Mixing Studies in Patients With Prolonged Activated Partial Thromboplastin Time or Prothrombin Time

Abstract

Background: Patients presenting for surgery may have isolated or combined prolonged activated partial thromboplastin time (aPTT) and/or prothrombin time (PT). In patients not receiving anticoagulants or with no identifiable cause for abnormal clot formation, a mixing study is performed. The index of circulating anticoagulant (ICA) has been used to predict the presence of an inhibitor, usually a lupus anticoagulant.

Methods: We retrospectively reviewed the results of mixing studies performed at Northwestern Memorial Hospital, between January 1, 2010 and February 29, 2012. We determined the number of samples that normalized or remained prolonged, the clotting factors associated with prolonged test results, and the presence of coagulation inhibitors. We calculated the ICA in the samples with prolonged aPTT and PT to determine its ability to predict a lupus anticoagulant. The primary comparison of interest was the diagnostic utility of the ICA at cutoff values of 11% for predicting the presence of lupus anticoagulant.

Results: There were 269 mixing studies performed: 131 samples with prolonged aPTT; 95 with prolonged PT; and 43 with both prolonged aPTT and prolonged PT. Of the samples with a prolonged aPTT, 55 of 131 (42%) normalized, 36 of 131 (27%) partially corrected, and 40 of 131 (31%) remained prolonged. Thirty-three of 95 samples (35%) with prolonged PT normalized, while 62 of 95 (65%) remained prolonged. Eight of 43 (19%) mixing studies of patients with prolonged PT and aPTT normalized; the aPTT normalized, but the PT remained prolonged in 17 of 43 (39%); the PT normalized, but the aPTT remained prolonged in 7 of 43 (16%); and both tests remained prolonged in 11 of 43 (26%) samples. Prolongations in the aPTT were primarily associated with low activities of CF XII, while the majority of the prolongations in PT were secondary to low activities in CF VII. Combined prolongations were secondary to deficiencies in both the intrinsic and extrinsic as well as the common pathways. An ICA >11% had 100% (95% CI, 59%-100%) sensitivity, 53% (95% CI, 35%-70%) specificity, and 77% (95% CI, 62%-92%) accuracy in predicting the presence of lupus anticoagulant in patients with prolonged aPTT.

Conclusions: Normalization of the aPTT and PT in a mixing study was associated with low clotting factor activity. The ICA may be helpful in predicting the presence of a lupus anticoagulant. As anesthesiologists take ownership of the perioperative surgical home, we need to understand the clinical implications of the results of mixing studies.