Necrobiosis Lipoidica: a Histopathological and Histochemical Study* Howard R
Total Page:16
File Type:pdf, Size:1020Kb
View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Elsevier - Publisher Connector THE JOURNAL OF INVESTIGATIVE DERMATOLOGY Vol. 44, No. 6 Copyright C 1965 by The Williams & Wilkins Co. Printed in U.S.A. NECROBIOSIS LIPOIDICA: A HISTOPATHOLOGICAL AND HISTOCHEMICAL STUDY* HOWARD R. GRAY, M.D., JAMES H. GRAHAM, M.D. AND WAINE C. JOHNSON, M.D. Urbach (1), in 1932, described necrobiosislipoidica diabeticorum and 17 biopsy specimens lipoidica diabeticorum as a complication offrom 14 patients with necrobiosis lipoidica were studied. Also, biopsy material from 25 consecutive diabetes nwllitus. Oppenheim (2)later inpatients with granuloma annulare was studied. 1932, reported another patient with diabetes Clinical data was obtained from clinical records, who had the same cutaneous involvement andquestionaires, personal interviews, and examina- used the term dermatitis atrophicans lipoidestion of the patient. All specimens were fixed in 10% neutral buffered formalin and most of the diabetica. Since then necrobiosis lipoidica dia-tissue was processed for routine paraffin-blocked beticorum has been generally the term used forsections. Multiple sections stained with hema- this skin disease associated with diabetes. Manytoxylin and eosin were examined from all patients. examples of necrobiosis lipoidica ot associatedSections of representative specimens from 5 pa- with diabetes have been reported (3—5) andtients each with necrobiosis lipoidica diaheticorum, necrobiosis lipoidica and granuloma annulare were the incidence of associated diabetes is variouslyprepared by the following methods: periodic reported from 25 to 87% (6—10). Necrobiosisacid-Schiff (PAS) reaction, with and without lipoidica diabeticorum occasionally precedesdiastase digestion; colloidal iron reaction (12), clinical evidence of diabetes (6, 9). The estab-with and without bovine testicular hyaluronidase lishment of definite histopathologic criteria fordigestion for 1 hour at 37° C.; Snook's reticulum stain; Movat's pentachrome I stain (13); Gomori's the differentiation of necrobiosis lipoidica in thealdehyde-fuchsin technic; and the alcian blue diabetic or potential diabetic patient from nec-method. The pH of the working solutions, and robiosis lipoidica in the non-diabetic patientthe methods used in the aldehyde-fuchsin and would be a valuable adjunct in the early diag-alcian blue techniques and detailed interpretation nosis and management of these patients. of the results were similar to those described by Johnson and Helwig (14) and Johnson, Graham Rollins and Winkelman (11) in '1960 re-and Helwig (15). Frozen, sections were prepared ported identical clinical appearance but differ-from the forinalin-fixed tissue from 1 patient with ent histopathologic patterns in necrobiosisnecrobiosis lipoidica diabeticorurn, 3 patients with lipoidica in diabetic and non-diabetic patients. necrobiosis lipoidica and 1 patient with granulorna annulare, and these were stained with the oil red The present paper is a report of histopatho-O stain for fat. With the exceptions given, the logical and histochemical observations of biopsyprocedures were carried out as outlined in the material of necrobiosis lipoidica from patients"Manual of Histologic and Special Staining Tech- with and without diabetes. nics" (16). Clinical Data MATERIALS AND METHODS The clinical appearance of the lesions india- betic and nondiabetic patients was similar and For purposes of this report the term necrobio-clinical differentiation could not be made (Fig. 1 sis lipoidica diabeticorum is used to designate theand 2). The lesions appeared as brownish-red cutaneous disease occurring in patients with dia-patches with a central yellowish hue and varied betes, and necrobiosis lipoidica refers to the erup-in size and shape. The involved skin appeared tion in patients without diabetes. Twenty-one bi-shiny, waxy, and atrophic and the superficial blood opsy specimens from 13 patients with necrobiosisvessels were usually telangiectatic. Most of the This investigation was supported in part by re-lesions were slightly scaly and a few showed search training grant no. 2A-5289 (C2), from theulceration and scarring. About half of the patients National Institute of Arthritis and Metabolichad symmetrical involvement of the anterior as- Diseases, U. S. Public Health Service, Bethesda,pect of the legs. A few patients had involvement Maryland 20014. of only one leg, ankle or dorsal surface of the Presented by title at the Twenty-fifth Annualfoot. One patient with diabetes had a solitary le- Meeting of The Society for Investigative Derma-sion of the lateral aspect of the right upper arm and tology, Inc., San Francisco, Calif., June 21—23, 1964. Received