ONE SIZE DOES NOT FIT ALL: THE IMPORTANCE OF PATIENT-SPECIFIC PREVENTION MEASURES FOR URETERAL INJURY DURING COLORECTAL PROCEDURES
Author(s)
Laila Rashidi, MD1, Naomi C. Hamm, PhD2, Yuki Liu, MS3, Feibi Zheng, MBA, MD4, Lee White, MD PhD5;
1MultiCare, Tacoma, WA, USA, 2Intuitive Surgical, Outcomes Researcher, Sunnyvale, CA, USA, 3Intuitive Surgical, Sunnyvale, CA, USA, 4Intuitive Surgical, Houston, TX, USA, 5El Camino Health, Mountain View, CA, USA
1MultiCare, Tacoma, WA, USA, 2Intuitive Surgical, Outcomes Researcher, Sunnyvale, CA, USA, 3Intuitive Surgical, Sunnyvale, CA, USA, 4Intuitive Surgical, Houston, TX, USA, 5El Camino Health, Mountain View, CA, USA
OBJECTIVES: Ureteral injuries (UI) are a rare but morbid complication of colorectal surgery. Reports vary on the efficacy of prophylactic UI measures such as indocyanine green (ICG) and ureteral stenting, potentially due to heterogeneity in patient- and procedure-specific UI risk factors and prophylactic practices. This study aims to identify homogeneous UI risk phenotypes in colorectal patients and assess the impact of ICG and ureteral stenting on UI incidence within each phenotype.
METHODS: We determined 90-day perioperative UI incidence for laparoscopic and robotic colorectal procedures in the Premier Health Database (2016-2023). We used latent class analysis, a statistical approach that clusters homogenous subgroups within a population, to phenotype patients based on patient and procedure-based UI risk factors. We assessed the effects of surgical modality (laparoscopic vs. robotic), prophylactic ureteral stenting, and ICG use on UI risk using logistic regression models. We controlled Type I error using a Benjamini-Hochberg adjustment.
RESULTS: Five patient UI risk phenotypes were identified. Phenotypes were primarily characterized by patient history of inflammatory conditions, procedure primary diagnosis, and procedure type. Other risk factors such as adhesions and admission type had minimal impact on phenotyping. Robotic modality was protective for 3/5 phenotypes. ICG use and ureteral stenting was associated with increased UI risk for 4/5 phenotypes. Using ICG and ureteral stents together was protective for patients receiving a sigmoidectomy for simple diverticulitis (OR=0.55, adjusted p-value=0.04). No other significant effects were found for simultaneous ICG and ureteral stent use.
CONCLUSIONS: The impact of ICG and ureteral stenting on UI incidence varied across UI risk phenotypes. Although protective effects were observed in a single phenotype when prophylactic measures were used concurrently, most phenotypes demonstrated no benefit or increased risk. This work highlights the importance of considering different patient UI risk phenotypes when assessing ICG and ureteral stenting efficacy for UI reduction.
METHODS: We determined 90-day perioperative UI incidence for laparoscopic and robotic colorectal procedures in the Premier Health Database (2016-2023). We used latent class analysis, a statistical approach that clusters homogenous subgroups within a population, to phenotype patients based on patient and procedure-based UI risk factors. We assessed the effects of surgical modality (laparoscopic vs. robotic), prophylactic ureteral stenting, and ICG use on UI risk using logistic regression models. We controlled Type I error using a Benjamini-Hochberg adjustment.
RESULTS: Five patient UI risk phenotypes were identified. Phenotypes were primarily characterized by patient history of inflammatory conditions, procedure primary diagnosis, and procedure type. Other risk factors such as adhesions and admission type had minimal impact on phenotyping. Robotic modality was protective for 3/5 phenotypes. ICG use and ureteral stenting was associated with increased UI risk for 4/5 phenotypes. Using ICG and ureteral stents together was protective for patients receiving a sigmoidectomy for simple diverticulitis (OR=0.55, adjusted p-value=0.04). No other significant effects were found for simultaneous ICG and ureteral stent use.
CONCLUSIONS: The impact of ICG and ureteral stenting on UI incidence varied across UI risk phenotypes. Although protective effects were observed in a single phenotype when prophylactic measures were used concurrently, most phenotypes demonstrated no benefit or increased risk. This work highlights the importance of considering different patient UI risk phenotypes when assessing ICG and ureteral stenting efficacy for UI reduction.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
CO192
Topic
Clinical Outcomes
Topic Subcategory
Clinical Outcomes Assessment
Disease
SDC: Gastrointestinal Disorders, SDC: Urinary/Kidney Disorders, STA: Surgery