Abstract: FR-PO534
Hospital Acquired Infections after Major Surgery among Patients with Clinical Comorbidities: The Stockholm Creatinine Measurements (SCREAM) Project
Session Information
- CKD: Epidemiology, Outcomes - Non-Cardiovascular - I
November 03, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Chronic Kidney Disease (Non-Dialysis)
- 304 CKD: Epidemiology, Outcomes - Non-Cardiovascular
Authors
- Ishigami, Junichi, Johns Hopkins School of Public Health, Baltimore, Maryland, United States
- Trevisan, Marco, Karolinska Institutet, Stockholm, Sweden
- Xu, Hong, Karolinska Institutet, Stockholm, Sweden
- Coresh, Josef, Welch Center for Prevention, Epidemiology & Clinical Research, Baltimore, Maryland, United States
- Matsushita, Kunihiro, Johns Hopkins Bloomberg School of Public Health, Baltimore, Alabama, United States
- Carrero, Juan Jesus, Karolinska Institutet, Stockholm, Sweden
Background
Whether incidence of hospital acquired infections (HAIs) after major surgery differs by types of clinical comorbidity has not been well characterized.
Methods
We evaluated 66,820 patients (mean age, 64 years; 58% female) undergoing four major types of surgery (neuro [n=1,922], cardiothoracic and vascular [n=6,403], abdominal [n=26,542], and orthopedic surgery [n=31,953]) with available serum creatinine measurement for estimating GFR up to 365 days prior to the hospitalization between 2007 and 2011 in Stockholm, Sweden, for the in-hospital incidence of four HAIs (pneumonia, urinary tract infections, surgical site infections, and bloodstream infections) across clinical comorbidities including low eGFR <60 ml/min/1.73m2, diabetes, coronary heart disease (CHD), congestive heart failure (CHF), stroke, chronic obstructive pulmonary disease (COPD), liver disease, and cancer.
Results
Among clinical comorbidities, cancer and low eGFR were most frequently identified (Table1). During hospital stay, 5.7% (n=3,784) had at least one type of HAIs. For each clinical comorbidity, the prevalence was disproportionally higher in patients with HAIs, and risk of HAIs was significantly higher in multivariable Poisson models (Table1). When population attributable fraction (PAF) was estimated, the PAF was highest for low eGFR (PAF, 0.13), followed by cancer, CHF, stroke and COPD (PAF, 0.11, 0.08, 0.06, and 0.05, respectively) (Table1). These findings were mostly consistent across the four types of infection.
Conclusion
For HAIs after major surgery, low eGFR posed the highest PAF among major clinical comorbidities, underscoring the importance of HAIs prevention measures for persons with low eGFR.
Table1: Prevalence of clinical comorbidity and relative risk of HAIs
| % in the overall population (n=66,820) | % among patients with HAIs (n=3,784) | Adjusted relative risk* (95%CI) | Population attributable fraction (PAF) | |
| eGFR <60 ml/min/1.73m2 | 18% | 35% | 1.62 (1.50-1.75) | 0.13 |
| Cancer | 21% | 34% | 1.45 (1.35-1.56) | 0.11 |
| CHF | 12% | 24% | 1.49 (1.36-1.62) | 0.08 |
| Stroke | 11% | 20% | 1.41 (1.29-1.53) | 0.06 |
| COPD | 12% | 18% | 1.33 (1.22-1.44) | 0.05 |
| Diabetes | 12% | 17% | 1.11 (1.02-1.21) | 0.02 |
| CHD | 8.8% | 14% | 1.12 (1.02-1.25) | 0.02 |
| Liver disease | 2.6% | 3.2% | 1.28 (1.07-1.53) | 0.01 |
*The model was adjusted for each clinical comorbidity, age, sex, and types for surgery
Funding
- Other NIH Support