Prognosis of Unclassified Eczema: A Follow-up Study

Objective  To investigate the outcome of unclassified eczema (UE). Eczema refers to a set of highly heterogeneous clinical conditions including contact dermatitis (CD) and several types of endogenous eczema. In clinics, however, the eczema of a large proportion of patients does not fit any known patterns and has been diagnosed as unclassified eczema. The prognosis of UE is largely unknown.

Design  Cohort study with 1-year follow-up.

Setting  Dermatology clinic in Peking University Third Hospital, Beijing, China.

Patients  All cases of UE in 655 consecutive patients with eczema patch tested with the European Standard series occurring during a 2-year period.

Interventions  At 1 year after patch testing, all patients were asked to revisit the clinic, and 599 patients returned (91.5%); of these, 192 had UE (32.1%).

Main Outcome Measures  The rate of symptom-free skin for at least 3 months without relapse before evaluation.

Results  During the follow-up period, only 15.1% of patients with UE had their lesions cleared, which was a much lower rate than for those with CD (50.4%) (P < .001; χ² test) but close to the rate of those with atopic dermatitis (7.1%). A total of 11.0% of patients with UE had atopic diathesis, and 72.4% of those with UE had a positive reaction to at least 1 contact allergen, which was a lower rate than that of patients with allergic CD (88.6%) but higher than that for patients with irritant CD (53.8%) (P < .001 and P = .004, respectively; χ² test); however, none of these results were relevant to eczema.

Conclusions  Unclassified eczema is a common type of eczema with a very poor prognosis. This important category of eczema should be recognized and studied further.

The term eczema is applied to specific inflammatory reactions of the skin and covers a set of etiologically highly heterogeneous clinical conditions.^1,2 Eczema is the most common type of dermatitis. Classification of eczema is largely empirical, and in most circumstances the diagnosis is based only on clinical findings. There are many types of eczema recognized clinically, including allergic contact dermatitis (ACD) and irritant contact dermatitis (ICD), atopic eczema (dermatitis), seborrheic dermatitis, nummular eczema, dyshidrotic eczema, and stasis eczema, to name a few. Recently, atopic eczema and CD have drawn much attention in the medical community because both of them are very common and have a great impact on the quality of life. The prevalence of atopic eczema is rising; in some areas it occurs in roughly 45% to 64% of preschool children, and in adults the incidence rate can be as high as 40%.³ Atopic eczema commonly predates the development of allergic rhinitis and asthma.³ In clinical practice, however, the eczema of large percentages of patients does not fit the criteria for any known types. The terms unclassified eczema (UE) and unclassified endogenous eczema have been applied to these conditions. Although UE is not uncommon, it has been seldom studied. We searched Medline for articles published from 1996 through 2006 and found only 8 studies.^4-11 Little is known about the prevalence and outcome of UE. The purpose of this study was to analyze the prognosis of UE in Chinese patients with eczema given a patch test (PT) in our department.

Eczema was diagnosed and classified according to standard textbooks.^1,12 Specific types of eczema, such as CD, atopic dermatitis (AD), asteatotic eczema, seborrheic dermatitis, and so forth, were clinically diagnosed accordingly, and the remaining unspecified eczema was diagnosed as UE. To diagnose UE, the following diagnostic criteria should be met: (1) the skin condition has to be first defined as eczema, (2) CD should be excluded, and (3) the clinical pattern of the eczema must not fit any types of known eczema.

Contact dermatitis was diagnosed clinically by the disease history and clinical examination based on a standard textbook.¹² Only patients with strong evidence of the condition supporting the diagnosis were included in the study. If the patient with suspected CD could reapply the suspected substance without any reactions, a diagnosis of CD was excluded. A diagnosis of CD was confirmed if patients with suspected CD also fulfilled the following criteria: (1) a positive usage test result, (2) a relevant positive PT reaction to allergens in the European standard series or a positive PT result with the suspected material as is, or (3) a positive repeated open application test result. The usage test was performed with the method similar to that reported by Bashir and Maibach,¹³ in which a patient thought to have CD used the suspected substance in the same way as when the dermatitis developed, for example, by applying suspected facial cream twice daily to a small area (1 cm × 1 cm) on the face for a week. If an eczematous skin reaction occurred during the test period, the test was considered positive and stopped. In a repeated open application test, test substances—either commercial products, as is, or special test substances (eg, PT allergen)—were applied twice daily to the upper arm on a 1 cm × 1 cm area for a week. If an eczematous skin reaction occurred in the test period, the test result was considered to be positive, and the test was stopped.¹² These tests were performed when CD was strongly suspected, but no relevant PT could be found. Similarly, if ACD was elicited clinically more than twice by the same contact allergen, a diagnosis of ACD was also considered clinically confirmed.

