Shrinkage: Fact or Fiction?

The diagnosis of giant cell arteritis (GCA) is confirmed by finding characteristic histological changes in the disease in an arterial segment, such as the superficial temporal artery. The length of the arterial specimen is crucial because of the presence of "skip lesions"¹ (areas of no inflammation) in GCA. Recently, there has been discussion concerning the required length of an arterial specimen to securely confirm or exclude the diagnosis of GCA.^2,3 It is generally accepted that 20 mm is an adequate specimen length for a unilateral biopsy. In a "Letter to the Editor" in the Journal of the American Medical Association, the issue has been raised as to whether this is a prefixation or postfixation measurement guideline and whether formalin fixation reduces the effective length of a temporal artery biopsy specimen.⁴ This is a potentially important question since the accuracy of the histological diagnosis of GCA is believed to be strongly correlated to the length of the fixed artery specimen that is available for pathological assessment. We report the first available data that directly address this question.

All patients undergoing temporal artery biopsy at Wills Eye Hospital (Philadelphia, Pa) between January 15 and July 1, 1999, were prospectively enrolled. Immediately upon excision of the temporal artery segment and prior to the placement of the specimen in 10% neutral buffered formalin fixative, the length of the specimen was measured in the operating room by the surgeon. It was later remeasured after fixation before sectioning by the pathologist, who was masked as to the initial measurement. Both sets of measurements were done using the same standardized millimeter rulers with measurements being made to the nearest tenth of a millimeter.

Twenty-eight temporal artery biopsies were performed. Fifteen were positive and 13 were negative for shrinkage. The average prefixation length was 28.4 mm (SD, 6.0 mm), whereas the average postfixation length was 26.0 mm (SD, 5.5 mm), representing a mean shrinkage of 2.4 mm (95% confidence interval, 1.6-3.1 mm; P<.001, 2-tailed test) or a mean reduction of about 8% (6%-13%). Twenty-two biopsy specimens were measured between 3 and 6 hours of being placed in fixation, and 6 were measured within 12 hours of fixation. No statistical difference was detected between these 2 groups.

Our study shows that shrinkage of temporal artery biopsy specimens does occur following formalin fixation. This observation suggests that clinicians may need to take slightly longer temporal artery biopsy specimens than previously recommended to insure that shrinkage due to chemical fixatives does not reduce diagnostic accuracy in the investigation of this treatable but potentially sight-threatening condition. While a 20-mm sample of temporal artery is generally considered to be an adequate biopsy specimen, because of the presence of skip lesions, the shorter the biopsy specimen, the more important the shrinkage factor is to diagnosis.

Corresponding author: Helen V. Danesh-Meyer, MD, Discipline of Ophthalmology, University of Auckland, Private Bag 92019, Auckland, New Zealand (e-mail: h.daneshmeyer@auckland.ac.nz).