MP14-17: Mortality Associated with Sepsis Post-Ureteroscopy Among a US-Based Commercial Population

Naeem Bhojani*, Montreal, Canada, Brian Eisner, Boston, MA, Manoj Monga, San Diego, CA, Rutugandha Paranjpe, Benjamin Cutone, Boston, MA, Ben H. Chew, Vancouver, Canada

Introduction: Real-world outcomes including mortality associated with sepsis post-ureteroscopy (URS) have not been previously reported. The objective of the current study was to evaluate all-cause inpatient mortality among patients who develop sepsis post-URS compared first to a non-sepsis cohort and second to a severe sepsis cohort.

Methods: Two retrospective claims analyses were conducted using IBM MarketScan commercial data. Patients aged 18+ were included if they had a URS procedure. A sepsis or severe sepsis event was defined as the first sepsis or severe sepsis event within 30-days post-URS. In the first analysis the sepsis cohort comprised of patients with sepsis or severe sepsis while in the second analysis the sepsis and severe patients were distinct. All-cause inpatient mortality was measured within 30-days of the sepsis or severe sepsis event and within 30-days of URS in the non-sepsis cohort. Descriptive analyses were conducted in the sepsis, severe sepsis and non-sepsis cohorts comparing all-cause inpatient mortality and time to all-cause inpatient death (days). Lastly, two multivariable cox regression models were conducted to evaluate the effect of sepsis and severe sepsis on all-cause inpatient mortality.

Results: The URS cohort comprised of 109,496 patients. The overall incidence of developing sepsis was 5.6%. Of these, 4.1% developed sepsis while 1.5% developed severe sepsis. The all-cause inpatient mortality among the severe sepsis, sepsis and non-sepsis cohort was 2.5%, 0.82%, and 0.03% respectively. The mean time to death in the sepsis cohort was significantly shorter than the non-sepsis cohort (9.6 days vs 16.7 days; p<0.001). Similarly, the mean time to death in the severe sepsis cohort was significantly shorter than the sepsis cohort (8.7 days vs 11.1 days; p<0.001). The first cox regression model determined that predictors of higher all-cause inpatient mortality after URS included sepsis (sepsis vs non-sepsis HR:17.2; 95% CI: 10.5-28.0; p<0.001) older age (55-64; p<0.024) and higher Elixhauser comorbidity index (p < 0.001). Lastly, patients who developed severe sepsis (model 2) were significantly associated with higher all-cause inpatient mortality as compared to patients who developed sepsis (HR: 8.2; 95% CI: 4.24-15.83; p<0.001).

Conclusions: The study findings demonstrated that both sepsis and severe sepsis post-URS are significant clinical events associated with significant morbidity and even mortality. Appropriate measures to prevent sepsis and severe sepsis post-URS are of utmost importance.

Source of Funding: Boston Scientific