Gamma Glutamyl Transferase Activity Has Limited Utility in Assessment of Alkaline Phosphatase Elevations

Abstract

Abstract Introduction As part of an ongoing effort to improve healthcare value for patients, laboratories increasingly implement test utilization review. Alkaline phosphatase (ALP) isoenzymes (hereafter: isoenzymes) testing distinguishes the various ALP isoforms to explain elevations in total serum ALP. Gamma glutamyl transferase activity (GGT) has served as a proxy for total ALP elevations attributable to the hepatic isoform given that both are membrane-bound proteins with a shared mechanism of release. We assessed the utility of GGT in evaluating isoenzymes requests. Methods We obtained 8 years of isoenzymes results and identified same-patient GGT measurements obtained within 7 days. We assessed the ability of GGT to predict elevations in hepatic, bone, intestinal, and nonhepatic ALP isoforms overall. We generated ROC curves and calculated sensitivity and specificity using our in-house reference limits for GGT. Results GGT as a predictor of hepatic isoform elevation had an area under the ROC curve (AUC) of 0.68, and GGT activity above the upper reference limit was 46.6% sensitive and 85.0% specific for hepatic ALP elevation. GGT activity as a predictor of nonhepatic isoform elevation had an AUC of 0.52, and GGT within reference limits was 59.8% sensitive and 46.4% specific for elevation in a nonhepatic ALP isoform. In 133 individuals with hepatic isoform elevations, 93 had a concurrent elevation in a nonhepatic ALP isoform. Conclusion GGT was reasonably specific but insensitive for hepatic ALP isoform elevation and was a poor predictor of ALP isoform elevation overall, suggesting that its usefulness in evaluating isoenzymes orders is limited.