Rash orientation in pityriasis rosea: a qualitative study

ARTICLE

The orientation of the characteristic rash in pityriasis rosea (PR) has been described in various terms such as Christmas-tree pattern [1-6], inverted Christmas-tree pattern, fir tree pattern [7], parallel to the ribs [1, 2, 7, 8] or along the skin cleavage lines [3-5, 9-11]. Many of these terms are imprecise (not having a well defined meaning), inaccurate (not describing the true facts) and can be confusing for students and trainee physicians.

Objectives

The objectives of this retrospective qualitative study are to analyse the orientation of the rash in patients with PR and to suggest the most appropriate description for such.

Materials and methods

We retrieved the clinical records of all patients definitely diagnosed as having PR over a two-year period (1 July 1999-30 June 2001) in a university teaching clinic. The diagnosis was made by a physician with training and qualifications in dermatology and in paediatrics. Our diagnostic criteria of PR were "an acute or subacute eruption of discrete circular or oval lesions with peripheral collarette scaling pattern and central clearance on some or all lesions". Neither a herald patch nor a truncal and proximal limb distribution was mandatory for our diagnosis, as atypical rash distribution is fairly common for PR. Patients with a suspected diagnosis of PR for which the diagnostic criteria were not fulfilled were not included in the study. Syphilis was excluded for all the patients by serological examination.

We retrieved all clinical photographs, if available, from these records. These clinical photographs were taken in the same examination chamber, under the same illumination, by the same photographer, and with the same camera.

We then studied these clinical photographs in a qualitative manner. We compared the rash orientation with the following patterns commonly described for PR: direction of branches of a fir tree, parallel to the ribs, and along Langer's cleavage lines. We also compared the rash orientation with two other rash orientation patterns: dermatomes and Blaschko's lines.

Results

We identified 25 patients with a definite diagnosis of PR in the two-year period. Of such, one patient had atypical PR features (vesiculo-bullous variant). Lesional histopathology of this patient revealed focal spongiosis with perivascular lymphocytic infiltrates compatible with PR. Lesional biopsy was not performed for the other 24 patients with typical features of PR. Of these 25 patients, clinical photographs of 11 patients were available for qualitative analysis. Investigation findings of some of these patients have been reported previously [12, 13].

Direction of branches of a fir tree

Fir trees and pine trees are most popular as Christmas trees. The orientation of branches of both is outwards and upwards. Represented as symbols, the fir tree is often drawn wrongly with its branches projecting outwards and downwards. This is part of the confusion. Another source of confusion is whether the front or the back of the patient is described. The Christmas tree pattern denotes the rash on the anterior trunk radiating medially and inferiorly. The rash on the back should be described as inverted Christmas tree pattern, i.e. radiating laterally and inferiorly.

This pattern is seen in eight of our 11 patients on the upper chest, and in six patients on the upper back. However, this pattern is present to some extent on the lower chest and abdominal wall on two patients only, and on no patient for the lower back.

Parallel to the ribs

Ribs are present on the chest and not on the abdomen. For the posterior and lateral chest walls, this description means that lesions have their long axis mainly in the transverse direction. This is seen in six of our 11 patients. This is a poor description for the lesions on the anterior chest, as the distal ends of the inferior ribs project medially and superiorly towards the costal cartilages and the sternum. No patient thus fits into this description for the anterior chest.

Along Langer's cleavage lines

Langer's lines exhibit three major patterns on the trunk. Firstly, on the upper chest and upper back, they are V-shaped lines mimicking a V-shaped collar. Secondly, adjacent to the shoulders and the hips, the lines are circumferential around the pectoral and pelvic girdles. Thus lines on the anterior axillae run laterally and inferiorly, lines inferior to the axillae run transversely, while lines on the posterior axillae run laterally and inferiorly. Thirdly, lines on the abdominal wall and lower back run transversely. The second and third patterns are quite distinct from the pine tree pattern.

All these three patterns are demonstrated by lesions of most of our patients. The first V-shaped pattern on upper chest and back is seen in nine out of 11 patients. Figures 1 and 2 show this pattern. The second circumferential pattern around the pectoral and pelvic girdles is seen in seven out of 11 patients. Figures 3 and 4 are examples of such around the shoulder joint. The third transverse pattern on the abdominal wall and lower back is seen in seven out of 11 patients, as demonstrated in Figures 5 and 6.

Dermatomes

Dermatomes on the trunk run transversely. This pattern is demonstrated by seven patients to some extent, especially lesions on the abdominal wall and lower back. Lesions on upper chests and upper backs of none of the patients follow this pattern.

Blaschko's lines

Blaschko's lines are complicated on the trunk, with wave-like whorls and arcuate lines. None of the patients has lesions following these lines.

Discussion

Christmas tree and inverted Christmas tree pattern are good descriptions for PR rash for the upper chest and the upper back only. They are poor descriptions of PR rash on the lower chest, abdominal wall, the axillary regions and the lower back. Adoption of these descriptive terms frequently causes confusion. Describing PR lesions as parallel to the ribs is fair for lesions on the posterior and lateral chest walls only. This is a poor description for the lesions on the anterior chest. PR lesions do not follow the direction of dermatomes or Blaschko's lines.

