (MCS) scores, from baseline to 16 weeks, favouring BTX-A. However, the improvement in PCS appeared small and may not be clinically significant. The two studies that used a bespoke unvalidated questionnaire73,80 both reported an improvement in quality of life following BTX-A treatment, although it is unclear whether or not this improvement was clinically significant.

Patient satisfaction

Three studies reported data on patient satisfaction, using HHIQ79 and a global assessment of treatment satisfaction questionnaire. 79,80,83 Patients’ global assessment of treatment satisfaction was measured on a scale ranging from –4 to 4, with negative values indicating worsening and positive values indicating improvement. A score of 0 indicates no change from baseline, +1 indicates slight improvement, +2 moderate improvement, +3 substantial improvement and +4 complete abolition of signs and symptoms. Patient satisfaction, as measured by global assessment score, ranged from moderate (mean score of 2.6)79,83 to substantial (mean score of 3.5)80 in patients receiving BTX-A and from unchanged (mean score of 0.3)79,83 to slight (mean score of 1.4)80 in patients receiving placebo at 4–16 weeks’ follow-up.

Adverse events

None of the studies reported serious or severe adverse events related to the intervention. The most common treatment-related events reported included injection site pain and CS. Four studies reported injection site pain events,73,75,77,80 with incidence rates ranging from approximately 1%75 to 12%. 77 Six studies reported higher rates of non-axillary/CS in patients/axillae receiving BTX-A,77,79,80,82,83,85 with event rates ranging from under 1%82 to 15%. 85 One event of increased facial sweating was also reported in one study. 75 There was no evidence to suggest an increase in incidence or severity of treatment-related adverse events in the two studies in which patients received repeated injections. 80,83

Further study details and results not included in the meta-analysis are reported in Appendix 2, Table 53.

Palmar hyperhidrosis

Seven studies evaluated the effectiveness of BTX for palmar hyperhidrosis. 74,76,81,84,88,90,92 Of those, five targeted the palm only74,76,84,88,90 and two targeted the palms and/or axillae. 81,92

Summary of evidence for subcutaneous injection of botulinum toxin

Twenty-three studies of BTX used as subcutaneous injection were evaluated. More than two-thirds of the studies focused on treatment for axillary hyperhidrosis exclusively and about a third focused on palmar hyperhidrosis.

In the case of axillary hyperhidrosis, BTX was compared with placebo in nine studies (eight RCTs,72,73,75,77,79,82,83,85 one open-label continuation study80), no treatment in three studies81,89,185 (non-RCTs) and curettage in four studies21,148,155,184 (one RCT, three non-RCTs). There was evidence that BTX, compared with placebo, resulted in a clinically and stastically significant reduction in sweat production (gravimetry) (five pooled RCTs, one open-label continuation study) and improvement in patient-reported symptoms (HDSS) (two pooled RCTs). There was evidence to suggest that reductions in sweat production (three pooled RCTs) and improvements in HDSS score (one RCT) were largely sustained at 16 weeks’ follow-up. There was also evidence of a statistically significant improvement in some measures of quality of life in BTX-treated patients compared with placebo-treated patients at 4 weeks’ (four RCTs)73,77,80,83 and 16 weeks’ follow-up (three RCTs, one open-label continuation study). 73,79,80,83 None of the studies of BTX for axillary hyperhidrosis reported serious or severe adverse events that were considered to be related to the intervention by the investigators. The most common treatment-related adverse events reported included injection site pain and CS. Three non-randomised half-side comparison studies (in which patients acted as their own control) of BTX and no treatment reported broadly similar results. 81,89,185

Of the four studies (one RCT, three non-RCTs) that compared curettage with BTX,21,148,155,184 one small half-side RCT21 found a difference in HDSS score (at 3 and 6 months’ follow-up) and gravimetric sweat measurement improvements (3 months’ follow-up) favouring BTX, although the difference was statistically significant only for HDSS score. Three non-randomised studies found no evidence of a significant difference in gravimetric sweat measurements at up to 26 weeks (one non-RCT),155 patient-reported sweating or patient satisfaction at 6 months (one non-RCT),148 HDSS score improvement at 6 weeks (one non-RCT)184 or in quality of life up to 26 weeks (one non-RCT),155 between curettage and BTX. When reported, the incidence of adverse events was higher with curettage than BTX (one RCT,21 two non-RCTs148,155).

For the treatment of palmar hyperhidrosis, BTX was compared with placebo in three RCTs, no treatment in two non-RCTs and iontophoresis in two studies (one RCT, one non-RCT). Compared with placebo, two studies reported a small but statistically significant reduction in sweating outcomes at short-term follow-up (3–13 weeks), measured either by gravimetry (one RCT)76 or by sweat area tests (one RCT),84 although one study found no statistically significant difference compared with placebo using the iodine starch test. 74 There was very limited evidence from a single study of a significant improvement in quality of life at 30 days’ follow-up favouring BTX. 74 One study (BTX-B 5000 U) found a high incidence of treatment-related adverse events. 74 Two non-randomised half-side comparison studies (in which patients acted as their own control) compared BTX with no treatment. 81,90 Both studies reported a difference in at least one measure of sweat reduction favouring the intervention, although the clinical significance of these results is unclear. There were no serious adverse events.

As stated previously (iontophoresis evidence summary), two studies (one RCT, one non-RCT)88,92 compared iontophoresis with BTX injections for palmar hyperhidrosis. One RCT88 found a statistically and clinically significant difference in treatment response (HDSS) and patient-reported symptoms between the two interventions, favouring BTX, at 4 weeks from baseline. One non-RCT92 reported a statistically and clinically significant improvement in patient-reported symptoms and willingness to undergo treatment at 1 month, but the difference was no longer statistically significant at 6 or 12 months. There was limited evidence suggesting that effect duration following discontinuation from iontophoresis was shorter than after BTX. Patients receiving BTX experienced pain more frequently than with iontophoresis.

Overall, there is moderate-quality evidence of a large effect of subcutaneous BTX on symptoms of axillary hyperhidrosis in the short and medium term (up to 16 weeks), and of a small to moderate positive effect on quality of life in the short term, compared with placebo. There is low-quality evidence to suggest that BTX is associated with higher patient satisfaction in the short to medium term, as well as a higher incidence of non-severe adverse events, notably injection site pain and CS, compared with placebo. There is very low-quality evidence regarding the relative effectiveness of BTX injections to the axillae compared with curettage and no evidence of a difference in longer-term effectiveness, and low-quality evidence suggesting a higher incidence of adverse events in patients undergoing curettage.

