Direct immunofluorescence testing in vasculitis A single institution experience with Henoch-Schönlein purpura

Transcription

1 Received: 11 August 2017 Revised: 21 September 2017 Accepted: 29 September 2017 DOI: /cup ORIGINAL ARTICLE Direct immunofluorescence testing in vasculitis A single institution experience with Henoch-Schönlein purpura Patrick Feasel Steven D. Billings Wilma F. Bergfeld Melissa P. Piliang Anthony P. Fernandez Jennifer S. Ko Departments of Pathology and Dermatology, Cleveland Clinic Foundation, Cleveland, Ohio Correspondence Jennifer S. Ko, MD, PhD, Departments of Pathology and Dermatology, Cleveland Clinic Foundation, L25, 9500 Euclid Avenue, Cleveland, OH koj2@ccf.org Background: Direct immunofluorescence (DIF) panels (IgG, IgA, IgM, C3 and fibrinogen) are ordered for clinically suspected vasculitis, with frequently negative results. Methods: Cases submitted for DIF and histology ( ) with vasculitis in the clinical data were examined, and the electronic medical record reviewed for clinical suspicion of Henoch-Schönlein purpura (HSP). Peri/intravascular IgA was considered positive, other reactants non-specific and no immunoreactivity negative. Results: Vasculitis was the given indication for 20% (258/1318) of DIF studies. HSP was clinically suspected in 36% (95/258). In this setting, leukocytoclastic vasculitis (LCV) was common (66%, 63/95) and DIF was positive in 43% (27/63). One hundred percentage of DIF+ had LCV+. In cases without HSP suspicion, 26% (42/163) were LCV+ and <1% DIF+. Of the 258 cases, LCV+ greatly enriched for DIF+ (105/258 LCV+ with 28/105 [27%] DIF+), captured 100% of HSP and included cases with non-specific DIF/etiologic findings. In LCV cases, DIF positivity was not seen, HSP was not diagnosed and non-specific DIF findings were common. Conclusions: LCV is an H&E-based histopathologic diagnosis that can have positive, negative and non-specific DIF results that are rarely contributory except in the setting of HSP, where DIF is best utilized with IgA as the sole immunoreactant. H&E-based triage of DIF orders is recommended. KEYWORDS cutaneous, direct immunofluorescence, Henoch-Schönlein purpura, vasculitis 1 INTRODUCTION Diagnosis of cutaneous vasculitis often requires biopsy confirmation, yet accurate classification is largely based on clinical and laboratory findings. The ultimate role of the pathologist is to confirm the presence of vasculitis, determine the nature of the inflammatory infiltrate and the size of vessels involved. 1,2 Cutaneous small-vessel vasculitis (herein referred to as leukocytoclastic vasculitis, LCV) refers to a small-vessel neutrophilic vasculitis affecting the superficial dermal plexus. LCV can be limited to the skin or additionally involve internal organs. LCV is linked to a wide variety of underlying triggers such as drug and infection, but it is idiopathic in 40% of cases. 3 Henoch-Schönlein purpura (HSP, IgA vasculitis) is a systemic vasculitis with IgA-dominant immune-complex deposits affecting small vessels. Based on European League Against Rheumatism criteria, classification as HSP requires the presence of purpura or petechiae with lower limb predominance plus 1 of the 4 criteria: (1) abdominal pain, (2) histopathology (IgA), (3) arthritis or arthralgia or (4) renal involvement. 4 Biopsies show typical LCV: a perivascular neutrophilic infiltrate with karyorrhexis and evidence of vascular damage, and DIF revealing granular deposits of IgA around both involved and uninvolved dermal vessels. 5 Direct immunofluorescence (DIF) panels (consisting of IgG, IgA, IgM, C3 and fibrinogen) are considered key diagnostic adjuncts for vesiculobullous diseases of the skin. They have also been historically 2017 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd 16 wileyonlinelibrary.com/journal/cup J Cutan Pathol. 2018;45:16 22.