for publication June 29, 1964. another had multiple lesions of the anterior aspects * From the Skin and Cancer Hospital of Phila-of the legs, left thigh and left hip. delphia, Department of Dermatology, Temple Necrobiosis Lipoidica Diabeticorun. The 13 University School of Medicine, Philadelphia, Pa.patients were all Caucasian. Twelve patients were 19107. female and one was male. The age of the pa- 369 370 THE JOURNAL OF INVESTIGATIVE DERMATOLOGY Fic. 1. Necrobiosis lipoidica diabeticorum in a 20 year old woman with diabetes for 6 years and skin lesions for 18 months. FIG. 2. Necrobiosis lipoidica in a 49 year old woman with skin lesions for 12 years tients at the time of the first biopsy examinationtients were Caucasian and 1 was a Negro. Thir- ranged from 18 to 75 years; the mean age wasteen patients were female and 1 was male. 44.7 years. The duration of necrobiosis lipoidicaThe age of the patients at the time of the first diabeticorum from onset to the time of the firstbiopsy examination ranged from 23 to 60 years. biopsy examination varied from 1 month to 5The mean age was 43.5 years. The duration of years; the mean duration was 21 months. Six pa-necrobiosis lipoidica prior to the first biopsy tients were known to have diabetes prior to theexamination varied from 9 months to 15 years; development of the ecrobiosis lipoidica diabeti-the mean duration was 7.9 years. All 14 patients comm. The duration of diabetes before develop-had negative laboratory tests for diabetes and ment of the skin lesions in 4 patients was knownmost patients had been tested at periodic intervals. and ranged from 1 to 14 years; the mean durationThe majority of patients had glucose tolerance was 6.1 years. Six patients developed cutaneoustests; a few had 2 hour post-prandial blood sugar lesions prior to the onset of clinical symptomsdeterminations; one patient not available for of diabetes mellitus, but laboratory evidence offurther investigation had only a urinalysis. Family diabetes was present at the time the diagnosis ofhistory from 11 patients was negative except for necrohiosis lipoidica diabeticorum was established. 1 patient whose maternal grandmother and mater- The duration of the skin lesion prior to the initialnal aunt had diabetes. diagnosis of diabetes ranged from 3 months to 11 Granuloma Annulare. Twenty-two patients were years; the mean duration was 2.8 years. The pa-Caucasian and 3 were Negro. Fifteen patients tient with skin lesions for 11 years had severalwere female and 10 were male. The age of the biopsy examinations performed before a histo-25 patients at the time of biopsy examination pathologic diagnosis of necrobiosis lipoidica dia-varied from 2 to 69 years. The mean age was beticorum was made. Nine years after onset a27.9 years. The disease had been present from borderline fasting blood sugar was reported, but1 week to 20 years prior to biopsy examination; not investigated. Two years later the patient wasthe mean duration was 2.4 years. Clinically, all studied and found to have a diabetic type ofthe patients had lesions characteristic of granu— glucose tolerance curve. Laboratory evidence ofloma annulare. The anatomical sites of the biopsy diabetes had probably been present several years.specimens in 23 patients were the lower extremi- One patient noted the onset of cutaneous lesionsties in 9; upper extremities in 10; buttocks in 2; at the time diabetes was initially detected, but theand the posterior aspect of the neck ad external diagnosis of necrobiosis lipoidica diabeticorumear in 1 each. was ot established until 4 years later. Family history revealed that 3 of 10 patients had close HISTOPATHOLOGIC OBSERVATIONS relatives with diabetes. Hematoxylin- and eosin-stained sections of Necrobiosis Lipoidica. Thirteen of the 14 pa-necrobiosis lipoidica generally showed similar NECEOBIOSIS LIPOIDICA 371 histopathologic changes in the diabetic andthe number of small blood vessels in the mid- non-diabetic patients, but certain differencescorium and deep corium was observed in 9 were observed. In lesions from both diabetic(47%) of the lesions. Groups of 3 to 6 capil- and non-diabetic patients the most prominentlaries were often observed associated with changes were in the coriurn with areas of nec-fibrosis and a perivascular infiltrate or inflam- robiosis, and a cellular infiltrate composed ofmatory cells (Fig. 5). variable numbers of lymphocytes, plasma cells, The superficial corium, mid-corium and deep histiocytes, epithelioid cells and giant cells.corium were all involved in 9 (69%) of the Capillary-endothelial proliferation with thick-13 specimens in which the entire thickness of ening of the vessel walls and narrowing of thethe coriuni was present for evaluation. Six lumina was frequently observed in the midbiopsy specimens were removed too superficially