Abstract
Objectives
To identify risk factors for abdominal pain and predictors of complicated abdominal pain in pediatric immunoglobulin A vasculitis (IgAV).
Methods
Retrospective multicenter study including children and adolescents (≤18 years) fulfilling the European League Against Rheumatism/Pediatric Rheumatology International Trials Organization/Pediatric Rheumatology European Society (EULAR/PRINTO/PRES) criteria for IgAV and followed at four tertiary centers in Brazil. Factors associated with abdominal pain in the overall cohort and with complicated abdominal pain (defined as abdominal pain associated with intussusception, intestinal bleeding, or acute abdomen) were evaluated during the first 3 months of the disease and at the end of follow‐up. Univariate analyses, a pre‐specified multivariable log‐binomial regression model, and receiver operating characteristic (ROC) curve analysis were performed.
Results
Among the 687 pediatric patients with IgAV, 420 (61.1%) developed abdominal pain within the first 3 months and 428 (62.3%) by the end of follow‐up (median follow‐up of 54.1 months [interquartile range 50.7–59.3]). Among patients with abdominal pain, 22.2% developed complications. Prolonged purpuric/petechial rash (>29 days; RRadj 1.78, 95% confidence interval [CI] 1.09–2.91) and leukocytosis (>13,500/mm3; RRadj 2.30, 95% CI 1.41–3.73) were independently associated with complications. Leukocytosis showed good discriminatory performance for predicting acute abdomen (area under the curve 0.813, 95% CI 0.704–0.923).
Conclusion
Prolonged purpuric/petechial rash and leukocytosis were independently associated with complicated abdominal pain. Leukocytosis may help identify children at higher risk of acute abdomen.
Keywords: abdomen, acute; gastrointestinal diseases; Henoch–Schonlein purpura; intussusception; leukocytosis
What is Known
Abdominal pain is self‐limiting in nature. However, a subset of patients may develop an acute abdomen, which can lead to serious or even fatal complications.
Previously reported risk factors for intestinal manifestations in immunoglobulin A vasculitis (IgAV) include diffuse purpura, renal involvement, leukocytosis, severe or prolonged abdominal pain, and thrombocytosis.
What is New
This pediatric cohort demonstrated that prolonged purpuric/petechial rash and leukocytosis were independently associated with complicated abdominal pain in children with IgAV experiencing abdominal pain.
Leukocytosis showed good discriminatory performance in predicting acute abdomen in patients with IgAV experiencing abdominal pain.
1. INTRODUCTION
Immunoglobulin A (IgA) vasculitis (IgAV) is the most common childhood vasculitis, with an incidence of 10–20 per 100,000. The age group primarily affected ranges from 2 to 12 years, with a peak between 4 and 7 years, and a rare occurrence before 2 years. 1
The disease primarily affects the skin, joints, gastrointestinal tract, and kidneys, and diagnosis is based on the validated pediatric classification criteria endorsed by the European League Against Rheumatism/Pediatric Rheumatology International Trials Organization/Pediatric Rheumatology European Society (EULAR/PRINTO/PRES). 2
Gastrointestinal (GI) manifestations are well described and affect 50%–75% of pediatric cases, predominantly involving the small intestine. Abdominal pain is the most common manifestation and has a good prognosis in the majority of cases. However, severe cases characterized by intestinal bleeding, intussusception, and intestinal perforation, which are capable of simulating a surgical acute abdomen, require additional care. 3 , 4
Although many studies have described the GI manifestations of IgAV, large pediatric cohorts are scarce among these reports. In addition, no clinical or laboratory markers can identify patients with potentially serious complications or recurrence. Therefore, this study aimed to identify the risk factors for the development of abdominal pain in pediatric IgAV and, among affected patients, to evaluate predictors of severe GI complications and clinical outcomes.
2. METHODS
2.1. Ethics statement
This study was approved by the Research Ethics Committee of the Hospital das Clínicas, Ribeirão Preto Medical School, University of São Paulo, Brazil (CAAE: 70992423.8.2002.5440) and by the local Research Ethics Committees of the participating centers. Given the retrospective nature of the study and the use of anonymized medical records, the requirement for informed consent was waived by the respective Research Ethics Committees.
2.2. Study design and protocol
We conducted a retrospective multicenter study including children and adolescents (≤18 years old) diagnosed with IgAV and followed at four university referral centers in Brazil. Patients diagnosed with IgAV between March 13, 1981, and November 24, 2023, were eligible for inclusion. IgAV was diagnosed according to the validated EULAR/PRINTO/PRES classification criteria. 2
An online investigator meeting defined the study protocol, including demographic data, clinical and laboratory examinations, treatments, and outcome parameters. We systematically collected data from medical records and managed them using Research Electronic Data Capture (REDCap) software from December 2023 to April 2025. All patients diagnosed with IgAV were included in this study. Exclusion criteria included a diagnosis of acute hemorrhagic edema of infancy.
2.3. Data collection and definitions
A database was built by retrospectively reviewing medical records at baseline (defined as the time of diagnosis and the following 3 months) and at the patient's last follow‐up visit. A group of investigators supervised data collection at each center and resolved discrepancies by conducting multiple rounds of queries to ensure accuracy. Follow‐up duration was calculated as the interval between the date of IgAV diagnosis and the date of the last available clinical evaluation recorded in the medical chart.
