CKS found no randomized, placebo-controlled trials of the efficacy and safety of aluminium chloride hexahydrate or any other aluminium salt for the treatment of primary focal hyperhidrosis. The recommendation to use 20% aluminium chloride hexahydrate in alcohol solution as first-line treatment of primary focal hyperhidrosis is based on poor quality evidence from two small, quasi-controlled trials [Rayner et al, 1980; Goh, 1990], four case series of between 12 and 65 highly selected subjects [Scholes et al, 1978; Ellis and Scurr, 1979; Jensen and Karlsmark, 1980; Glent-Madsen and Dahl, 1988], and expert opinion, including two published consensus statements [Hornberger et al, 2004; Solish et al, 2007], two guidelines [Lowe et al, 2003; International Hyperhidrosis Society, 2008b], and an evidence-based review [DTB, 2005]. Satisfaction with treatment varied between 63% and 99%, and discontinuation due to irritation varied between 2% and 20%. However, these findings should be interpreted with caution in the absence of any randomized, controlled trials, and they are not readily generalizable to a UK primary care population.

• In one quasi-controlled trial, published as a letter in the British Medical Journal, 38 people with axillary hyperhidrosis on a waiting list for surgery were treated with topical 20% aluminium chloride hexahydrate, and a placebo, applied on alternate nights for 14 nights [Rayner et al, 1980]. After 14 days, whichever solution relieved symptoms (it is not clear how this was decided) was used on the opposite axilla on alternate nights for a further 14 days, and when necessary to the axilla where relief had already been obtained, and presumably thereafter for 6 months, although this is not clear.
• 24 people obtained considerable relief with the active compound, two obtained considerable relief with the placebo, four obtained equal relief with both solutions, six showed no improvement, and two defaulted.
• Five of the 24 people experiencing considerable relief with the active compound opted for surgery because the treatment was not acceptable.
• After 6 months, only six of the original 19 people experiencing considerable relief and finding the treatment acceptable, had sustained relief. In all, 26 out of 38 people still opted for surgery.
• In another quasi-controlled trial, with assessor blinding only, 12 people with idiopathic palmar hyperhidrosis were instructed to apply aluminium chloride hexahydrate to one palm every night for 4 weeks, with the other palm acting as control [Goh, 1990].
• All subjects reported an improvement in symptoms on the treated side.
• Four people experienced stinging, which settled in three people. One person dropped out because of this adverse effect.
• In the remaining 11 people, treatment was associated with a significant reduction in skin water vapour loss compared with no treatment (p = 0.004).
• Four case series have been published:
• Sixty-four out of 65 people with axillary hyperhidrosis, treated in UK dermatology clinics with 20% aluminium chloride hexahydrate in absolute alcohol, achieved 'excellent control of sweating' at 12 months [Scholes et al, 1978]. Twenty-nine people experienced some irritation, 28 of whom said that it was readily relieved by applying 1% hydrocortisone cream on the morning after treatment.
• Forty-two people with axillary hyperhidrosis on a waiting list for surgery were treated with topical 20% aluminium chloride hexahydrate in absolute alcohol [Ellis and Scurr, 1979]. At 3 months, eight people had defaulted, 27 people described the treatment as highly successful, and seven people stopped using the treatment (three because they could not manage it, and only three because of severe irritation or soreness).
• Sixteen people with palmar or plantar hyperhidrosis, or both, were treated with aluminium chloride hexahydrate 25% in absolute ethyl alcohol. In 12 out of 13 people with palmar hyperhidrosis, control of sweating was achieved after 3–4 weeks of daily treatment [Jensen and Karlsmark, 1980]. In 10 out of 11 people with plantar hyperhidrosis, control of sweating was achieved after 5–6 weeks of daily treatment. Five people complained of itching.
• One study compared the efficacy of aluminium chloride hexahydrate 25% in ethanol alone and combined with an additional topical treatment, triethanolamine [Glent-Madsen and Dahl, 1988]. The results of the aluminium chloride only arm alone can be viewed as a case series. Out of 30 people treated, the median for sweat secretion (authors stated 'reduction', presumably incorrectly) decreased to 25% of the value before treatment. Six people (20%) discontinued treatment because of irritation.
• Two other studies which essentially amount to case series, suggest that good efficacy may be achieved by using a different vehicle for aluminium chloride hexahydrate, or by using a different aluminium salt. Neither are currently available for use in the UK [BNF 56, 2008].
• Out of 238 people with localized hyperhidrosis treated on 332 anatomical sites with aluminium chloride hexahydrate (of varying concentrations), in a salicylic acid gel base [Benohanian et al, 1998]:
• Good or excellent outcomes were reported for over 90% of treatments to axillae, 60% of treatments to hands, and over 80% of treatments to feet.
• Subjects self-reported outcomes. The duration of treatment was not specified.
• Twenty people with primary or idiopathic hyperhidrosis were treated with a new foam formulation containing 20% of an aluminium salt (sesquichlorhydrate) [Innocenzi et al, 2005]:
• On average, sweating halved after 15 days' treatment, but quality of life improved only for people with axillary, not palmar, hyperhidrosis.
• CKS found no randomized, placebo-controlled trials of foot powder containing an aluminium salt for people with primary focal hyperhidrosis. An RCT was published that compared a preparation referred to as Zeasorb^® with placebo, but the constituents differed markedly from the current preparation, and did not include an aluminium salt [Helfand, 1963].