The utility of screening for deep venous thrombosis in asymptomatic, non-ambulatory neurosurgical patients

Objectives: Decisions regarding deep venous thrombosis (DVT) prophylaxis are complicated in neurosurgical patients because of the potential for catastrophic bleeding complications. Screening with venous duplex ultrasound (VDUS) may improve outcomes, but can strain hospital resources. Since there is little data to guide VDUS surveillance, we investigated the utility of a comprehensive VDUS screening program in neurosurgical patients. Methods: Medical records of patients admitted to the neurosurgical service at a university-affiliated hospital from October 2007 through January 2010 who underwent weekly VDUS of the lower extremities until ambulatory or discharged were retrospectively reviewed. Demographics, comorbidities, interventions, and use of DVT prophylaxis were recorded. All patients in this study were asymptomatic for clinical evidence of DVT. When DVT was identified, VDUS reported its location and progression. Results: One hundred seventy-four consecutive patients were screened according to the established protocol. They had 312 VDUS studies, 68 (21.8%) of which were positive in 40 (23%) unique patients; 10 were bilateral and two catheter-related. There were no documented pulmonary emboli in this series. Seventeen patients (37.7%) had isolated calf DVT, four of which were bilateral (totaling 21 thrombi), and 9 (20%) had coexistent thrombi in calf and proximal veins. Of the 21 isolated calf DVTs, 15 had follow-up studies and two progressed to the popliteal or ileofemoral vein on follow-up (13.3%). Mechanical prophylaxis was uniformly utilized, but chemical prophylaxis varied based on surgeons' assessment of bleeding risk. DVT developed in 19.3% (28/145) of patients receiving prophylactic medication (unfractionated heparin or low-molecular weight heparin) and 41.4% (12/29) receiving no chemoprophylaxis (P < .001). The only patient characteristic that correlated with DVT risk was a body mass index <30 (9.1% vs 29.4%, P = .01). Conclusions: Despite the uniform application of mechanical DVT prophylaxis and the use of chemoprophylaxis in a majority of patients, we found a 23% incidence of DVT in these hospitalized, nonambulatory, neurosurgical patients. No patients with isolated calf DVT had an embolic complication but 13.3% progressed proximally in short-term follow-up. While chemical prophylaxis significantly reduced DVT risk, no factor was sufficiently predictive to exclude patients from screening. These data substantiate the importance of full leg VDUS screening and maximizing DVT prophylaxis in this high risk population. (J Vase Surg 2011;53:1309-15.)