Demographic variables included age at diagnosis, sex, number of days from symptom onset to diagnosis, follow‐up duration, and auxological parameters (weight, height, and body mass index). Laboratory data included hemoglobin, white blood cell (WBC) count, platelet count, serum IgA and immunoglobulin G (IgG) levels, C‐reactive protein (CRP), erythrocyte sedimentation rate (ESR), antistreptolysin O (ASO) titer, C3 complement test, serum urea and creatinine levels, and urinalysis results. Reference ranges for laboratory parameters were defined according to the standards provided by the clinical laboratory of each participating center. Laboratory tests were obtained during the clinical evaluation of abdominal pain as part of the routine investigation of GI involvement.
The treatment data included use of corticosteroids, ranitidine (used until its effective discontinuation in April 2020), proton pump inhibitors, antiproteinuric agents, immunosuppressants (methotrexate, azathioprine, cyclosporine, mycophenolate, and/or cyclophosphamide), biological agents (rituximab, adalimumab, and infliximab), intravenous immunoglobulin, plasmapheresis, dialysis, and renal transplantation.
Abdominal pain associated with IgAV was defined as acute, diffuse, and colicky. In addition, abdominal pain was classified as complicated abdominal pain when associated with intussusception, GI bleeding, or acute abdomen. Acute abdomen was defined as sudden onset of severe abdominal pain associated with clinical signs of peritoneal irritation (e.g., guarding or rebound tenderness) and/or imaging findings suggestive of a surgical abdomen requiring urgent surgical evaluation. Patients were classified according to the presence or absence of pain and whether the abdominal pain was complicated or uncomplicated.
The presence of skin purpura or petechiae was a mandatory inclusion criterion. The appearance of new lesions after complete resolution characterized recurrence, while lesions lasting 6 weeks or longer defined persistence. 5 The presence of joint swelling or pain, along with limitation of movement, defined the presence of arthritis, while isolated pain without swelling or restriction of movement characterized arthralgia. Arthritis lasting 6 weeks or longer was defined as persistent, while recurrent arthritis referred to new episodes occurring after complete resolution of the initial episode. 5
The renal assessment included serum urea and creatinine, as well as screening for hematuria, proteinuria, and urinary casts. 2 Systolic and/or diastolic values ≥95th percentile for sex, age, and height characterized arterial hypertension. 6 Additional evaluations for the occurrence of nephrotic syndrome, nephritic syndrome, and renal failure were also performed.
Other clinical aspects, including orchitis and subcutaneous edema, were also investigated systematically.
2.4. Statistical analysis
Descriptive statistics included medians and their respective minimum and maximum values for continuous variables, as well as relative and absolute frequencies for categorical variables. Comparisons between groups were made using Fisher's exact test for independent categorical variables and the Wilcoxon–Mann–Whitney test for continuous variables, with a significance level of 5%. A pre‐specified multivariable log‐binomial regression model was constructed including clinically relevant covariates selected a priori based on biological plausibility and prior literature: age at diagnosis, sex, renal involvement, duration of purpuric/petechial rash, leukocytosis, and thrombocytosis.
We used a conditional inference tree as an exploratory method to identify the potential cutoffs of continuous variables.
Receiver operating characteristic (ROC) curve analysis was performed to evaluate the discriminatory ability of leukocytosis to identify acute abdomen. Given the limited number of acute abdomen cases, this analysis was considered exploratory.
Due to the retrospective design of the study, not all clinical and laboratory variables were available for all patients; therefore, the analyses were performed using available‐case (complete‐case) data without imputation.
The analyses were conducted using SAS (Version 9.4) and R (Version 4.5.0) software.
3. RESULTS
A total of 687 patients fulfilled the EULAR/PRINTO/PRES criteria for IgAV. Of these, 420 (61.1%) developed abdominal pain within the first 3 months after diagnosis, increasing to 428 patients (62.3%) by the end of follow‐up.
Of the patients with abdominal pain, 95 (22.2%) developed complicated abdominal pain. Notably, GI manifestations occurred predominantly early in the disease course: abdominal pain developed at diagnosis or within the first 3 months in 420 of the 428 affected patients (98.1%), and 93 of the 95 cases of complicated abdominal pain (97.9%) were identified during the same period.
The proportion of missing values for key clinical and laboratory variables is summarized in Table S1. To explore the potential impact of incomplete immunologic testing, we compared patients with and without available serum IgA measurements among those presenting with abdominal pain. The frequency of complicated abdominal pain did not differ between groups (25.5% vs. 22.6%, p = 0.58), although patients with available IgA measurements were older at diagnosis (Table S2).
Table 1 compares the clinical and laboratory data of children with IgAV who did and did not present with abdominal pain at diagnosis. Of the total cohort, 51.3% of patients with abdominal pain were male, with a median age at diagnosis of 5.9 years, a median interval of 1 week between symptom onset and diagnosis, and a median follow‐up of 54.1 months (interquartile range 50.7–59.3). A history of tonsillitis was more frequent in the group without abdominal pain (58.4%), and no other relevant triggers were identified. Clinically, children with abdominal pain exhibited diffuse purpura (48.3% vs. 29.5%, p < 0.01) and petechiae (48.2% vs. 39.7%, p = 0.03), more frequently and had a longer duration of purpura (median 15 vs. 10 days, p < 0.01). In this group, most patients also presented with arthralgia and/or arthritis, although this manifestation was less frequent among those with abdominal pain (78.2% vs. 86.9%, p < 0.01). Renal involvement occurred in one‐third of patients with abdominal pain (33.6%) compared with 19.5% of those without abdominal pain (p < 0.01). Laboratory data showed higher median IgA levels (214 mg/dL), WBC counts (10,000/mm3), and hemoglobin levels (12.5 g/dL) in the abdominal pain group, whereas no relevant differences were observed for the remaining laboratory parameters.
