regulation of water, electrolyte and acid–base balance

 excretion of the products of protein and nucleic acid metabolism: e.g. urea, creatinine and uric acid.

Serum creatinine

The first step in urine formation is the filtration of plasma at the glomeruli (Fig 14.1). The glomerular filtration rate (GFR) is defined as the volume of plasma from which a given substance is completely cleared by glomerular filtration per unit time. This is approximately 140 mL/min in a healthy adult, but varies enormously with body size, and so is usually normalized to take account of this. Conventionally, it is corrected to a body surface area (BSA) of 1.73 m² (so the units are mL/min/1.73 m²).

Historically, measurement of creatinine in serum has been used as a convenient but insensitive measure of glomerular function. Figure 14.3 shows that GFR has to halve before a significant rise in serum creatinine becomes apparent – a ‘normal’ serum creatinine does not necessarily mean all is well. By way of example, consider an asymptomatic person who shows a serum creatinine of 130 µmol/L:

Creatinine clearance

Simultaneous measurement of urinary excretion of creatinine by means of a timed urine collection allows estimation of creatinine clearance. The way this is worked out is as follows. The amount of creatinine excreted in urine over a given period of time is the product of the volume of urine collected (say, V litres in 24 hours) and the urine creatinine concentration (U). The next step is to work out the volume of plasma that would have contained this amount (U × V) of creatinine. This is done by dividing the amount excreted (U × V) by the plasma concentration of creatinine (P):

This is the volume of plasma that would have to be completely ‘cleared’ of creatinine during the time of collection in order to give the amount seen in the urine. It is thus known as the creatinine clearance. Although it is more sensitive than serum creatinine in detecting reduced GFR it is inconvenient for patients and imprecise, and has now been largely superseded by the so-called prediction equations which estimate GFR.

Estimated GFR (eGFR)

The relatively poor inverse correlation between serum creatinine and GFR can be improved by taking into account some of the confounding variables, such as age, sex, ethnic origin and body weight. The formula developed by Cockcroft and Gault in the 1970s, and the four-variable equation derived more recently from the Modification of Diet in Renal Disease (MDRD) Study are the most widely used of these prediction equations. These are compared in Table 14.1.

Limitations of eGFR

Although prediction equations are an improvement on serum creatinine, and creatinine clearance, they are merely estimates of GFR and should be interpreted with caution. For example, they are more likely to be inaccurate in subjects with relatively normal GFR. For this reason, many hospital laboratories do not report a specific result when the GFR is greater than 60 mL/min/1.73 m². Other patient groups where eGFR is less accurate include those with abnormal body shape or mass, e.g. muscle wasting, amputees. Finally, there is evidence that the GFR estimated by the MDRD formula is affected by consumption of meat.

eGFR – additional observations

It is worth putting the limitations of eGFR outlined in the previous section into their proper context. Estimates of GFR e.g. the four-variable MDRD formula, are undoubtedly better than serum creatinine on its own at identifying reduced glomerular function, simply because they take some of the confounding variables into account (see Table 14.1). The Cockcroft–Gault formula requires weight in addition to age and sex (and creatinine) in order to be applied. It is therefore much easier to apply the MDRD formula which incorporates age, sex and ethnicity, but not weight. This is one of the reasons why the MDRD equation is widely used. However, Cockcroft–Gault is still widely used to calculate drug dosages.

Reduced glomerular function, e.g. eGFR 50–60 mL/min/1.73 m², is known to be associated with cardiovascular risk and subsequent progression to more severe renal failure, but much remains to be clarified about this group of patients, e.g. the time-course of progression. This is an area of active research.

Other measures

Cystatin C is a low-molecular-weight protein, the serum concentrations of which, like creatinine, correlate inversely with GFR. However, unlike creatinine, the concentration of cystatin C is independent of weight and height, muscle mass, age (>1 year) or sex and is largely unaffected by intake of meat or non-meat-containing foods. Thus it has been studied as a potential alternative method of assessment.

Various other markers may be used to estimate clearance, but are too costly and labour-intensive to be widely applied: their use is mainly limited to research or specialized nephrology settings such as screening potential kidney donors. They include inulin, iothalamate, iohexol and radioisotopic markers such as ⁵¹Cr-EDTA. The latter is commonly used in paediatric oncology units for estimation of renal function prior to chemotherapy dose calculation.

Proteinuria

Another aspect of glomerular function is its ‘leakiness’. This is dealt with separately on pages 34–35.