Different from most drug administration regimens, creatinine is constantly produced and released into plasma by the body muscle mass. Because many drugs are partially or totally eliminated by the kidney, an accurate estimation of renal function is an important component in the application of pharmacokinetics to designing drug therapy regimens. Creatinine clearance as determined by a urine collection and corresponding plasma sample is considered by many clinicians to be the most accurate test of renal function. In the clinical setting, the time delay and the difficulty in obtaining the 24-hour creatinine collection limit the utility of the 24-hour urine collection. In addition, all too often, the urine collection is inaccurate because a portion is accidentally discarded or the time of collection is shorter or longer than requested. Perhaps, the most common error is an incomplete collection, which will result in an underestimation of renal function. Because decisions with regard to drug dosing must often be made quickly, several authors have suggested a variety of methods by which creatinine clearance (ClCr) can be estimated using a serum creatinine value. The most accurate of these equations include serum creatinine, body weight or size, age, and gender.

**Creatinine Pharmacokinetics**

The pharmacokinetics of creatinine is presented in far more detail elsewhere, but a brief overview is necessary. Creatinine is a metabolic by-product of muscle, and its rate of formation (RA) is primarily determined by an individual’s muscle mass or lean body weight. It varies, therefore, with age (lower in the elderly) and gender (lower in the females). For any given individual, the rate of creatinine production is assumed to be constant. Once creatinine is released from muscle into plasma, it is eliminated almost exclusively by renal glomerular filtration. Any decrease in the glomerular filtration rate ultimately results in a rise in the serum creatinine level until a new steady state is reached and the amount of creatinine cleared per day equals the rate of production. In other words, at steady state, the rate in must equal the rate out. Since the rate of creatinine production remains constant even when renal clearance diminishes, the serum creatinine must rise until the product of the clearance and the serum creatinine again equals the rate of production.

Creatinine, RA = RE

**Estimating Creatinine Clearance from Steady-State Serum Creatinine Concentrations**

**Basic Rationale**

The degree to which a steady-state serum creatinine rises is inversely proportional to the decrease in creatinine clearance. Therefore, the new creatinine clearance can be estimated by multiplying a normal ClCr value by the fractional change in the serum creatinine: normal SCr/patient’s SCrss. For the 70-kg man, it can be assumed that the normal SCr is 1.0 mg/dL and that the corresponding ClCr is 120 mL/min.

**New ClCr = (120 mL/min) [1 mg/dL / SCr ss]** (**Equation 1**)

On the basis of this concept, one can see that each time the serum creatinine doubles, the creatinine clearance falls by half and that **small changes in the serum at low concentrations are of much greater consequence than equal changes in the serum creatinine at high concentrations**. To illustrate, if a patient with a normal serum creatinine of 1.0 mg/dL is reported to have a new steady-state serum creatinine of 2 mg/dL, the creatinine clearance has decreased from 120 to 60 mL/min. However, if a patient with chronic renal dysfunction has a usual serum creatinine of 4 mg/dL (ClCr = 30 mL/min), a similar 1.0 mg/dL increase in the serum creatinine to 5 mg/dL would result in a small drop in the ClCr (6 mL/min) and a new clearance value of 24 mL/min. However, at some point even small changes in ClCr can be physiologically significant to the patient. As an example, for a patient with a creatinine clearance of 100 mL/min to have their renal function decline by 10 mL/min is of very little consequence, but for a patient with a creatinine clearance of 15 mL/min, a 10 mL/min decrease would probably change their clinical status from a patient with very poor renal function to a patient who would require dialysis.

The estimation of ClCr from SCr ss alone is reasonably satisfactory as long as the patient’s daily creatinine production is average (i.e., 20 mg/kg/day); the patient weighs approximately 70 kg and the serum creatinine is at steady state (i.e., not rising or falling). These conditions are usually present in the young healthy adult, but young healthy adults are not the typical patients for whom pharmacokinetic manipulations are most useful.

**Adjusting to Body Size: Weight or Body Surface Area**

To account for any changes in creatinine production and clearance that may result from a difference in body size, **Equation 1** can be modified to compensate for any deviation in BSA from the 70-kg patient (1.73 m2):

The patient’s BSA can be obtained from a nomogram, estimated from **Equation 2**:

BSA in m2 = [(Patient’s Weight in Kg / 70 kg)^0.7]*(1.73 m2)

or calculated from the following equation:

BSA in m2 = (W^0.425)(H^0.725)*0.007184

where BSA is in meters squared (m2), W is weight in kilograms, and H is the patient’s height in centimeters.

