Overall, 81 patients were identified and included in the analysis. IVIG was administered over 3–5 days per cycle every 4 weeks. Mean age at commence of IVIG was 62.6 (± 17.3, range: 21–90) years; 48 (59.3%) patients were female and 33 (41.7%) were male. In 28 patients (34.6%), more than one part of the body was affected. By the time of first administration of IVIG, two ulcers (4.3%) measured less than 5.0 cm2, one ulcer (2.1%) measured between 5.1 and 10.0 cm2. All remaining ulcers (93.6%) were greater than 10.0 cm2. Associated comorbidity was present in 24 (29.6%) patients. Most frequent conditions were hemato-oncologic disorders (seven patients, 8.6%), Crohn’s disease, rheumatoid arthritis and solid malignancies (six patients, 7.4% each). Details are provided in Table 1.

Patients had received a mean of 2.1 (+/− 1.14) previous treatments (Table 2). Therapies prior to IVIG administration were (methyl-)prednisolone (n = 79; 97.5%), mycophenolate mofetil (n = 26; 32.1%), cyclosporine A (n = 16; 19.8%), anti-TNF-α therapy (n = 12; 14.8%; infliximab, n = 10, 12.3% and/or adalimumab, n = 4, 4.9%), azathioprine (n = 11; 13.6%), methotrexate (n = 10; 12.3%), dapsone (n = 10, 12.3%), ustekinumab (n = 2; 2.5%), canakinumab (n = 1; 1.2%) and secukinumab (n = 1; 1.2%). One patient (1.2%) received IVIG as a first-line treatment.

Patients received an average of 4.75 cycles of a mean IVIG dosage of 1.35 g per kg bodyweight (range 0.8–2.0 g per kg bodyweight). In 77 (95.1%) cases, IVIG were used as adjunct therapy together with (methyl-) prednisolone (n = 72, 93.5%), mycophenolate mofetil (n = 24, 31.2%), dapsone (n = 8, 10.4%), azathioprine (n = 6, 7.8%), cyclosporine A (n = 5, 6.5%), methotrexate (n = 4, 5.2%), infliximab (n = 2, 2.6%) or adalimumab (n = 1, 1.3%).

Response data were available for 75 patients 1 month after starting treatment. A total of 16 (21.3%) showed a complete response (healing tendency), 21 (28.0%) showed a partial response (stopping the progression of the lesion) and 38 (50.7%) showed no response (worsening of the lesion; Fig. 1). At 3 months after initiation (n = 75), 19 (25.3%) and 28 (37.3%) patients showed a complete or partial response, respectively, and 28 (37.3%) showed no response. After 6 months (n = 69), 13 (18.8%) and 27 (39.1%) of patients had a complete or partial response, respectively, and 28 (40.6%) had no response. Only one patient (1.5%) had healed completely within 6 months after treatment initiation.

Response to IVIG therapy after 1,3, and 6 months

Altogether 14 (17.3%) potentially treatment-related adverse events could be identified in ten patients (12.3%; Table 3). Of which three patients (3.7%) had a thromboembolic event (deep vein thrombosis in two cases, both rated moderate, myocardial infarction in one case, rated severe), three patients (3.7%) had an acute kidney injury (two mild, one moderate) and two patients (2.5%) suffered from infusion reactions. There was one death (1.2%; the same patient had suffered from myocardial infarction before). Other reported adverse events were headache, fatigue, abdominal pain, diabetic ketoacidosis and pneumonia (one case, 1.2% each). The latter two were unlikely to be related to the treatment.

There was no statistically significant difference in sex (female OR 0.72, 95% CI 0.25–2.12) and no effect by age (OR 0.98, 95% CI 0.95–1.01). A higher dose tended to be more efficient, but was not statistically significant (OR 2.69, 95% CI 0.84–8.63). A higher PARACELSUS score was associated with a better response, although the effect was small (OR 1.31, 95% CI 1.05–1.64). The number of previous therapies tended to be negatively associated with the response rate (OR 0.66, 95% CI 0.37–1.16). Patients with solid malignancies or hemato-oncological disease had a tendency for a higher response rate (OR 4.36, 95% CI 0.79–23.91; n = 6 and OR 1.57, 95% CI 0.28–8.92; n = 7; respectively). Comorbidities associated with a significantly higher treatment response were thyroid disease (OR 6.64, 95% CI 1.01–43.57, n = 5) and diabetes mellitus (OR 3.49, 95% CI 1.13–10.80, n = 21). Chronic venous insufficiency tended to be associated with worse response (OR 0.25, 95% CI 0.05–1.21, n = 24; Fig. 2, Table S1).

Odds ratio (OR) and confidence interval (CI) of comorbidity. Values greater than 1 show a better response rate than average, values lower than 1 are associated with worse response to IVIG therapy