2 FEASEL ET AL. 17 TABLE 1 Clinical features and pathologic diagnosis of patients grouped according to suspicion for HSP (n = 258) recorded as present or absent based on review of clinician notes and laboratory results (ie, urinalysis) in electronic medical records. Demographics Suspicion HSP No suspicion HSP Cases (%) 95 (36) 163 (64) Age (range) 49 (1-84) 56 (5-89) Sex (male, %) 53 (56) 67 (41) Site H&N 0 2 (1) Trunk 12 (13) 21 (13) Extremity 83 (87) 140 (86) Arm 8 (8) 23 (14) Hand 4 (4) 4 (3) Thigh 28 (29) 30 (18) Leg 38 (40) 65 (40) Foot 5 (5) 18 (11) LCV present 63 (66) 42 (26) DIF positive 27 (28) 1 (paraneoplastic IgAV), 1 (linear IgA disease) HSP diagnosis after biopsy 28 (29) 0 Abbreviations: DIF, direct immunofluorescence; HSP, Henoch-Schönlein purpura; LCV, leukocytoclastic vasculitis. utilized in autoimmune connective tissue disease and vasculitis. Standard panels typically cost $350 to $450 in the United States, require separate tissue submission with additional work of processing and 3 RESULTS 3.1 Significant numbers of DIF studies are performed for vasculitis in the absence of clinical suspicion for HSP Of 1318 skin biopsies with DIF performed and H&E reviewed, vasculitis was the given indication for DIF studies in 20% of all cases (258/1318). Medical records of these 258 patients were reviewed for clinical suspicion for HSP. The symptoms and signs leading clinicians to rule out HSP included the presence of a petechial rash and at least one of the following: laboratory evidence of renal involvement (ie, abnormal urinalysis, elevated creatinine, etc.), GI symptoms/pain, joint involvement/pain or preceding infection. Table 1 summarizes clinical features and pathologic results. Overall, HSP was suspected in 95/258 (36%) patients. The mean age was 49 years in the HSP suspected group, compared to 56 years when HSP was not suspected, although both groups contained a broad age range (1-84 and 5-89 years old, respectively). The gender distribution in the HSP suspected group was 56% men vs 41% women, and body site distribution was predominantly lower extremities in both groups. Correlation of DIF findings with clinical symptoms in patients with suspected HSP is discussed later. frequently yield negative results. Within the vasculitides, DIF can determine the presence or absence of immune-complex deposition, with most cases of LCV being positive/non-specific (82%-100% C3 and/or Ig within vessels). 3 Given the non-specific nature of these findings, we have assessed the use and utility of DIF testing in the setting of vasculitis, focusing on HSP. 2 MATERIALS AND METHODS With Institutional Review Board (IRB) approval, a pathology database search was performed to identify cohort cases over a 5-year period ( ) which contained the qualifiers skin and DIF in diagnosis/comment sections. Only cases with both hematoxylin and eosin (H&E) and DIF performed were included in the study. Additional inclusion criteria included vasculitis indicated on the requisition form. Clinical data were obtained via electronic medical records. Review of clinical notes from time of biopsy was performed for each individual case to assess for true clinical suspicion for HSP. Suspicion was confirmed when consideration for HSP was stated in the clinical note. Histologic diagnoses of LCV were rendered based on the presence of at least focal fibrinoid necrosis of blood vessels and at least focal neutrophil transmigration and leukocytoclasis/karyorrhexis. DIF was considered positive (DIFpos) when there was perivascular reactivity for IgA, non-specific (DIFns) when there was reactivity for IgG, IgM, complement C3 and fibrinogen, but not for perivascular IgA and negative (DIFneg) when reactivity for all immunoreactants was absent. Presence of systemic involvement (renal, gastrointestinal [GI] and joint) was 3.2 Clinical suspicion for HSP enriches for LCV+ and DIF+ cases LCV was more likely to be diagnosed on H&E-stained slides when there was clinical suspicion for HSP (63/95, 66%) vs