There is very low-quality evidence suggesting that BTX injections have a small positive effect on palmar hyperhidrosis symptoms compared with placebo or no treatment, although there is some very low-quality evidence to suggest a high incidence of adverse events with BTX-B (5000 U). The evidence for an effect of BTX injections for palmar hyperhidrosis on quality of life is insufficient.

As stated previously, there is very low-quality evidence to suggest that iontophoresis is less effective than BTX injections at reducing palmar hyperhidrosis symptoms in the short term and that the duration of effect is shorter than with BTX.

Curettage compared with botulinum toxin

Four studies compared BTX with curettage (see Appendix 2, Table 55). 21,148,155,184 Study results are reported in Botulinum toxin.

Comparison of curettage with other minor surgical interventions

Five studies (three RCTs, two non-RCTs)24,147,149,150,154 compared suction curettage with other types of minor surgery that are more radical (see Appendix 2, Table 58): radical skin excision, radical skin excision with Y-plasty closure, Shelley’s procedure (skin-sparing technique) and curettage with aggressive manual shaving.

Bechara et al. 150 included a total of 40 patients who were randomised to one of three surgical interventions: liposuction-curettage (n = 15), radical skin excision with Y-plasty closure (n = 14) and a skin-sparing technique (n = 11). At 1-year follow-up, all three groups experienced significant mean reductions in sweat rates from baseline and there were no statistically significant differences between the three interventions. Aesthetic outcome was assessed by patients on a scale of 1 (very good) to 5 points (not good at all). The mean aesthetic outcome score was 3.2 (range 2–5) in the skin excision group, 2.3 (range 1–3) in the skin-sparing technique group and 1.5 (range 1–2) in the curettage group. The mean aesthetic outcome score was significantly better in the curettage group than in the other groups (p < 0.05). The incidence of haematoma was similar across groups, although one moderate to severe hematoma occurred in the curettage group that resolved over 4 weeks. No other severe adverse events were reported.

Jemec149 carried out a non-randomised parallel group trial comparing curettage with radical skin excision in a total of 41 participants. Patient satisfaction and adverse events were the only outcomes reported. Follow-up data were reported between 6 and 9 months post intervention. Levels of patient satisfaction were comparable between the two interventions. It was reported that there were no abscesses, haematomas or wound defects. No other adverse event data were reported.

The study by Bechara et al. 154 was a non-randomised half-side comparison trial in which patients acted as their own control, comparing curettage with or without aggressive manual shaving. The trial was stopped early after treatment of only four patients because of extensive skin damage associated with aggressive shaving.

Tronstad et al. 147 compared tumescent suction curettage with curettage alone in a half-side randomised comparison involving 22 patients. Sweating was assessed using gravimetry and skin conductance tests. Patients also assessed their sweating using a VAS for each side at follow-up (no further details reported). Quality of life was assessed using the DLQI questionnaire. All measurements were performed at 3, 6 and 12 months’ follow-up. The reduction in sweating (both objective and subjective measurements) was statistically and clinically significant at 6 and 12 months compared with baseline in both treatment groups (p < 0.05), and significantly greater with tumescent suction curettage than with curettage alone (p < 0.05). Three-month follow-up results were similar, although the difference between interventions was not statistically significant for gravimetry. Quality-of-life results were not reported. One patient from the tumescent suction curettage group experienced postoperative neuropathic pain lasting through the observation period. There were no haematomas or infections requiring antibiotics at 1 week and no other adverse events were reported.

Curettage compared with energy-based sweat gland destroying (‘destructive’) technologies

Leclere et al. 24 (see Appendix 2, Table 59) conducted a RCT that included 100 patients and compared four interventions: laser alone with a radiation of 975 nm, laser alone with simultaneous wavelengths of 924 nm and 975 nm, laser (924/975 nm) followed by curettage, and curettage alone. Effectiveness was measured using the Starch-Iodine Scale Improvement, a four-point single-item questionnaire measuring changes in sweating, with scores ranging from 0 points (no evidence of sweating by starch test) to 3 points (big dark area without changes), and HDSS. At both 1 and 12 months’ follow-up, the largest reduction in mean HDSS and sweating scores was achieved in the the laser plus curettage group, followed by the laser 924/975 nm only group, the curettage only group and the laser 975 nm group. Although the results suggest a clinically relevant difference between laser plus curettage and the other interventions, the authors did not state whether or not differences between the interventions were statistically significant. Aesthetic outcomes were measured using the Global Aesthetic Improvement Scale (GAIS), a five-point scale rating global aesthetic improvement, compared with baseline, as judged by the investigator. At 1 and 12 months’ follow-up mean scores on the GAIS were highest in the laser plus curettage patients, followed by laser 924/975 nm, curettage alone and laser 975 nm groups. There were few adverse events in all groups, and all had resolved within 1 month of follow-up.

Evidence summary for curettage

In summary, nine studies (four RCTs,21,24,147,150 five non-RCTs148,149,154,155,184) evaluated curettage for axillary hyperhidrosis. All were rated as being at a high risk of bias.

Four studies (one RCT, three non-RCTs) compared curettage with BTX. 21,148,155,184 One small half-side RCT21 found a difference in HDSS score (at 3 and 6 months’ follow-up) and gravimetric sweat measurement improvements (3 months’ follow-up) favouring BTX compared with curettage, although the difference was statistically significant only for HDSS score. Three non-randomised studies found no evidence of a significant difference in gravimetric sweat measurements at up to 26 weeks (one non-RCT),155 patient-reported sweating or patient satisfaction at 6 months (one non-RCT),148 HDSS score improvement at 6 weeks (one non-RCT)184 or quality of life up to 26 weeks (one non-RCT),155 between curettage and BTX. When reported, the incidence of adverse events was higher with curettage than with BTX (one RCT, two non-RCTs).

Among the five studies (three RCTs, two non-RCTs) that compared suction curettage with other surgical interventions, one RCT150 found no difference in sweating (gravimetry) at 1 year of follow-up between liposuction curettage, radical skin excision and a skin-sparing technique (Shelley’s procedure), although patients undergoing curettage reported significantly greater satisfaction with aesthetic outcomes. One non-RCT149 reported comparable levels of patient satisfaction and safety between curettage and radical skin excision at 6–9 months’ follow-up. One small non-randomised half-side comparison trial (in which patients acted as their own control), evaluating curettage with and without aggressive manual shaving, was stopped early because of extensive skin damage associated with aggressive shaving. 154 One RCT147 evaluated tumescent suction curettage against curettage alone in a half-side comparison. Reduction in sweating (gravimetry, skin conductance and patient reported) was statistically significantly greater with tumescent suction curettage than with curettage alone at 6 and 12 months’ follow-up, although, owing to limited reporting, it is not clear whether or not these results were clinically significant. One event of neuropathic pain lasting throughout the 12 months’ follow-up was reported in the tumescent suction curettage group. One four arm RCT24 found a clinically significant benefit in mean HDSS and sweating, as well as better aesthetic outcomes, with laser plus curettage compared with two different regimens of laser alone, or curettage alone. Few adverse events were reported in all groups.

