Does Elevated Initial Creatinine Clearance Predict Infections in Critically Ill Trauma Patients?

Author(s):
Sarah Eidelson; Michelle Mulder; Rishi Rattan; SriGita Madiraju; Charles Karcutskie; Carl Schulman; Nicholas Namias; Kenneth G Proctor

Background:

Infections pose a major risk of morbidity and mortality following trauma, despite early use of powerful antibiotics.  One possibility is augmented renal clearance (ARC= creatinine clearance (CL_Cr) > 130 mL/min), which frequently occurs early after trauma and might alter antibiotic pharmacokinetics.

Hypothesis:

We test the hypothesis that CL_Cr significantly differs in those that develop infection.

Methods:

In 158 consecutive trauma ICU patients with admission serum creatinine (S_Cr) ≤1.65 mg/dL, 24 hr CL_Cr was measured within 5 days of admission to the ICU, and correlated with three traditional clinical estimates of CL_Cr: Cockroft-Gault (CG), modification of diet in renal disease (MDRD), and chronic kidney disease epidemiology (CKD-EPI).  Data are expressed as M±SD if parametric, or median [interquartile range] if not, and compared with univariate analysis and multivariate logistic regression.

Results:

The population was 44±20 years, 67% male, 10% burn, 71% blunt, and 20% penetrating mechanism of injury.  Admission S_Cr was 0.97±0.24mg/dL, CL_Cr was 154±76 ml/min and the ARC incidence was 58.2%.  Length-of-stay was 17[10-33] days, infection rate was 43.7% and mortality was 7.6%.  The mean CL_Cr was 140 ml/min patients with positive blood, urine or bronchoalveolar lavage cultures vs 165 ml/min (p=0.044) in those without positive cultures.  Of 17 potential risk and protective factors, including age, body mass index, admission vital signs, Abbreviated Injury Scale chest/abdomen/head >2, injury mechanism, transfusion and catheter history, mechanical ventilation was the only independent predictor of infection (OR 3.17 [1.187-8.479], p=0.021), while elevated CL_Cr was the only factor protective from infection (OR 0.99 [.983-.997], p=0.009.)   Additionally, clinical estimates CKD-EPI, MDRD, and CG underestimated CL_Cr by 29%, 23% and 14%, respectively (all p<0.01).

Conclusions:

These data confirm that ARC is present in the majority of severely injured trauma ICU patients and that traditional clinical estimates of CL_Cr fail to detect this phenomenon.  More importantly, this is the first demonstration that, counterintuitively, CLcr is significantly higher in those that resist infection.  We speculate that augmented CLcr either provides some protective mechanism for patients at risk of infection, and/or is merely a manifestation of an overall beneficial compensatory survival response after trauma.   Thus, CLcr could play a potential role for targeted therapy. Further studies are warranted.