Despite advanced infection control practices including preoperative antibiotic prophylaxis, surgical site infection (SSI) remains a challenge. This study aimed to test whether local administration of a novel prolonged-release Doxycycline-Polymer-Lipid Encapsulation matriX (D-PLEX) before wound closure, concomitantly with standard of care (SOC), reduces the incidence of incisional SSI after elective abdominal colorectal surgery. This was a phase 3 randomized, controlled, double-blind, multinational study (SHIELD 1) between June 2020 to June 2022. Patients with at least one abdominal incision length >10cm were randomized 1:1 to the investigational arm (D-PLEX+SOC) or control (SOC) arm . The primary outcome was a composite of incisional SSI, incisional reintervention, and all-cause mortality. A total of 974 patients were analyzed, of whom 579 (59.4%) were male. The mean age (±SD) was 64.2±13.0 years. The primary outcome occurred in 9.3% of D-PLEX patients versus 12.1% (SOC) (risk difference estimate [RDE], -2.8%; 95% CI [-6.7%, 1.0%], P=0.1520). In a pre-specified analysis by incision length, a reduction in the primary outcome was observed in the >20cm subpopulation: 8% (D-PLEX) versus 17.5% (SOC) (RDE, -9.4%; 95% CI [-15.5%, -3.2%], P=0.0032). In the >10 to ≤20cm subgroup, no reduction was observed: 9.9% versus 7.9% (RDE, 2.0%; 95% CI [-2.8%, 6.7%], P=0.4133). Exploratory post-hoc analyses of patients with increased SSI risk (≥1 patient-specific comorbidity) indicated a reduction in the incidence of the primary outcome: 9.0% (D-PLEX) versus 13.7% (SOC) (RDE, -4.8%; 95% CI [-9.5%, -0.1%], P=0.0472). The D-PLEX safety profile was good (no difference in treatment-emergent adverse events between the groups). The SHIELD-1 study did not meet its primary outcome of reduced incisional SSI, incisional reinterventions, or all-cause mortality. Pre-specified and post-hoc analyses suggested that D-PLEX may reduce the incidence of the primary outcome event in patients with increased SSI risk, including lengthy incisions.
Read full abstract