Pyoderma Gangrenosum-Like Presentation of Herpetic Panniculitis in a Patient With Immunosuppression

The nonspecific clinical presentation of panniculitides requires clinical pathological correlation to determine the etiology. Infection-induced panniculitis is often seen in the immunocompromised patient and presents as 1 or more fluctuant nodules that ulcerate on the lower extremities.¹ Infection with herpes virus usually presents as an easily distinguishable vesicular rash. In the patient with immunosuppression, however, herpes virus infection can manifest with uncommon clinical findings.² The 2 subtypes of infection-induced panniculitis are primary, that is, a direct infection into the fat tissue, and secondary, which occurs through hematogenous spread of the pathogen.^3,4 However, the presence of ulcerative nodules may also be suggestive of the presence of other underlying etiologies, such as medium-sized vasculitis and vasculopathies (eg, calciphylaxis), factitial etiology, or inflammatory ulcerations (eg, pyoderma gangrenosum [PG]).  

A 79-year-old male with a history of diabetes mellitus and rheumatoid arthritis undergoing treatment with methotrexate 15 mg per week and prednisone 7.5 mg per day presented with a 3-month history of a mostly nontender chronic ulcer on the left buttocks that was refractory to wound care and antibiotic therapy. Physical examination revealed a 1.4 cm × 2.7 cm deep skin ulcer with violaceous borders. Superficial bacterial wound cultures were negative. Ultrasonography revealed no fluid collections. According to the diagnostic criteria proposed by Su et al⁵ for PG, the patient’s progressive skin ulcer with irregular undermining and violaceous border qualified as major criteria and the history of an autoimmune condition qualified as a minor criterion. As the diagnosis of PG was suspected, the ulcer was treated with intralesional triamcinolone. 

Two weeks later, the ulcer had not improved. Instead, it had become tender, increased in size, and developed accentuated, bizarre borders and surrounding small nodules suggestive of possible pathergy. To complete the workup of this atypical ulcer, a punch biopsy of the edge of the ulcer was done, which revealed extensive necrosis within the subcutis as well as a diffuse infiltrate of lymphocytes, neutrophils, plasma cells, and histiocytes suggestive of an infectious or inflammatory etiology. Staining for herpes simplex virus 1 and 2 highlighted many cells in areas of fat necrosis (Figure). Staining for varicella-zoster virus (human herpesvirus 3) was negative. Results of additional, special stains and microbial cultures for fungal (periodic acid-Schiff technique), bacterial (Gram stain), and acid-fast mycobacterial (Fite method) organisms were negative. These findings supported the diagnosis of herpetic panniculitis. Polymerase chain reaction testing of a subsequent herpetic flare on the anterior leg confirmed the diagnosis of herpes simplex virus 2. The patient had an unknown history of shingles and herpes complex infection. 

A 4-week course of valacyclovir 500 mg twice daily was started based on kidney function. The lesion on the left buttock began improving but had not completely resolved by 4-week follow-up. An infectious disease specialist was consulted, who recommended an extended course of valacyclovir 500 mg twice a day to complete up to 8 weeks of treatment, transitioning to prophylaxis dosing with valacyclovir 500 mg for 1 year. At the 3-month follow-up visit, the lesions were improved. Time to complete healing was approximately 6 months. 

This case is comparable to the only previously published case (to the authors’ knowledge) of herpes virus leading to panniculitis, including that both patients received immunosuppressive therapy.⁶ In the patient reported on by Tomasini and Ribero,⁶ however, the ulcer completely regressed after 1 month of treatment with acyclovir. In comparison, although the patient in this case had shown some improvement with after 4 weeks of valacyclovir 500 mg daily, new lesions continued to develop, which prompted an increased dose to 1000 mg daily. This patient began showing improvement of all ulcers after a 10-day course of the increased dose. 

Importantly, the patient in this case was initially and erroneously given a clinical diagnosis of PG; subsequent pathology results led to the final diagnosis of herpetic panniculitis. Pyoderma gangrenosum is a rare condition that commonly presents as a full-thickness ulcer with bluish purple undermining borders and can have white cell debris that looks like pus but often shows no growth on culture.⁷ There have been prior reports of herpes virus infection mimicking PG.^8-12 However, none of these cases resulted in panniculitis. Distinguishing early PG from panniculitis can be challenging in clinical practice.¹³

Pyoderma gangrenosum is a clinical diagnosis arrived at only after infection, malignancy, and trauma have been ruled out.¹⁴ The pathology of PG is nonspecific, but biopsies can help rule out similar clinical presentations, such as vasculitides and malignant abnormalities.¹⁴ Other causes of cutaneous ulcers that may mimic PG include arterial and venous disease, vascular occlusion, hematologic causes, calciphylaxis, drug-induced ulceration, hypertension (Martorell ulcer), and inflammatory disorders such as cutaneous Crohn's disease.¹⁵

This case provides another example of herpes virus induced panniculitis in an immunocompromised patient and highlights a not uncommon clinical scenario of misdiagnosis of an atypical ulcer. The chance for misdiagnosis also highlights the importance of performing biopsy on all patients with presumed PG and maintaining a high index of suspicion for rare causes of cutaneous ulcers in those who are immunocompromised. Additionally, this case suggests that not all cases of herpes virus induced panniculitis resolve quickly even when treated with appropriate antivirals and that long term and prophylactic treatment may be indicated in some cases. 

Given that this is a single case study, the strength of the conclusions is limited. Several limitations led to an initial misdiagnosis of PG and, ultimately, a delay in appropriate treatment for herpetic panniculitis. The patient’s autoimmune disease coupled with the wound morphology, for example, the violaceous wound border, was suggestive of PG according to the Su et al⁵ criteria. The diagnosis of PG was made prior to performing immunohistochemical staining and obtaining tissue culture results. However, given the incidence of PG (3 to 10 patients per million population per year) and the rarity of herpetic panniculitis, empiric treatment for PG was warranted.¹⁶ The lesions of the patient reported herein drastically worsened when treated empirically with intralesional triamcinolone, which prompted further workup described in the case. 

Atypical presentations of herpes virus can manifest in patients who are immunocompromised and may result in infection-induced panniculitis. Additionally, skin biopsies should be considered in patients with chronic ulcers (especially in such patients who are immunocompromised) that are not responsive to standard of care treatment to rule out ulcerative conditions of infectious etiology. Treatment should consist of antiviral therapy that is active against herpes virus. Long-term treatment may be needed. 

Authors: Victoria E. Orfaly, BS; Erika L. Hagstrom, MD; Kevin P. White, MD; and Alex G. Ortega-Loayza, MD, MCR

Affiliation: Oregon Health & Science University Hospital, Portland, OR

Correspondence: Victoria E. Orfaly, BS, Medical Student, Oregon Health & Science University Hospital, Dermatology, Center for Health & Healing Building, 3303 S Bond Avenue, 16th Floor, Portland, OR 97239; orfaly@ohsu.edu

Disclosure: The authors disclose no financial or other conflicts of interest.