Can skin biopsy reports for adult Henoch Schnlein purpura identify patients at risk for renal involvement |
Can skin biopsy reports for adult Henoch-Schönlein purpura identify patients at risk for renal involvement?

Acta Eruditorum

Abby Van Voorhees

Dr. Van Voorhees is the physician editor of Dermatology World. She interviews the author of a recent study each month.

Bookmark and Share

In this month’s Acta Eruditorum column, Physician Editor Abby S. Van Voorhees, MD, talks with David A. Wetter, MD, about his recent Journal of the American Academy of Dermatology article, “Histopathology and correlates of systemic disease in adult Henoch-Schonlein purpura: a retrospective study of microscopic and clinical findings in 68 patients at Mayo Clinic.”

Dr. Van Voorhees: Let’s start with a refresher about Henoch-Schonlein purpura. Remind us about what is known about this process.

Dr. Wetter: Henoch-Schonlein purpura (HSP) is a type of cutaneous small-vessel vasculitis (CSVV) that typically manifests as palpable purpura of the lower extremities and buttocks. In addition to the skin, HSP can commonly affect the kidney, gastrointestinal tract, and joints. Microscopically, HSP demonstrates leukocytoclastic vasculitis (LCV) on routine histology and IgA-predominant vascular deposits on direct immunofluorescence microscopy (DIF) (although other conjugates in addition to IgA may also be seen on DIF specimens of patients with HSP). Infections and medications can cause HSP, although patients may have idiopathic disease, and rarely HSP can be due to an underlying malignancy. Although recent literature suggests that IgA vascular deposits are the defining pathophysiologic feature of HSP, it should be noted that there are several entities that can cause non-HSP-related IgA vasculitis, including multiple myeloma, IgA paraproteinemia, cryoglobulinemia, systemic lupus erythematosus, and inflammatory bowel disease, among others, and thus the classification of a patient with IgA vasculitis as HSP-related or non-HSP-related depends on the presence or absence of characteristic extracutaneous involvement (i.e. joints, kidney, gastrointestinal tract) as noted above. [pagebreak]

Past studies have sought to determine clinical prognostic factors of renal disease in patients with HSP. Clinical factors associated with renal disease in adults in various studies include recent infectious history, older age, pyrexia, elevated inflammatory markers, anemia and hematuria at presentation, and onset of disease during summer months. The significance of lesions above the waist is less certain (Poterucha et al, J Am Acad Dermatol 2012;67:612-6). A recent study (Byun et al, Am J Dermatopathol 2012;24:139-44) found that older age, hematuria, and severity of leukocytoclasis were associated with a relapsing course of HSP. Other studies have suggested that cases of HSP related to medications may have decreased rates of renal involvement.

Dr. Van Voorhees: What has been known up until now about the potential to use the histological findings as a predictor of systemic involvement in HSP? Are there other diseases where this has previously been possible?

Dr. Wetter: To our knowledge, very little existed in the literature regarding potential associations between histopathologic findings and systemic involvement in adult HSP, and that gap served as the main impetus for our study. In a study of patients with CSVV, tissue eosinophils were found to be increased in patients with CSVV associated with drugs (Bahrami et al, Arch Dermatol 2006;142:155-61), and other studies have suggested that drug-induced vasculitis has a low rate of progression to renal disease. A few other studies have also tried to look for correlations between histopathologic findings of CSVV and systemic involvement of vasculitis, although the findings have not been conclusive: Sanchez et al (Arch Dermatol 1985;121:220-224) noted that vasculitis extending to the deep reticular dermis or subcutis was more often associated with malignancy or connective tissue disease, although Cribier et al (Am J Dermatopathol 1999;21:532-536) found that the severity of histopathologic changes did not predict extracutaneous involvement of vasculitis. [pagebreak]

Dr. Van Voorhees: Tell us about your study. What did you find?

Dr. Wetter: We performed a retrospective review of 68 adult HSP patients (age >18 years) seen at Mayo Clinic between 1992-2011, and analyzed their clinical information and pathology slides in order to assess whether various histopathologic markers were associated with renal or other systemic involvement in adult HSP. Inclusion criteria for our study included:

  1. diagnosis of HSP with clinical findings of palpable purpura;
  2. leukocytoclastic vasculitis on histopathologic evaluation;
  3. presence of IgA vascular deposits on DIF; and
  4. skin biopsy slides available at our institution to formally review.

