Estimated GFR and Hospital-Acquired Infections Following Major Surgery

Abstract

Rationale & objective: Low estimated glomerular filtration rate (eGFR) increases infection risk, but its contribution to hospital-acquired infection following major surgery is unknown.

Study design: Retrospective cohort study.

Setting & participants: Residents of Stockholm, Sweden, 18 years or older with at least 1 recorded serum creatinine measurement, no recent diagnoses of infection, and hospitalized for orthopedic, abdominal, cardiothoracic and vascular, or neurologic surgery between January 2007 and December 2011. EXPOSURES: eGFR<60mL/min/1.73m² (termed low eGFR) and other clinical comorbid conditions at admission: cancer, cerebrovascular disease, chronic obstructive pulmonary disease (COPD), coronary heart disease, diabetes, heart failure (HF), and liver disease.

Outcomes: Incidence and population-attributable fractions of 4 major types of hospital-acquired infections: pneumonia, urinary tract infection, surgical-site infection, and bloodstream infection.

Analytical approach: Logistic regression analysis.

Results: 66,126 patients with a history of orthopedic (n=31,630), abdominal (n=26,317), cardiothoracic and vascular (n=6,266), or neurologic (n=1,913) surgery were studied. Cancer (21%) and low eGFR (18%) were the most prevalent comorbid conditions at admission, followed by diabetes, HF, and COPD (12%). Overall, 3,617 patients (5.5%) had at least 1 type of hospital-acquired infection (1,650 cases of urinary tract infection, 1,196 cases of pneumonia, 635 cases of surgical site infection, and 411 cases of bloodstream infection). The OR of hospital-acquired infection was highest for low eGFR (1.64; 95% CI, 1.51-1.77), followed by HF (1.60; 95% CI, 1.46-1.76), cerebrovascular disease (1.47; 95% CI, 1.34-1.61), cancer (1.43; 95% CI, 1.33-1.55), and COPD (1.37; 95% CI, 1.25-1.50). Low eGFR demonstrated the highest population-attributable fraction for hospital-acquired infections (0.13), followed by cancer (0.10), HF (0.09), and cerebrovascular disease (0.06). When assessed by type of infection, low eGFR particularly contributed to pneumonia (0.15) and urinary tract infection (0.10).

Limitations: Outcome ascertainment based on diagnostic codes.

Conclusions: These findings highlight the association between low eGFR and hospital-acquired infection and may inform evidence-based hospital-acquired infection prevention policies following major surgery.

Keywords: chronic kidney disease; chronic renal insufficiency; estimated glomerular filtration rate; hospital acquired infection; population attributable fraction; reduced kidney function; surgery.