Topic Article:
Hand Eczema Extent and Morphology- Association and Influence on Long-term Prognosis
Brigitta Medling, Karin Wrangsjo, Bengt Jarvholm
Journal of Investigative Dermatology (2007) 127, 2147–2151; doi:10.1038/sj.jid.5700841

Hand Eczema Prognosis
Robert S. Kirsner¹ and Shasa Hu¹
Journal of Investigative Dermatology (2007) 127, 2072; doi:10.1038/sj.jid.5701044

Hand eczema is a common, prevalent condition (Meding and Jarvhl, 2002). In many patients it assumes a chronic and relapsing state that persists over long periods of time. In a 15-year follow-up study of individuals with hand eczema, Meding and colleagues (2005) previously identified several factors of importance in predicting which patients are likely to have a chronic course. Among the factors that negatively impacted prognosis were extent of eczema involvement at initial presentation, history of childhood eczema, and being younger than 20 years of age at onset of disease. In this issue of the Journal of Investigative Dermatology, Meding et al. (2007) further examine factors that may influence long-term prognosis of patients with hand eczema. Using 868 of the original patients examined in 1983, and evaluating patients’ initial extent of involvement and the number of different morphologic lesions they had at presentation, they determined the effect of these parameters on long-term prognosis. The authors found that the extent of involvement correlated with a polymorphic presentation, but having many types of lesions did not significantly affect the ability to predict patient outcome. Through the following questions we will delve into this paper in greater detail.

QUESTIONS

1. What disease is being studied in this article?

2. “HiEx” and “LoEx” as well as “HiMo” and “LoMo” are used to dichotomize the population studied. How and why were these chosen?

3. The maximum possible score with respect to extent is 74 yet the mean was 5.2. Is it typical to have such ‘mild’ (low mean score compared to maximum score) disease when these types of scoring systems are used?

4. What is the outcome of interest in this study? What other outcomes could have been used to test the same hypothesis?

5. How might the results of this study impact clinical practice?

ANSWERS

1. The title of the article states the disease studied is “hand eczema”. Hand eczema, or hand dermatitis, is multi-factorial, may be due to atopic dermatitis, dyshidrotic eczema, nummular dermatitis, irritant or allergic contact dermatitis (Coenraads, 2007), palmar psoriasis, or rare diseases such as cutaneous T-cell lymphoma. Therefore, “hand eczema” per se is not a specific disease, but rather a broad reaction pattern or a “condition”. Additionally, a person with hand eczema may have one or more underlying diseases (such as allergic contact dermatitis superimposed on atopic dermatitis). The study patients did not receive biopsies for histological examination; thus, hand eczema is a broad category with heterogeneous and unknown causes in this article.

Many benefits may be associated with studying and drawing conclusions about a broad reaction pattern. From an epidemiologic perspective, it is easier to study conditions that do not require a high degree of expertise or complicated algorithm to diagnose. For example, epidemiologic studies of atopic dermatitis in children often are diagnosed based on longstanding pruritus of the flexures beginning in childhood (Williams, 1999). Secondly, the conclusions made from studying a reaction pattern, as opposed to a specific disease, may have broader or more practical implications, as a reaction pattern is more easily recognized and/or accepted. While a non-dermatologist may readily make the assessment of “hand eczema”, classifying it into underlying causes such as atopic dermatitis or allergic contact dermatitis or psoriasis requires more advanced knowledge and work up (e.g. biopsy, patch testing).

While this article does not specify underlying causes of hand eczema in the study population, earlier publications from the same group based on the same study population imply that a majority of the study patients have irritant dermatitis, allergic contact dermatitis, dyshidrosis or atopic dermatitis (Meding 1990, 2005).