Table 1.
Demographic data and clinical, laboratory characteristics in 687 IgA vasculitis patients with and without abdominal pain at diagnosis.
| Variables at diagnosis | With abdominal pain n = 420a (61.1) | Without abdominal pain n = 267a (38.9) | p |
|---|---|---|---|
| Demographic data; median (min–max) | |||
| Age at diagnosis, years | 5.9 (0.1–16) | 5.7 (0.2–17.4) | 0.50 |
| Male; n (%) | 215/419 (51.3) | 125/267 (46.8) | 0.27 |
| Days between symptom onset and diagnosis | 7 (0–731) | 7 (0–374) | 0.47 |
| Body mass index, kg/m2 | 16.3 (7.1–44) | 16.5 (12.6–31) | 0.47 |
| Triggers and penicillin prophylaxis; n (%) | |||
| Triggers | 158/419 (37.7) | 100/267 (37.4) | 1.00 |
| Tonsillitis | 56/128 (43.7) | 52/89 (58.4) | 0.04 |
| Penicillin prophylaxis use | 22/414 (5.3) | 18/266 (6.7) | 0.50 |
| Purpura and/or petechiae | |||
| Diffuse purpura (not only lower limbs); n (%) | 200/414 (48.3) | 78/264 (29.5) | <0.01 |
| Persistent; n (%) | 42/413 (10.2) | 16/262 (6.1) | 0.07 |
| Recurrent, n (%) | 69/405 (17.3) | 30/255 (11.8) | 0.07 |
| Duration; days (min–max) | 15 (1–90) | 10 (1–120) | <0.01 |
| Petechiae; n (%) | 202/419 (48.2) | 106/267 (39.7) | 0.03 |
| Arthritis and/or arthralgia | 327/418 (78.2) | 232/267 (86.9) | <0.01 |
| Persistent; n (%) | 7/318 (2.2) | 4/230 (1.7) | 0.76 |
| Recurrent; n (%) | 18/321 (5.6) | 8/230 (3.5) | 0.31 |
| Duration, days; median (min–max) | 7 (1–100) | 7 (1–113) | 0.10 |
| Subcutaneous edema; n (%) | 135/419 (32.2) | 83/267 (30.7) | 0.80 |
| Renal involvement; n (%) | 141/419 (33.6) | 52/267 (19.5) | <0.01 |
| Arterial hypertension | 27/331 (8.1) | 10/191 (5.2) | 0.28 |
| Nephrotic syndrome | 10/66 (15.5) | 5/33 (15.5) | 1.00 |
| Acute kidney injury | 25/137 (18.2) | 6/52 (11.5) | 0.38 |
| Leukocyturia | 61/135 (45.2) | 15/51 (29.4) | 0.06 |
| Urinary casts | 7/134 (5.2) | 3/50 (6) | 1.00 |
| Hematuria | 113/138 (81.9) | 44/52 (84.6) | 0.83 |
| Proteinuria | 68/138 (49.3) | 33/51 (67.7) | 0.07 |
| Orchitis; n (%) | 44/235 (18.7) | 14/139 (10) | 0.02 |
| Laboratory parameters; median (min–max) | |||
| Hemoglobin (g/dL) | 12.5 (6.7–17.7) | 12.3 (7.6–15.2) | 0.04 |
| WBC (cells/mm3) | 10000 (392–91,000) | 9570 (336–26,500) | 0.03 |
| Platelet count (×103/mm3) | 379 (153–842) | 369 (150–1960) | 0.22 |
| ESR (mm/first hour) | 22 (1–349) | 20 (2–98) | 0.81 |
| CRP (mg/dL) | 2.4 (0–333) | 1.4 (0–241) | 0.12 |
| Creatinine (mg/dL) | 0.43 (0.05–0.9) | 0.5 (0.16–1.3) | 0.01 |
| Immunologic markers; median (min–max) | |||
| IgA (mg/dL) | 214 (59–511) | 163 (40–1002) | <0.01 |
| IgG (mg/dL) | 1065 (459–3529) | 1140 (120–2580) | 0.38 |
| ASO (IU/mL) | 232 (0–15,000) | 263 (19–3823) | 0.99 |
| Reduced C3, n (%) | 3/125 (2.4) | 2/78 (2.5) | 1.00 |
Note: Bold values indicate statistically significant associations (p < 0.05).
Abbreviations: ASO, antistreptolysin O; CRP, C‐reactive protein; ESR, erythrocyte sedimentation rate; IgA, immunoglobulin A; IgG, immunoglobulin G; IU, International Unit; n, number; WBC, white blood cells.
Values are the number of patients with positive findings/patients with available information (%), unless indicated otherwise.
Among the 428 patients who experienced abdominal pain, 22.2% developed complicated abdominal pain (Table 2). Within this group, GI bleeding occurred in 85 patients, acute abdomen in 16, and intussusception in 2, with partial overlap between these manifestations.
