Kidney Week

Abstract: SA-PO0534

Postoperative Sepsis and Hospital-Acquired Infection by Level of Kidney Function: A Population-Based Cohort Study

Session Information

• CKD: Clinical - Therapies, CKM Health, Special Populations, and More
  October 24, 2026 | Location: Exhibit Hall A, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: CKD (Non-Dialysis)

• 2202 CKD (Non-Dialysis): Clinical, Outcomes, and Trials

Authors

• Laube, Mikayla, University of Calgary, Calgary, Alberta, Canada

Mikayla Laube, MD

• Harrison, Tyrone, University of Calgary, Calgary, Alberta, Canada

Tyrone Harrison, MD, PhD

• Scory, Tayler D., University of Calgary, Calgary, Alberta, Canada

Tayler D. Scory, MSc

• James, Matthew T., University of Calgary, Calgary, Alberta, Canada

Matthew T. James, MD, PhD

• Ruzycki, Shannon M., University of Calgary, Calgary, Alberta, Canada

Shannon M. Ruzycki, MD

• Ronksley, Paul E., University of Calgary, Calgary, Alberta, Canada

Paul E. Ronksley, PhD

• Hemmelgarn, Brenda, University of Alberta, Edmonton, Alberta, Canada

Brenda Hemmelgarn, MD, PhD

• Agarwal, Arnav, University of Alberta, Edmonton, Alberta, Canada

Arnav Agarwal, MD

Background

People with advanced kidney disease have increased risk of adverse postoperative outcomes. Though postoperative infectious complications have been identified as key research priorities, differences in risk across the chronic kidney disease (CKD) spectrum remain unclear. Our objective was to quantify the association between preoperative kidney function and postoperative sepsis and hospital-acquired infection.

Methods

We conducted a population-based retrospective cohort study of adults in Alberta, Canada undergoing surgery between April 2008 and February 2021. The exposure was preoperative kidney function, categorized into eight groups based on estimated glomerular filtration rate (eGFR), and including dialysis and kidney transplant. The primary outcome was 30-day postoperative sepsis, identified using two validated administrative algorithms with different reported sensitivity. The secondary outcome was 30-day hospital-acquired infection. Associations were estimated using unadjusted and adjusted generalized estimating equation models with adjustment for demographics, comorbidity, and surgical factors.

Results

We identified 3,344,202 surgeries in 1,589,154 participants, of which 73% occurred in ambulatory settings. The median age was 58.5 years and 51% were female. Using the more sensitive algorithm, 66,895 (2.0%) sepsis events were identified versus 39,844 (1.2%) with the less sensitive algorithm. Both algorithms demonstrated progressively higher odds of sepsis with lower eGFR, with the highest odds observed in the G5 Dialysis (G5D) group (OR 4.58, 95% CI 4.27-4.90). After adjustment, effect estimates were attenuated but remained significant for G5ND (G5 no dialysis), G5D, and transplant groups. There were 38,926 (1.2%) hospital-acquired infection events identified. Adjusted odds were increased only among transplant recipients (adjusted OR 1.35, 95% CI 1.13-1.61).

Conclusion

Lower kidney function was associated with higher odds of postoperative sepsis and hospital acquired infection. The risk of sepsis was attenuated after adjustment, though persistence of higher odds in the G5ND, G5D, and transplant categories may help guide appropriate perioperative risk counseling. This study is congruent with current literature where patients with CKD have higher postoperative mortality and morbidity including infections. Future studies should explore interventions to mitigate this risk.