Overall, there is very low-quality evidence regarding the relative effectiveness and safety of curettage compared with other minor surgical interventions for axillary hyperhidrosis. Compared with the more radical excision techniques, there is no clear evidence of a difference in sweat reduction, patient satisfaction or safety. However, aggressive manual shaving was associated with extensive skin damage in one very small study, which was stopped early. There is insufficient evidence to conclude whether or not sweating symptoms are significantly more reduced with tumescent suction curettage than with curettage alone in the medium term (up to 12 months).

There is low-quality evidence suggesting that curettage combined with laser treatment is more effective at reducing hyperhidrosis symptoms and has better aesthetic outcomes than laser or curettage alone, and no evidence of a difference in safety between these therapies.

As stated previously, there is very low-quality evidence regarding the relative effectiveness of BTX injections to the axillae compared with curettage and no evidence of a difference in longer-term effectiveness, and low-quality evidence suggesting a higher incidence of adverse events with curettage than with BTX.

Laser epilation compared with no treatment

Two RCTs comparing laser epilation with no treatment were included (see Appendix 2, Table 60).

The studies by Bechara et al. 25 and Letada et al. 26 were both half-side RCTs that evaluated the efficacy of a long-pulsed laser. Bechara et al. 25 found a significant reduction in sweat rate measured by gravimetry on the laser-treated side and the untreated contralateral side, but no significant difference in reduction between treated and non-treated sides at 12 months’ follow-up (p = 0.10). Participant satisfaction with reduction in sweating was measured using a VAS score ranging from 0 to 10 (0 = not satisfied at all, 10 = absolute satisfaction). Patients reported a mean score of 5.9 after the last laser treatment and 4.1 at follow-up. No serious adverse events were reported.

Letada et al. 26 reported visibly reduced sweating in the laser-treated axilla, as measured by iodine starch test, in all six patients compared with the contralateral side at 1 month follow-up. All patients at 1 month, and two out of the three patients with results at 3 months, reported good or excellent improvement in Global Assessment Questionnaire. No adverse events were reported.

Laser epilation compared with laser epilation with curettage

The results of the Leclere et al. study,24 which compared different laser interventions with curettage, are reported in Curettage and in Appendix 2, Table 58.

Evidence summary for laser epilation

Two small RCTs of half-side comparisons between laser epilation and no treatment were included; both were rated as being at high risk of bias. 25,26

One study25 found no significant difference in sweat reduction measured by gravimetry between the treated and non-treated sides at 12 months. The other study26 found that sweating was visibly more reduced at 1 month according to the iodine starch test in the treated axillae. Both studies reported no serious adverse events.

Overall, there is insufficient evidence to suggest that laser epilation alone is more effective than no treatment at reducing hyperhidrosis symptoms or improving quality of life. There was insufficient evidence regarding the safety of laser epilation, although no evidence was found to suggest the intervention was not safe. As stated previously, there is low-quality evidence suggesting that curettage combined with laser treatment is more effective at reducing hyperhidrosis symptoms and has better aesthetic outcomes than laser or curettage alone, and no evidence of a difference in safety between these therapies.

Evidence summary for fractionated microneedle radiofrequency

Overall, there is insufficient evidence regarding the relative effectiveness and safety of fractionated microneedle radiofrequency for axillary hyperhidrosis.

Evidence summary for microwave

Only one study of a microwave device for axillary hyperhidrosis was included. 27 Overall there is very low-quality evidence suggesting that microwave therapy is more effective than placebo at reducing patient-reported disease severity, although there was no evidence of a significant difference in sweat reduction at up to 6 months. The evidence regarding the safety of microwave therapy is insufficient.

Evidence summary for ultrasound

Two small RCTs of microfocused ultrasound were included. 29 Overall, there was insufficient evidence regarding the safety and effectiveness of ultrasound therapy for axillary hyperhidrosis.

Botulinum toxin for hyperhidrosis of the axilla

There is sufficient evidence demonstrating the clinical effectiveness of BTX for hyperhidrosis of the axilla and, therefore, no more trials of BTX compared with placebo in the axilla are warranted. Future trials of interventions for hyperhidrosis of the axilla should use BTX as an active comparator.

Iontophoresis for palmar hyperhidrosis

There are three studies of tap water iontophoresis compared with placebo37,40,42 and two studies of dry-type iontophoresis compared with no treatment. 43,44 Although they have methodological limitations, they all consistently show that iontophoresis is more effective than placebo/no treatment for hyperhidrosis of the palm; therefore, no further trials of iontophoresis compared with placebo/no treatment are warranted. Iontophoresis is currently standard practice for palmar hyperhidrosis in many dermatology units.

Iontophoresis compared with botulinum toxin for palmar hyperhidrosis

There is limited evidence for the effectiveness of BTX for palmar hyperhidrosis. Although the studies of iontophoresis for palmar hyperhidrosis are of poor quality, there is sufficient evidence to indicate a real, albeit limited, effect of iontophoresis. To date there have been two reasonably small, poor-quality, studies88,92 comparing BTX with iontophoresis (one of which also prescribed topical aluminium chloride lotion to the iontophoresis group88) for palmar hyperhidrosis. BTX looks promising for palmar hyperhidrosis, but the current evidence is not sufficiently reliable to draw firm conclusions; therefore, further research is required. A well-conducted, adequately powered RCT of BTX (with anaesthesia) compared with iontophoresis for palmar hyperhidrosis is warranted. The cost of BTX plus anaesthesia is considerably higher than the cost of iontophoresis; therefore, cost-effectiveness should also be assessed.

Microwave, laser, fractionated microneedle radiofrequency and ultrasound

These new, energy-based, ‘destructive’ technologies offer the prospect of the benefits of curettage for hyperhidrosis of the axilla, but with fewer risks. Some small, poor-quality, studies have shown promising results for these interventions compared with placebo or no treatment; however, these studies cannot be considered reliable, so there may not be sufficient promise to warrant further research. Studies of microwave, laser, fractionated microneedle radiofrequency and ultrasound therapies are ongoing, but the numbers of participantsare low (≤ 40) and the interventions are not being ompared against other active treatments or placebo. If the results of this ongoing research are consistent with the existing study results, then a trial comparing these new energy-based technologies with BTX for hyperhidrosis of the axilla may be warranted.

