Alan Segal, MD

Radiocontrast agents can adversely affect the kidney, and the risk of acute injury is higher in patients with underlying renal dysfunction or unstable renal function. For decades, clinicians have tried to use the plasma creatinine concentration to assess renal function. Plasma creatinine concentration is not linearly related to renal function, however; in the absence of relevant anthropomorphic information, the use of plasma creatinine concentration alone can lead to significant errors and misinterpretations. Improved methods to calculate the estimated glomerular filtration rate (eGFR), along with pertinent clinical information, can help radiologists assess the potential risk of contrast-induced nephropathy (CIN) in patients scheduled to receive intravenous contrast. Emerging methods for calculating and interpreting the eGFR will help radiologists assess renal function and the potential risk of CIN.

The use of intravenous contrast greatly improves the quality and diagnostic value of many radiological studies. The contrast itself, however, can injure the kidney, leading to acute hemodynamic and/or nephrotoxic acute renal failure.1,2 In fact, CIN is reported to be the third most common cause of acute renal failure among hospitalized patients.3 Although acute renal failure itself is typically reversible, this complication prolongs hospitalization and significantly increases the risk of morbidity and mortality.

Well-known factors that increase the risk of contrast injury include baseline abnormal renal function (especially in patients with diabetic nephropathy), defined as an eGFR of less than 60 mL per minute per 1.73 m2 of body-surface area (BSA); hypovolemia; and the presence of other drugs that may adversely affect the kidney (such as nonsteroidal anti-inflammatory drugs, aminoglycosides, angiotensin-converting enzyme inhibitors, and angiotensin-receptor blockers). As a first step in minimizing the risk of CIN, eGFR should be obtained for most patients prior to contrast exposure. Screening tools that rely on medical history may be used to reduce the number of patients for whom an estimate of renal function is necessary. For example, an estimate is recommended in all individuals more than 70 years of age or having a history of diabetes, hypertension, renal disease, or cardiac disease. Individuals with one or more of these conditions are at increased risk for renal injury from radiocontrast.

Plasma Creatinine Concentration

A steady-state level of creatinine (a stable plasma creatinine concentration) is achieved when the creatinine excretion rate is equal to the creatinine production rate. Creatinine is produced in muscle by the metabolism of creatine phosphate at a rate of about 20 mg/kg per day and is generally higher in men (20 to 25 mg/kg per day) than in women (15 to 20 mg/kg per day) because of gender-associated differences in muscle mass.

Although plasma creatinine concentration has been (and probably remains) the most commonly used test by which renal function is judged, it is nonlinearly related to GFR and notoriously insensitive to actual changes in GFR. Given the variability in patients’ muscle mass and the laboratory determination (and reported normal range) of the blood test, using plasma creatinine concentration to assess renal function is difficult, to say the least. Consider, for example, a 25-year-old, healthy, 75-kg man with a plasma creatinine concentration of 1 mg/dL and a 75-year-old man who weighs 65 kg and has the same plasma creatinine concentration. Whereas the younger man has an eGFR of 115 mL per minute per 1.73 m2, the older man has an eGFR of only 62 mL per minute per 1.73 m2, nearly 50% less than the younger man.

Estimation of Renal Function

Table 1. (Click the images for a larger version.)

The kidney performs myriad physiological functions, including regulation of salt and water homeostasis, acid-base balance, calcium metabolism, and production of erythropoietin. Although there is no single test that accurately reflects overall renal function, the GFR of the kidney has always been used as the major renal performance index. GFR peaks in young people and decreases linearly with age at a rate of about 0.75 mL per minute per 1.73 m2 per year between the ages of 20 and 704 (see Table 1). The decline in GFR with age is small and linear in the healthy population, which helps make GFR easy to interpret and clinically relevant.

These numbers are not adjusted for factors such as weight, gender, race, and diet. Normalization to BSA, however, may indirectly and partially adjust for some other factors. The average BSA is 1.9 m2 for men and 1.6 m2 for women. The standard value of BSA for normalization is 1.73 m2. There are several formulae in the literature for estimating BSA based on weight and height, the first of which was published by Dubois and Dubois5 in 1916:

BSA=[71.84x(kg)0.425x(cm)0.725]/10,000

It should be noted that this equation was based on only nine patients. In 1970, another equation based on many more patients was formulated by Gehan and George6:

BSA=[0.0235x(kg)0.51456x(cm)0.42246]

Table 2. (Click the images for a larger version.)

As a quick comparison, consider a 70-kg person who is 172 cm tall. The Dubois-Dubois equation yields a BSA of 1.825 m2, while the Gehan-George equation yields 1.84 m2. Here, the Gehan-George equation will be used to calculate BSA.

Equations for Estimating GFR

Several equations have been developed to calculate either the estimated creatinine clearance or eGFR for patients with completely stable renal function. The equation most often used for estimated creatinine clearance was developed by Cockcroft and Gault.7,8 The Cockcroft-Gault equation, based on blood and urine collections from 249 men with creatinine clearances ranging from 30 to 130 mL per minute, is:

estimated creatinine clearance for a man=[(140-age)xkg/(72xplasma creatinine concentration)]

estimated creatinine clearance for a woman=[(140-age)xkg/(84.7x plasma creatinine concentration)]

The Cockcroft-Gault equation yields an estimate of the absolute creatinine clearance that is not normalized to BSA. Moreover, due to proximal tubular secretion of creatinine,9 the creatinine clearance overestimates the GFR by 5% to 10%. This error increases as the true GFR decreases.

