Nodular Vasculitis: Retrospective Study of an Uncommon Disease in a Non-Tuberculosis Endemic Country with Focus on Treatment Modalities and Efficacy

Nodular vasculitis (NV) is a form of panniculitis frequently associated with vasculitis typically manifesting as painful or ulcerating nodules on the dorsal lower legs primarily in females. Erythema induratum of Bazin (EIB) is considered by some authors to be a cutaneous tuberculin hypersensitivity presenting as NV [1]. EIB was first described in 1855 by Ernest Bazin, a French dermatologist who observed the clinical picture “on the legs of female laundresses and in young and plump, well-nourished women with the typical phenotype of those with scrofula” [2].

These lesions were later associated with tuberculosis (TB) when Mycobacterium tuberculosis was discovered in 1882 and found in cervical lymph nodes of patients with benign erythematous tuberculids [1]. However, the mycobacteria can only be cultured or detected by PCR very rarely in EIB lesions. Thus, EIB is considered a hypersensitivity reaction to M. tuberculosis antigens in the subcutaneous adipose tissue presenting as NV.

Other etiologies that cause the NV have been described including associations with Mycobacterium avium [3], hepatitis B and hepatitis C virus infection, Crohn’s disease, BCG vaccination, and anti-tumor-necrosis factor treatment [4‒11]. Overall, NV is thought to be a hypersensitivity reaction to a chronic inflammatory disease.

Several retrospective studies have confirmed that predominantly middle-aged women (male:female ratio = 1:10, mean age of onset of 57.2, 44.1, 50.7, 36.7 years is reported) are affected by EIB [7, 12‒14]. Patients typically present with subcutaneous violaceous, brownish-erythematous usually painful, sometimes ulcerative nodules and plaque-like infiltrations predominantly on the posterior or anterolateral regions of the lower legs (Fig. 1).

The histopathologic pattern of NV is highly dependent on the age of the lesions. In early stages, lobular panniculitis with granulomatous and lymphocytic inflammatory infiltrates and small necrosis foci with nuclear debris are typical. In later stages, fibrosis with thickened subcutaneous septa and predominantly lobular granulomatous panniculitis occurs. Extended necrotic areas, also involving the epidermis, can be detected [15]. Vasculitis of small- and medium-caliber vessels and even thrombotic occlusion is possible, but it is not a histopathological requirement for the diagnosis [1]. The detection of mycobacteria in the biopsy specimens is extremely difficult. The reported positivity rates of positive PCR are very variable, between 0 and 77% [16‒18]. There are currently no defined standards for the treatment of NV. This study aimed to retrospectively analyze all patients diagnosed with NV over 20 years in a large university hospital in Switzerland with a focus on demographics, clinical presentation, histopathological and laboratory findings, and efficacy of different treatment modalities.

After ethical approval was obtained (BASEC No.: 2021-02207), the electronic hospital database was searched for patients ≥18 years old, treated at the Department of Dermatology at the University Hospital Zurich over a period of 20 years (January 1, 2001, to December 31, 2021). The following keywords were used to identify all patients diagnosed with NV or EIB: “Erythema induratum”, “Bazin”, “Bazin-Syndrom”, “Nodular vasculitis”, “Vaskulitis nodularis”, “nodöse Vaskulitis”, “noduläre Vaskulitis”, “Nodularvaskulitis”, “Hypodermitis nodularis subacuta saltans”, “nodöses Tuberkulid”, “Phlebiitis nodularis”, “Tuberculosis cutis indurativa”, und “Tuberculosis indurativa cutanea et subcutanea”. Initially, 116 patients were identified. Fifty-four patients had to be excluded from the study because they did not meet the inclusion criteria: (1) histopathologically confirmed diagnosis of NV/EIB, (2) availability of clinical and demographic information, (3) age>18 years. The following clinical and histopathological information was extracted from the electronic database: Patient characteristics (sex, age, ethnicity, travel history, comorbidities), clinical characteristics (localization, number and size of the lesions, presence of ulceration, presence of pain, potentially triggering drug, disease duration), histopathologic findings including special stains (Ziehl-Neelsen, periodic acid-Schiff [PAS], Brown-Brenn) and PCR for mycobacteria, laboratory findings (QuantiFERON test), treatment modalities, and treatment efficacy (presence of relapse).