Patients were excluded if they had viral hepatitis, connective tissue disease, hematologic disorders, coexisting malignancy, cryoglobulinemia, systemic lupus erythematosus, or other autoimmune disease. Clinical data was obtained and then all available histopathology slides of the 68 patients were reviewed by two dermatopathologists who were blinded to the clinical data of the patients. Statistical analyses were then performed to look for associations between histopathologic features and the presence of systemic involvement.

We found that the mean age of our patient cohort was 45.8 years, with 41 (60 percent) male. Renal involvement by HSP occurred in 30 patients (44 percent), joint in 32 (47 percent), gastrointestinal tract in 27 (40 percent), and any systemic involvement in 52 (76 percent). The most interesting finding of our study was that those who were older than 40 years and had LCV with an absence of eosinophils on skin biopsy had higher rates of renal involvement that those who did not have both of these features (18 of 24 patients, 75 percent; versus 12 of 44, 27 percent; P <.001). Patients with LCV with an absence of histiocytes had higher rates of gastrointestinal involvement (P = .03). In addition, patients age 40 years or younger had an increased risk of gastrointestinal involvement (P = .004) and a nonsignificant trend for joint involvement (P = .06). Skin ulceration and the other histopathologic variables that were analyzed were not found to be significantly associated with the presence of systemic involvement of HSP. [pagebreak]

Dr. Van Voorhees: Does being able to predict the risk of renal involvement amongst the elderly patients suggest clinical approaches for these patients? What about for the younger patients without eosinophils what does this mean for these patients?

Dr. Wetter: Our study found that lack of eosinophils on histopathologic evaluation (P = .008), as well as increased patient age (>40 years, P = .03) were associated with renal disease. When both age and presence/absence of eosinophils were taken into account, the only patients we found with an increased risk of renal involvement were those who both had LCV without eosinophils on skin biopsy evaluation and were older than 40 years of age. Because a definite cause (e.g. drug or infection) could not be ascertained in the majority of adult HSP patients in our study, we were unable to determine whether the presence of eosinophils was a marker of drug-induced HSP and therefore might have portended a better renal prognosis for this reason.

Nonetheless, our findings suggest that a simple point-of-care approach that takes into account (1) the age of the patient and (2) whether the skin biopsy showed LCV without eosinophils may help clinicians to stratify the risk of renal involvement in the adult HSP patients that they are evaluating. In particular, our data suggest that patients over 40 years of age who lack eosinophils on their skin biopsy should be followed very closely for the development of renal disease. While the presence of eosinophils in those under age 40 is not a suggestive variable for renal disease, all patients remain at increased risk for renal involvement and it is important that they be screened for it. [pagebreak]

Dr. Van Voorhees: Did this study provide as clear an association between gastrointestinal disease or arthritis and the histopathological findings?

Dr. Wetter: We found that patients with LCV and an absence of histiocytes had higher rates of gastrointestinal involvement (P = .03), but otherwise no other statistically significant associations between histopathologic features and either gastrointestinal disease or joint involvement were found in our study. We are uncertain of the clinical significance of the finding of an increased risk of gastrointestinal tract involvement in patients with skin biopsies showing LCV with an absence of histiocytes. We were unable to determine the age of lesions biopsied in our cohort of 68 patients; since histiocytes become more common as lesions age, information regarding the age of each lesion and correlating it with the absence or presence of histiocytes would be important in determining whether this particular finding in our study has any clinical significance.

Dr. Van Voorhees: Will you change your clinical practice as a consequence of what you have found?

Dr. Wetter: Yes, I have changed the way I manage adult HSP patients based upon our study’s findings. In particular, I specifically talk to the dermatopathologist reading the patient’s skin biopsy specimen and ask about the presence or absence of perivascular eosinophils in association with LCV. If the patient is older than 40 years and does not have eosinophils on skin biopsy, I will then follow the patient (with the assistance of a nephrologist) more closely over time for the development of renal disease. Further studies are needed to determine whether more aggressive treatment of adult HSP patients older than 40 years of age and without eosinophils on skin biopsy could decrease the development of renal sequelae.

Dr. Wetter is associate professor of dermatology at the Mayo Clinic. His article was published in the Journal of the American Academy of Dermatology; J Am Acad Dermatol. 2013 Mar;68(3):420-424.e3. doi: 10.1016/j.jaad.2012.08.011.