when there was not (42/163, 26%) (Table 1). Additionally, clinical suspicion for HSP enriched for DIF positivity (27/95, 28%), and this was increased again when LCV was observed on histopathology (27/63, 43%). When there was no clinical suspicion for HSP, not only was LCV less commonly diagnosed (26%), but DIF was almost never positive (1/163, <1% perivascular IgA). The single positive case was paraneoplastic IgA vasculopathy and LCV was not present on H&E slides. The single case in which HSP was diagnosed with negative skin DIF (Table 1) occurred in a 31-year-old man with biopsy-proven cutaneous LCV who developed nephropathy and had renal DIF and H&E revealing IgA nephropathy 1 year later. 3.3 Diagnosed HSP is typically suspected clinically, shows LCV on H&E and shows additional non-specific DIF findings Table 2 more closely examines those patients (n = 63) where HSP was clinically suspected and H&E slides showed LCV, and an example of such a patient is showed in (Figure 1A-D). As compared with DIFnegative patients, those with perivascular IgA were 6 years younger on average and more likely to be male. In addition to IgA, positivity for other immunoreactants was also observed, most commonly

3 18 FEASEL ET AL. TABLE 2 Clinical features and DIF results of cases with suspicion for HSP and LCV present on biopsy (n = 63) Demographics Pos DIF Neg/NS DIF Cases (%) Age (range) 47 (8-79) 53 (19-78) Sex (male, %) 17 (63) 18 (50) Site Head & neck 0 0 Trunk 1 3 Extremity Arm 3 3 Hand 2 0 Thigh 8 12 Leg Foot 2 0 DIF positive 27 (100) 0 IgA 27 (100) 0 IgG 1 (4) 3 (8) IgM 8 (30) 4 (11) C3 16 (59) 17 (57) Fibrinogen 14 (52) 24 (80) Renal 13 (48) 2 (6) Gastro-intestine 15 (56) 2 (6) Joint 2 (7) 1 (3) Infection 5 (19) 5 (14) Malignancy 3 (11) 0 (0) Follow up Resolved 15 (56) 27 (75) Without 7 (47) 18 (67) treatment With 8 (53) 9 (33) treatment Persistent disease 7 (26) 8 (22) (1 HSP, 1 cryoglobilunemic vasculitis, 1 GPA) Died of disease 1 (4) 0 (0) Died other causes 2 (7) 0 (0) No follow up 2 (7) 1 (3) HSP diagnosis 27 (100) 1 (3) Abbreviations: DIF, direct immunofluorescence; HSP, Henoch-Schönlein purpura; LCV, leukocytoclastic vasculitis. complement C3 (59%) and fibrinogen (52%) in a perivascular pattern. Both of these were not uncommonly found in IgA-negative cases that had LCV on H&E slides (57% complement C3 and 80% fibrinogen). IgG reactivity was rare in both groups (4% HSP+ and 8% HSP ). IgM reactivity was more common in HSP patients (30% vs 11%). In patients diagnosed with HSP, extra-cutaneous manifestations most commonly involved the GI and renal systems (56% and 48%, respectively) followed by joint involvement (7%). This was compared to 6%, 6% and 3% GI, renal and joint involvement, respectively, for IgA-negative cases. A higher percentage of patients with non-hsp LCV had resolution of symptoms (75% vs 56%); and they were less likely to receive treatment (67% non-hsp without treatment vs 47% HSP). 3.4 Relevance of non-iga immunoreactants in patients with suspected HSP and LCV+ biopsies Table 3 details the specific DIF immunoreactants observed in the various clinical categories of patients with suspected HSP and histologically confirmed LCV. All DIFpos patients had extracutaneous manifestations (27/27,100%), whereas the majority of DIFneg had only cutaneous manifestations (29/36, 81%). Overall, 19/27 (70%) HSP patients showed DIF immunoreactivity with IgA only, and IgM was the next most commonly seen antibody (8/27, 6/27 IgA + IgM+ and 2/27 IgA + IgM + IgG+). IgG reactivity was not seen in the absence of IgA and IgM reactivity. HSP patients with concomitant IgM reactivity were slightly less likely to have resolution of symptoms (3/8, 38% not resolved vs 5/19, 26% not resolved with IgA only). Overall, IgM and/or IgG immunoreactivity was more common in DIFpos cases (8/27, 30%) than in DIFneg/ns cases (5/36, 14%), a finding mostly attributable to the increase in IgM reactivity seen in HSP compared to non-hsp cases. In DIFneg/ns cases, the presence of IgG