The efficacy of the treatment was evaluated using information from the patient’s medical records. Patients were retrospectively assessed for relapses after 3 months. Treatment efficacy was calculated as the ratio of the number of no relapses over the number of patients treated with the drug (number of no relapses/patients treated with a specific drug). Treatment efficacy based on clinical evaluation was assessed retrospectively 12 weeks and, if available, 24 weeks after treatment initiation. In all patients, follow-up was available at least for 12 weeks.

All histopathology specimens were reviewed by a board-certified dermatopathologist (I.K.) and a second investigator (C.G.). Special stains (PAS, Ziehl-Neelsen, Brown-Brenn) and PCR for Mycobacteria were performed if sufficient paraffin-embedded biopsy material was available.

Statistical analyses were conducted using SPSS 29.01. A descriptive analysis with presentation of absolute and relative (%, in parentheses) frequencies was performed. Correlations between two dichotomous variables were calculated using odds ratios (OR) and confidence intervals. Based on observed cell frequencies, significance was determined using Fisher’s exact test (for one-sided distribution according to Hedderich [19]). Differences between metric-scaled independent samples would be determined by Mann-Whitney U test (for two-sided distribution). A p < 0.05 was considered statistically significant.

A total of 62 patients with histopathologically confirmed NV and sufficient clinical information on follow-up and treatment were included in this study. Forty-seven patients (76%) were female and 15 (24%) male. The mean age was 52 years (range: 19–85 years). The most common localization was the lower extremities (46, 74%). Lesions were found on one body site in 44 cases (71%), and in 15 cases NV affected multiple body sites. Localization of NV was unknown in 3 cases (5%). The total disease duration ranged from several months to >20 years with <1 year being the most common. Thirty-eight patients (61%) reported pain. Ulceration of the lesions was present in only 5 cases (8%). Patient characteristics and clinical findings are summarized in Table 1.

A predominantly lobular panniculitis was found in all cases (64, 100%), small vessel vasculitis was observed in 21 cases, and vasculitis of medium-sized vessels in 11 cases. Granuloma formation was present in 40 cases and foci of necrosis in 41 cases. Plasma cells were found in 10 biopsies. Epidermal changes were found in 5 histopathological specimens.

Sufficient histopathological material for further staining was available in 32 cases. Ziehl Neelsen, PAS, and Brown-Brenn stains were negative in all available samples (32, 100%).

QuantiFERON test was performed in 37 cases and positive in 22 (59%). Mycobacteria PCR was carried out in 27 biopsies and positive in 2 (7.4%) samples. Table 2 summarizes the histopathologic and laboratory findings regarding the presence of (latent) TB.

Thirty-five patients (56%) had at least one comorbidity. Several different comorbidities were noted: obesity (17 patients, 7%), arterial hypertension (9 patients, 15%), diabetes mellitus (5 patients, 8%), thyroid disease (hyperthyroidism or hypothyroidism) (6 patients, 10%) autoimmune disease (9 patients, 15%), had an chronic inflammatory skin disease (2 patients, 3%), chronic inflammatory bowel disease (2 patients, 3%), hepatitis B or C (6 patients, 10%), and sarcoidosis (1 patient, 2%).In 4 cases (7%), NV manifested shortly after initiation of hepatitis C treatment with interferon (2 cases), interferon and ribavirin therapy (1 patient), and with adalimumab (1 patient). Comorbidities are summarized in Table 3.

In 17 patients (26.5%), no information regarding treatment was available. These patients were excluded from further analysis. In 47 patients, information regarding treatment was available and thus were considered for further analysis (100%).

A total of 15 (31.9%) patients received high potency topical corticosteroids (TCS) as first-line treatment, in 7 (14.9%) patients combined with the use of compression stocking. Topical steroids were generally prescribed in a degressive manner over 12 weeks.