Table 2.
Demographic data and clinical, laboratory characteristics in 428 patients with IgA vasculitis and abdominal pain, uncomplicated and complicated at follow‐up.
| Variables | Complicatedb n = 95a (22.2) | Uncomplicated n = 333a (77.8) | p |
|---|---|---|---|
| Demographic data; median (min–max) | |||
| Age ≤ 7 years, n (%) | 62 (65.2) | 216 (64.9) | 1.00 |
| Age at diagnosis, years | 5.9 (1.7–16) | 5.9 (0.1–15.7) | 0.99 |
| Male, n (%) | 49 (51.6) | 170 (51) | 1.00 |
| Follow‐up, months | 60 (0–239) | 32 (0–405) | <0.01 |
| Body mass index, kg/m2 | 19.9 (13.9–44.1) | 17.7 (11.5–36.9) | <0.01 |
| Obesity/overweight, n (%) | 36/84 (42.8) | 82/263 (31.2) | 0.02 |
| Gastrointestinal involvement | |||
| Recurrent abdominal pain, n (%) | 30/95 (31.6) | 56/327 (17.1) | <0.01 |
| Persistent abdominal pain, n (%) | 5/94 (5.3) | 7/327 (2.1) | 0.15 |
| Abdominal pain duration, days (min–max) | 12 (1–129) | 5 (1–90) | <0.01 |
| Purpura and/or petechiae, n (%) | |||
| Diffuse purpura (not only lower limbs) | 51/95 (53.7) | 156/329 (47.4) | 0.29 |
| Persistent and/or recurrent | 40/95 (42.1) | 101/329 (30.6) | 0.04 |
| Duration, days (min–max) | 22 (3–423) | 15 (1–142) | <0.01 |
| Petechiae | 44/95 (46.3) | 163/332 (49.1) | 0.64 |
| Arthritis and/or arthralgia, n (%) | 76/95 (80) | 260/332 (78.3) | 0.78 |
| Persistent or recurrent | 13/76 (17.1) | 27/256 (10.5) | 0.16 |
| Subcutaneous edema, n (%) | 26/95 (27.3) | 111/333 (33.3) | 0.32 |
| Renal involvement, n (%) | 50/95 (52.6) | 133/333 (39.9) | 0.03 |
| Persistent and/or recurrent | 33/48 (68.7) | 76/132 (57.7) | 0.22 |
| Arterial hypertension | 12/85 (14.1) | 21/276 (7.6) | 0.08 |
| Nephrotic syndrome | 3/29 (10.3) | 9/74 (12.2) | 1.00 |
| Acute kidney injury | 10/48 (20.8) | 26/132 (19.6) | 0.83 |
| Chronic kidney disease | 2/48 (4.6) | 1/131 (0.8) | 0.17 |
| Orchitis, n (%) | 15/58 (25.9) | 29/185 (15.7) | 0.11 |
| Laboratory parameters, median (min–max) | |||
| Hemoglobin (g/dL) | 12.3 (9.7–17.1) | 12.5 (6.7–17.7) | 0.47 |
| WBC (cells/mm3) | 12,200 (4290–91,000) | 9600 (1600–28,000) | <0.01 |
| Platelet count (×103/mm3) | 413 (158–842) | 370 (153–1960) | 0.03 |
| ESR (mm/first hour) | 22 (2–56) | 21 (1–349) | 0.84 |
| CRP (mg/dL) | 2.3 (0–98) | 2.7 (0–333) | 0.49 |
| Creatinine (mg/dL) | 0.41 (0.2–0.9) | 0.44 (0.05–0.9) | 0.29 |
| Immunologic markers; median (min–max) | |||
| IgA (mg/dL) | 227 (74–511) | 190.3 (59–488) | 0.25 |
| IgG (mg/dL) | 1065 (599–1561) | 1074 (459–3529) | 0.59 |
| Reduced C3, n (%) | 1/29 (3.4) | 2/95 (2.1) | 0.55 |
| Increased ASO (>200 IU/mL), n (%) | 14/30 (46.6) | 44/82 (53.6) | 0.53 |
| Treatment, n (%) | |||
| Prednisone | 78/95 (82) | 183/333 (54.9) | <0.01 |
| Intravenous methylprednisolone | 29/95 (30.5) | 46/327 (14) | <0.01 |
| Ranitidine | 57/95 (60) | 110/331 (33.2) | <0.01 |
| Proton pump inhibitor use | 12/95 (12.6) | 34/330 (10.3) | 0.57 |
| ACE inhibitors | 18/95 (18.9) | 35/332 (10.5) | 0.03 |
| Immunosuppressive agents | 7/95 (7.3) | 18/322 (5.6) | 0.46 |
| Intravenous immunoglobulin | 2/95 (2.1) | 1/329 (0.3) | 0.13 |
Note: Bold values indicate statistically significant associations (p < 0.05).
Abbreviations: ACE, angiotensin‐converting enzyme; ASO, antistreptolysin O; CRP, C‐reactive protein; ESR, erythrocyte sedimentation rate; IgA, immunoglobulin A; IgG, immunoglobulin G; IU, International Unit; n, number; WBC, white blood cells.
Values are the number of patients with positive findings/patients with available information (%), unless indicated otherwise.
Defined as intussusception, intestinal bleeding, or acute abdomen.