Atopic dermatitis was diagnosed by using the United Kingdom diagnostic criteria.¹⁴ Atopic diathesis was considered when allergic rhinitis, allergic asthma, or AD could be found in the patient's personal or family history. If multiple factors were involved in 1 patient, the final diagnosis was based only on the major cause of the eczema. For example, if a patient's dermatitis fulfilled the diagnostic criteria of AD and was also found to have reacted to some allergens, but the ACD was temporal and could not explain the whole skin condition, the final diagnosis was AD.

A total of 655 patients with eczema who were referred by other dermatologists for PT and who underwent testing at Peking University Third Hospital from September 2003 to May 2005 were evaluated. There were 187 men and 468 women, with a mean (SD) age of 40.5 (16.0) years. The ethnicity of all patients was Chinese Han. The local ethics committee approved this study, and the participants gave oral consent.

The suspected causal agent was investigated by using a questionnaire.¹⁵ In the questionnaire, the patient's personal data, history of the present illness (patient's description, date of onset, effects of weekends and vacation on dermatitis, and previous therapy), contactants that existed at work and in clothes, toiletries, household contact and treatment medications, atopic diathesis, and medications used (current use or within the past 3 months) were included.

Patch testing was performed with a modified European standard series of allergens and an improved quality chamber supplied by Chemotechnique Diagnostics, Malmö, Sweden (Table 1). Allergens were applied to the patient's upper back, and the results were recorded at 2 and 3 days according to International Contact Dermatitis Research Group grading.¹² If patients could supply the suspected substances, PT with their own products was also performed, using published methods.¹² After testing, the information on the sources of the positive test allergens was given to the patient, and complete avoidance was suggested. If the dermatitis definitely improved with the avoidance of the given allergen, the PT was considered relevant. In patients with positive PT reactions related to past ACD, those with no improvement after avoidance of the given allergen, or those who denied any exposure to the suspected allergen, the PT result was considered not relevant.

After PT, all patients were treated in our clinic with well-accepted methods.^1,12 Briefly, localized eczema was treated with topical corticosteroids and if necessary, skin moisturizers or antibiotics. Systemic corticosteroids were given to patients with widespread eczema to control the disease. We also educated patients to follow the following instructions: (1) to do their best to find and avoid any predisposing and exacerbating factors; (2) to bath properly and use skin moisturizers if necessary; (3) to eat simple and fresh food, avoiding spicy food or any food for which intolerance had been identified; and (4) to wear loose-fitting cotton clothing. One year after PT, all patients were asked to revisit our clinic, and their prognosis was evaluated according to the following categories: (1) clearance (the patient was completely free of skin symptoms and had had no recurrence for more than 3 months before the evaluation), (2) relapse (skin lesions had resolved but reappeared before evaluation), and (3) improvement or no improvement (assessed globally by the observers). The data were processed by χ² test and rank sum test with SPSS statistical software (version 11.0; SPSS Inc, Chicago, Illinois).

Of 655 consecutive patients with eczema, 599 (91.5%) finished the study; 262 (43.7%) were diagnosed as having CD (hereinafter, CD group), 192 (32.1%) were diagnosed as having UE (hereinafter, UE group), 131 (21.9%) had other forms of eczema, and 14 (2.3%) had AD (hereinafter, AD group). The age, sex distribution, concurrence of atopic diathesis and ichthyosis, and drug allergy history for patients with UE, ACD, ICD, and AD are shown in Table 2. All patients in the AD group were adults. There were no differences in mean age among each group. However, males were notably a minority in the ACD group. The male to female ratio was 1:1.5 and 1:4.4 in the UE and ACD groups, respectively (P < .001; χ² test). Unlike the AD group, in which 92.9% of cases were atopic, the rate of atopic diathesis in the UE group was lower (11.0%; P < .001, Yates corrected χ² test). There were no differences in the concurrence of ichthyosis and drug allergy among the UE, ACD, ICD, and AD groups. The outcomes for UE, ACD, ICD, and AD are shown in Table 2. The clearance rate of UE was much lower than those of ACD (15.1% vs 49.7%; P <.001; χ² test) and ICD (15.1% vs 44.9%; P < .001; χ² test) but similar to that (7.1%) of AD (P = .41; χ² test). The clearance rate of CD was 50.4%, which was also much higher than that of UE (P < .001; χ² test).