2. Definitions:
LoEx = low extent (score 1-5);
HiEx = high extent (score ≥6);
LoMo = low morphology (1-2 morphologic signs);
HiMo = high morphology ( ≥3 morphologic signs)

In an effort to predict who with hand eczema will have a prolonged and refractory course, the study investigators evaluated the correlation between baseline characteristics of the study patients and long-term prognosis of their hand eczema. Previous studies by Meding et al found that extent of eczema at initial presentation, history of childhood eczema and being younger than 20 years of age at initial presentation were negative predictive factors of both persistence and recurrence of hand eczema (Meding, 2005). In this study the investigators sought to improve a clinician’s ability to predict outcome by examining the association between eczema extent and morphology with long-term prognosis. The extent of hand eczema was calculated based on anatomical locations involved, with a maximum possible score of 74 for both hands per individual. A cut-off at 5 was likely chosen to dichotomize the population into “LoEx” and “HiEx” because the mean score at the time of initial clinical examination (1983) for extent of involvement was 5.2.

Morphologic rating was calculated from how many categories of clinical signs each patient had at the initial examination. The six possible categories were erythema, papules, vesicles/pustules, scaling, fissures and edema/infiltration. The article did not specify the mean value of morphologic rating of the study cohort. It is possible that the cut-off at 2 may represent the mean or median value, or it may be an arbitrary choice. Interestingly, it has been shown that interobserver agreement on clinical signs/morphologies (using the same categories) of hand eczema is lower than that of extent (Held et al, 2005), which may be the reason Meding et al chose to dichotomize the populations based on the NUMBER of morphologies rather than the nature of morphologies (such as grouping papules, vesicles/pustules together versus fissures and scaling). Additionally, while in certain diseases different morphologic presentations may portend prognosis, in other diseases this may not be the case. For instance, a nodular or tumoral cutaneous T cell lymphoma (CTCL) denotes stage IIb or higher disease which has poorer prognosis than patch or plaque type CTCL given similar lymph node status. Erythrodermic CTCL implies even worse prognosis, as it is often seen with Sezary syndrome. In a more common disease such as acne, cystic acne is both clinically more severe and more resistant to therapy than comedonal or papular acne. However, in hand eczema or hand dermatitis, it is unclear whether patients with vesicles/pustules will have worse clinical course and/or prognosis than those with fissures or scale. Thus, the authors probably lack a clear rationale for dividing the study cohort based on morphologic features.

Statistical analyses were then performed using the “LoEx”/ “HiEx” and “LoMo”/ “HiMo” separation of the study cohort at baseline. The authors concluded that HiEx was correlated with HiMo and LoEx with LoMo, but that adding polymorphology (HiMo) to extent of involvement did not substantially improve one’s ability to predict long-term prognosis of a patient’s hand eczema.

3. It might seem unusual that a disease severity score system is designed such that the majority of the patients have low score compared to the possible maximum score. Interestingly, this is the typical feature of severity scoring systems. In fact, several studies on atopic dermatitis and psoriasis severity scoring scales have used such large scales, with patients affected by severe disease having a score considerably lower than the highest possible score. For example, the hand ezcema severity index evaluated by Held et al ranges from 0 to 360 points; most of the fifteeen patients evaluated scored below 100 (Held et al, 2005). The Psoriasis Area Severity Index (PASI) is a validated and widely used scale system that has a maximum possible score of 72. Most clinical trials on moderate to severe psoriasis enroll patients with a PASI score of 20. By having a large scale, one can more accurately differentiate and account for even small differences in disease severity between any 2 individuals, which ultimately allows for more accuracy in the scoring system. In other words, when the scale is larger, there is more allowance for intermediate readings, thus making the scale more sensitive. In fact, validity scales for hand eczema do exist, with the 5-fold higher maximum scores, but such complex scales may have limited real-life clinical utility.

4. The outcome of interest is the answer to a single question on the follow-up questionnaire sent 15 years after the original cohort was examined. The question was whether, according to the patient, the patient had had eczema on any occasion during the past 12 months.

Using this as the outcome of interest has a number of limitations, one of which is the subjective nature of the question, since a “yes” answer to the question does not necessarily equate to a clinical confirmation and diagnosis of persistence and/or recurrence of hand eczema. Based on the outcome of the original examinations performed in 1982-83, it is likely that an 11% error rate might exist in answering this question (in 1982-83, 1385 individuals who attended the examination thought they had hand eczema; hand eczema was diagnosed in 89% or 1238 of those individuals; thus 11% of examined patients thought they had hand eczema but did not).