We believe that Langer's cleavage lines is the most appropriate description for the orientation of the characteristic rash in PR. All three components of these lines on the trunk, i.e. V-shaped pattern on upper chest and upper back, circumferential pattern around the shoulders and hips, and transverse pattern on lower anterior trunk and lower back, are demonstrated by most patients. We advocate describing the rash orientation in PR as along the lines of cleavage. Confusing and imprecise descriptions such as Christmas tree pattern and inverted Christmas tree pattern, and inaccurate descriptions such as parallel to direction of the ribs should best be abandoned.

One may argue that it is just logical that the rash in PR follows the lines of cleavage, as the resistance for expansion is greater in the direction crossing the lines than parallel to the lines. However, should the mechanism be true, it would be expected that other annular rashes, especially those with mainly epidermal pathologies, should also follow Langer's lines. Annular rashes with mainly dermal pathologies such as erythema annulare centrifugum, erythema marginatum, granuloma annulare and subacute cutaneous lupus erythematous do not follow Langer's lines. Even tinea corporis, an annular rash with a mainly epidermal pathology, does not follow these lines. Kaposi's sarcoma, which does follow Langer's cleavage lines [14] has mainly dermal pathology. Thus with the present state of knowledge, the mechanism for PR rash following Langer's lines is best considered unknown.

A limitation of this study is the relatively small number of patients. A higher number of patients would have revealed more significant data. Another limitation is that clinical photoxgraphs were used instead of direct and structured clinical examination. Structured clinical examination would have provided a more adequate impression than photographs.

Addendum

Regarding the article by van Steensel MAM et al.: Mal de Meleda without mutations in the ARS coding sequence, published in EJD 2, 2002, pp. 129-32, we have received the following additional information from the authors:

Recently, we ascertained an 18-year old Dutch male suffering from classical mal de Meleda. His parents, who were unaffected, denied consanguinity. We sequenced the ARS gene in the patient and his parents using the conditions and primer sets described in this paper. We found no deleterious mutations in the patient's ARS sequence, just heterozygous polymorphisms. This confirms our previous findings that suggest that mal de Meleda may be genetically heterogeneous.

CONCLUSION

Based on a qualitative analysis of clinical photographs of 11 patients with PR, we found that describing the rash orientation as following Langer's cleavage lines is most precise, accurate and appropriate. The underlying mechanism for this orientation pattern is best considered unknown. We advocate abandoning other descriptions which can cause confusion to students and trainee physicians.

REFERENCES

1. Hunter JAA. Disease of the skin. In: Edwards CRW, Bouchier IAD, Haslett C, Chilvers ER, eds. Davidson's Principles and Practice of Medicine. 17th ed. Edinburgh: Churchill Livingstone, 1995: 941-76.

2. Champion RH, Burton JL, Burns DA, Breathnach SM. Viral rashes. In: Rook's Textbook of Dermatology. 6th ed. Oxford: Blackwell Sciences, 1998: 1092-5.

3. Darmstadt GL, Lane A. Diseases of the epidermis. In: Nelson WE, Behrman RE, Kliegman RM, Arvin AM, eds. Nelson Textbook of Pediatrics. 15th ed. Philadelphia: WB Saunders, 1996: 1863-9.

4. Goldstein BG, Goldstein AO. Papulosquamous disease. In: Practical Dermatology. St Louis: Mosby-Year Book, 1992: 143-68.

5. Cohen BA, Davis HW, Mallory SB, Zitelli JA. Dermatology. In: Zitelli BJ, Davis HW, eds. Atlas of Pediatric Physical Diagnosis. 3rd ed. St Louis: Mosby-Wolfe, 1997: 211-64.

6. Imamura S, Ozaki M, Oguchi M, Okamoto H, Horiguchi Y. Atypical pityriasis rosea. Dermatologica 1985; 171: 474-7.

7. Rogers M, Barnetson RSC. Diseases of the skin. In: Campbell AGM, McIntosh N, eds. Forfar & Arneil's Textbook of Pediatrics. 5th ed. Edinburgh: Churchill Livingstone, 1998: 1616-48.

8. Burton JL. Viral skin infections. In: Essential Dermatology. 2nd ed. Edinburgh: Churchill Livingstone, 1985: 114-25.

9. Steigleder GK, Maibach HI. Erythematous squamous diseases. In: Pocket Atlas of Dermatology. 2nd ed. New York: Thieme Medical Publishers, 1993: 18-28.

10. Cavanaugh RM Jr. Pityriasis rosea in children. A review. Clin Pediatr (Phila) 1983; 22: 200-3.

11. Hsu S, Le EH, Khoshevis MR. Differential diagnosis of annular lesions. Am Fam Physician 2001; 64: 289-96.

12. Chuh AAT, Chiu SSS, Peiris JSM. Human herpesvirus 6 and 7 DNA in peripheral blood leukocytes and plasma in patients with pityriasis rosea by polymerase chain reaction - a prospective case control study. Acta Derm Venereol 2001; 81: 289-90.

13. Chuh AA, Peiris JS. Three cases of pityriasis rosea in children with no evidence of human herpesvirus 7 infection. Pediatr Dermatol 2001; 18: 381-3.

14. Rendon MI, Roberts LJ, Tharp MD. Linear cutaneous lesions of Kaposi's sarcoma: a clinical clue to the diagnosis of AIDS. J Am Acad Dermatol 1988; 19: 327-9.