A new retrospective analysis has found 1.5-2 cm to be the optimal length of a temporal artery biopsy for detecting giant cell arteritis. Longer lengths did not yield enough improvement in diagnosis to justify the increased risk of complications. The length calculation accounts for post-fixation shrinkage.
The study, published Aug. 20 in Lancet Rheumatology, represents an “important contribution” to help with the diagnosis of giant cell arteritis when a decision has been made to perform a temporal artery biopsy, according to authors of an editorial accompanying the study.
Giant cell arteritis is an inflammatory condition of medium and large arteries, usually affecting the aorta and proximal aorta. Diagnosis includes a combination of clinical presentation and imaging or histology via a temporal artery biopsy, but the optimal tissue length for a biopsy has not been established. Longer lengths were initially considered best because inflammation can be non-uniform, and a shorter length could therefore raise the risk of a false negative if it contained few signs of inflammation.
Studies in the 1990s and early 2000s concluded that biopsies 2-5 cm in length were optimal. But later studies determined that a minimum of just 0.5 cm was necessary. The European League Against Rheumatism updated its recommendations in 2018 and the British Society for Rheumatology followed suit in 2020, both with a suggested minimum length of 1.0 cm. Despite these guidances, the optimal biopsy length beyond 1 cm remains unknown.
For the study, first author Raymond Chu, MD, of the University of Alberta Hospital, Edmonton, reviewed electronic medical records of all patients who underwent temporal artery biopsies in Alberta between Jan. 1, 2008, and Jan. 1, 2018. A single pathologist reviewed all positive findings to ensure uniformity of pathological interpretation. When the reviewer disagreed with the initial diagnosis, researchers removed the result from the analysis.
The study included 1,203 biopsies from 1,176 patients at 22 institutions. A total of 13 positive biopsies were removed following pathologist review. The median biopsy length was 1.3 cm. Median erythrocyte sedimentation rate (ESR) was 41 mm/hour, and median C-reactive protein (CRP) level was 14.7 mg/L. Univariate analyses found associations between positive biopsy and increased age (75.3 vs. 71.3 years; P < .0001), increased ESR (57 vs. 36 mm/hour; P < .0001), lower CRP (12.1 vs. 41.8 mg/L; P < .0001), and longer biopsy length (1.6 vs. 1.2 cm; P = .0025).
In a multivariate analysis, the only variables associated with a positive biopsy were age (adjusted odds ratio [aOR], 1.04; P = .0001), lower CRP levels (aOR, 1.01; P = .0006), and biopsy length (aOR, 1.22; P = .047). The researchers then stratified the sample by biopsy length, using categories of < 0.5 cm, 0.5-1.0 cm, 1.0-1.5 cm, 1.5-2.0 cm, 2.0-2.5 cm, and ≥ 2.5 cm. They identified the two top change points according to the Akaike information criterion as 1.5 cm and 2.0 cm, but only 1.5 cm was statistically significant (≥ 1.5 versus < 1.5; OR, 1.57; P = .011).
Accounting for an average 8% contraction following excision, the researchers recommend an optimal pre-fixation biopsy length of 1.5-2.0 cm.
Some previous studies had suggested no association between increased sample length and false negatives, but they were based on small sample sizes. The current study is limited by its retrospective design and lack of treatment data. The lack of marked inflammation in the sample population suggests that patients were frequently treated empirically with glucocorticoids, and this could have increased the frequency of false negative biopsies, the researchers said.
Dr. Frank Buttgereit
In the accompanying editorial, Frank Buttgereit, MD, of Charité University Medicine in Berlin and Christian Dejaco, MD, PhD, of the Medical University of Graz (Austria) point out that ultrasound is now often used for the diagnosis of giant cell arteritis, following clinical examination and laboratory testing. When it has been determined that biopsy is necessary, they said that it is imperative that the harvest be carried out by an experienced physician, and the new study provides a useful contribution through its clear recommendation for biopsy length.
Dr. Christian Dejaco
The authors of the editorial also point out the importance of experienced pathologists, but interpretation is subject to inter- and intraobserver variability, as shown in a previous study that found that ultrasound and histology have similar reliability.
The study received no funding. Several authors reported receiving personal fees from Hoffmann-LaRoche and serving as site primary investigators for industry-sponsored vasculitis trials.
SOURCE: Chu R et al. Lancet Rheumatol. 2020 Aug 20. doi: 10.1016/S2665-9913(20)30222-8.