Clinical outcomes associated with opioid use disorder in hospitalized patients with venous thromboembolism Free

Introduction

Opioid use disorder (OUD) is an emerging significant risk factor for inpatient complications, including thrombi formation, though data is limited. Few studies have suggested that OUD may alter inflammatory and coagulation factors, which may contribute to this finding. Venous thromboembolism, such as deep vein thrombosis (DVT) and pulmonary embolism (PE), are known adverse complications of hospitalized patients. However, clinical outcomes of venous thromboembolism in patients with OUD are not well established in the literature. Through this study, we aimed to characterize complications, interventions, and mortality in this high-risk patient population.

Methods

We conducted a cross-sectional study using the 2018 National Inpatient Sample database to assess the effects of OUD among adult (age > 18 years) hospitalizations on clinical outcomes in DVT and/or PE. OUD and clinical outcomes were identified using ICD-10-CM codes, and mechanical thrombectomy was used with ICD-10 procedure code. The charlson comorbidity index was used to compare comorbidities. Multivariable logistic regression was used to evaluate the association between OUD and clinical outcomes, adjusting for demographics, hospital region, and comorbidities. Descriptive data was used to compare urban and rural hospitals with respect to mechanical thrombectomies, acute right heart strain, and rates of heparin induced thrombocytopenia (HIT). Descriptive data was used to compare rates of mechanical thrombectomies and HIT. Significance of descriptive data was determined by chi square tests. National estimates were generated by the addition of survey weights.

Results

Among 822,695 weighted hospitalizations of patients with DVT and/or PE, 26,760 (3.2%) involved OUD. Patients with OUD had higher rates of Medicaid insurance (48.5% vs. 13.1%) and length of stay (8 vs 5 days) compared to non-OUD hospitalizations (all p<0.0001). After logistical adjustment, OUD was independently associated with increased odds of mechanical thrombectomy of pulmonary artery (aOR 2.950, 95% CI: 1.457-5.970, p=0.003), acute right heart stain/cor pulmonale (aOR 1.915, 95% CI: 1.543-2.375,p <0.0001), heparin-induced thrombocytopenia (aOR 1.694, 95% CI: 1.053-2.723, p=0.03), and hospital mortality (aOR 1.183, 95% CI: 1.015-1.379, p = 0.031). OUD was more common in urban settings (p<0.0001), mechanical thrombectomies were more likely to occur in urban settings (p=0.019), rates of HIT were not significantly different (p=0.769), rates of acute right heart strain were more likely to occur in rural settings (p<0.0001). Among patients who had mechanical thrombectomies, 3% had HIT verses 0.5% of patients without the procedure (P<0.001).

Conclusions

Our study demonstrates that OUD, in patients with DVT and/or PE, was independently associated with significantly increased odds of mechanical thrombectomy of pulmonary artery(s), acute right heart stain/cor pulmonale, and hospital mortality. The high rates of mechanical thrombectomies at urban hospitals may hinder our ability to associate OUD as an independent risk factor. However, even after adjusting for the region of hospital, OUD still increased the odds of mechanical thrombectomy. Additionally, higher rates of acute right heart strain were found in rural settings, yet OUD was associated with increased risk of this complication. Thus, it appears that acute right heart strain may disproportionately affect OUD patients in urban settings, strengthening our association. HIT was significantly more common in patients who had mechanical thrombectomies, thus it appears to be a confounder.

Overall, this data reveals significant adverse clinical outcomes associated with OUD and VTE. A further dive into the underlying mechanisms of these findings is warranted. Opioid induced microvascular damage and endothelial dysfunction may play a role in the pathogenesis of venous thrombosis. Further, in patients that inject opioids, endocarditis and septic embolic may contribute to complication development. Lastly, on a systemic level, reduced access to care and delayed disease presentation may add to the outcomes we described.