Abstract: TH-PO414
Estimated Glomerular Filtration Rate and Incidence of Major Surgery: A Population-Based Cohort Study
Session Information
- CKD: Risk Scores and Translational Epidemiology
November 07, 2019 | Location: Exhibit Hall, Walter E. Washington Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: CKD (Non-Dialysis)
- 2101 CKD (Non-Dialysis): Epidemiology, Risk Factors, and Prevention
Authors
- Harrison, Tyrone, University of Calgary, Calgary, Alberta, Canada
- James, Matthew T., University of Calgary, Calgary, Alberta, Canada
- Zarnke, Kelly B., University of Calgary, Calgary, Alberta, Canada
- Ma, Zhihai, University of Calgary, Calgary, Alberta, Canada
- Ruzycki, Shannon M., University of Calgary, Calgary, Alberta, Canada
- Hemmelgarn, Brenda, University of Calgary, Calgary, Alberta, Canada
Background
Approximately 1 in 9 Canadian adults have a surgical procedure each year, however whether the rates of major surgical procedures vary by level of estimated glomerular filtration rate (eGFR) is unknown. We aimed to quantify the incidence of major surgery by varying degrees of impaired kidney function.
Methods
We identified 1,455,565 adults in Alberta, Canada that had at least one outpatient serum creatinine measure or were in receipt of chronic dialysis between 2008 and 2009. As in prior studies, incident major surgical procedures were identified and categorized into 13 major surgical subtypes using procedure codes and physician claims; major surgery was defined by requiring hospital admission for at least one day. Surgical subtypes included musculoskeletal, intra-abdominal, lower urologic/gynecologic, head and neck, vascular, skin and soft tissue, cardiac, breast, neurosurgery, retroperitoneal, thoracic, anorectal, and ophthalmologic. Only the first event per surgical subtype was included. Patients were followed from January 1 2010 to December 31 2016, and were censored at kidney transplantation, outmigration or death. Incidence rates by eGFR strata were estimated using negative binomial regression and adjusted for age, sex, income, location, proteinuria, and comorbidity.
Results
The median age of the cohort was 51.6 years (IQR 39.4, 63.4) with the majority female (57.0%). Most patients had an eGFR ≥60mL/min/1.73m2 (92.2%), with 0.1% having an eGFR <15mL/min/1.73m2 not on dialysis. Over a median follow up of 7.00 years, musculoskeletal surgeries were the most common across all eGFR strata, with adjusted incidence rates between 9.48 per 1000 person-years (95%CI: 9.39, 9.56) for those with eGFR ≥60mL/min/1.73m2 and 74.81 per 1000 person-years (95%CI: 65.60, 85.32) for dialysis patients. Similar trends of increasing surgical incidence for those with lower eGFR were noted for all surgical types except for breast and lower urologic/gynecologic.
Conclusion
In a large population-based cohort, incidence rates of major surgical procedures increased with decreasing eGFR, and were highest among dialysis-dependent patients. Further research is needed to investigate whether differences exist in perioperative outcomes and costs for those with varying degrees of kidney function.