Henoch-Schönlein purpura with joint involvement: Analysis of 71 cases

A total of 71 patients with HSP who attended the Department of Dermatology at Dongying Hospital, Shandong Province, between January 2010 and March 2012 were included in this study. The patients were evaluated retrospectively. The diagnosis of HSP was based on European League against Rheumatism (EULAR) endorsed consensus criteria for HSP classification [7]. All cases developed skin purpura, acute arthritis or acute arthralgia, accompanied with: (1) abdominal pain: diffuse colicky pain. Some patients had gastrointestinal bleeding and emesis. (2) histopathology: proliferative glomerulonephritis with predominant IgA deposit or typically leucocytoclastic vasculitis with predominant IgA deposit. (3) renal involvement: proteinuria > 30 mmol/mg of urine albumin/creatinine ratio on a spot morning sample or >0.3 g/24 h; presence of red blood cell casts or haematuria; red blood cells casts in the urinary sediment or > 5 red blood cells/high power field or ≥ 2+ on dipstick. Arthritis was defined as arthralgia with limited movement or joint swelling. Arthralgia was defined as joint pain without joint limited movement or swelling.

The inclusion criteria were as follows: (1) age < 14 years old; (2) no history of allergic or hematological disorders; (3) not diagnosed with systemic lupus erythematosus, renal or any other immunological disorders; (4) no active infections such as tuberculosis, hepatitis B and other contagious diseases. The exclusion criteria were: (1) patients with incomplete information; (2) patients unable to comply with the treatment; (3) patients with severe heart, liver, lung, kidney or other organ system diseases.

A total of 131 HSP patients were admitted to our hospital between January 2010 and March 2012. Of these, 79 cases of HSP had joint involvement and were thus considered for inclusion in this study. Among them, two patients were found to be older than 14 years old (41 and 16 respectively), one case had hypothyroidism, and four patients had incomplete data (of these, one was admitted to another hospital, and three patients were lost to follow up). Thus, a total of 71 cases were finally included in the study. The study was approved by the Ethics Committee at the Dongying Hospital.

All patients underwent laboratory tests such as antinuclear, anti-DNA, anti-ENA antibodies, routine blood and urine tests, fecal occult blood, coagulation profile, anti-streptolysin O titers, C-reactive protein, levels of C3, IgA, IgG, IgM, D-dimer, BMI value, blood pressure, liver function (ALT, AST), kidney function (SCr, BUN) and joint x-ray.

Continuous variables were expressed as mean ± standard deviation (SD) or median (interquartile range). Categorical variables were expressed as percentages. Inter-group differences on univariate analysis were evaluated by χ ² test. Multivariate logistic regression analysis was performed to determine odds ratio at 95 % confidence interval (CI). All statistical analyses were performed using SPSS 16.0. P < 0.05 was considered statistically significant.

A total of 71 cases including 40 males and 31 females were included. The ratio of male to female was 1.29:1. The average age was 8.55 ± 2.13 years and median age was eight. Sixty- one cases (86 %) had peak age for disease onset between six and 11 years, with male to female ratio of 1.22:1 (Fig. 1). As shown in Fig. 2, disease onset occurred during all the seasons, with the highest incidence being reported during autumn and winter seasons (70 %). Respiratory tract infection was the most frequent predisposing factor followed by vigorous exercise, with 66 cases (93 %) being aware of the trigger factors (Fig. 3).

Main clinical features of the 71 HSP children were summarized in Table 1. All 71 cases had typical skin manifestations such as petechiae and ecchymosis. In 64 cases (90 %), skin symptoms were the initial manifestations. Skin rash was most commonly observed in the lower limbs and foot, followed by thigh, upper limbs, abdomen, waist-hip, scrotum and face. Twenty-seven cases (38 %) had rash in more than three affected areas (Fig. 4).

Petechiae and ecchymosis were distributed symmetrically interspersed with mild raised skin bumps spots. The colour was bright crimson to begin with, and then turned dark and faded over a period of three to ten days. Three children developed wheal-like rash, necrotizing purpura, blisters or bullae. 52 (73 %) patients experienced recurrent skin symptoms. Twenty-nine cases (41 %) relapsed within 1 month, 15 cases recurred within 1 to 3 months (21 %), six cases (8 %) recurred in 3 months to half a year, and two cases (3 %) relapsed after 8 months and 14 months respectively. 22 cases (31 %) had a recurrence of skin lesions on more than three occasions.

Most children developed joint symptoms simultaneously with rashes or within 11 days of appearance of the rash. Forty-six cases (65 %) developed joint symptoms after having had the rash for 1 to 3 days. Nineteen cases (27 %) experienced the symptoms between 4 and 7 days, while four cases (6 %) experienced symptoms in the period between 8 and 11 days.

The joint manifestations included arthritis and arthralgia. Some children had sudden onset of symptoms during resting state. The arthralgia was worsened on extra weight bearing. A total of 71 (100 %) patients had arthralgia and discomfort, including 59 patients (83 %) with joint swelling, three patients (4 %) limited mobility in the absence of swelling; nine patients (13 %) had arthralgia only. Distending pain (pain accompanied by a distending sensation) was reported in 29 cases (41 %), sore pain (pain with sore and tired discomfort) in 22 cases (31 %), hot pain (hot and painful) [9] in 13 cases (18 %) and seven cases had indescribable nature of pain (10 %). Multiple joints were often involved (Fig. 5), and the pain was migratory and recurring in nature. In 46 cases (65 %), the joint involvement was unilateral, while 25 cases (35 %) had bilateral involvement. A total of 26 cases (37 %) experienced recurrence; 11 cases (15 %) recurred within 1 month, seven cases (10 %) recurred within 1 to 3 months, five cases recurred (7 %) within 3 to 6 months, three cases (4 %) relapsed in 6 to 17 months. Joint symptoms occurred simultaneously or successively with rash during relapse. Thirteen patients (18 %) relapsed on more than two occasions. However, the recurrence rate of joint involvement was relatively low as compared to that of rash. Joint symptoms usually improved within 1 to 5 days (median, 3 days); 29 cases improved within 1 to 2 days (41 %), 37 cases (52 %) in 3 to 4 days and five cases (7 %) improved in 5 days.