Curettage compared with botulinum toxin for hyperhidrosis of the axilla

Botulinum toxin has been shown to be effective for hyperhidrosis of the axilla. However, the procedure has to be repeated at regular intervals (at least annually). Curettage offers the prospect of a single treatment. There are currently a few small and/or non-randomised trials comparing curettage with BTX for hyperhidrosis of the axilla. The short- to medium-term evidence suggests that these two interventions are comparable in terms of effectiveness, or that BTX may be superior to curettage. In view of the ongoing research into less invasive, energy-based, technologies (microwave, laser, fractionated microneedle radiofrequency and ultrasound), a trial comparing BTX with curettage for hyperhidrosis of the axilla may not be warranted at this time. When further evidence is available on the newer energy-based technologies, then it will be clearer whether or not further research on curettage is warranted.

Comparison of different oral/topical medications: propantheline bromide, glycopyrrolate, oxybutynin, methantheline bromide and newer medications

The oral medications have not been well studied in trials; there are no placebo-controlled trials of oral glycopyrrolate (thought to be the most effective) or of propantheline bromide (the cheapest available). There are no direct comparisons of the different oral/topical anticholinergic medications for hyperhidrosis. However, there are ongoing and recently completed trials of new oral and topical formulations; therefore, it may not be relevant to undertake further research of these specific drugs at this time. Although different medications may work better for some patients than others, it may be difficult to power a study to find any statistically significant differences between treatments.

Population

Patients with primary hyperhidrosis (including adults and children). Patients with hyperhidrosis secondary to other conditions, such as overactive thyroid or spinal cord injury, or social anxiety disorder, were not eligible for inclusion.

Outcomes

Any quality-of-life-related outcome measure was included. Quality-of-life outcomes can be measured using a variety of tools: disease specific, such as the HDSS and HHIQ; discipline specific (dermatology), such as the DLQI or Skindex-29; or more general health or utility, such as the Short Form questionnaire-36 items (SF-36).

Study designs

Any study design was eligible for inclusion in this review.

Hyperhidrosis Disease Severity Scale

The HDSS was identified as the most commonly used tool. The HDSS was used in 63 studies in total: for surgical research in 27 studies,17,24,27–29,159–161,164,167,171,174–176,179,181,207–217 for dermatological research in 27 studies,2,45,51–53,59,61,77,91,97,102,106–108,119,125,131,136,139,145,180,218–223 for surgery and dermatology combined in two studies,21,224 for three quality-of-life-only studies225–227 and in four economic evaluation studies228–231 for measuring quality of life in patients with hyperhidrosis. The HDSS was often used in combination with the DLQI, with 18 studies using both tools (10 dermatology studies,61,77,102,106,107,119,139,221–223 five surgical studies,176,208,210,214,217 two quality of life studies225,227 and one economic assessment230).

Studies that were found to use the HDSS tool were conducted globally, including the USA and Canada (n = 2121,27,29,52,77,107,108,139,159,160,164,175,176,179,180,208,209,218,225,226,229), Spain (n = 6167,207,210,224,230,231), the UK (n = 591,97,125,222,232), Germany (n = 52,131,145,161,227), Italy (n = 5102,106,136,214,223), France (n = 359,61,211), South Korea (n = 351,171,174), Australia (n = 253,221), Egypt (n = 2213,219), Iran (n = 228,220), the Netherlands (n = 2215,217), Turkey (n = 2212,216), Austria (n = 117), Europe (n = 124), Greece (n = 1119), Pakistan (n = 145) and Russia (n = 1181). The majority of studies were case series (n = 2717,45,61,97,102,106,107,119,125,131,136,139,145,159–161,164,167,171,174–176,181,207–210). There were 15 RCTs,21,24,27,29,51–53,59,77,87,108,179,180,218,220 and the remainder were non-RCTs, retrospective observational studies or reviews (n = 222,28,91,211–217,219,221–227,229–232). The HDSS was used in studies to measure quality of life in patients with sweating in various areas of the body; studies of axillary sweating were most common (n = 362,21,24,27–29,52,53,77,91,97,107,108,125,136,139,159–161,164,167,171,174–176,179–181,209,212,214,218,219,221,223,229), with an additional seven studies102,106,119,131,145,207,213 examining sweating of the palm and axilla combined. Eleven studies17,45,61,208,211,215,222,224,225,230,231 were identified of the foot and/or palm combined and nine of the face, palm and/or foot or generalised hyperhidrosis. 51,59,210,216,217,220,226,227,232

The HDSS is a disease-specific tool considered important for diagnostic use in clinical practice and for research to identify and quantify the severity of disease in patients with hyperhidrosis and also to assess treatment effects over time. 12,205 The HDSS allows researchers to measure the impact of hyperhidrosis on those suffering from excessive sweating by using a four-point scale:

1. My sweating is never noticeable and never interferes with my daily activities.

2. My sweating is tolerable but sometimes interferes with my daily activities.

3. My sweating is barely tolerable and frequently interferes with my daily activities.

4. My sweating is intolerable and always interferes with my daily activities.

Information on the original design of the HDSS is lacking, although it was possibly created by Kowalski in 2004 (Lisa Pieretti, International Hyperhidrosis Society, 5 June 2016, personal communication). Efforts have been made to justify the use of the tool in hyperhidrosis research; for example, Kowalski et al. 205 presented work that examined the validity and reliability of the HDSS as a conference abstract in 2004. 205 The study presented data from a longitudinal clinical trial and a cross-sectional national survey of 150,000 US households. The HDSS score at 4 weeks post treatment was found to correlate well with the DLQI and relevant activity items from the HHIQ (r = 0.35 to 0.77; p < 0.001). Kowalski et al. 205 also presented data on construct analysis, which suggested that a reduction (improvement) in the HDSS score correlated well with a reduction in gravimetric results, a clinical test used to measure quantity of sweating in patients with hyperhidrosis, and that patients reported an increase in HDSS score with increases in limited daily activity due to sweating. The test–retest reliability coefficient was also found to be within an acceptable range (r = 0.82; p < 0.05). 205

Studies of hyperhidrosis were found to use the HDSS more frequently than any other method for measuring quality of life. The tool’s simple design has raised questions about its value as a tool to measure patient-reported quality of life (Lisa Pieretti, personal communication), and a consensus exercise by the Canadian Hyperhidrosis Advisory Committee considered the HDSS more appropriate as a measure of disease severity. 12 Furthermore, Panhofer et al. 233 published a survey and validation guide for tools used to measure quality of life in patients undergoing surgical interventions for hyperhidrosis and the HDSS was not considered as part of this work. This reflects the fact that the HDSS is not viewed purely as a quality-of-life tool; it is also possible the patient may complete the tool based on an improvement in quality of life or simply a perceived reduction in sweating (with an assumption that this will improve their quality of life).

