Assessing the Role of Prophylactic Antibiotics in Preventing Clinically Relevant Bacteremia in Neutropenic Patients Undergoing GI Endoscopy

Introduction: Patients with neutropenia are at increased risk for bacteremia and sepsis after endoscopy. According to the American Society for Gastrointestinal Endoscopy (ASGE), there is insufficient evidence to recommend for or against administration of prophylactic antibiotics prior to routine endoscopic procedures in patients with neutropenia. This study was conducted to assess the safety of gastrointestinal endoscopic procedures in patients with neutropenia and to compare outcomes between patient who received periprocedural prophylactic antibiotics to those who did not. Methods: We studied neutropenic patients who underwent endoscopic procedures from 2012 through 2022. Neutropenia was defined as an absolute neutrophil count (ANC) < 1500 cells/mL which was further sub-classified into mild (ANC < 1500), moderate (ANC < 1000) and severe (ANC < 500) neutropenia. Multilevel logistic regression models were used to assess factors associated with clinically relevant bacteremia. Results: We identified 102 neutropenic patients who underwent gastrointestinal endoscopies; 45% (N=46) of patients received periprocedural prophylactic antibiotics. 16% (N=9) of patients who did not received prophylactic antibiotics were started on antibiotics within 3 days following the procedure due to sepsis. A similar proportion of patient 15% (N=7) who received periprocedural antibiotics developed sepsis within 3 days of endoscopy and required either resumption or broadening of antibiotic regimen. Subsequently, blood cultures that were drawn on these septic patients did not grow any organisms. Poor performance status was associated with increased risk of infectious adverse events. No association was observed between low ANC or use of steroid with infectious adverse events (p > 0.6) (Table). Conclusion: This study showed a low rate of mortality and clinically relevant bacteremia following endoscopy in neutropenic patients. The functional status of the patient, in the absence of the need for urgent or necessary endoscopic interventions, should be considered when deciding to perform endoscopy. The ANC did not seem to affect outcomes, neither did the use or periprocedural prophylactic antibiotics. At this interim analysis, our study is underpowered to detect significant differences and further data collection will be carried out to ensure appropriate statistical power. Table 1. - Distribution of sample characteristics and outcomes stratified by periprocedural antibiotic use during endoscopy No antibiotics (N=56)N (%) Periprocedural antibiotics(N=46)N (%) Age - Mean (SE) Overall 60.4 (1.51) 64.3 (1.95) 55.7 (2.18) Gender Male 26 (46.4) 30 (53.6) Female 30 (65.2) 16 (34.8) Type of cancer / Transplant Solid organ 17 (94.4) 1 (5.6) Hematologic malignancy 7 (28) 18 (72) status post solid organ transplant 32 (54.2) 27 (45.8) ANC count ANC < 250 3 (23.1) 10 (76.9 ANC < 500 10 (66.7) 5 (33.3) ANC < 1000 21 (61.8 13 (38.2) ANC < 1500 22 (55) 18 (45) Performance status ECOG 1-2 18 (64.3) 10 (35.7) ECOG 3-4 4 (57.1) 3 (42.9) Unknown 34 (50.7) 33 (49.3) Blood Cultures Collected Within 3 Days of Endoscopy Yes 10 (17.9) 7 (15.2) No 46 (82.17) 39 (84.8) Patient required new antibiotic started within 3 days of procedure Yes 10 (17.9) 8 (17.4) No 46 (82.1) 38 (82.6) Adverse events related to sepsis (Hypotension, fever, hypoxia, ICU admission) Any 6 (10.7) 8 (17.4) None 50 (89.29) 38(82.6)