Is there a reliable index of glomerular filtration rate in critically ill patients?

Assessment of renal function in critically ill patients is important for appropriate individualization of dosage regimens and nutrition, but is complicated by a high incidence of acute renal failure (ARF). The most common cause of ARF in intensive care unit (ICU) patients is hypoperfusion. Other causes of ARF include intrinsic injury, nephrotoxicity, and postrenal obstruction. ARF is associated with a decreased glomerular filtration rate (GFR), reduced or maintained urine output (nonoliguric renal failure), and alterations in other commonly obtained urinary indices. Twenty-four-hour or shorter urinary creatinine clearance studies may overpredict GFR as creatinine is both filtered and secreted. The use of serum creatinine in empiric predictive equations is impaired in ICU patients because of decreases in creatinine production due to immobilization and malnutrition or increases in creatinine production due to catabolic illnesses. Adjustment of empiric methods by employing lean body weight, ideal body weight, or corrected serum creatinine values has not been evaluated against uncorrected values in this population, but is routinely performed in clinical practice. Inulin and radiolabeled substances are not practical for routine clinical use and may overpredict GFR in ARF due to backleak of large molecular-weight substances through the tubules. Comparative clinical trials have shown essentially equivalent performance of empiric methods relative to 24-hour urinary creatinine clearance in adults. No studies have compared these methods to a reference method for determination of GFR. Until conclusive data become available, clinicians should cautiously compare results from at least two independent methods of assessment to estimate renal functional impairment in ICU patients.