Disease-specific health-related questionnaire for hyperhidrosis (Amir et al.186)

The disease-specific health-related questionnaire for hyperhidrosis was designed and validated by a research team in Israel, Amir et al. ,186 to assist with clinical decision-making and to measure the efficacy of surgical interventions in reducing sweat production in patients with hyperhidrosis. The questionnaire was designed and validated with a patient population who were considering surgery for the treatment of their excessive sweating.

The Amir et al. 186 questionnaire was found in five studies identified for the review of quality-of-life measures: the original development and validation work, integrated with a surgical study with a focus on quality of life,186 and four additional studies, all from the same team in Brazil (three surgical studies234–236 and one dermatology study237).

The Amir et al. 186 tool development was conducted in two stages. First, 10 patients aged between 15 and 35 years were interviewed by two psychology students using a semistructured interview method with questions that focused on the ways in which hyperhidrosis limits the daily life of patients. The responses were analysed by the same psychology students, a clinical psychologist and a surgeon. Participants were encouraged to discuss the effect on their leisure activities, social interactions, work or interpersonal relationships. From the responses, four domains were identified (functional, social, interpersonal and emotional). Functional included driving, writing and sports; social included shaking hands or being in public places; personal involved being intimate with a partner; and emotional was split between the patient’s own feelings about their condition (emotional self) and how they think others view them (emotional other). The authors note that not all patients would find all questions relevant (e.g. those who do not drive).

The questionnaire was designed with 35 questions separated into the five domains with a 7-point Likert scale for each response, where a score of 6–7 indicated a very low quality of life, 3–5 a medium level of quality of life and 1–2 a high level of quality of life. The validation found a high level of internal reliability (Cronbach’s α = 0.84).

The results suggest that patients felt impaired in many facets of life, both intimate and practical, and that the effect of heat and stress played a major part in the patients’ quality of life. The authors found no differences between the body area (hands, feet and axilla) and quality of life. Analysis of variance for the quality of life domains in relation to sex and duration since onset found that females reported a lower quality of life in all domains except emotional other. In addition, patients in whom onset of hyperhidrosis occurred during childhood had a lower quality of life. A limitation of the validation work noted by the authors is that only patients waiting for surgery were used in the survey and these may be patients whose symptoms are at the more severe end of the spectrum. 186 Another point to mention is that the tool was validated in Israel and reported in studies conducted in Brazil, both countries with a very hot climate that could have an impact on the patient population and subsequent outcome.

Hyperhidrosis Quality-of-Life Questionnaire

The Hyperhidrosis Quality of Life Questionnaire (HQLQ) was designed by de Campos et al. 238 in 2003 as a disease-specific tool to assess the effect of surgical intervention for patients with hyperhidrosis. The design built on the previous validation work of Amir et al. 186 and tested the tool on a large patient population (n = 378), with the aim of replacing more generic measures.

The HQLQ was used in 31 studies included in the quality-of-life review. Seventeen studies, including the validation work, had a surgical intervention,234,236,238–252 13 dermatology studies55,57,60,62–67,237,253–255 and one quality-of-life review. 227 The tool was used in Austria (n = 6240–242,248,249,251), Brazil (n = 2157,60–67,234,236–239,243,244,246,247,252–255), Germany (n = 1227), Italy (n = 1245), South Korea (n = 155) and India (n = 1250).

de Campos et al. 238 used their questionnaire to evaluate preoperative HRQoL and postoperative improvement in symptoms of hyperhidrosis in all patients undergoing surgical intervention between 1995 and 2002. The development study included 386 hyperhidrosis patients who underwent video-assisted thoracic surgery. 238 The survey response rate was 91% (344 patients). Patients included in the study suffered with either palmar and plantar (57.4%), palmar, axillary and plantar (25%), axillary (15.7%) or facial (6.5%) sweating.

The HQLQ consists of a single question to start ‘How would you rate your quality of life before and after treatment?’, to which the patient is asked to respond with a score between 1 and 5, with 5 being very poor/much worse. This is followed by 20 questions selected for relevance from the 36 items taken from the Amir et al. 186 questionnaire, again scored between 1 (excellent) and 5 (very poor), and further separated into domains (again as described previously by Amir et al. 186), including (1) functional or social life, (2) personal life, (3) emotional self and (4) special circumstances (such as climate or stress). Therefore, the final score for quality of life can range from 20 (excellent/much better after surgery) to 100 (very poor/much worse after surgery). No validation or reliability statistics or cross-validation with other tools was reported.

Keller Hyperhidrosis Scale

The Keller Hyperhidrosis Scale256 was cited in 10 studies in total: seven surgical studies,240–242,248,249,251,257 one dermatology study55 and two quality-of-life reviews. 227,256 The tool was designed by an American team, although, apart from the one South Korean dermatology study,55 all studies using the Keller tool appear to come from the same team in Austria and all studies except one257 used both the Keller and the de Campos et al. HQLQ238 methods combined to measure quality of life.

The Keller scale was designed by Keller et al. 256 to measure preoperative and postoperative quality of life in patients receiving bilateral ETS for palmar and plantar hyperhidrosis and appears to be a popular tool in the field. The validation work compared patient score against the SF-36 and validation work reported a strong level of reliability (Cronbach’s α = 0.89). The Keller Hyperhidrosis Scale measures quality of life on a scale of 0 (mild) to 10 (severe). A total of 71 patients were entered into the study with a mean preoperative Keller score of 6.85, which decreased significantly (p < 0.001) to 1.79, 1.53 and 1.91 at 2 weeks, 6 months and 1 year post surgery, respectively. Keller et al. 256 reported no significant MD between preoperative and postoperative scores for the SF-36 (p = 0.70) and concluded that standard quality-of-life instruments do not accurately assess the benefits of surgery to patients with palmar and plantar hyperhidrosis.

Hyperhidrosis Impact Questionnaire

The HHIQ was designed by Teale et al. 258 in 2002 to assess the impact of hyperhidrosis on the daily lives of patients and measure the effect of treatment. The HHIQ development and validation were performed by a team of researchers funded by Allergan Inc. (Irvine, CA, USA), a large international pharmaceutical company and significant producer of BTX-A for the treatment of hyperhidrosis. The relatively high frequency of use of this tool reflects its use in Allergan-sponsored research. The validation work was not available as a full publication, but does appear as a conference abstract presented at the ninth Annual Conference of the International Society for Quality of Life Research (ISOQOL) in 2002. 258

Eight studies selected for inclusion into the quality-of-life review used the disease-specific HHIQ for their quality-of-life assessment or review: the original validation work by Teale et al. 258 in 2002 and seven additional studies that examined the effect of BTX-A on axillary or palmar hyperhidrosis. 76,78,80,86,88,107,139 The study by Rajagopal and Mallya88 was conducted on behalf of the Indian armed forces and focused on the use of BTX to treat palmar hyperhidrosis and appears independent of the other six studies, all of which were conducted in either Europe or the USA/Canada and were funded or supported by Allergan Inc.