Patients with complicated abdominal pain were followed up for a longer median duration (60 vs. 32 months, p < 0.01) and had higher median body mass index (19.9 vs. 17.7 kg/m2, p < 0.01). Overweight or obesity was more frequent in the complicated group (42.8 vs. 31.2%, p = 0.02). There were no differences in sex or age between complicated and uncomplicated groups.
Recurrent abdominal pain (31.6% vs. 17.1%, p < 0.01), persistent and/or recurrent purpura or petechiae (42.1% vs. 30.6%, p = 0.04), and renal involvement (52.6% vs. 39.9%, p = 0.03) occurred more frequently in the complicated abdominal pain group. In addition, these patients had a longer duration of both abdominal pain (median 12 vs. 5 days, p < 0.01) and purpuric/petechial rash (median 22 vs. 15 days, p < 0.01). Diffuse purpura, extending beyond the lower limbs, was more common in the complicated abdominal pain group, although the difference was not statistically significant. Other clinical manifestations, including arthritis or arthralgia, subcutaneous edema, orchitis, arterial hypertension, nephrotic syndrome, and renal injury, did not differ between groups. In the laboratory assessment, the median WBC count (12,200 vs. 9600/mm3, p < 0.01) and platelet count (413,000 vs. 370,000/mm3, p = 0.03) were higher in patients with complicated abdominal pain, whereas no differences were observed for other inflammatory or biochemical markers. Regarding treatment, most patients with complicated abdominal pain received prednisone (82% vs. 54.9%, p < 0.01), methylprednisolone pulses (30.5% vs. 14%, p < 0.01), ranitidine (60% vs. 33.2%, p < 0.01), and angiotensin‐converting enzyme inhibitors (18.9% vs. 10.5%, p = 0.03).
Table 3 presents the results of multivariable analyses. After adjustment for predefined clinical and laboratory covariates, leukocytosis (>13,500/mm3) (adjusted relative risk [RRadj] 2.30; 95% confidence interval [CI] 1.41–3.76) and purpuric/petechial rash duration >29 days (RRadj 1.78; 95% CI 1.09–2.91) remained independently associated with complicated abdominal pain.
Table 3.
Multivariate analysis of predictors of complicated abdominal pain during follow‐up in pediatric IgA vasculitis cohort.
| Parameter | Abdominal pain complicated | |||
|---|---|---|---|---|
| RR | 95% CI | Adjusted RR | 95% CI | |
| Age at diagnosis ≤ 7 years | 0.97 | 0.67–1.42 | 0.78 | 0.46–1.33 |
| Sex | 1.01 | 0.71–1.45 | 1.12 | 0.70–1.81 |
| Purpura/petechial duration > 29 days | 1.98 | 1.34–2.93 | 1.78 | 1.09–2.91 |
| Renal involvement | 1.48 | 1.04–2.12 | 1.57 | 0.96–2.57 |
| WBC count a > 13,500/mm3 | 2.33 | 1.63–3.34 | 2.30 | 1.41–3.76 |
| Plateletsa > 613 × 103/mm3 | 2.62 | 1.72–3.98 | 1.67 | 0.80–3.50 |
Note: Multivariable model included age, sex, renal involvement, purpura/petechial rash duration, leukocytosis, and thrombocytosis. Bold values indicate statistically significant associations (p < 0.05).
Abbreviations: CI, confidence interval; IgA, immunoglobulin A; RR, relative risk; WBC, white blood cell.
Cut‐offs were identified using conditional inference tree analysis.
We constructed an ROC curve to evaluate the predictive performance of leukocytosis, as shown in Figure 1. For the identification of abdominal pain at diagnosis, the area under the ROC curve (AUC) was 0.653 (95% CI 0.586–0.720), indicating limited predictive ability. In turn, leukocytosis demonstrated good accuracy for identifying acute abdomen among patients with abdominal pain, with an AUC of 0.813 (95% CI 0.704–0.923). The optimal cutoff value was 14,150/mm3, yielding a sensitivity of 75% and a specificity of 83%. The corresponding positive predictive value (PPV) was 0.16 (95% CI 0.09–0.26), whereas the negative predictive value (NPV) was 0.98 (95% CI 0.97–0.99).
Figure 1.
ROC curves evaluating leukocytosis as a predictor of abdominal pain among patients with IgA vasculitis (A) and as a predictor of acute abdomen among patients with IgA vasculitis and abdominal pain (B). The AUC for abdominal pain was 0.653 (95% CI 0.586–0.720), and the AUC for acute abdomen was 0.813 (95% CI 0.704–0.923). AUC, area under the curve; CI, confidence interval; IgA, immunoglobulin A; ROC, receiver operating characteristic.
4. DISCUSSION
We observed abdominal pain in 61.1% of our patients with IgAV, which is in line with previously reported frequencies of 50%–75%, 4 , 7 including adult cohorts. 3 When considering abdominal pain with complications, we observed an incidence of 22.2%, which is comparable to 18% reported in an Italian pediatric cohort. 8 In a Chinese cohort, 50% of patients hospitalized with IgAV exhibited GI involvement. Of these, 48.9% had severe intestinal involvement. 1 Thus, the frequency of complicated intestinal manifestations in IgAV can vary widely from 13.5% to 63.1%. 8 , 9 , 10 This variability in the incidence of complicated GI manifestations may reflect the lack of standardized definitions. Some studies consider only surgical complications as indicators of severity, whereas others also include involvement of additional gastrointestinal organs.
