Atopic Dermatitis: A Review of Current Diagnostic Criteria and a Proposed Update to Management

March 2020 | Volume 19 | Issue 3 | Original Article | 244 | Copyright © March 2020

Published online February 4, 2020

Matthew Reynolds , Joe Gorelick , Matthew Bruno

aArkansas Dermatology and Skin Cancer Center, Little Rock, AR bCalifornia Skin Institute, San Jose, CA cDermatology and Skin Cancer Surgery Center, Allen, TX

systemic treatment should depend not only on the disease severity but also on the psychologic needs of the patient and the risk–benefit ratio of the systemic therapy involved.23 Surprisingly however, apart from dupilumab, the majority of systemic treatments most commonly used to treat AD in the US (methotrexate, cyclosporine, mycophenolate, and azathioprine) do not carry an approved label for that indication.23 Moreover, despite its approval by the US Food and Drug Administration as the first biologic medication indicated for moderate-to-severe AD, dupilumab has yet to be listed on insurance plans as a first-line agent for these patients over the off-label alternatives.

In addition to the inclusion of newer treatments, the AAD treatment guidelines should be updated to reflect more accurately the risk–benefit profiles of the various recommended systemic immunomodulators. A recent study, for example, comparing methotrexate with cyclosporine in the treatment of moderateto- severe AD revealed that despite its inferior efficacy (at lower doses only), methotrexate displayed a significantly favorable safety profile compared with cyclosporine.24

Once updated, physicians should be urged to review the guidelines regularly. This is particularly pertinent as a recent study assessing the adherence to evidence-based guidelines of care for AD in the US described an educational gap in the implementation of the current, albeit outdated, guidelines.25

Proposed Diagnostic Criteria for AD
Of all the existing diagnostic criteria, none are considered wholly or mutually exclusive in the diagnosis of AD; the clinician ultimately makes the final diagnosis. This lack of standardization has obvious implications on the comparability of research findings and was highlighted in a recent systematic review of the diagnostic criteria used in randomized controlled trials of AD.26 In 212 trials examined, 10 different sets of criteria were used, with the Hanifin–Rajka criteria most widely used (in 41% of studies) and the AAD criteria used in only 3.8% of studies. Interestingly, no criteria were specified in 37.3% of the studies, and the authors concluded that there was a need for a harmonized set of diagnostic criteria.

We propose a simplified model consisting of the essential features of the more widely used diagnostic criteria, clarifying the duration and age at onset of the features. Importantly, these criteria include the newer features of AD such as conjunctivitis, periorbital dermatitis, and prurigo/prurigo nodularis, that were previously regarded to have relatively little significance in the context of the overall disease. Our model also proposes clear definitions for the acute (≤6 weeks), subacute (≥6 weeks but not longer than 6 months), and chronic (≥6 months) classifications of AD. For comparison, our proposed 2-plus-1 model is shown in Table 1 alongside the diagnostic criteria currently recommended by the AAD,5 and in Figure 1. We believe that these simplified and updated criteria for the diagnosis of AD will aid dermatology and non-dermatology providers in establishing the diagnosis earlier, leading to a lower rate of misdiagnosis, earlier treatment, and fewer criteria requirements for payers to agree with providers in initiating systemic therapy options for patients.


In recognition of the need for the diagnostic criteria for AD to be updated, we propose a novel, simplified 2-plus-1 model that comprises essential and important features of the most widely used criteria while also emphasizing features previously considered to be of less clinical significance. This model also clarifies