A review of the literature and interviews with key stakeholders (patients and physicians in the UK and Germany) informed the design of the tool; a pilot study with the same stakeholders then tested the validity and linguistic equivalence of the questionnaire. 258 The questionnaire contained four sections: (1) disease and treatment background; (2) direct impact on medical and non-medical resource utilisation; (3) indirect impact on employment and productivity; and (4) intangible impacts on emotional status. Forty-one questions measured baseline impact of the disease and a furtehr 10 questions were included in the follow-up assessments. The final design of the HHIQ was validated against the SF-12 health survey and the DLQI, using a population of 345 patients and 145 non-hyperhidrosis controls. A test–retest of the 10 follow-up questions using a cohort of clinical patients found consistent reliability and responsiveness of the HHIQ. The authors conclude that the tool is a valid and reliable instrument for measuring disease and treatment effects in patients with hyperhidrosis.

Hyperhidrosis Questionnaire

The Hyperhidrosis Questionnaire is a disease-specific instrument identified from a quality-of-life-only study published in 2004 by Kuo et al. 259 describing the tool’s design and validation. Although the development and design of this tool appears to be quite comprehensive, the tool has been used in only one other study selected for the quality-of-life review, a review paper of quality-of-life tools for hyperhidrosis by Rzany et al. 227

The development study reported a review of the literature, to inform the elements to include in the instrument’s design, followed by interviews with patients, nursing staff and clinicians. The pilot questionnaire content was then validated by a panel of patient experts before being tested. The pilot questionnaire contained 34 questions answered using a Likert scale of 1 (least disturbance) to 5 (most disturbance). The final questionnaire contained 29 questions across five factors (domains): (1) functional, (2) psychological, (3) social, (4) affective and (5) physical.

The study included 85 patients suffering from a combination of plantar, palmar, axilla or generalised hyperhidrosis attending a thoracic surgery outpatient clinic in southern Taiwan between April 2002 and March 2003. Internal reliability and construct validity was reported (Cronbach’s α = 0.95, range 0.71–0.94 across domains). There was no investigation of whether or not the scale reflects changes in disease severity, nor was any cross-validation with other scales reported.

Hyperhidrosis Quality of Life Index

The HidroQoL was not identified in any of the studies selected for the quality-of-life review apart from those published to describe its design and validation, probably because these are relatively recent (2012 and 2015). 204,260,261

The tool was developed as a disease-specific aid to both clinical practice and research to assist with hyperhidrosis patient and clinician communication. The HidroQoL was not developed for any specific body site or intervention. The HidroQoL tool was created using responses to interviews and focus group discussions. In 2012, Kamudoni et al. 204 recruited an online cohort of 71 patients from a number of social networking sites to participate in initial interviews. From this, a prototype of the HidroQoL was created. It contained 47 questions with a six-point patient response scale: (1) not at all, (2) a little, (3) somewhat, (4) quite a bit, (5) very much and (6) not relevant. A panel of experts, including patients suffering from focal or generalised hyperhidrosis (n = 7) and dermatology clinicians (n = 5), was recruited to validate the content of the prototype tool using a questionnaire administered by e-mail. Further work in 2015261 used modern test theory to examine differential item functioning, which allowed the researchers to assess if the tool would function in different ways for different groups or test-takers.

The second stage of validation involved a cross-sectional cohort of 595 patient volunteers recruited using prespecified inclusion criteria (age > 18 years, self-reported hyperhidrosis, HDSS score of ≥ 2 points and disease onset prior to early adult years). The participants completed a number of questionnaires for comparison (HDSS, DLQI and Skindex-17). Construct validity, reliability and responsiveness were assessed using an online longitudinal study (n = 260 patients), for which patients were required to complete the tool on three occasions over a period of time (7–21 days), using the final version of the questionnaire. Construct validation found that the HidroQoL correlated well with the DLQI (r = 0.6; p < 0.01) and HDSS (r = 0.59; p < 0.001), and showed correlation to the Skindex-16 scale, but to a lesser extent. The change in HDSS score recorded by participants over time was also indicated in the HidroQoL scores from the longitudinal study, suggesting that the tool could identify slight change or small responses to treatment over a period of time. Reliability, tested using baseline measures and a test–retest method, showed strong reproducibility (internal consistency, Cronbach’s α overall scale = 0.89; test–retest reliability, intraclass correlation = 0.93).

Dermatology Life Quality Index (includes children’s version)

Described as a practical technique for routine clinical use, this questionnaire is a short and concise tool (10 questions) for use in the management of chronic skin disorders. 262 Developed by Finlay and Khan262 to provide a patient-centred method for comparison between different types of skin disease, the questionnaire records the impact each disease has on a patient’s quality of life and the relative effectiveness of treatment.

The DLQI was found in 48 studies for reporting quality-of-life measures in patients with hyperhidrosis. The study selection process found 28 dermatology studies,54,56,61,69,77,101–103,106,107,116,117,119–121,128,134,139,221–223,263–269 11 surgical studies,147,158,170,173,176,208,210,214,217,270,271 two dermatology/surgery combined,272,273 two economic assessments230,274 and five studies or reviews of quality of life. 225,227,261,262,275 As mentioned previously (see Hyperhidrosis Disease Severity Scale), the DLQI was often used alongside the HDSS in dermatology studies (n = 1061,77,102,106,107,119,139,221–223), surgical studies (n = 5176,208,210,214,217), studies of quality of life only (n = 2225,227) and also economic assessments (n = 1230).

