Venous Thromboembolism in Pancreatic Cancer: Unraveling the Paradox of Mortality and Hospitalization Outcomes

Abstract

Background: Pancreatic cancer is known for its aggressive nature and poor prognosis, with a five-year survival rate of less than 10%. Patients diagnosed with pancreatic cancer face a myriad of complications, one of the most significant being venous thromboembolism (VTE). VTE primarily includes deep venous thrombosis (DVT) and pulmonary embolism (PE), and it represents a major cause of morbidity and mortality in this patient population. The increased risk of VTE in pancreatic cancer patients is attributed to multiple factors. Malignancy itself is a well-recognized hypercoagulable state, largely due to the secretion of pro-coagulant factors by tumor cells. Additionally, pancreatic cancer can induce a systemic inflammatory response, leading to widespread activation of pro-coagulant factors and platelets. This hypercoagulable state is further exacerbated by factors such as immobilization, surgery, chemotherapy, and the presence of central venous catheters, which are common in these patients. Several studies have highlighted the prevalence of VTE in pancreatic cancer patients, suggesting that up to 20-30% of these patients may develop VTE at some point during their disease course. The management of VTE in pancreatic cancer patients is complex, as it requires a balance between preventing thrombosis and minimizing the risk of bleeding, particularly in the context of anticoagulant therapy. Methods: We analyzed real-world evidence from the 2021 Nationwide Inpatient Sample database the Agency of Healthcare Research and Quality (AHRQ). The study included all hospitalized pancreatic cancer patients with both a primary diagnosis of venous thromboembolism. We conducted a comparative analysis of demographic characteristics and clinical outcomes between a cohort of pancreatic cancer patients with VTE as a primary diagnosis and those without VTE. Results: In 2021, there were 76,315 patients with a discharge diagnosis of pancreatic cancer. 2,580 patients (3.38%) had a relevant diagnosis of VTE or PE at the time of hospitalization. The mean age for patients in both the VTE and non-VTE groups was 68 years. 51.1 percent of patients in the VTE group were female compared to 46.5 percent of patients in the non-VTE group (p<0.0001). The mortality rate was 8.62% in the non-VTE group, compared to 6.78% in the VTE group (p<0.0001). Furthermore, patients with pancreatic cancer who developed venous thromboembolism (VTE) were found to have no significant difference in mortality compared to those without VTE during the index hospitalization (p<0.179) after adjusting for sex, race, and age. The odds ratio for length of stay in the VTE group was 1.57 (95% CI: 0.75-3.32, p = 0.230). This suggests that there is no statistically significant difference in hospital stay duration between patients with pancreatic cancer who develop VTE and those who do not. The mean total hospitalization charges for the VTE group was $59,556 and $74,249 for the non-VTE group. Discussion: The presence of VTE in hospitalized pancreatic cancer patients is a critical factor that influences clinical management but does not necessarily worsen the short-term prognosis or increase hospitalization duration. Differences in clinical management between the two groups could also contribute to our findings. Patients who develop VTE may receive more intensive monitoring and treatment, potentially improving their overall outcomes. Conversely, variations in the use of prophylactic anticoagulation and differences in the aggressiveness of VTE treatment could impact mortality and length of stay, independent of the presence of VTE. The non-significant difference in length of stay between the VTE and non-VTE groups may reflect hospital practices and discharge criteria rather than patient outcomes. Hospitals with varying policies on discharge timing and criteria can influence length of stay without directly correlating with mortality. Furthermore, differences in post-hospitalization care and follow-up may play a role, as patients discharged sooner might receive better outpatient care, which is not captured in the dataset. Further research should aim to include a sensitivity analysis and more covariates related to disease severity, co-morbidities, and treatment specifics. This can improve the adjustment for confounding factors and provide a more comprehensive understanding of the factors influencing mortality and length of stay.