Validating the Clinical Value of Temporal Artery Biopsy.

Background: Giant cell arteritis (GCA) is a potentially devastating disease that may require treatment with high-dose steroids. Traditionally, diagnosis requires patients to meet at least 3 of 5 clinical criteria, one of which is a positive temporal artery biopsy (TAB). Vascular surgeons are often asked to perform TAB though it is not necessarily required for diagnosis or management. This study aimed to determine if TAB results altered management of patients with a concern for GCA by changing steroid use postoperatively in our health care system.

Methods: A retrospective review at a single-center tertiary care hospital was performed between 2007 and 2018. The inclusion criteria were patients greater than 18 years old with complete steroid treatment records who underwent a temporal artery biopsy due to concern for GCA. Steroid use and duration of treatment both pre- and post-operative were collected and analyzed.

Results: Eighty-three of 117 cases reviewed met inclusion criteria. Ninety-one percent (76) of patients had a negative biopsy. Twenty-nine percent (23) of negative biopsies met criteria for GCA prior to biopsy. Of those with a negative biopsy, steroids were continued in 68% (52) of patients after 30 days, 49% (37) after 90 days and 45% (34) after 180 days. Steroids were never started in 11% (6). One patient with a positive biopsy was discontinued on steroids due to intolerance. There was no statistically significant difference in duration of steroids between those with a positive and negative biopsy (average 610 and 787 days respectively; P = 0.682). Average follow up was 33 months.

Discussion: The duration of steroid use for patients with concern for GCA was not found to be altered by the performance of a TAB at our institution. Given the extremely low yield and absence of impact on steroid duration, TAB is not a useful diagnostic test at our institution. Similar reviews are recommended to determine the utility of TAB at other institutions that may differ in patient population or prescribing practices.