Thrombocytosis in patients with biopsy-proven giant cell arteritis

Abstract

Objective: To compare the platelet counts, complete blood counts, and Westergren sedimentation rates (WESR) of patients with a biopsy positive for giant cell arteritis (GCA) with those of patients with negative biopsies.

Design: Retrospective, case-control series.

Method: The medical records of 91 consecutive patients who underwent temporal artery biopsy for possible GCA, over an 8-year period, were reviewed.

Main outcome measures: Values obtained for the complete blood counts, including platelet counts, WESR, and biopsy results of 91 consecutive patients undergoing temporal artery biopsy were analyzed.

Results: The mean platelet count of 47 patients with positive temporal artery biopsies (433 x 10(3)/microl) was significantly higher than that of 44 patients with negative temporal artery biopsies (277 x 10(3)/microl), P < 0.0001. There was no statistically significant difference in the mean WESR between the biopsy-positive (82 mm/hour) and biopsy-negative (70 mm/hour) groups, P = 0.12. The sensitivity of an elevated WESR for biopsy-positive patients was greater (79%; 95% confidence interval [CI], 64%-89%) than that of an elevated platelet count (57%; 95% CI, 42%-72%). However, the specificity (91% [95% CI, 78%-97%] versus 27% [95% CI, 15%-43%]), positive predictive value (87% [95% CI, 70%-96%] versus 54% [95% CI, 41%-66%]), and negative predictive value (67% [95% CI, 53%-78%] versus 55% [95% CI, 32%-76%]) favored an elevated platelet count compared with WESR, or to the combination of platelets and WESR, as a better test for diagnosing GCA in the 91 patients studied. The area under the receiver operating characteristic function for platelets (0.72) was greater than that for WESR (0.59) or the combination of platelets and WESR (0.65).

Conclusions: In patients suspected of having GCA, an elevated platelet count greater than 400 x 10(3)/microl is a useful marker of a positive temporal artery biopsy.