Although intestinal bleeding, intussusception, and acute abdomen may appear clinically heterogeneous, these events likely represent a spectrum of intestinal small‐vessel vasculitis. The inflammatory process affecting the intestinal microvasculature leads to submucosal edema, hemorrhage, and ischemia, which may manifest clinically as abdominal pain, GI bleeding, intussusception, or, in severe cases, acute abdomen. 7 , 11 , 12 Pathological findings such as leukocytoclastic vasculitis, lamina propria hemorrhage, and edema support this shared pathophysiological mechanism. 13 , 14 Furthermore, the partial overlap among these manifestations observed in our cohort reinforces the notion that they represent progressive or coexisting complications of a single underlying vasculitic process.
Identifying clinical or laboratory markers that help predict pediatric GI complications is essential for improving patient management and longitudinal care. Previous studies have shown that a prolonged and diffuse purpuric rash is associated with GI symptoms in both adults 15 and children, 1 , 16 and these findings are consistent with our present data. Similarly, joint symptoms and renal involvement have been linked to GI involvement, as shown here. Adults with joint manifestations were twice as likely to develop GI symptoms, while exhibiting elevated interleukin‐1β (IL‐1β) levels, suggesting a shared inflammatory pathway between articular and GI involvement. 17 The pediatric evidence remains heterogeneous, with some reporting lower incidence of arthralgia or arthritis among patients with GI involvement, 18 while others described an opposite scenario. 8 The link between renal and GI involvement has also been described. Children whose disease begins with GI symptoms show a markedly increased risk of renal impairment, with a 2.45‐fold increase in the first month and up to a six‐fold increase during follow‐up. 15
Progression to severe GI complications represents a critical determinant of outcome in IgAV. Prolonged or severe skin rash is predominantly associated with complicated abdominal pain. 8 , 19 , 20 In the current study, complicated abdominal pain occurred approximately twice as often in patients whose skin rash persisted for more than 29 days.
Prolonged abdominal pain and renal involvement have been associated with complicated abdominal pain in IgAV, which is consistent with our present findings. Intestinal bleeding studies have identified prolonged abdominal pain as an important risk factor, 21 , 22 and recent data support an association between GI and renal involvement, although the severity of GI manifestations does not parallel renal disease activity. 23 Conversely, a previous meta‐analysis suggested an increased risk of renal complications in those with severe GI involvement. 24 Differences in population and geographic characteristics may account for this discrepancy. For example, studies from China report GI complications in nearly 50% of patients, 1 whereas lower frequencies have been observed in other regions. 8
We also observed a higher frequency of obesity or overweight among patients with complicated abdominal pain in the unadjusted analysis. Obesity has been increasingly recognized as a pro‐inflammatory condition characterized by elevated levels of inflammatory mediators such as CRP 25 and tumor necrosis factor‐α, 26 which may amplify systemic inflammation in rheumatic diseases. 27 , 28 However, the association between adiposity and complicated abdominal pain in IgAV has not been consistently demonstrated. Therefore, this observation should be interpreted with caution, and further studies are needed to clarify whether excess adiposity contributes to disease severity.
Among the GI complications of IgAV, the most concerning is acute abdomen, given the potential for unnecessary laparotomy. A reliable laboratory marker to anticipate this outcome would therefore be clinically valuable. In contrast to immunologic markers such as serum IgA, which have shown inconsistent associations with GI involvement in previous studies, 15 , 22 , 29 inflammatory hematologic parameters appeared to be more informative in our cohort. Leukocytosis and thrombocytosis were associated with a higher risk of complicated abdominal pain in univariate analyses, and leukocytosis remained independently associated with complications after adjustment. These associations likely reflect the systemic inflammatory response and endothelial activation characteristic of IgAV. 13 , 30 Cytokines released from the injured endothelium, such as tumor necrosis factor alpha (TNF‐α) and interleukin 8 (IL‐8), amplify inflammation and promote neutrophil recruitment. 31 , 32 Neutrophil infiltration of the intestinal mucosa contributes to edema, bleeding, intussusception, and, in severe cases, the development of an acute abdomen. 4 Thrombocytosis, driven by interleukin 6 (IL‐6)–mediated platelet production, may further increase the risk of microthrombi and intestinal ischemia, potentiating the risk of complications such as acute abdomen. 18
ROC analysis demonstrated that leukocytosis had limited ability to discriminate abdominal pain overall but showed better discriminatory performance for identifying acute abdomen among patients with abdominal pain. The exploratory cutoff value of 14,150/mm3 yielded a sensitivity of 75% and a specificity of 83%. However, this analysis should be interpreted cautiously because the number of acute abdomen events in our cohort was relatively small. The PPV at this threshold was modest (0.16), whereas the NPV was high (0.98), reflecting the low prevalence of acute abdomen in the cohort. These findings suggest that leukocytosis may be more useful for identifying patients at low risk of acute abdomen rather than confirming its presence. In this context, leukocytosis may contribute to clinical risk stratification and help clinicians identify children with IgAV who require closer monitoring during episodes of abdominal pain, particularly in the early phase of the disease.
