The Striking Incidence of Venous Thromboembolism in Hospitalized Patients with Necrotizing Pancreatitis
James Lopes*, Nicholas J. Zyromski, Yuriy Zhukov, Attila Nakeeb, Henry a. Pitt, Keith D. Lillemoe, C. Max Schmidt, Thomas J. Howard
Surgery, Indiana University, Indianapolis, IN
Background: Acute pancreatitis is a substantial medical problem, accounting for more than 220,000 hospital admissions yearly in the US. Twenty percent of these patients will develop necrotizing pancreatitis (NP), which is characterized by pancreatic and peripancreatic necrosis. Venous thromboembolism (VTE) is also a major national health concern, with an incidence in all hospitalized patients approaching 10%. Surprisingly few data are available regarding VTE in NP. The aim of this study was to determine the incidence of VTE in patients with NP.
Methods: Retrospective review (1992-2007) of patients with NP treated at a University Hospital. The diagnosis of VTE was confirmed using standard radiographic criteria for duplex ultrasonography (U/S) or helical Computed Tomography (CT). Chi square was used to compare discrete variables, accepting p<0.05 as statistically significant.
Results: 171 patients with NP were identified. Of these, 96 patients (56%) had radiographic evidence of VTE. Fifty-one (53%) of patients with VTE had thrombi identified in multiple locations for a total of 152 DVT identified in this series. Four patients (2%) had documented pulmonary emboli. DVT was found more frequently in abdominal vasculature (n=121/78%) as compared to the extremities (n=31/20%) (p<0.001). Fourteen out of sixteen (88%) patients with an upper extremity DVT had an associated intravenous catheter. The following table details DVT location.
Conclusion: These data demonstrate that: 1) 56% of patients with NP have associated DVT and that 53% of these patients had more than one thrombus, 2) Patients with NP are more likely to develop abdominal DVT (78%) compared to extremity DVT (20%), and 3) 88% of UE DVT are associated with catheter placement. From these data, we conclude that VTE is a significant problem in patients with NP and that aggressive VTE prophylaxis should be implemented.
LOCATION | Splenic | SMV/PV | UE | LE | Catheter Associated | PE |
# (%) | 63 (40) | 58 (37) | 16 (10) | 15 (10) | 14 (88*) | 4 (2) |
SMV/PV - superior mesenteric/portal vein, UE- upper extremity, LE- lower extremity, PE - pulmonary embolus
*88% of UE DVT were associated with an intravenous catheter.