NAPLES, FLA. – Upper extremity deep vein thromboses accounted for a surprising 64% of 316 DVTs reported in a prospectively screened cohort of critically ill surgical and trauma patients.
Contrary to expectations, anticoagulation did not affect outcomes. In 77 patients with 123 upper extremity DVTs (UEDVTs), the clot resolution rate was 59% with no anticoagulation, 60% with prophylactic anticoagulation, and 61% with therapeutic anticoagulation (P = .976), Dr. Darren Malinoski said at the annual meeting of the Eastern Association for the Surgery of Trauma.
The American College of Chest Physicians (ACCP) recommends therapeutic anticoagulation for UEDVT, and recommends against the removal of a necessary, patent, central venous catheter. If a catheter is removed, however, the duration of anticoagulation should not be decreased.
Dr. Malinoski, director of the surgical ICU at the University of California Medical Center in Irvine, and his colleagues prospectively followed all surgical ICU patients from January 2008 to May 2010. A standardized DVT-prevention protocol was utilized, and screening duplex ultrasonography exams were obtained within 48 hours of admission and then weekly.
In all, 316 DVTs were identified in 198 patients, of which 201 (64%) occurred in the upper extremities of 129 patients, he said. Data from at least one follow-up duplex ultrasound was available in 77 patients with 123 UEDVTs, and these results formed the basis of the analysis.
The average number of UEDVTs in the 77 patients was 1.6 (range, 1-5), and average time to diagnosis was 19 days after admission. The internal jugular was the most common site and 72% were nonocclusive. Some 70% of UEDVTs occurred in men, 35% in trauma patients, and 11% in those with a history of cancer; 2% were associated with fractures. In all, 64% of the UEDVTs were catheter associated.
A total of 60% resolved prior to discharge, and only 2.4% embolized, Dr. Malinoski said. Notably, 46% of the catheter-associated UEDVTs were associated with double- and triple-lumen catheters, 30% with percutaneously inserted central catheters (PICCs), and 24% with Cordis or hemodialysis catheters.
A comparison of catheter-related vs. noncatheter-related DVTs found that line removal was significantly associated with clot improvement on the next duplex (55% vs. 17%; P = .04), he said.
In multivariate analysis, independent predictors of clot resolution on final duplex were DVT location in the arm (with an odds ratio of 4.1, compared with the internal jugular), and time between first and final duplex exam (OR, 1.05 per day), Dr. Malinoski said.
Invited discussant Dr. Susan Brundage of New York University Langone Medical Center asked how clinicians should reconcile the data with the ACCP guidelines and medical/legal pressures, and expressed surprise at the high rate of DVTs associated with PICC lines.
"We are using PICC lines much more often, as we think of them as having less morbidity and mortality," she said. "Given that one of your PEs [pulmonary embolisms] was actually associated with a PICC line, should we change our screening practices and our diligence in these PICC lines we are placing in our critically ill patients?’’
Dr. Malinoski responded that he does not like PICC lines for infection reasons, and recommended that surgical patients with upper extremity central venous catheters be screened for the presence of a clot.
"When you find a clot, I think the only thing you need to do is remove the catheter, as opposed to treating them with massive anticoagulation in this era and with these types of patients," he said.
Dr. Malinoski pointed out that ACCP’s recommendation for full anticoagulation is based on patients’ requiring long-term catheters, typically to deliver chemotherapy. Removing and replacing a catheter isn’t feasible in these patients, and also could put them at risk for a reclot.
Dr. Malinoski and Dr. Brundage disclosed no relevant conflicts of interest.