The DLQI was identified in studies conducted globally, including in the UK (n = 9103,232,261,262,267,268,275), the USA and Canada (n = 877,107,139,173,176,208,222,263), Italy (n = 7101,102,106,214,223,266,272), Germany (n = 569,158,227,270,271), Sweden (n = 4114,116,117,128), Greece (n = 3119–121), Ireland (n = 2265,274), Iran (n = 2170,264), Spain (n = 2210,230), Australia (n = 1221), Austria (n = 1273), France (n = 161), India (n = 1269), the Netherlands (n = 1217) and Norway (n = 1147). The studies included 24 case series,54,56,61,101–103,106,107,114,116,117,119–121,128,139,158,170,173,176,208,210,214,266 nine retrospective observational studies,217,221,222,230,263,265,268,270,271 six quality-of-life or non-experimental studies227,232,261,262,274,275 and three RCTs;69,77,223 the remainder were non-RCTs (n = 3264,269,272) or surgical reviews or reviews of surgery compared with dermatology (n = 3147,267,273). The DLQI was used in studies to measure quality of life on various areas of the body; studies of axillary sweating combined with other parts of the body, such as palms or feet, were most common (n = 2654,56,69,77,101,103,106,107,114,119–121,139,147,158,170,173,176,214,221,223,263,266,273,274). Studies were identified of the palm or palm and foot combined (n = 1161,128,208,222,230,264,265,269–272) and of other combinations including the face, palm and/or foot or generalised hyperhidrosis (n = 11116,117,210,217,227,232,261,262,267,268,275).

The DLQI was validated by Finlay and Khan262 using a survey of 128 patients aged 15–70 years who attended a dermatology clinic. Patients were asked to document how certain aspects of their skin disease affected their life, for example any effects the condition had on their work life, social life or personal relationships. The results were combined and a questionnaire developed based on the 10 most commonly reported aspects of impaired quality of life. The answers to each of the 10 questions were scored from 0 to 3, where a total maximum score of 30 and minimum score of 0 could be achieved: the higher the score, the greater the impairment in quality of life. Once designed, the questionnaire was tested for use on 200 outpatients and 100 healthy volunteers and reliability was further validated using a cohort of 53 patients by means of measuring the correlation of test–retest results over a 10-day period.

A recent review of the DLQI reported that repeatability, internal consistency and sensitivity to change have all been demonstrated for this tool and it has been cross-validated against a number of other dermatology tools (though mainly for psoriasis and acne). 276 The DLQI instrument is licensed for use and, therefore, permission is required from the authors to use the tool.

Skindex: a quality-of-life measure for people with skin disease

The Skindex suite of tools includes the original Skindex questionnaire, Skindex-29, Skindex-17 and Skindex-16. The tools are dermatology-specific HRQoL instruments, originally based on the DLQI. The original tool was a 61-question survey refined to a 29-item questionnaire (Skindex-29),277 to reduce the time needed to complete the questions and to improve the tool’s evaluative properties. 278 Further refinement included using patients’ perspective responses to create a 16-item questionnaire (Skindex-16)279 for use in longitudinal research to measure changes over time in patient quality of life in addition to reducing the tool to one page. Finally, a study in 2006 created Skindex-17. A Rasch-reduced tool was created using a response theory model to address issues such as response order and differential item functioning,280 and resulted in a reduction of the Skindex-29 item tool to a 17-item instrument. The questions in the Skindex-16 and Skindex-17 differ; however, a simple comparison conducted by Nijsten et al. 280 found that the different reduction approaches (patient opinion vs. statistical modelling) resulted in very similar instruments.

The Skindex-29 tool was used in five of the six studies identified for the group of Skindex instruments by the hyperhidrosis quality-of-life review. 143,227,281–283 The Lessa et al. 284 study used Skindex-16. The studies included two quality-of-life reviews, a RCT, a single case series, a case report and a surgical observational study. Two studies were carried out in Germany and the other four studies in the Netherlands, Turkey, Spain and the USA.

The Skindex-61 item questionnaire was first created by Chren et al. 278 in 1996; the validation of the tool was later published in 1997. Cronbach’s α (0.71–0.94) was calculated to measure the reliability of the psychometric test. 285 The tool’s design was based on findings from a review of the literature and focus group interviews with patients and clinicians to construct the initial framework for the ways in which patients are affected by skin disease. The tool has eight scales (cognitive, social, discomfort, limitations, depression, fear, embarrassment and anger) within three domains (symptoms, emotions and functioning). Patients are asked to respond to the questions with one of five responses (never, rarely, sometimes, often, all the time), and the score is reported as three values, one for each domain, calculated by averaging the patient’s responses from each of the questions within each of the three domains.

In a study of 201 patients, scale scores were reproducible after 72 hours and were internally consistent (Cronbach’s α = 0.76–0.86). Construct validity was also demonstrated. However, physicians’ judgement of disease severity did not consistently correspond with Skindex scores and Skindex does not appear to have been cross-validated against other quality-of-life measures.

The authors suggest that the Skindex-29 tool is more suited to research for which a comprehensive overview of HRQoL is required and the goal may be to understand the response to treatment. 285 The Skindex-16 was designed to serve longitudinal data collection. As patients may be required to complete the same survey many times over the course of a study, it was considered that it would be better if this was a shorter exercise. The Skindex instruments are copyrighted and, therefore, permission is required from the authors to use either of the tools.

VQ-Dermato scale: a French-language scoring instrument validated for chronic skin diseases – Grob et al.286

The VQ-Dermato scale was designed and validated to provide a French-language dermatology-specific instrument for routine use to assess the quality of life of patients with ‘chronic skin disorders’. 286 The scale was identified in one article selected for inclusion into the quality-of-life review, a retrospective analysis of a 142 patient study to examine the long-term (> 5 years) effects of ETS on patient satisfaction and quality of life for patients with palmar hyperhidrosis between 1994 and 2005. 211

The VQ-Dermato scale is a self-administered 28-item instrument developed from interviews with patients. Grob et al. 286 tested the tool on a population of 231 hospital and private practice patients in France suffering from chronic skin conditions and reported the instrument was validated in terms of content, construct, reliability/reproducibility and internal consistency. A strong correlation was reported by authors between the VQ-Dermato and the SF-36 (Cronbach’s α = 0.67–0.88).

Freiburg Life Quality Assessment

The Freiburg Life Quality Assessment was designed and validated as a set of dermatology-specific modules. The first module addresses the core issues of all skin diseases. The additional questions are more specific to distinct diseases. The tool was found to have satisfactory discriminatory power and validation data were published in 2004. 287 This tool was found to be used in only one hyperhidrosis quality-of-life review. 288

Patient Benefit Index

The Patient Benefit Index created by Augustin et al. 289 is an instrument to identify patient-reported needs and benefits of dermatology research and treatment. Assessment is a two-step process; the first captures data on the patients’ needs prior to treatment, and this is followed by assessment of improvement after treatment. The result is an index of patient benefit in response to treatment. The measure was validated in 2009 by the authors using a large cohort of patients (n = 500) with many different skin diseases, including some with hyperhidrosis (n = 50). 289 Use of this tool was identified in only one quality-of-life review of patients with hyperhidrosis. 288

Illness Intrusiveness Rating Scale

The Illness Intrusiveness Rating Scale (IIRS) was developed by Devins290 in 1994 to measure the impact of illness based on its intrusion on valued lifestyle activities. Devins produced a follow-up publication in 2010 describing the use of the tool and validating how it can contribute to the understanding of HRQoL in chronic disease. 291 The IIRS was identified in only one other study from 1999. 292 This study of 68 patients found that total IIRS score correlated well with other disease severity measures (r = 0.61; p < 0.001) and reflected improvements post surgery. The tool is a 13-item survey scored on a Likert scale from 1 (not very much) to 7 (very much). The tool can be used as a 13-item scale or divided into three subsets with a score calculated for each subset. The subsets are relationships and personal development (may include family or social relationships), intimacy (with partner) and instrumental considerations (such as work).