The clearance rates of UE, ACD, and ICD decreased with the increase of disease duration before PT (Table 3). Because all of our patients with AD were adults, their disease duration before testing was longer than 24 months.

The PT results are shown in Table 1. Of the UE group, 72.4% (139 of 192 patients) reacted to at least 1 allergen in the European standard series, which was lower than that of the ACD group (88.6% [148 of 167]) but higher than that of the ICD group (53.6% [37 of 69]) (P < .001 and P = .004, respectively; χ² test); however, none of the PT results were found to be relevant to the eczema. The PT results in 28 of 139 patients with positive results in the UE group were related to the past incidences of ACD. There were no significant differences among the positivity rates to each allergen in the ACD, UE, and ICD groups.

The term unclassified eczema describes a group of conditions that do exist but are seldom studied.^4-11 According to the criteria of the International Statistical Classification of Diseases, 10th Revision (ICD-10),¹⁶ our cases of UE do not correspond to any types of classified eczema. This study further demonstrated that a high percentage of cases of UE (32.1%) was found in Chinese patients with eczema who underwent PT. In our study, the percentage of AD was only 2.5%. This may be partly because all of our patients were adults and partly because patients with AD did not always undergo PT in our clinics. The low prevalence of AD (0.70% in students aged 6-20 years) in China may also have contributed to this finding.¹⁷

The 1-year outcome of the UE group was very poor. The clearance rate of UE was only 15.1% ,which was much lower than the clearance rates of the ACD and ICD groups but similar to that of the AD group (7.1%).

The mechanism of UE is still unknown. In 2001, we reported¹¹ that 49.0% of patients with UE reacted to aeroallergen intradermal testing. Other research⁵ also suggested that UE might be a kind of AD because 12 of 34 patients had an elevated IgE level (> 240 μg/L; to convert IgE to milligrams per liter, multiply by 0.001). However, none of these patients could be diagnosed as having AD based on any current diagnostic criteria of AD. A high positive rate of PT in our study suggested that these patients may have had undiagnosed ACD; however, none of these positive results were found to be related to the eczema, and ACD could not be diagnosed. Based on these findings, we strongly recommend that the medical community should pay more attention to UE and that this important category of eczema should be recognized and further studied.

It has been reported that if more allergens are tested, more clinical relevance is found. For example, the North American Contact Dermatitis Group¹⁸ recommends 65 allergens for PT. Some patients with UE in our care might have had undiagnosed ACD owing to the relatively small number of allergens tested in this study. Therefore, we suggest that to explore the possible mechanism of UE in the future, patients should undergo PT with more contact allergens, and detection serum IgE that specifically responds to common environmental allergens is necessary to exclude the possible allergic reactions to environmental allergens first.

Correspondence: Lin-feng Li, MD, PhD, Department of Dermatology, Peking University Third Hospital, 49 N Garden Rd, Haidian District, Beijing 100083, China (zoonli@sina.com).

Accepted for Publication: February 25, 2007.

Author Contributions: Dr Li had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Li and Liu. Acquisition of data: Li, Liu, and Wang. Analysis and interpretation of data: Li and Liu. Drafting of the manuscript: Li, Liu, and Wang. Critical revision of the manuscript for important intellectual content: Li. Statistical analysis: Liu. Obtained funding: Li. Administrative, technical, or material support: Li and Wang. Study supervision: Li.

Funding/Support: This study was supported in part by the national “211 Project” of the Peking University Evidence-Based Medicine Group (No. 93000-246156071) and a special fund for the promotion of education (Dr Li) from the Ministry of Education, China.

Financial Disclosure: None reported.

Role of the Sponsors: The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; or in the preparation, review, or approval of the manuscript.

Additional Contributions: We are indebted to the “211 Project” of Peking University and the Ministry of Education for their support.