and/or IgM immunoreactivity was associated with granulomatosis with polyangitis (IgG+) and cryoglobulinemic vasculitis (IgG+, IgM+). Two patients with IgM immunoreactivity only had cutaneous manifestations only, and they subsequently resolved. Patients with extracutaneous manifestations and cutaneous LCV often had non-hsp systemic illness, including granulomatosis with polyangitis, cryoglobulinemic vasculitis secondary to lymphoma, cryoglobulinemic vasculitis not associated with lymphoma, ulcerative colitis, Crohn s disease and lupus erythematosus; and they largely showed C3 immunoreactivity+/ fibrinogen only on DIF examination. Compared to HSP, patients without HSP were slightly less likely to have malignancy (2/36, 6% vs 3/27, 11%) or infection (4/36, 11% vs 7/27, 26%). 3.5 Description of non-specific DIF immunoreactivity observed in HSP-negative cases The majority of cases referred for DIF studies were negative for perivascular IgA reactivity, and yielded either non-specific (139/258, 54%) or negative (90/258, 35%) DIF patterns (229/258, 89% negative, Table 4). Immunoreactivity for non-iga immunoglobulins was infrequent among all groups. In the presence or absence of LCV, fibrinogen positivity +/ C3 was the most common finding. This was more frequently seen in cases with LCV (76% HSP not suspected; 67% HSP suspected). However, even in the absence of LCV, nonspecific DIF patterns were commonly observed (45% HSP not suspected; 47% when HSP suspected), demonstrating the non-specific nature of most DIF findings in the absence of histologically confirmed LCV. 3.6 Final pathologic diagnoses for all cases submitted for DIF as vasculitis Table 5 summarizes the final pathologic diagnoses found in cases included in the study. LCV was present in 105/258 (41%), including 6 cases of urticarial vasculitis (5.7%). HSP represented 27/105 (26%)

4 FEASEL ET AL. 19 FIGURE 1 A, Palpable purpuric rash coalescent into plaques with overlying bulla shown on bilateral ankles and feet. B, Punch biopsy demonstrating perivascular neutrophilic infiltrate with extravasation of erythrocytes and fibrinoid necrosis of blood vessels within the dermis (H&E, 40). C, Higher magnification showing neutrophil karyorrhexis and vascular wall damage (H&E, 200). D, Direct immunofluorescence showing granular deposition of IgA within vessel walls (H&E, 400) of cases with LCV. Most LCV cases were idiopathic and rendered non-specific DIF findings (72/105, 62.6%). Vasculitis mimics (LCV /DIFneg/ns groups) represented 152/258 (59%) of cases and included dermal hypersensitivity/urticarial reactions (27, 18%), spongiotic dermatitis (23, 15%), perivascular dermatitis (19, 12.5%), vasoocclusive disease (16, 10.5%), non-inflammatory purpura (14, 9%) and pigmented purpuric dermatosis (13, 9%). 4 DISCUSSION The histologic diagnosis of cutaneous vasculitis is performed when inflammatory infiltrates within vessel walls are accompanied by fibrin deposition and endothelial cell injury. The term cutaneous leukocytoclastic angiitis endorsed by the Chapel Hill Consensus Conference (CHCC) is used to describe patients with cutaneous LCV without systemic disease. 4 Depending on the age of the lesion, skin biopsies of leukocytoclastic vasculitis can show deposition of IgG, IgA, IgM,or complement C3 by DIF, which has provided historical evidence of immune-complex-mediated pathogenesis. 3 DIF-negative cases may reflect sampling error, immune-complex consumption in older lesions or other vasculitis categories, including anti-neutrophil cytoplasmic antibody (ANCA)-associated or pauci-immune groups. While certain DIF patterns may suggest specific entities, most DIF tests yield nonspecific patterns of reactivity, with the exception of HSP. In this study, we have investigated clinical and pathologic findings of all specimens submitted over 5 years for DIF studies where an accompanying formalin-fixed, paraffin embedded biopsy for H&E evaluation was also available. A global schematic of our methods and findings is found