Regarding antitubercular treatment, a total of 22 patients had a positive QuantiFERON test, of which 21 (44.6%) received antitubercular treatment. In 17 (36.1%) cases, the treatment regimen of antitubercular therapy included a tuberculostatic 4-drug therapy with isoniazid, rifampicin, pyrazinamide, and ethambutol for 2 months followed by a 2-drug therapy with isoniazid and rifampicin for 4 months. In 3 (6.4%) cases, NV was treated with isoniazid only for 6–9 months. The follow-up of 1 patient was lost.

A total of 7 (14.9%) patients received systemic steroids in combination with topical steroids with or without nonsteroidal inflammatory drugs as a first-line treatment. The standard treatment regimen with systemic steroids lasted 6 weeks with a starting dose of 1.0 mg/kg per day with a dose reduction to 0.5 mg/kg per day. Two (4.3%) patients received oral potassium iodide in combination with TCS as first-line treatment.

In patients treated with TCS with or without compression stockings, 7 (14.9%) patients showed a complete and lasting response after tapering off TCS. Eight (17%) patients experienced a relapse within 24 weeks. One (2.1%) patient was switched to systemic steroids, one (2.1%) to potassium iodide, and one (2.1%) to hydroxychloroquine. Five (10.6%) patients underwent another treatment cycle of TCS. After cessation of antitubercular treatment, 11 (52%) out of 21 patients relapsed within 24 weeks (2 (9.5%) patients on isoniazide monotherapy and 9 (42%) on a multidrug treatment).

Of the 7 patients treated with systemic steroids, 4 relapsed after 24 weeks. One patient was switched to potassium iodide, one to colchicine, and one to methotrexate, followed by colchicine and then hydroxychloroquine. Out of the 2 patients treated with potassium iodide, one had a relapse after approximately 20 weeks and was then lost to follow-up.

We studied the treatment efficacy of each modality based on the respective relapse rates. The treatment efficacy was calculated as the ratio of the number of patients with no relapses to the total number of patients treated with the drug. The results are summarized in Table 4. Our study showed that all treatments (topical steroids, antitubercular therapy, systemic steroids, or potassium iodide) showed favorable treatment outcomes in half of the patients treated.

Furthermore, possible risk factors for relapse were statistically determined. The factors female sex (OR: 1.075, p = 0.607), localization of the nodes on the upper half of the body (OR: 1.64, p = 0.454), presence of comorbidities (OR: 1.95, p = 0.345), obesity (OR: 3.71, p = 0.109), hypertension (OR: 3.58, p = 0.240), and diabetes (OR: 1.48, p = 0.607) as comorbidities were associated with a higher relapse rate but not statistically significant. Multiple nodes (OR: 0.67, p = 0.463), node size >5 cm (OR: 0.38, p = 0.407), autoimmune disease (OR: 0.64, p = 0.436), and hepatitis B or C (OR: 0.67, p = 0.503) as comorbidities and a positive QuantiFERON test (OR: 0.75, p = 0.527) were associated with more remissions than relapses but also not statistically significant. Age was not risk factor for relapse (p = 0.257).

This retrospective study aimed to analyze the characteristics and treatment outcomes of patients with NV in a university hospital in Switzerland. This is the first study on NV in a non-TB endemic country with a predominantly Caucasian population and the first study to systematically analyze treatment efficacy in NV with and without TB association. Knowledge of treatment modalities and efficacy in NV is very limited, with only small studies and case reports on therapeutic strategies available.

Over 20 years (January 1, 2001, to December 31, 2021), we identified a total of 62 patients. With approximately 70,000 outpatient visits in our department each year, this low number underlines the rarity of NV in our patient population. As expected, NV was predominantly found in middle-aged female patients. The varying disease duration, which ranged from several months to over 20 years, emphasizes the chronic and recurring nature of NV [20].

Overall, 61% of patients experienced pain, which highlights the potential impact of NV on patients’ quality of life and the need for effective management strategies to alleviate symptoms. It is worth noting that NV primarily affected a single body site in most cases.