Regarding the therapeutic aspects, in our cohort, patients with complicated abdominal pain more frequently received corticosteroids and intravenous methylprednisolone pulses. Currently, there are no formal guidelines recommending the use of corticosteroids in patients with IgAV and GI manifestations. However, observational studies and expert consensus statements recommend their use in cases of severe manifestations, 8 , 33 and this difference likely reflects greater disease severity at presentation rather than a direct treatment effect, as the clinical manifestations guided therapeutic decisions.
Our present study has notable strengths, including a large multicenter cohort of 687 pediatric patients with IgAV, which provides broad clinical and demographic representation across specialized healthcare settings.
Nevertheless, several limitations should be acknowledged. The retrospective design resulted in incomplete availability of some clinical and laboratory data. Missing data may have introduced some degree of selection bias. However, a supplementary analysis comparing patients with and without available serum IgA measurements showed no difference in the frequency of complicated abdominal pain, suggesting that missing IgA data were unlikely to substantially bias the main findings. The laboratory measurements obtained during the clinical evaluation of abdominal pain limit the ability to establish the temporal relationship between laboratory abnormalities and the subsequent development of complications. In addition, the cutoff values identified for leukocytosis and thrombocytosis were derived using conditional inference tree analysis within the same dataset and should therefore be considered exploratory. External validation in independent cohorts is required before these thresholds can be applied in clinical practice. Finally, the inclusion of tertiary referral centers may limit the generalizability of our findings to other healthcare settings. Despite these limitations, our findings provide clinically relevant insights into factors associated with severe GI involvement in pediatric IgAV and may help inform risk stratification in children presenting with abdominal pain.
5. CONCLUSION
Our present findings involving a large Latin‐American cohort concluded that prolonged purpuric/petechial rash and leukocytosis were associated with an almost two‐fold increased risk of complicated abdominal pain in pediatric IgAV. Leukocytosis demonstrated discriminatory performance for identifying patients with abdominal pain who were at a higher risk of developing acute abdomen, supporting its use as a valuable marker in clinical risk stratification. Future prospective multicenter studies with standardized definitions of GI complications and longitudinal data collection are needed to validate these findings and further clarify clinically useful predictors of severe GI involvement.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
Supporting information
Supplementary Table S1. Availability of key clinical and laboratory variables (N = 687).
Supplementary Table S2. Comparison between patients with and without available serum IgA measurements among those with abdominal pain.
ACKNOWLEDGMENTS
Luciana Martins de Carvalho received financial support from the Fundação de Apoio ao Ensino, Pesquisa e Assistência do Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto da Universidade de São Paulo (FAEPA‐HCFMRP‐USP). The funding source had no role in study design, data collection, analysis, interpretation of the data, or preparation of the manuscript.
REFERENCES
- 1. Li Y, Zhang X, Liu H, et al. Severe gastrointestinal involvement in pediatric IgA vasculitis: a retrospective single‐center cohort study in China. Front Pediatr. 2023;11:1194214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Ozen S, Pistorio A, Iusan SM, et al. EULAR/PRINTO/PRES criteria for Henoch–Schönlein purpura, childhood polyarteritis nodosa, childhood Wegener granulomatosis and childhood Takayasu arteritis: Ankara 2008. Part II: Final classification criteria. Ann Rheum Dis. 2010;69(5):798‐806. [DOI] [PubMed] [Google Scholar]
- 3. Audemard‐Verger A, Pillebout E, Amoura Z, et al. Gastrointestinal involvement in adult IgA vasculitis (Henoch–Schönlein purpura): updated picture from a French multicentre and retrospective series of 260 cases. Rheumatology. 2020;59(10):3050‐3057. [DOI] [PubMed] [Google Scholar]
- 4. Mu J. Clinical characteristics and risk factors of IgA vasculitis with intussusception and intestinal perforation. J Paediatr Child Health. 2024;60(1):5‐11. [DOI] [PubMed] [Google Scholar]
- 5. Roman C, Dima B, Muyshont L, Schurmans T, Gilliaux O. Indications and efficiency of dapsone in IgA vasculitis (Henoch–Schonlein purpura): case series and a review of the literature. Eur J Pediatr. 2019;178(8):1275‐1281. [DOI] [PubMed] [Google Scholar]
- 6. Flynn JT, Kaelber DC, Baker‐Smith CM, et al. Clinical practice guideline for screening and management of high blood pressure in children and adolescents. Pediatrics. 2017;140(3):e20171904. [DOI] [PubMed] [Google Scholar]
- 7. Ebert EC. Gastrointestinal manifestations of Henoch–Schonlein purpura. Dig Dis Sci. 2008;53(8):2011‐2019. [DOI] [PubMed] [Google Scholar]
- 8. Rubino C, Monacelli C, Marrani E, et al. Gastrointestinal involvement in IgA vasculitis: a single‐center 11‐year study on a cohort of 118 children. Clin Rheumatol. 2021;40(12):5041‐5046. [DOI] [PubMed] [Google Scholar]