The IIRS validation was informed by a qualitative review of literature examining the results of studies in which it has been used, and describes the value of the tool for studies measuring patient-reported quality of life in 36 chronic disease groups. 291 Hyperhidrosis was considered in the review, with data on 223 patients provided by Claudio Cina. 291 Cronbach’s α coefficient for the total scale was 0.97, for relationships and personal development was 0.94, for intimacy was 0.92 and for instrumental considerations was 0.93.

Although the results presented by Devins290 suggest that the IIRS is potentially useful and valid for measuring quality of life in patients with hyperhidrosis, it is possible that it is too general to be popular for measuring quality of life in studies of hyperhidrosis, as reflected by its limited presence in such studies since 1999.

Medical Outcomes Study short forms (Short Form questionnaire-36 items and Short Form questionnaire-12 items)

The Medical Outcomes Study (MOS) was a large US cross-sectional study (n = 22,462 patients) conducted in 1989 to examine the impact that different systems of care and clinician expertise may have on patient outcomes, with the aim of developing better tools to monitor patients in clinical care. 293 The study worked with a conceptual framework considering the structure of care, the process of care and the subsequent outcome of care. Patient-reported outcomes contributed significantly to the study and ‘short-form’ surveys (MOS short forms) were designed to facilitate data collection.

Everyday Life Questionnaire or Fragebogen des täglichen Lebens

The Everyday Life Questionnaire (EDLQ) was designed by a German research team, Bullinger et al. 306 The EDLQ (or Fragebogen des täglichen Lebens) was created from responses to open-ended questions given to patients and healthy control subjects as a process to measure behavioural disease-specific quality of life. Questions consider issues such as mental health, physical well-being, social life and medical care. The authors suggest that it is more useful if the tool is used alongside the Munich Life quality Dimensions List. 307 The review found that the EDLQ was used in only one surgical study of palmar hyperhidrosis. 308

State–Trait Anxiety Inventory

The State–Trait Anxiety Inventory was originally developed in the early 1960s to measure anxiety in healthy populations, but then became useful as a tool for measuring anxiety disorders in patients. The review found use of the State–Trait Anxiety Inventory in only one surgical study of generalised hyperhidrosis. 305 The tool is designed to measure anxiety caused by certain situations (state anxiety) and also the tendency to perceive situations as threatening (trait anxiety). 309

International quality of life assessment (EuroQol-5 Dimensions, five-level version) or EuroQol-5 Dimensions

The international quality of life assessment (EuroQol-5 Dimensions, five-level version; EQ-5D-5L), or EuroQol-5 Dimensions (EQ-5D), is a HRQoL measure developed in 1990 by the EuroQoL group310 from the results of a postal survey conducted in England, the Netherlands and Sweden. The EuroQoL was designed to be used alongside other quality-of-life measures as a health status index for reference purposes. The group has since expanded from 23 academic contributors to 75 and has become an international multilingual and multidisciplinary team producing a suite of copyright-standardised generic instruments for describing HRQoL. Use of the EuroQoL tool was identified only in one study, an economic assessment of the burden of hyperhidrosis on the patient, by Kamudoni et al. 228 in 2014.

Leibowitz Social Anxiety Scale

The Leibowitz Social Anxiety Scale206 measures fear, anxiety and avoidance of social situations before and after treatment. 311 The questionnaire is completed by a clinician asking the patient a list of questions: 11 questions relating to social interaction and 13 questions related to public performance. It has also been validated in a modified version as a patient (self-) reported outcome measure. 312 Use of this tool was identified in only one study of hyperhidrosis, a surgical case series examining sweating problems in patients in Finland. 313

University of California, Los Angeles, Loneliness Scale

The University of California, Los Angeles, Loneliness Scale was developed in 1978 by Russell et al. 314 in the USA. The purpose of the scale is to measure loneliness in patients participating in social psychological research where they may suffer from limited social interaction or feel lonely as a result of their disease. The search identified its use in a single dermatology study from Greece,121 which used both the University of California, Los Angeles, Loneliness Scale and the DLQI together to measure quality of life in patients receiving BTX to treat palmar and axillary hyperhidrosis.

Nottingham Health Profile

The Nottingham Health Profile was first designed in 1975 at Nottingham University using survey responses from a cohort of 700 people asked to describe typical ill health. The tool was developed as a generic and standardised HRQoL tool to measure the health status of a population. The tool can also be used as an outcome measure for interventional management alongside a clinical assessment. 315 The Nottingham Health Profile was discussed in a review of quality of life by Rzany et al. 227 in 2012, along with 12 other tools for measuring quality of life in patients with hyperhidrosis, but was not found in any other studies.

The Questionnaire of Quality of Life adapted from the Caregiver Questionnaire

The Questionnaire of Quality of Life is an adaptation of the Caregiver Questionnaire which was developed by Geabler-Spira and adapted by Schneider et al. 316 The Caregiver Questionnaire is a disease-specific measure of HRQoL developed for children with spastic quadriplegic cerebral palsy who were to undergo treatment at the Rehabilitation Institute of Chicago, IL, USA. 316 Developed as a HRQoL and functional outcome measure for children, particularly those with disabilities, this tool was adapted by Coutinho dos Santos et al. 105 to measure quality of life in children and adolescents treated with BTX-A for palmar hyperhidrosis.

Step 1

The likely distribution of possible responses (a reduction in HDSS score of 0, 1, 2 or 3 points) depends on the skewness of the distribution across that range. If the population were evenly distributed across the possible responses, then the mean change would be 1.5. The distance of the mean change from 1.5 is an indication of the skewness of the distribution across the possible range of distributions.

There is no reason to think that there are differences for distinct populations in how they are likely to respond to treatment, so a unimodal distribution was assumed. This was imposed as a constraint in the simulation model by assuming that if the mean change is < 1.5 then:

where E_x represents the number of events with X points of reduction on the HDSS scale. If the mean change is > 1.5 then: