Evaluation and therapeutic outcome of palpable purpura

Introduction

Allergic or hypersensitivity vasculitis is the most common variety of necrotizing vasculitis involving the post capillary venules.[1],[2],[3],[4] It has a multifactorial etiology with diverse clinical manifestations, involving organ systems other than the skin. Histopathologic picture is that of leukocytoclastic vasculitis which is characterized by fibrinoid necrosis of vessel wall, extravasation of RBC′s, presence of perivascular neutrophilic infiltrate with leukocytoclasis (nuclear dust).[1] This study has described the etiology, laboratory findings, associated clinical features, histology and therapeutic prognosis of palpable purpura.

Patients and Methods

Thirty-six consecutive patients who presented with palpable purpura to the Dermatology department of Kasturba Hospital, Manipal between August 1996 and November 1998 were studied. Pityriasis lichenoides et varioliformis acuta, pigmented purpuric dermatoses and patients on warfarin and heparin were excluded. A detailed history regarding duration of purpura, constitutional and systemic symptoms, infections, drug intake, food allergy,malignancy and collagen vascular disease was taken. Blood counts, C-Reactive protien, antistreptolysin titres (ASO), urea, creatinine, urinary proteins, microscopy, 24 hour urinary proteins, stool occult blood and chest X ray were done in all patients. Abdominal ultrasound and gastroscopy were done whenever indicated. Two biopsies were taken from the purpuric lesions for histopathology (H & E) and direct immunofluorescence (DIF). DIF studies were performed with fluorescein conjugated antisera to IgG, IgA, IgM, C3 and fibrinogen. All patients were followed up at one, three and six monthly intervals for a maximum period of 2 years. Microscopic hematuria was defined as> 3 RBC/hpf in adults and> 10/hpf in children.

Results

The study group consisted of 20 males (55.6%) and 16 females (44.4%) aged 3-66 years (mean, 31.3 years). Duration of purpura ranged from 2-40 days (mean, 12.4 days) with lower extremities being the commonest site of involvement (44.4%). Generalized lesions were seen in 8 patients (22.2%). In addition 7 had erythematous macules, 3 had blisters and in 4 ulceration was noted.

Clinical or laboratory evidence of systemic involvement was seen in 30 cases (83.3%) [Table - 1]. Joint pain mostly involving the knee was the commonest symptom accompanying purpura seen in 26 cases (72.2%) followed by gastrointestinal tract involvement with findings of colicky abdominal pain in 21 (58.3%). One had severe intestinal haemorrhage. None suffered from neurological symptoms. 47.2% (17 cases) showed abnormal urinalysis with microscopic haematuria and / or proteinuria, indicating renal involvement. Other features included scalp edema and scrotal swelling in one patient each.

In 26 cases (72%) a precipitating factor was found based on history and investigations [Table - 2].

Patients with a triad of joint pain, gastrointestinal (GI) symptoms and purpura were considered as Henoch - Schonlein purpura. In 50% of our HSP patients, purpura was the first manifestation of the illness. None of our patients had systemic lupus erythematosus or malignancy. A definite history of upper respiratory tract infection occurring within 3 weeks of the onset of symptoms was elicited in 12 (33.3%) cases. Nine had non-specific fever, 1 patient had hepatitis C and another had HIV confirmed by ELISA.

The most frequent laboratory abnormality was an elevated ESR found in 29/36 cases. 17 cases had urine abnormalities of which 6 had microscopic haematuria and proteinuria and 11 had only proteinuria but significant proteinuria (>1 gm/day or dipstick of > 3+) was noticed in 5 patients only [Table - 3]. A raised 24 hour urinary protein (> 180 mg/day) was noticed in 9. Most patients giving a history of preceding upper respiratory tract infection had elevated ASO titres [Table - 3].

Histological evidence of vasculitis was seen in 94.4% cases. Lesions were grouped into 3 stages as proposed by Grunwald et al.[5] 11 patients were in the early stage, 19 in the fully developed and 4 patients had late stage lesions. In add-ition to neutro-phils and mono-nuclear cells, eosinphilis were seen in the infiltrate in 7 patients.

DIF revealed deposition of immunoreactants in 35/36 cases (97.2%) with fibrinogen and C3 occurring most frequently followed by IgA, IgM and IgG [Table - 4]

IgA and C3 were usually seen as g r a n u l a r deposits. three of the 36 patients were lost for follow up, the details of the treatment and follow up are shown in [Table - 5] and [Table - 6].

Discussion

Palpable purpura is often localised to the lower extremities and constitutes the most common manifestation of allergic vasculitis.[4],[6],[7],[8] A more generalized eruption was reported to be predictive for renal involvement[9] but we found no significant correlation. An age of onset of < 20 years is an important criteria in the diagnosis of HSP,[19] 50% of our patients with HSP were over 20 years of age. Systemic involvement was noted in 30 (83.3%), other studies have reported an incidence of 40-90%.[6],[7],[8] Comparative results are shown in [Table - 1]. Musculoskeletal system was the most commonly involved in our study (72.2%) and renal involvement was noted in 47.2% in comparison to previous authors.[6],[10] Majority of our cases were HSP although collagen vascular disease has been reported earlier as a major causative factor [Table - 2].

Inspite of biopsies being performed after 24 hours majority of the lesions were in the fully developed state with a predominant neutrophilic infiltrate; 4 were in the !ate stage with a lymphocytic infiltrate probably favouring the theories of the dynamic nature of inflammatory infiltrate.[11],[12],[13] The presence of eosinophils in the infiltrate was noted in 7 patients, this has earlier been reported as a rare finding.[1] DIF was positive in 97.2% cases. The same high percentage was reported by others with fibrin and C3 occurring most frequently.[5],[7],[10] The age of lesion is reported to effect immunofluorescence finding with immune deposits being absent in older lesions.[10],[11],[14] Contradictory to this we found immune deposits in lesions older than 24 hours as all our biopsies were performed on > 24 hours to 5 day old lesions, although IgA deposits tended to be slightly weaker in some samples.

IgA deposits were present in 28/36 (77%) of which 17 were cases of HSP, 3 had drug induced allergic vasculitis,2 had rheumatoid arthritis and 6 were in the idiopathic group. 15/28 patients who had IgA deposits showed either proteinuria and or microscopic hematuria and 6 patients who had persistent proteinuria till the end of their follow up (6m-2years) had IgA deposits. one of them developed acute renal failure but recovered subsequently following dialysis.

The incidence of !gA deposits was 77.7% in our study in comparison to a 40% incidence found by Mackel and Jordan and 17% by Grunwald et al. More than 50% of our cases (17/28) showing IgA deposits were clinically diagnosed as HSP suggesting that IgA can be of diagnostic value in cases of HSP. Mackel and Jordan found IgA commonly in the drug related and idiopathic groups as well.

In 11/36 cases purpura subsided spontaneously hence no treatment was instituted, two of these were lost to follow up. Untreated patients with HSP had recurrent episodes. Eighteen were treated with dapsone alone and 7 with a combination of steroids and dapsone. Steroids were given in 6 for severe gastrointestinal symptoms and in one for glomerulonephritis. On follow up only 8 of the 24 cases on dapsone developed purpura. It is said to benefit the patients in whom the vasculitis is limited to the cutaneous capillaries as it inhibits neutrophil cytotoxicity.[15] Winter[16] reported that steroids may affect the resolution of abdominal pain in HSP although the value in preventing gastrointestinal bleeding was difficult to prove. Four of our 7 patients on steroids had no recurrence and in 2 the GI symptoms were milder. The role of corticosteroids in renal disease is controversial.[17] One patient developed acute renal failure during the first attack in spite of steroids but subsequently recovered with dialysis. In most of our cases symptoms subsided without complications. None developed nephritis or renal failure during the follow up period of 2 years although 6 had persistent proteinuria.

We found that administration of dapsone in early stages of purpura can enhance the resolution of the lesions and prevent relapse. Steroids can prove beneficial in severe gastrointestinal and renal involvement.