Improving povidone‐iodine and iodine preparations for patch testing

Allergy evaluation by patch testing with povidone‐iodine (PVP‐I) or iodine remains challenging, because current patch test preparations frequently lead to false‐positive or irritant skin reactions.


| INTRODUCTION
Iodine compounds have been used as antiseptics at least since the 18th century. 1 The inhibitory effects of iodine on bacteria, viruses, and fungi led to large-scale applications of iodine in virtually all medical fields. Free iodine is not easily dissolved in water 1 ; further disadvantages include its skin-irritating properties, unstable chemical preparations, and high reactivity in oxidation processes. These negative properties of iodine may be overcome by supplementation with polyvinylpyrrolidone (commonly called povidone [PVP]), a free-iodine binding iodophor. The combination of the water-soluble polymer PVP with iodine results in a stable solution. 2 Thus PVP-iodine (PVP-I) is used almost exclusively, instead of iodine alone, owing to its low irritancy and toxicity. 3 German PVP-I products usually contain 10% releasable iodine in the form of PVP-I. For the diagnosis of iodine contact allergy, patch tests must be performed. However, the well-known irritant properties of iodine 3,4 may complicate evaluation of patch test results, particularly with this compound. 5,6 In an earlier study, 500 patients were patch tested with different preparations of 1% PVP-I diluted in ethanol, glycerol, or water, 7 and 14 of these individuals exhibited positive patch test reactions to PVP-I 1%. However, in additional repeated open application tests (ROATs), only two reactions were found to be relevant. In another study from 2005, various iodine preparations were investigated in patch tests in 24 healthy volunteers without suspicion of iodine allergy. 3 Here, the tests were performed with iodine in petrolatum (0.5%, 1%, 5%, and 10%), iodine in 70% isopropyl alcohol (0.5%, 0.75%, and 1%), and PVP-I in water (1%, 5%, 7.5%, and 10%). With regard to the petrolatum-based test preparations, iodine 0.5% pet.
was mostly nonirritating. Iodine 1% pet. revealed some skin irritation, whereas 5% and 10% iodine in pet. showed strong irritating properties. Patch tests with the water-based test preparations with PVP-I 1% and 5% were completely negative. Among all volunteers, only one individual demonstrated a skin reaction to PVP-I 7.5% aq. and 10% aq. without suspected allergy.
In a previous study, we investigated 79 patients with suspected contact dermatitis caused by antiseptics. 5 Patch tests were performed with PVP-I 0.1%, 1%, and 10% aq., and with iodine 0.5% pet. The relevance of positive tests was assessed with ROATs. Testing with PVP-I 0.1% aq. and 1% aq. proved to be unsuitable, because no positive reactions were identified at these low test concentrations. Strong positive reactions to PVP-I 10% aq. and positive reactions to iodine 0.5% pet. correlated well in this former study. However, testing with PVP-I 10% aq. also revealed some false-positive reactions. This test concentration is well known to frequently result in irritant skin reactions, which are sometimes difficult to distinguish from weak-positive reactions. 6 Less-experienced allergologists frequently diagnose iodine allergy in patients showing these results.
Owing to the high number of irritant and therefore potentially falsepositive reactions, patch testing with iodine preparations remains highly unsatisfactory. In our previous study, we concluded that patch testing with PVP-I 2% aq., 5% aq., and iodine 0.5% pet. appears to be more promising than PVP-I 10% aq., which is currently included in the German patch test series for disinfectants. From this starting point, we continued our search for the best possible test concentration and the optimal vehicle. In addition, we investigated whether late readings on day (D) 7 might add any value to the results of patch tests with iodine preparations. PVP-I (Caelo) was prepared at test concentrations of 2%, 5%, and 10% in water. All patients except one were tested with these different vehicles and concentrations of iodine and PVP-I in parallel. Patch tests were performed with 48 hour exposure through the use of Finn Chambers (8 mm inner diameter) on Scanpor tape (Smart-Practice, Greven, Germany) and read at D2, D3, and D7, according to the current guidelines of the German Contact Dermatitis Research Group. 8,9 This retrospective study was reviewed and approved by the local ethics committee at the University Medical Center Göttingen.  Regarding the test preparations, all patients except one were tested simultaneously with iodine 0.5% pet. and PVP-I 2% aq., 5% aq., and 10% aq., and the reactions were read on D2 and D3. One third (31/95) demonstrated an irritant or doubtful reaction on D2 and/or D3 and/or D7 to one or more test preparations (Table 1)  We expected the test preparations PVP-I 2% aq. and iodine 0.5% pet. to be best suited in terms of sensitivity and specificity, on the basis of our previous work. 5 Comparison of these two test preparations is depicted in Table 2. Only two patients revealed positive skin reactions to both test preparations, and only the patient with contact allergy to iodine had an extreme positive (+++) reaction to iodine 0.5% pet. as well as a (+) positive reaction to PVP-I 2% aq. In a second analysis using the most widely used test preparation, PVP-I 10% aq., we compared the patch test results with this patch test preparation with those with PVP-I 2% aq. (Table 2) Table 3). The late reading did not provide added value. In addition, no increasing reactions were observed from D3 to D7, a finding that would have been relevant in the evaluation of weak or borderline positive skin reactions on D3.

| Annual consumption data of skin disinfectants reveals a strong increase in octenidine use from 2010 to 2019
To correct for an indirect effect on iodine allergy due to a possible change in the use of antiseptics, we analyzed the annual sales of skin T A B L E 1 Patch test results on day 3 (D3) in 95 patients tested with iodine and povidone-iodine (PVP-I) Note: One patient was not tested to PVP-I 10% in aqua. Only one patient among all tested patients (n = 95) was diagnosed with iodine allergy and demonstrated positive reactions to all tested preparations (as depicted in Figure 1  Because recent analyses revealed that skin disinfection with chlorhexidine solutions compared with iodine significantly reduces the risk of surgical site infections, 14,15 we hypothesized that the frequency of iodine use might have declined in recent years. We therefore investigated real-life data on the sales of PVP-I as well as octenidine-and chlorhexidine-containing disinfectants, dispensed from German pharmacies over the past 10 years (2010-2019). However, iodine consumption has been largely stable over the years, whereas sales of octenidine have strongly increased. A major advantage of octenidine is its effectiveness against methicillin-resistant Staphylococcus aureus, thus potentially explaining its increased use. 16 Chlorhexidine-containing products are mainly preparations for oral use.
In summary, in contrast to the number of suspected cases, the literature and our own experience suggest a low frequency of actual iodine allergy. We conclude that PVP-I 2% aq. is the best possible test preparation for patch tests that can currently be achieved. Evaluating this test preparation in adequately powered studies is strongly recommended.