Pre-admission Intravenous Drug Abuse is Associated with Reduced Index Morbidity but Higher Post-Discharge Hospital Utilization following Necrotizing Soft Tissue Infections

Authors:
Jasmine Kelley, Clark Ingram, Courtney Collins, Brooke Davis, Holly Baselice, Jinwei Hu, Jon Wisler, Anahita Jalilvand

Body of Abstract:
Introduction: While pre-admission intravenous drug use has been associated with certain types of necrotizing soft tissue infections (NSTIs), its impact on post-operative outcomes and healthcare utilization is unclear. The primary objective of this study was to characterize the impact of pre-admission IVDU on index hospitalization and 90-day outcomes following admission for NSTIs.

Methods: We reviewed 625 NSTI patients at a single tertiary care institution (2013-2023). A chart review was conducted to obtain detailed baseline characteristics, operative, microbial, and antibiotic data, and complete 90-day outcomes. All data pertaining to 90-day ED visits/readmissions were documented if captured in our statewide electronic medical record. Comparisons were made between the IVDU cohort (n=65) and non-IVDU patients (n=569). A p<0.005 was considered statistically significant. Results: Compared to the control group, IVDU patients were younger (42(33-49) vs 56 (46-64), p<0.005), had a lower median BMI (26.6 (23.3-30.3) vs 34.9 (28.7-43.6), p<0.005), were less likely to have a PCP (31% vs 60%, p<0.005), had lower Charlson Comorbidity Index (1 (0-3) vs 4 (2-6), p<0.005) and SOFA scores (1(0-3) vs 2(1-4), p = 0.004). Patients with IVDU history had higher prevalence of extremity NSTIs (74% vs 38%, p<0.005) with Group A Strep (23% vs 9%, p<0.005) but lower incidence of amputations (9% vs 5%, p=0.03) or gram negative culture growth (45% vs 67%, p<0.005). The IVDU cohort was associated with decreased ICU stay (37% vs 66%, p<0.005), mechanical ventilation (22% vs 41%, p<0.005), vasopressors use (28% vs 42%, p=0.003) and dependent discharge (43% vs 54%, p=0.003). Compared to the control group, IVDU patients had higher overall 90-day readmissions (52% vs 35%, p=0.01), increased incidence of 2+ readmissions (24% vs 12%, p=0.02), and NSTI-related readmissions (43% vs 19%, p<0.005). Additionally, they were readmitted sooner following discharge (13 days (4-25) vs 25 days (9-47), p=0.005), had less post discharge follow-up (31% vs 60%, p=0.007), and trended towards having shorter readmission stay (<24 hours) (33% vs 18%, p =0.1). IVDU was an independent predictor of having an NSTI-related readmission within 90-days (OR 2.1, 95th CI: 1.1-4.0), after controlling for age, NSTI location, transfer status, time to operation, admission SOFA, sex. Discussion: Despite reduced index hospital morbidity, patients with IVDU history had higher overall 90-day NSTI-related readmissions and were readmitted sooner than non-IVDU patients. Given the timing and cause of these visits, these data support optimizing follow-up practices with earlier clinic visits and/or utilization of alternative modalities (ex. Telehealth) to mitigate potentially modifiable readmissions for this population.