in Figure 2. We found that vasculitis was the clinical indication submitted for 20% of these cases (258 patients). Given vasculitis is an H&E diagnosis and specific DIF findings in this setting are mostly limited to HSP, we reviewed clinical charts of these 258 patients to determine whether real clinical suspicion existed for HSP at time of biopsy. Somewhat surprisingly, in only 36% of cases (95 patients) was there true clinical suspicion for HSP. That said, true clinical suspicion for HSP increased the likelihood of finding LCV on biopsy (66% vs 26%), and the yield of DIF was highest in this group (suspected HSP/LCV+) with 43% positive DIF studies. While this context (suspected HSP) appears to be the most appropriate for DIF

5 20 FEASEL ET AL. TABLE 3 Clinical features and DIF patterns of patients with suspected HSP and LCV+ biopsies (n = 63) No f/u Did not receive No. of patients HSP Clinicopathologic dx other than LCV Rash only Renal GI Joint Infection Malignancy Resolved DIF Immunoreactants IgA Positive/specific DIF (n = 27) IgA + IgM IgA + IgG 0 0 IgA + IgM + IgG IgG 1 0 Granulomatosis with polyangitis 1 (also diffuse alveolar hemorrhage) Negative/nonspecific DIF (n = 36) IgM IgG + IgM 2 0 Cryoglobulinemic vasculitis (2) 0 2 b C3+/ fibrinogen 16 0 Cryoglobulinemic vasculitis, ulcerative colitis, Crohn s disease, lupus erythematosus Negative 15 1 a Abbreviations: DIF, direct immunofluorescence; GPA, Granulomatosis with polyangiitis; HSP, Henoch-Schönlein purpura; LCV, leukocytoclastic vasculitis; SLE, systemic lupus erythematosus. a HSP recurrent LCV without systematic sx, but RBS casts in urine (GPA vs SLE), LCV (1/16/12), kidney bx (5/10/13) glomerulonephritis IgA+, skin, joints, in remission prednisone and plaquenil at 30 mo. b Both cases were monoclonal gammopathy in the setting of the non-hodgkin lymphoma. testing, it represented only 10% of DIF orders in our series, highlighting the potential overutilization of DIF when considering a vasculitic process. While the clinical significance of vascular IgA deposition is well documented, this pattern is not necessarily specific for HSP, 6 8 and one study found perivascular IgA deposition to be more often associated with drug-induced hypersensitivity vasculitis rather than HSP, 9 again questioning the utility of DIF when HSP is not in the clinical differential diagnosis. 10,11 In a 10-year review of IgA-associated small-vessel vasculitis, Linskey et al observed a specificity of 83%, sensitivity of 81%, positive predictive value (PPV) of 84%, and negative predictive value (NPV) of 81%. 11 The authors concluded that there is a strong association between IgA deposits on DIF and HSP, but also highlighted a subset of HSP cases in which vascular IgA deposition in the skin appears negative. Larson et al found similar sensitivity (86%) and specificity (84%) of vascular IgA for diagnosis of HSP, however, the PPV was 48% and NPV was 97%. 10 In this study at our institution, all patients eventually diagnosed with HSP were clinically suspected to have HSP, based on incorporation of clinical and laboratory findings. Every one of these patients had +LCV on H&E-stained sections, and the vast majority (27/28) had IgA+ DIF. Hence, clinical suspicion appears to be highly sensitive (although not specific) and greatly enriches for LCV+ and IgA+ DIF in our experience. Likewise, IgA+ DIF in the setting of LCV is highly specific for HSP in our hands. Overall, our findings are congruent with that in the literature showing IgA as the most frequently reactive immunoglobulin in LCV, and IgG and IgM reactivity as very rare. Additionally, similar to other reports, we found that IgA deposition corresponded to higher rates of systemic involvement and lower rates of disease resolution without treatment. Because IgA+ DIF is the only specific DIF finding observed in LCV, the utility of other immunoreactants warrants examination. In a study by Alalwani et al, which looked at various causes of clinical vasculitis, the combined findings of LCV and IgA immunoreactivity by DIF selected for a group of patients who were more likely to have internal organ involvement. 