Article Text
PDF
Abstract
Background Refractory intestinal infections, encompassing known and unknown pathogens, constitute a pressing global health concern. This study aimed to assess the efficacy and safety of washed microbiota transplantation (WMT), a novel fecal microbiota transplantation method, in patients with antibiotic-refractory intestinal infections.
Methods This was a nationwide retrospective study based on the prospective cohorts of patients who underwent WMT from September 2015 to February 2023 in China. The patients diagnosed with refractory intestinal bacterial infections and then underwent WMT were included. The primary outcome was the clinical cure rate at one-month post-WMT, with secondary outcomes including clinical response rate at one week and one month and factors influencing efficacy and safety.
Results The analysis involved 221 patients (IDDF2024-ABS-0351 Figure 1), comprising 127 with Clostridioides difficile infection (CDI) and 94 with refractory intestinal infections beyond CDI (IDDF2024-ABS-0351 Figure 2 (a)). The types of antibiotics administered prior to rescue WMT are illustrated in IDDF2024-ABS-0351 Figure 2 (b), and the median number of antibiotic classes prescribed within one month before WMT was 2 (2-4). The overall clinical cure rate at one-month post-WMT was 66.1% (146/221). Specifically, clinical cure rates were 55.0% (11/20) for patients with intestinal active multi-drug resistant organisms’ infections and 48.1% (38/79) for those with undetermined bacterial infections (IDDF2024-ABS-0351 Figure 2 (c)). Risk factors affecting WMT efficacy included World Health Organization performance status of 4 (OR = 2.753, 95% CI: 1.269-5.973; p = 0.010), concomitant enterovirus infection (OR = 9.629, 95% CI: 3.004-30.871; p = 0.001) and antibiotics use within one or two weeks after WMT (OR = 2.428, 95% CI: 1.223-4.821; p = 0.011) (IDDF2024-ABS-0351 Figure 3). Concomitant enterovirus infection emerged as the sole independent factor significantly influencing the safety of WMT (OR = 3.127, 95% CI: 1.131-8.560; p = 0.028) (IDDF2024-ABS-0351 Figure 4). A machine learning classifier Random Forest, incorporating ten baseline characteristics (IDDF2024-ABS-0351 Figure 5 (a)), achieved an area under the curve of 0.819 for predicting clinical cure at one-month post-WMT (IDDF2024-ABS-0351 Figure 5 (b)).
Conclusions The favorable clinical outcomes observed in this study provide cohort-based evidence for using WMT to treat refractory intestinal bacterial infections beyond CDI.