How can the diagnostic value of head and neck biopsies be increased in Wegener's granulomatosis: a clinicopathologic study of 49 biopsies in 21 patientsHead and neck biopsies usually have a low diagnostic value in Wegener's granulomotosis (WG). On the basis of 49 biopsies obtained from 21 WG patients at diagnosis from various sites, i.e. nose (29), paronosal sinus (7), oral cavity (4), larynx (4), conjunctiva (3) and external ear (2), we described the suggestive histological features and studied the diagnostic potential of the biopsy size, anaesthesia method (general (GA) or local (LA)), anatomic region of the biopsy, number of sections, and presenting macroscopic manifestations. Associated granulamatous inflammation (scattered giant cells, 28.5% of biopsies; poorly-formed granulomas, 28.5%), necrosis (neutrophilic microabscesses, 16.3%; geographic necrosis, 18.3%), angiitis (leakocytoclastic, 10%, necrotizing, 12%; and granulomatous, 6%) which confirmed the diagnosis were only present in 18.3% of the biopsies (28.5% of the patients). We think it is possible to propose a "WG-compatible" diagnosis when at least one of these histological features is present (24.5% of biopsies, 26 % of patients in our study).We found that it was always better to perform biopsies targeted on macroscopic lesions. When there was no lesion, samples from paranasal sinuses obtained under GA had the highest diagnostic value in the head and neck region, whereas 90% of nasal systematic biopsies performed under LA were nonspecific. Moreover, we demonstrated that performing two further sections increased the sensitivity of histological examination by 7%.

Head and neck biopsies usually have a low diagnostic value in Wegener's granulomatosis (WG). On the basis of 49 biopsies obtained from 21 WG patients at diagnosis from various sites, i.e. nose (29), paranasal sinus (7), oral cavity (4), larynx (4), conjunctiva (3) and external ear (2), we described the suggestive histological features and studied the diagnostic potential of the biopsy size, anaesthesia method (general (GA) or local (LA)), anatomic region of the biopsy, number of sections, and presenting macroscopic manifestations. Associated granulomatous inflammation (scattered giant cells, 28.5% of biopsies; poorly-formed granulomas, 28.5%), necrosis (neutrophilic microabscesses, 16.3%; geographic necrosis, 18.3%), angiitis (leukocytoclastic, 10%; necrotizing, 12%; and granulomatous, 6%) which confirmed the diagnosis were only present in 18.3% of the biopsies (28.5% of the patients). We think it is possible to propose a "WG-compatible" diagnosis when at least one of these histological features is present (24.5% of biopsies, 26% of patients in our study). We found that it was always better to perform biopsies targeted on macroscopic lesions. When there was no lesion, samples from paranasal sinuses obtained under GA had the highest diagnostic value in the head and neck region, whereas 90% of nasal systematic biopsies performed under LA were nonspecific. Moreover, we demonstrated that performing two further sections increased the sensitivity of histological examination by 7%.