5 Indeed, IgA deposition was the only immunoglobulin associated with renal and GI involvement in this study, which included numerous HSP patients. Similarly, Barnadas et al found IgA to be the most frequently detected immunoglobulin in LCV patients overall, and showed its association with renal involvement. 6 Nevertheless, in 2 previously reported studies, IgM deposition was found to be associated with a higher risk of renal involvement, primarily in adult patients 12,13 ; however, these findings were not reproducible in children. 14 Importantly, in this study, the presence of immunoreactants other than IgA in the setting of histologically confirmed LCV did not seem to predict systemic or persistent disease, although the number of cases with IgG or IgM immunoreactivity was low. This further suggests these reagents may not be needed/useful in the setting of vasculitis. The clinical significance of IgM DIF positivity in the setting of LCV and IgA DIF positivity may warrant further investigation in a larger cohort of patients. 15 As expected, non-specific DIF findings (DIFns) were more common in the setting of LCV (83%) compared to LCV cases (49%). Similarly, negative DIF (DIFneg) findings were more common in the absence of LCV (51%) compared to LCV+ cases (17%). With the

6 FEASEL ET AL. 21 TABLE 4 DIF findings in IgA/HSP-negative cases (DIF-neg/ns) (n = 229/258, 89%) HSP suspected (n = 95) Neg/NS DIF (n = 68) HSP not suspected (n = 163) Neg/NS DIF (n = 161) LCV+ (n = 36) LCV (n = 32) LCV+ (n = 41) LCV (n = 120) Totals (229 cases) Negative DIF 8 (22) 15 (47) 5 (12) 62 (52) 90 (39) Non-specific DIF 28 (78) 17 (53) 36 (88) 58 (48) 139 (61) IgG (2) 0 1 (.4) IgM 1 (3) 0 1 (2) 1 (1) 3 (1) C3 14 (39) 7 (22) 14 (34) 16 (13) 51 (22) IgG + C3 1 (3) 0 1 (2) 1(1) 3 (1) IgM + C3 1 (3) 0 2 (5) 6 (5) 9 (4) IgG + IgM + C3 1 (3) 0 2 (5) 1(1) 4 (2) IgG + IgM 1 (3) (.4) Fibrinogen +/ Ig/C3 24 (67) 15 (47) 31 (76) 54 (45) 124 (54) Abbreviations: DIF, direct immunofluorescence; HSP, Henoch-Schönlein purpura; LCV, leukocytoclastic vasculitis. TABLE 5 Final diagnoses for all 258 cases with clinical suspicion for vasculitis sent for DIF testing HSP suspected (n = 95) HSP not suspected (n = 163) DIFpos = 27 DIFneg/ns = 68 (DIFneg = 23, DIFns = 45) LCV+ 27 (27 HSP) 36 (34 LCV, 2 UV) 1 paraneoplastic IgAV DIFpos = 2 DIFneg/ns = 161 (DIFneg = 67, DIFns = 94) Total 41 (37 LCV, 4 UV) 105/258 (41%) LCV a 1 LIGA 120 b 153/258 (59%) Abbreviations: HSP, Henoch-Schönlein purpura; IgAV, IgA vasculitis; LCV, leukocytoclastic vasculitis; LIGA, linear IgA disease; UV, urticarial vasculitis. a Non-inflammatory purpura (8), dermal hypersensitivity (5), perivascular dermatitis (5), Pigmented purpuric dermatoses (PPD) (4), spongiotic dermatitis (3), interface dermatitis (2), vaso-occlusive (2) and other (3). b Other (22), dermal hypersensitivity/urticaria (22), spongiotic dermatitis (20), perivascular dermatitis (14), vaso-occlusive (14), PPD (9), non-inflammatory purpura (6), interface dermatitis (5), neutrophilic dermatosis (4), cryoglobulinemia (1), EN (1), LV (1) and MVV (1). FIGURE 2 Flowchart illustrating overall schematic of the study, including diagnostic classifications and final diagnoses based on clinical suspicion for Henoch-Schönlein purpura, presence of leukocytoclastic vasculitis on H&E-stained slides and direct immunofluorescence results in 258 cases