Recently, Levey et al10 developed several equations to estimate GFR in mL per minute per 1.73 m2. Because the population used was from the Modification of Diet in Renal Disease (MDRD) Study Group, these formulae are referred to as the MDRD equations. Among nephrologists, the simplified version of the MDRD equation has become the method of choice to estimate GFR in patients with abnormal, yet stable, renal function.9,11

It is important to note that the MDRD equations have been validated in mostly white patients with chronic kidney disease (with a mean GFR of about 40 mL per minute per 1.73 m2) who did not have diabetic nephropathy or renal transplants, and have not been validated in people with normal renal function.

Unstable Renal Function

The empirical equations for estimating GFR require stable (steady-state) renal function. Of course, this is not the case for a number of hospitalized patients, especially those who are critically ill. Surprisingly little work has been done to develop reliable methods to estimate the GFR in patients with acute renal injury. In fact, a key piece of information that radiologists need to know is whether a patient scheduled to receive contrast has unstable renal function. If so, it is strongly recommended that a nephrology consultation be obtained prior to the study.

Comparison of the Modification of Diet in Renal Disease Study Group equation (black squares) and the Cockcroft-Gault equation (red circles) for a 60-year-old man who weighs 70 kg and has a plasma creatinine value ranging from 0.8 to 5 mg/dL. (Click the images for a larger version.)

Recently, a tool has become available that may be helpful in patients with fluctuating renal function.12 At Fletcher Allen Hospital, Burlington, Vt, we have developed a user-friendly Microsoft Excel® program that would allow clinicians to estimate GFR in such patients, and we are currently validating the model in patients exposed to radiocontrast. This model is in excellent agreement with the MDRD equation for the steady-state case, which is a boundary condition that must be met for such a tool to be robust.

Identifying CIN

Estimating renal function from the plasma creatinine concentration when renal function is changing is important for defining CIN. The injury induced by contrast results both in alterations in renal tubular cell function and in a decrease in GFR. Only the latter is currently accepted as defining CIN. Ideally, individuals who suffer a loss of GFR that is clinically significant would be identified. While a loss of 25 mL per minute per 1.73 m2 would be well tolerated by someone starting with a GFR of 125 mL per minute per 1.73 m2, it might result in the need for dialysis in someone with a GFR of 40 mL per minute per 1.73 m2. For this reason, the absolute loss of GFR does not correlate with outcomes very well and needs to be interpreted in the context of the baseline GFR.

Using a relative loss of 25% or 50% of baseline GFR is more likely to reflect levels of renal injury with similar clinical consequences. A relative increase in plasma creatinine concentration of more than 25% has often been used to define CIN. This change in plasma creatinine concentration has been correlated with clinical outcomes, including mortality.13 This relative increase in plasma creatinine concentration, however, does not translate into a loss of 25% of GFR. Creatinine secretion increases as plasma creatinine concentration increases and, therefore, GFR can fall with only a minimal increase in plasma creatinine concentration. In general, the increase in plasma creatinine concentration underestimates the true change in GFR.

Some work has mistakenly used an absolute increase in plasma creatinine concentration of 0.5 mg/dL over baseline to define CIN. Simple mathematics demonstrates that any individual with a baseline plasma creatinine concentration of less than 2 mg/dL will need to suffer a greater renal injury to achieve a 0.5 mg/dL increase than to achieve an increase in plasma creatinine concentration of 25%, thus raising the bar for defining CIN. Indeed, most clinical trials of contrast media are conducted in individuals with plasma creatinine concentration of less than 2 mg/dL, and the mean reported incidence of CIN is usually lower for that reason.

Alan Segal, MD, associate professor of medicine, University of Vermont School of Medicine and staff physician, division of nephrology, Fletcher Allen Health Care, Burlington, Vt.

References

  1. Barrett BJ, Parfrey PS. Clinical practice. Preventing nephropathy induced by contrast medium. N Engl J Med. 2006;354:379-386.
  2. Solomon R. Contrast-medium-induced acute renal failure. Kidney Int. 1998;53:230-242.
  3. Nash K, Hafeez A, Hou S. Hospital-acquired renal insufficiency. Am J Kidney Dis. 2002;39:930-936.
  4. Coresh J, Astor BC, Greene T, Eknoyan G, Levey AS. Prevalence of chronic kidney disease and decreased kidney function in the adult US population: Third National Health and Nutrition Examination Survey. Am J Kidney Dis. 2003;41:1-12.
  5. DuBois D, DuBois EF. A formula to estimate the approximate surface area if height and weight are known. Arch Intern Med. 1916;17:863-871.
  6. Gehan EA, George SL. Estimation of human body surface area from height and weight. Cancer Chemother Rep. 1970;54:225-235.
  7. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16:31-41.
  8. Gault MH, Cockcroft DW. Letter: creatinine clearance and age. Lancet. 1975;II:612-613.
  9. Poggio ED, Wang X, Greene T, Van Lente F, Hall PM. Performance of the modification of diet in renal disease and Cockcroft-Gault equations in the estimation of GFR in health and in chronic kidney disease. J Am Soc Nephrol. 2005;16:459-466.
  10. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med. 1999;130:461-470.
  11. Coresh J, Stevens L. Kidney function estimating equations: where do we stand? Curr Opin Nephrol Hypertens. 2006;15:276-284.
  12. Jelliffe R. Estimation of creatinine clearance in patients with unstable renal function, without a urine specimen. Am J Nephrol. 2002;22:320-324.
  13. Gruberg L, Mintz GS, Mehran R, et al. The prognostic implications of further renal function deterioration within 48 h of interventional coronary procedures in patients with pre-existent chronic renal insufficiency. J Am Coll Cardiol. 2000;36:1542-1548.