Forty patients (56 %) had gastrointestinal symptoms including abdominal pain, nausea, vomiting, with diarrhea or melena on few occasions. Children tended to have one or more of the above symptoms; 35 children (49 %) had abdominal pain, 21 (30 %) had vomiting, and 13 cases (18 %) had gastrointestinal bleeding. There were no cases with intussusception.

A total of 37 children (52 %) (23 males and 14 females with ratio at 1.64:1) had signs of renal involvement, mostly onset within one month of appearance of rash. In 19 cases (27 %), the renal involvement manifested within 15 days of appearance of the rash; in nine cases (13 %) onset from 16 days and 1 month, in five cases (7 %) between 1 and 3 months, in three cases (4 %) between 3 months and 1 year, and in one case (1 %) between 1 and 3 years. Renal involvement tended to manifest after skin symptoms. Proteinuria (11 %) was detected in eight cases, hematuria in six cases (8 %), concomitant hematuria and proteinuria in 13 cases (18 %), acute nephritis in 31 cases (44 %), nephrotic syndrome in 11 cases (15 %), and chronic nephritis in two cases (3 %).

Other symptoms were as follows. Leg edema was observed in 12 cases (17 %), fever in seven cases (10 %) (body temperature, 37.5 °C–38.7 °C). Five (7 %) patients suffered from testicular swelling and pain. Orchitis and testicular torsion was excluded by ultrasound examination. Three cases (4 %) had headache and dizziness. Cranial CT examination showed no abnormalities. In two cases, there were visible lung shadows which resolved after treatment.

In two cases (3 %), joint symptoms appeared 6 and 13 days, respectively, prior to the onset of the rash and were treated in orthopedics department with immobilization and anti-infectives. In five (7 %) children abdominal pain was the first symptom and treated by anisodamine and cimetidine but with little symptom relief. A few days later, blood spots and petechiae with arthralgia appeared. One case (1 %) presented as acute abdomen pain suggestive of gastrointestinal obstruction. The rash appeared when the patient was being prepared for surgery. After consultation with our department, treatment of 80 mg (2 mg/kg per day) methylprednisolone was administered once a day for 3 days and the symptoms subsided.

General treatments included bed rest, avoidance of high protein diet and strenuous exercise. Temporary fasting and nutritional support therapies were advised for patients with severe gastrointestinal symptoms. Oral anti-histamine drugs, such as loratadine or cetirizine was administered. Patients with bacterial infections were given antibiotics, and leflunomide or tripterygium was administered in patients with renal involvement.

All 71 patients were treated with corticosteroids, such as methylprednisolone at 1–2 mg/kg per day as suggested by Joel et al. [10]. The treatment lasted from 1 to 4 weeks and the dosage was gradually reduced after 3 to 5 days, depending on the alleviation of symptoms. During the corticosteroids treatment, we monitored the patients for changes in rash, joint symptoms, and gastrointestinal manifestations. The average time for rash dissipation, reduction of joint swelling and pain, and gastrointestinal symptoms recovery (symptoms of abdominal pain, nausea under control, fecal occult blood test turned negative) were 5.76 ± 1.72, 2.65 ± 1.08, and 3.25 ± 1.28 days, respectively. Clinical and laboratory parameters, including BMI, blood pressure, routine blood test (WBC, RBC, PLT), liver function (ALT, AST) and kidney function (SCr, BUN) were measured before, during, and up to 4 weeks after corticosteroid treatment. Some patients had improvement in these values during corticosteroids treatment, though not significant compared to either pre-treatment or post-treatment (P > 0.05). Also, there was no significant difference in these values between pre-treatment and post-treatment periods (P > 0.05). In contrast to previous reports [11], patients who were treated with corticosteroids had faster recovery from rashes (t = −10.00, P = 0.00) and joint swelling and pain (t = −34.62, P = 0.00) than patients who did not have corticosteroids treatment.

The patients were followed up for 3 years. Fifty-two cases (73 %) had recurrence of rash. Kidney symptoms, joint symptoms, gastrointestinal symptoms were recurred in 29 cases (41 %), 26 cases (37 %) and 14 cases (20 %), respectively. Recurrence of abdominal pain, arthralgia and other symptoms tended to last for a short time. Some children experienced recurrence after withdrawal of treatment, episodes of upper respiratory tract infection or strenuous activity. Among the 37 cases of HSP nephritis, 19 cases (27 %) of proteinuria and urine occult blood turned negative within 1 week to 1 month of incidence, of which urine abnormality was detected in 13 cases (18 %) during first the 2 weeks; in another 15 cases (21 %) urine became normal within 1 month to 1 year and a half; in three cases (4 %) microscopic hematuria, and proteinuria continued to persist in small amounts.

On univariate analysis, gastrointestinal symptoms, increased levels of CRP and D-dimer and scrotal involvement were found to be significantly associated with renal impairment in HSP children with joint manifestations (Table 2). On multivariate logistic regression analysis, scrotal involvement, gastrointestinal symptoms, raised D-dimer levels were independently associated with renal impairment in these children (Table 3).