7 22 FEASEL ET AL. exception of Type 1 essential cryoglobulinemia, none of the non-lcv diagnoses in our patients are known to have specific DIF findings. It is worth noting that of 17 cases with IgM immunoreactivity, none were because of the presence of cryoglobulins, while 5/16 cases with vaso-occlusive disease were positive for IgM (all 16 showed intravascular thrombi on H&E). In summary, a high percentage of DIF biopsies are performed for vasculitis. LCV is common when HSP is suspected, and the contribution of DIF is high in this setting. DIF often yields negative/nonspecific results regardless of H&E findings when HSP is not clinically suspected. When HSP is clinically suspected and LCV is histopathologically confirmed, the yield of DIF is relatively high, and IgA +/ IgM may be used as the sole immunoreactant(s). Given that clinical sensitivity for LCV was 41%, and histologic sensitivity 100%, DIF utilization could be decreased by 60% by first triaging cases with routine histopathologic examination. As the analytic sensitivity and specificity of DIF is dependent on timing of biopsy, we recommend the following in addition to H&E evaluation of lesional tissue: (1) a single peri-lesional biopsy specimen is procured and sent to the laboratory in saline or Michels solution, (2) frozen and held until pathologist review of the H&E-stained slides of lesion tissue and (3) processed for DIF testing based on the presence of LCV. Realtime discussions with clinicians following routine microscopic examination, with triage of potential LCV/HSP cases should ensue. Our data strongly support that this work flow process would safely result in better utilization of laboratory resources and significant savings. ORCID Steven D. Billings Jennifer S. Ko 4. Jennette JC, Falk RJ, Bacon PA, et al revised international Chapel Hill consensus conference nomenclature of vasculitides. Arthritis Rheum. 2013;65(1): Alalwani M, Billings SD, Gota CE. Clinical significance of immunoglobulin deposition in leukocytoclastic vasculitis: a 5-year retrospective study of 88 patients at Cleveland Clinic. Am J Dermatopathol. 2014; 36(9): Barnadas MA, Perez E, Gich I, et al. Diagnostic, prognostic and pathogenic value of the direct immunofluorescence test in cutaneous leukocytoclastic vasculitis. Int J Dermatol. 2004;43(1): Magro CM, Crowson AN. A clinical and histologic study of 37 cases of immunoglobulin A-associated vasculitis. Am J Dermatopathol. 1999;21(3): Magro CM, Roberts-Barnes J, Crowson AN. Direct immunofluorescence testing in the diagnosis of immunobullous disease, collagen vascular disease, and vascular injury syndromes. Dermatol Clin. 2012; 30(4): viii. 9. Al-Mutairi N. Spectrum of cutaneous vasculitis in adult patients from the Farwaniya region of Kuwait. Med Princ Pract. 2008;17(1): Larson AR, Granter SR. Utility of immunofluorescence testing for vascular IgA in adult patients with leukocytoclastic vasculitis. Am J Clin Pathol. 2014;142(3): Linskey KR, Kroshinsky D, Mihm MC Jr, Hoang MP. Immunoglobulin- A-associated small-vessel vasculitis: a 10-year experience at the Massachusetts General Hospital. J Am Acad Dermatol. 2012;66(5): Belli AA, Dervis E. The correlation between cutaneous IgM deposition and renal involvement in adult patients with Henoch-Schonlein purpura. Eur J Dermatol. 2014;24(1): Takeuchi S, Soma Y, Kawakami T. IgM in lesional skin of adults with Henoch-Schonlein purpura is an indication of renal involvement. JAm Acad Dermatol. 2010;63(6): Tirado-Sanchez A, Bonifaz A, Ponce-Olivera RM. IgM in lesional skin is indicative of renal involvement in adults with Henoch-Schonlein purpura but not children. J Am Acad Dermatol. 2011;64(6): Demirkesen C. Approach to cutaneous vasculitides with special emphasis on small vessel vasculitis: histopathology and direct immunofluorescence. Curr Opin Rheumatol. 2017;29(1): REFERENCES 1. Chen KR, Carlson JA. Clinical approach to cutaneous vasculitis. Am J Clin Dermatol. 2008;9(2): Carlson JA, Ng BT, Chen KR. Cutaneous vasculitis update: diagnostic criteria, classification, epidemiology, etiology, pathogenesis, evaluation and prognosis. Am J Dermatopathol. 2005;27(6): Carlson JA, Chen KR. Cutaneous vasculitis update: small vessel neutrophilic vasculitis syndromes. Am J Dermatopathol. 2006;28(6): How to cite this article: Feasel P, Billings SD, Bergfeld WF, Piliang MP, Fernandez AP, Ko JS. Direct immunofluorescence testing in vasculitis A single institution experience with Henoch-Schönlein purpura. J Cutan Pathol. 2018;45: