Iodine-based compounds have been used for centuries as antiseptics and disinfectants. However, it is only in the past 20 years that their use has become widespread in the Spanish health service, after they became the surgical antiseptics of choice in place of the mercury derivatives, which were also known during the 20th century to be contact sensitizers. The iodine compound most widely used at the present time is povidone-iodine (polyvinylpyrrolidone-iodine [PVP-I]).

Contact dermatitis due to PVP-I has never been considered common,1 particularly in view of such widespread use. There are isolated reports of irritant2,3 and allergic4–9 contact skin reactions, but it is often difficult to discriminate between the two types of contact dermatitis because of the lack of consensus on the concentration of the iodinated compounds (PVP-I and/or iodine) and the vehicles to be used when performing diagnostic patch tests.10–14

We describe 7 patients referred to the skin allergy unit of the dermatology department for the diagnosis and treatment of contact dermatitis after undergoing surgery in various operating rooms in our hospital. In 5 of the patients the dermatitis developed in the surgical field 24 to 48 hours after the operation, but in 2 patients it appeared at distant skin sites. The antiseptic used in all cases was Betadine solution, which, in Spain, contains the following components: PVP-I, glycerol, disodium phosphate, citric acid, sodium hydroxide, nonoxynol-9, and water.

The diagnosis of contact dermatitis due to PVP-I was based on the clinical manifestations, a history of exposure, the site of the lesions, and, most relevantly, the results of the skin tests.

The 7 patients were diagnosed with contact dermatitis, for which they received symptomatic treatment. Additional tests were performed some weeks after resolution of the dermatitis.

• 1.

  Patch tests were performed using Finn-chambers aluminum patches on Scanpor held in place by Oper-Tape. The following allergens, supplied by the company Marti Tor, were used:

  • a)

    the standard series of the Spanish Contact Dermatitis and Skin Allergy Research Group (GEIDAC); b) an antiseptics series (Marti Tor) that included 0.5% iodine in petrolatum and to which we added 2% nonoxynol-9 in water, as it is one of the components of Betadine; and c) povidone-iodine (Fragon Ibérica, CAS number: 25655-41-8), prepared by Marti Tor at 3 different concentrations (10%, 5%, and 1%) both in water and in petrolatum.

• 2.

  The dressings used in the different operating rooms in our hospital—Hartman Cosmopore E, Oper dres IHT, and 3M Health Care Tegaderm Film—were applied to healthy skin for 48 hours.

• 3.

  Repeated open application test (ROAT) with Betadine, as supplied, by open application that was repeated twice a day for 7 consecutive days to the volar surface of the forearm.

In order to determine the irritant potential of PVP-I, we studied a control group of 30 adult volunteers (9 men and 21 women) with no previous history of skin reactions to antiseptic products. These individuals underwent patch testing with PVP-I at 10%, 5%, and 1%, both in petrolatum and in water.

In both groups, application of the patch and reading of the results was performed in accordance with GEIDAC recommendations, taking 2 readings, one at 48 hours and one at 96 hours.

Contact dermatitis caused by solutions containing PVP-I is a more common adverse reaction after surgery than previously thought and may be underdiagnosed. It should be suspected when dermatitis develops around the surgical wound or in the operating field, although only patients with more intense manifestations tend to be referred to a dermatologist. This dermatitis is typically limited to the area occluded by the surgical dressing or covered by the drapes used during the operation, and is usually more intense and with more clearly defined borders where there is greater pressure and occlusion; it does not necessarily or commonly affect the surgical wound itself. Dermatitis arising on other areas of the skin, at a distance from the operating field, can cause diagnostic doubt, even for the specialist. In these cases, the dermatitis may be aggravated by occlusion for hours against absorbent pads, sheets, or the surface of the operating room stretcher or table, or under the return electrode or cables of the electric scalpel, drainage tubes, etc., all of which can become impregnated or covered with PVP-I solution when it is applied generously before the operation.

Our 7 patients were initially diagnosed with irritant contact dermatitis due to PVP-I, as the open test (ROAT) with Betadine solution on healthy skin was negative. The clinical appearance in some patients, with well-defined borders of the dermatitis, also suggested this diagnosis. However, we were interested by the type of individual susceptibility or idiosyncrasy of the skin of these patients that could cause them to develop such intense irritant reactions, as, under the same conditions, the majority of surgical patients in our hospital tolerate PVP-I well.

This observation led us to consider the possibility that contact dermatitis caused by solutions of PVP-I after surgery involved not only an irritant but also an allergic mechanism, and that PVP-I becomes a complete antigen only after it becomes associated with transport proteins in the skin of these patients, and this requires occlusion to be maintained for a certain length of time.

Although in general we agree with Lachapelle14 that ROAT on healthy skin is the most suitable tool to confirm (or exclude) allergic sensitization to a given product, this is not always the case. Experience has shown us that there may be exceptions and that ROAT is not necessarily positive, for example, in patients sensitized to volatile allergens, such as ethyl alcohol.15 We believe that the same may occur with the aqueous PVP-I solution, though for a different reason. This compound may require a sealed environment (which occurs with patch testing but not with ROAT) to trigger contact dermatitis in sensitized patients.

A number of studies have given rise to controversy regarding the solvents that should be used and the concentrations of PVP-I and iodine that should be analyzed in the patch tests. Whilst Lee et al.10considered that a 10% concentration in water should be used, Lachapelle14 considered that 10% PVP-I in water is still irritant. Kouzuka12 recommended patch testing with 10% PVP-I in water, but allowing the solution to dry before applying the patch. Based on the results obtained in our controls with the patches of 10% and 5% PVP-I in aqueous solution, we consider that this solvent should be excluded. We agree that PVP-I dissolved in water is irritant, but not PVP-I dissolved in petrolatum.

The combination of irritation and allergy in the case of patches with PVP-I in water would explain the greater intensity of the positive patch tests at 48 hours in our patients (with pustular blisters in one of them) compared with the positive tests with PVP-I in petrolatum. At 96 hours, the reactions of PVP-I in water were less intense, whereas the reactions to PVP-I in petrolatum were more intense, as usually occurs only in allergic reactions.

In conclusion, we consider that humidity and occlusion are necessary for contact dermatitis to Betadine to develop after surgical interventions. This therefore leaves many specialists with diagnostic, clinical, and pathogenic doubts, as, in our opinion, this reaction is due to a dermatitis that is both irritant and allergic.

The open test (ROAT) with this antiseptic, which appears to lose its irritant and allergenic properties as it dries on the skin, is usually negative in these patients.

When using patch tests to diagnose this type of contact dermatitis, we suggest using 10% PVP-I in petrolatum, as the results obtained with PVP-I in aqueous solution in patients and in some of the controls lead us to exclude this solvent due to the demonstrated irritability.

The authors declare that they have no conflicts of interest.