Estimation of Creatinine Clearance in Morbidly Obese Patients

Jasmina A. Demirovic; Amy Barton Pai; Manjunath P. Pai


Am J Health Syst Pharm. 2009;66(7):642-648. 

In This Article


The common use of the Cockcroft-Gault equation suggests that it will remain the method of choice for drug dosing in patients with renal impairment. In contrast, the MDRD4 equation is likely to remain a clinical tool for staging chronic kidney disease. The reliance of the Cockcroft-Gault equation on weight as a key variable has reduced its accuracy in patients with obesity. Improved approaches to body-size description using LBW provide an opportunity to improve this equation.[12]

However, measuring actual LBW is not simple. In contrast, the IBW equations, which serve as a surrogate of LBW, are simple to use but were not derived by rigorous scientific means.[14] As a result, IBW equations have been consistently documented to underestimate CLcr in obese patients.[7] Correction of this IBW bias has been attempted through the use of ABW for specific drugs and has improved estimates of CLcr.[8] However, the generalizability of this approach is unknown, and the most appropriate correction factor for estimation of ABW arouses further debate. Given these constraints, improved estimates of LBW remain the most meaningful method of generating a universal body-size descriptor to aid in drug dosing. Most importantly, development of a simple method of estimating LBW increases its chances of clinical applicability.

According to Green and Duffull,[12] LBW appears to be the most useful approach to estimate clearance and volume of distribution for most drugs. Duffull and colleagues[22] developed a patient normalized weight (PNWT) equation for patients with a BMI of > 30 kg/m2. This equation provided better predictions of gentamicin clearance than the use of TBW -- a most relevant finding, given that gentamicin clearance approximates GFR. Unfortunately, the equations for PNWT are not suitable for patients with a BMI of > 52 kg/m2. In the current study, PNWT was not used because that BMI applied to 16 (30%) of our patients. Recognizing this limitation, Janmahasatian et al.[17] used a two-stage semimechanistic model to derive a LBW function based on height, TBW, and sex. The model used a rigorous scientific approach to estimate LBW using DXA and BIA methods in two separate populations. The resultant LBW estimate was used in the current study.

We demonstrated that the MDRD4 and Salazar-Corcoran equations provided biased estimates of CLcr without improvement of precision or accuracy of estimates relative to measured values. Similarly, body-size descriptors such as TBW, IBW, and ABW were biased and inaccurate when applied in the Cockcroft-Gault equation. In contrast, FFW estimates using BIA and LBW estimates using the outlined approach were unbiased. These body-size descriptors also resulted in more precise and accurate estimates of CLcr. The current study validates the findings of Janmahasatian and colleagues,[23] which indicated that LBW is the body-size descriptor that best predicts GFR measured by inulin clearance.

The growing obesity epidemic presents continued challenges for pharmacists who are faced with drug dosing considerations. An estimated 1 in 20 Americans is morbidly obese, and this number is expected to more than double by 2015.[24] The potential adverse events that can result from underdosing or overdosing patients with obesity have not been well characterized. However, emerging data suggest that worse pharmacologic outcomes occur among obese patients relative to normal weight patients. Standardization of body-size description is a critical point that must be addressed in future pharmacokinetic studies in order to improve dosing recommendations in this population.[12] Surrogates of LBW have predated the Cockcroft-Gault equation, and most of these values used in pharmacy practice (e.g., IBW, ABW) are likely applied without a scientific basis. The results of the current study validate the concept that LBW estimation through FFW using BIA provides improved estimates of CLcr in obesity. However, the availability and feasibility of BIA as a routine clinical tool remain unlikely. Consequently, estimation of LBW through improved equations such as those proposed by Janmahasatian and colleagues[17] seems the most viable approach.

The current study is limited in that alternative estimation methods of CLcr were not studied. An additional 10 equations could have been pursued as viable options. In addition, the current data set could have been used to derive a new equation to estimate CLcr in this extremely morbidly obese population (31% of patients with a BMI of > 50 kg/m2). Both options were considered but dismissed as futile approaches to changing pharmacy practice, given engrained curricula, FDA guidelines, and NKDEP recommendations.[5] The current study also included a limited sample of patients with chronic kidney disease and did not include a healthy outpatient population. We utilized stringent criteria to eliminate variables known to influence CLcr. However, potential unmeasured confounding variables associated with hospitalized patients may have inadvertently affected our results. Finally, GFR was represented by 24-hour measured CLcr instead of more accurate measures (e.g., inulin or 125I-sodium iothalamate clearance). Despite these limitations, the current study included a population that has not been evaluated systematically using CLcr and GFR equations.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.