A disadvantage of using only weight or BSA is that the elderly or emaciated patients who have a reduced muscle mass do not have a “normal” creatinine clearance of 120 mL/min/1.73 m2 with a serum creatinine value of 1.0 mg/dL. For this reason, it may be erroneous to assume that a SCr of 1.0 mg/dL is indicative of a creatinine clearance of 120 mL/min/1.73 m2 in these individuals.

On average, as patients age, their muscle mass represents a smaller proportion of their total weight and creatinine production is decreased (Table 5). There are a number of equations that consider age, gender, body size, and serum creatinine when calculating or estimating creatinine clearance for adults. Although all these methods are similar and equivalent in clinical practice, the most common method used by clinicians is probably the one proposed by Cockcroft and Gault.

ClCr for males (mL/min) = (140 – Age)(Weight) / [(72)(SCr ss)] [**Equation 3**]

ClCr for females (mL/min) = (0.85)(140 – Age)(Weight) / [(72)(SCr ss)] [**Equation 4**]

where age is in years, weight is in kg, and serum creatinine is in mg/dL. Equation 3 and 4 calculate creatinine clearance as mL/min for the patient’s characteristics entered into the equation.

The two most critical factors to consider when using Equation 3 and 4 are the assumptions that **the serum creatinine is at steady state** and **the weight, age, and gender of the individual reflect normal muscle mass**. For example, when estimating a creatinine clearance for an obese patient, an estimate of the non-obese or ideal body weight (IBW) should be used in Equation 3 and 4. This estimate can be based on IBW tables or the following equations.

**TBW Significantly Larger than IBW**

Ideal Body Weight for males in kg = 50 + (2.3)(Height in Inches > 60) [**Equation 5**]

Ideal Body Weight for female in kg = 45 + (2.3)(Height in Inches > 60) [**Equation 6**]

It should be pointed out, however, that an IBW derived from a patient’s height, as in Equation 5 and 6, may not represent the actual non-obese weight of a patient. Although there are some potential flaws in estimating the non-obese weight from height, the IBW is usually preferable to using the actual weight [total body weight (TBW)] when a patient is markedly obese. **As a clinical guideline, one approach is to make an adjustment for IBW if the patient’s actual weight is > 120% of their IBW**.

**There are studies indicating that TBW overestimates and IBW underestimates renal function in the morbidly obese patient**. It has been suggested that an adjusted body weight between IBW and TBW be used to estimate renal function in obese individuals. While this adjustment factor is variable, 40% of the excess weight is commonly used:

Adjusted Body Weight = IBW + (0.4)(TBW – IBW) [**Equation 7**]

where IBW is the patient’s ideal body weight in kg as calculated from Equation 5 and 6, and TBW is the patient’s total body weight in kg.

There are other factors not considered in these equations for IBW and Adjusted Body Weight. As an example, **in patients with extensive ****spacing of fluid** (i.e., edema or ascites), the liters (kilograms) of excess third-space fluid should probably not be included in the patient’s estimate of TBW. As an example, consider a 5-foot 4-inch male patient weighting 75 kg and having an estimated 15 kg of edema and ascitic fluid. Using the patient’s height (64 inches) and weight (75 kg) might suggest that the patient is more than 120% over his IBW and therefore “clinically obese” for the purposes of doing pharmacokinetic calculations.

For this patient, IBW = 59.2, TBW/IBW = 127%. However, the patient is not obese but rather has a significant amount of interstitial fluid accumulated. This is obvious if we subtract the excessive third-space fluid weight of 15 kg from his total weight of 75 kg, resulting in a weight of 60 kg. Clearly, the difference between the “non-excess third-space fluid weight” of 60 kg and the estimated IBW of 59.2 kg is so small that the patient would not be considered clinically obese.

Likewise, when calculating an Adjusted Body Weight, it would be the patient’s weight minus any significant third-space fluid weight that would be used in Equation 7. The excessive third-space fluid weight may or may not be important to consider in making pharmacokinetic calculations. As an example, significant third-space fluid does contribute to the apparent volume of distribution for some drugs, but **is unlikely to be an important contributor to volume of distribution if the apparent volume of distribution is large or if there is significant plasma protein binding**.

**Third-space fluid weight is unlikely to contribute to and should not be used when initial estimates of clearance are made**. However, **while not directly influencing clearance, it is possible that the presence of ascites or edema may indicate the presence of a disease process that is known to alter clearance**.