A second limitation is that the hand eczema experienced during the past 12 months may not be the same type of eczema the patients had 15 years earlier. In addition, a patient may be eczema-free for most of the 15 years but then develop an acute episode or eczema from different underlying etiology at the time of follow-up survey. So again, a “yes” response may not equate to having the same disease in the past 12 months as that of 15 years prior.

Additionally, the wording of the question does not capture any degree of severity of the hand eczema. Whether the hand eczema led patients to use treatment, whether they sought medical care, whether it affected their quality of life, or whether it led to lost days of work, are not addressed. Neither did the question capture the duration of hand eczema during the past 12 months. Living with hand ezcema 1 day or 365 days certainly shows very different chronicity and severity. While the question does dichotomize those with any hand eczema from those without, one might construct the exact same study but envision a series of questions aimed at detecting not solely the presence of the eczema but other features of disease severity to better define the long-term prognosis of the study cohort. Interestingly, the follow-up survey did ask about sick leave due to eczema, disability or pension due to hand eczema, and job change due to eczema (Meding, 2005). It may be of greater clinical value if the extent and morphology of initial presentation of hand eczema can be correlated with answers to these more specific questions.

5. While we hope any individual paper will have direct impact on clinical practice, more often than not it does not. At best it is often a body of work, either by a single group of investigators or several groups trying to answer the same clinical question, which begins to build a story or answer a question.

The broad clinical question asked in this paper is which factors improve the accuracy of predicting long-term prognosis of patients with hand eczema. From a practical standpoint, using information from this study we learn that one does not need to render a specific diagnosis to answer the question. The authors conclude that a simple assessment of hand eczema extent can predict long-term prognosis, while morphology (in particularly polymorphism) does not contribute to prognosis. Based on this, hypothetically speaking, a clinician can advise a patient with erythema and scale of bilateral partial palms and two volar pads, a HiEx (6 points for extent)/LoMo (2 morphologies) category, that based on the extent alone, his or her hand eczema will likely have more prolonged course, before any diagnostic tests are performed. A majority of non-dermatologist practitioners will be able to quickly and easily use such classification in the office setting, while a “splitter” or more inquisitive physician may not find the process satisfying. Most dermatologists will likely perform additional diagnostic tests to investigate the underlying etiology of hand dermatitis.

Based on the conclusion of the study, which places greater importance on extent than morphology, can one infer that reduction in the extent of eczema is a superior clinical response than changes/reduction in morphology? Alternatively, from a patient’s perspective the answer might be different. For example, a secretary may appreciate resolutions of fissures and/or vesicles to erythema on her volar pads more than reduction of eczema on her dorsal hand. Thus, whether in practice improvement in morphologic features equals improvement might depend on the perspective of the one judging the improvement.

REFERENCES

Meding B, Jarvhl B (2002) Hand eczema in Swedish adults—changes in prevalence between 1983 and 1996. J Invest Dermatol 118:719–23

Meding B, Wrangsjo K, Jarvholm B (2005) Fifteen-year follow-up of hand eczema: predictive factors. J Invest Dermatol 124:893–7

Meding B, Wrangsjo K, Jarvholm B (2007) Hand eczema extent and morphology—association and influence on long-term prognosis. J Invest Dermatol 127:2147-2151

Coenraads PJ (2007) Hand eczema is common and multifactorial. J Invest Dermatol 127:1568-70

Held E, Skoet R, Johansen JD, Agner T (2005) The hand eczema severity index (HECSI): a scoring system for clinical assessment of hand eczema. A study of inter- and intraobserver reliability. Br J Dermatol 152:302-7

Meding B, Swanbeck G (1990) Occupational hand eczema in an industrial city. Contact Dermatitis 22:13-23

Meding B, Wrangsjo K, Jarvholm B (2005) Fifteen-year follow-up of hand eczema: predictive factors. J Invest Dermatol 124:893-7

Williams HC (1999) Diagnostic criteria for atopic dermatitis: where do we go from here? Arch Dermatol 135:583-6

¹Department of Dermatology and Cutaneous Surgery, University of Miami Miller School of Medicine, Miami, Florida, USA