The presence of comorbidities in over half of the patients, such as obesity, arterial hypertension, diabetes mellitus, thyroid disease, autoimmune disease, and hepatitis B or C, highlights the potential role of underlying health conditions in contributing to the development or exacerbation of NV. The occurrence of NV following therapy for hepatitis C, including interferon, interferon and ribavirin therapy, and adalimumab, indicates the possibility of drug-induced NV in some cases [11].

In diagnosing NV, skin biopsies were routinely performed. Ziehl-Neelsen stain was negative in all available cases, whereas the QuantiFERON test was positive in 22 of 37 performed tests. Mycobacteria PCR was only positive in 2 of 27 cases. Accordingly, 21 patients received antitubercular treatment (either isoniazide for 6–9 months or a multidrug combination), which however only led to a permanent remission of NV in 10 patients; 11 patients relapsed after the cessation of antitubercular treatment.

This outcome is interesting as it is lower than in previous studies, which reported a cure rate of 73% and 87.5%, respectively [21, 22]. In our cohort, both multidrug regimens and monotherapy led to relapses in half of the patients.

In the absence of latent TB, most patients received potent TCS as the initial treatment, followed by systemic steroids and potassium iodide. Nevertheless, in approximately 50% of these cases, relapse occurred within 12–24 weeks, leading to a switch to a second-line therapy.

We attempted to identify risk factors for relapse using Fisher’s exact test. Unfortunately, we were not able to detect any significant risk factors; however, we did see that female sex, involvement of the upper half of the body, and the presence of comorbidities such as obesity and diabetes tended to have a higher recurrence rate.

A significant limitation of our study is its retrospective, monocentric design and the absence of a control cohort. Furthermore, the overall number of patients is low, and there is a high variability in the treatment regimens and treatment outcomes, which prevents us from obtaining robust statistical results. It is likely, that, patients with complete remission or well-controlled disease were lost to follow-up and therefore are not represented in this study.

Nevertheless, we suggest a conservative approach with potent topical steroids and compression stockings as first-line treatment. Systemic treatment should be considered if this strategy is not effective. If systemic treatment is necessary, we would most likely recommend systemic steroids initially, followed by an immunomodulatory treatment, such as potassium iodide.

Our study provides valuable insights into the demographics and treatment outcomes of patients with NV in a non-TB endemic country with a predominantly Caucasian population. The rarity of this disease is emphasized by the low number of patients identified in our cohort over a 20-year period. Our findings are consistent with previous literature, supporting the diagnosis of NV based on characteristic histopathological features and comorbidities.

Based on our results, we cannot definitively recommend a specific treatment. Topical steroids, antitubercular therapy, systemic steroids, and potassium iodide all showed similar response rates. In the absence of latent TB, potent topical steroids for several weeks should be considered as first-line treatment, if possible, in combination with the use of compression stockings. In case of insufficient response to topical therapy, immunomodulatory or immunosuppressive systemic therapy should be evaluated. Tuberculostatic therapy upon detecting latent TB is recommended. Further larger and, above all, prospective studies are necessary to confirm our findings.

This study was conducted ethically in accordance with the guidelines for human studies and in accordance with the World Medical Association Declaration of Helsinki. The study protocol was approved by the institute’s committee on human research (Kantonale Ethik-Kommission Zürich, BASEC No.: 2021-02207). Written informed consent was obtained from all participants if diagnosed after 2016. An exemption to retrospectively collect anonymized clinical data was made by the abovementioned Ethics Committee if the diagnosis was made before 2016 and patients could not be contacted (BASEC No.: 2021-02207).

All authors have no COI to declare.

This study was not supported by any sponsor or funder.

C.G.: conception and design, acquisition/analysis, and interpretation, drafting of the manuscript, final approval, and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. I.K.: conception and design, data interpretation, drafting of the manuscript, final approval, and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. X.H.: conception and design, acquisition/analysis, and interpretation and drafting of the manuscript. A.S.: conception and design, data interpretation, critical revision for important intellectual content, final approval, and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. P.S.G.: conception and design, data interpretation, critical revision for important intellectual content, final approval, and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

The data supporting the results of this study are not publicly available because they contain information that could compromise the privacy of research participants but are available from I.K. upon reasonable request. The raw data that support the findings of this study are available on request from the corresponding author [I.K.].