**TBW Significantly Smaller than IBW**

Patients who weigh significantly less than their IBW or are emaciated also require special consideration when estimating renal function. While it may seem counterintuitive, a creatinine clearance calculated for an emaciated subject using the patient’s weight also tends to over predict the patient’s creatinine clearance. This is because patients who are emaciated tend to have a disproportionally greater loss in muscle mass than TBW. Consequently, serum creatinine in the denominator of Equation 3 and 4 decreases more than the weight in the numerator, resulting an overestimate of creatinine clearance. For this reason, **if the patient’s actual weight is less than their IBW, the actual weight should be used when calculating creatinine clearance in emaciated subjects**. **Even then, the creatinine clearance is likely to be overestimated**.

**Low Serum Creatinine Level**

In addition, it has been suggested that when serum creatinine values are < 1.0 mg/dL, more accurate predictions of creatinine clearance can be obtained if these levels are upwardly adjusted or normalized to a value of 1.0 mg/dL. This suggestion is based on the assumption that low serum creatinine values are related to small muscle mass and a decreased creatinine production rather than to an unusually large creatinine clearance. It is a common practice for clinicians to normalize serum creatinine values < 1 to 1 mg/dL. However, there is evidence suggesting that using the actual serum creatinine values of < 1 mg/dL result in more accurate estimates of creatinine clearance. Because of this continuing controversy and the difficulty in estimating creatine clearance accurately, it is important to use clinical judgement in evaluation the risk versus the benefit of drug therapy. When a serum creatinine of < 1mg/dL is used in Equation 3 and 4, most clinicians would recommend setting an upper limit for creatinine clearance. As an example, a 50-year-old man weighing 60 kg with a serum creatinine of 0.5 mg/dL would have a calculated creatinine clearance of 150 mL/min if the serum creatinine of 0.5 mg/dL is used. And a value of 75 mL/min if the serum creatinine is normalized to 1 mg/dL.

Even if the first method is used, many clinicians would suggest that an upper limit for a calculated creatinine clearance should be set at somewhere near 120 mL/min. Of course in specific situations (e.g., very large, non-obese, young healthy male patient), a creatinine clearance of more than 120 mL/min might be appropriate to consider. Therefore, whether to normalize a patient’s serum creatinine and whether there should be some upper limit for the calculated value of creatinine clearance should be dictated by clinical judgement rather than a specific rule.

**Estimating Time to Reach a Steady-State Serum Creatinine Level**

All the above methods for estimating ClCr require a steady-state serum creatinine concentration. When a patient’s renal function suddenly changes, some period of time will be required to achieve a new steady-state serum creatinine concentration. In this situation, it is important to be able to estimate how long it will take for the SCr to reach steady state. **If a rising serum creatinine is used in any of the previous equations, the patient’s creatinine clearance will be overestimated**.

As presented earlier, half-life is a function of both the volume of distribution and clearance. If the volume of distribution of creatinine (0.5 to 0.7 L/kg) is assumed to remain constant, the time required to reach 90% of steady state in patient with normal renal function is less than 1 day. As an example, the average 70-kg patient with a creatinine clearance of 120 mL/min (7.2 L/hr) with a volume of distribution for creatinine of 45.5 L (0.65 L/kg) would be expected to have a creatinine t1/2 of 4.4 hours.

Under these conditions, 90% of steady state should be achieved in approximately 15 hours (3.3 t1/2s). However, if the same patient had a creatinine clearance of 10 mL/min (0.6 L/min), the creatinine t1/2 would be 52.5 hours and more than a week would be required to ensure that steady state had been achieved. **One useful approach, that helps clinicians to make relatively rapid assessments of SCr, is to remember that as a drug (in this case creatinine) concentration is accumulating toward steady state, half of the total change will occur in the first half-life**. Therefore, two serum creatinine concentrations obtained several hours apart (8 to 12 hours) that appear to be similar (i.e., not increasing or declining significantly) and that represent reasonably normal renal function probably represent steady-state conditions. As renal function declines, proportionately longer intervals between creatinine measurements are required to assure that steady-state conditions exits.

In clinical practice, patients occasionally have a slowly increasing serum creatinine. As an example, a patient might have the following serum creatinine concentrations on 4 consecutive days: 1, 1.2, 1.6, and 1.8 mg/dL. First, it should be recognized that the increase in serum creatinine from day 1 to day 2 could be due to assay error alone, as the **absolute error for most creatinine assays is +- 0.1 to 0.2 mg/dL**. Also, given that the t1/2 of creatinine at concentrations in the range of 1 to 2 mg/dL is approximately 4 to 8 hours, steady state should have been achieved in the first day. Therefore, the continued increase in serum creatinine probably reflects ongoing changes in creatinine clearance over the 4 days. The difficult clinical issue is not what the creatinine clearance is on each of the 4 days, but rather what it will be tomorrow, what is the cause, and how to prevent or minimize the ongoing renal damage.