Prediction tool for thrombi associated with peripherally inserted central catheters.

The upper extremity deep vein thrombosis rate is increasing at the same time that the rate for insertions of peripherally inserted central catheters is on the rise. There is little information on whether the established risk factors for lower extremity deep vein thromboses are effective to predict the occurrence of upper extremity deep vein thrombosis. The purpose of this study was to identify patients at highest risk for upper extremity deep vein thrombosis in order to initiate effective prophylaxis. A retrospective review was undertaken of medical records of all patients with peripherally inserted central catheters inserted in a 6-month period at a Midwestern US hospital. Of the 233 charts reviewed, 17 (7.3%) recorded an upper extremity deep vein thrombosis during the patient's hospital stay. Of the multiple factors identified with deep vein thrombosis in the literature, a weighted risk factor measure, the upper extremity deep vein thrombosis prediction tool, was developed. Sensitivity of the instrument for upper extremity deep vein thrombosis is high (88%), as are its specificity (82%) and negative predictive value (99%), whereas the positive predictive value is low (28%). The total percentage of cases correctly classified is 82%. Further testing is indicated on a larger sample to extend the validity of this instrument.