- 9. Liao CH, Tsai M, Yang YH, Chiang BL, Wang LC. Onset age is a risk factor for refractory pediatric IgA vasculitis: a retrospective cohort study. Pediatr Rheumatol. 2020;18(1):86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Ekinci RMK, Balci S, Melek E, et al. Clinical manifestations and outcomes of 420 children with Henoch Schönlein purpura from a single referral center from Turkey: a three‐year experience. Modern Rheumatol. 2020;30(6):1039‐1046. [DOI] [PubMed] [Google Scholar]
- 11. Du L, Wang P, Liu C, Li S, Yue S, Yang Y. Multisystemic manifestations of IgA vasculitis. Clin Rheumatol. 2021;40(1):43‐52. [DOI] [PubMed] [Google Scholar]
- 12. Liu C, Du LN, Zhao Q, et al. Immunoglobulin A vasculitis with intussusception in children. Am Surg. 2024;90(6):1298‐1308. [DOI] [PubMed] [Google Scholar]
- 13. Louie CY, Gomez AJ, Sibley RK, Bass D, Longacre TA. Histologic features of gastrointestinal tract biopsies in IgA vasculitis (Henoch–Schönlein purpura). Am J Surg Pathol. 2018;42(4):529‐533. [DOI] [PubMed] [Google Scholar]
- 14. Song Y, Huang X, Yu G, et al. Pathogenesis of IgA vasculitis: an up‐to‐date review. Front Immunol. 2021;12:771619. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Hočevar A, Tomšič M, Jurčić V, Perdan Pirkmajer K, Rotar Ž. Predicting gastrointestinal and renal involvement in adult IgA vasculitis. Arthritis Res Ther. 2019;21(1):302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Rigante D, Guerriero C, Silvaroli S, et al. Predictors of gastrointestinal involvement in children with IgA vasculitis: results from a single‐center cohort observational study. Children. 2024;11(2):215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Delapierre A, Terrier B, Pillebout E, et al. Clinical phenotype and cytokine profile of adult IgA vasculitis with joint involvement. Clin Rheumatol. 2022;41(5):1483‐1491. [DOI] [PubMed] [Google Scholar]
- 18. Cao D, Liu X. Analysis of clinical features and high‐risk factors of gastrointestinal involvement in children with IgA vasculitis. Clin Pediatr (Phila). 2026;65:e99228251389785. [DOI] [PubMed] [Google Scholar]
- 19. Sestan M, Kifer N, Frkovic M, et al. Gastrointestinal involvement and its association with the risk for nephritis in IgA vasculitis. Ther Adv Musculoskelet Dis. 2021;13:1759720X211024828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Choong C, Beasley S. Intra‐abdominal manifestations of Henoch–Schönlein purpura. J Paediatr Child Health. 1998;34(5):405‐409. [DOI] [PubMed] [Google Scholar]
- 21. Yang X. Analysis of the influencing factors of abdominal Henoch–Schonlein purpura in children with gastrointestinal bleeding and the clinical value of PLR. Am J Transl Res. 2024;16(8):3867‐3874. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Su D, Yang M, Wang X, Li G, Hong S. Risk factors for gastrointestinal bleeding in children with Henoch–Schönlein purpura. Front Pediatr. 2025;13:1587535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Sağlam MK, Yıldırım S, Ergüven M, Sungur MA. Gastrointestinal features of pediatric iga vasculitis and their association with renal complications: an observational study. Eur J Pediatr. 2025;184(5):320. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Chan H, Tang YL, Lv XH, et al. Risk factors associated with renal involvement in childhood Henoch–Schonlein purpura: a meta‐analysis. PLoS One. 2016;11(11):e0167346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Cook DG, Mendall MA, Whincup PH, et al. C‐reactive protein concentration in children: relationship to adiposity and other cardiovascular risk factors. Atherosclerosis. 2000;149(1):139‐150. [DOI] [PubMed] [Google Scholar]
- 26. Caballero AE, Bousquet‐Santos K, Robles‐Osorio L, et al. Overweight Latino children and adolescents have marked endothelial dysfunction and subclinical vascular inflammation in association with excess body fat and insulin resistance. Diabetes Care. 2008;31(3):576‐582. [DOI] [PubMed] [Google Scholar]
- 27. Held M, Sestan M, Jelusic M. Obesity as a comorbidity in children and adolescents with autoimmune rheumatic diseases. Rheumatol Int. 2023;43(2):209‐219. [DOI] [PubMed] [Google Scholar]
- 28. Sener S, Arslanoglu Aydin E, Batu ED. Cardiac involvement and cardiovascular risk factors in pediatric primary systemic vasculitides. Clin Rheumatol. 2023;42(3):673‐686. [DOI] [PubMed] [Google Scholar]
- 29. Yang L, Guo J, Xiong F. Immunological features of children with gastrointestinal bleeding due to Henoch–Schonlein purpura. Pediatric Health Med Ther. 2024;15:59‐66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Xu L, Li Y, Wu X. IgA vasculitis update: epidemiology, pathogenesis, and biomarkers. Front Immunol. 2022;13:921864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Besbas N, Saatci U, Ruacan S, et al. The role of cytokines in Henoch Schonlein purpura. Scand J Rheumatol. 1997;26(6):456‐460. [DOI] [PubMed] [Google Scholar]
- 32. Yang YH, Lai HJ, Huang CM, Wang LC, Lin YT, Chiang BL. Sera from children with active Henoch–Schönlein purpura can enhance the production of interleukin 8 by human umbilical venous endothelial cells. Ann Rheum Dis. 2004;63(11):1511‐1513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Trapani S, Rubino C, Indolfi G. Gastrointestinal involvement in childhood vasculitides. Acta Paediatr (Stockholm). 2020;109(11):2226‐2236. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table S1. Availability of key clinical and laboratory variables (N = 687).
Supplementary Table S2. Comparison between patients with and without available serum IgA measurements among those with abdominal pain.