Estimated GFR calculator

Calculation method

eGFR = 175 × [creatinine × 0.011312]^-1.154 × [age]^-0.203 × [1.212 if afro carribean race] × [0.742 if female]

Notes

• eGFR is the preferred NHS measure of renal function but will give unreliable estimates in acute renal failure (as will serum creatinine).
• The eGFR figure is normalised in units ml/min/1.73 m² and chronic kidney disease is suggested by a GFR of less than 60 ml/min/1.73m² present for at least 3 months. eGFR estimates of 90 ml/min/1.73m² or less should not be considered as evidence of kidney disease unless there is other evidence of renal damage.
• For prescribing purposes the BNF defines mild renal impairment as a GFR of 20-50 ml/min, moderate renal impairment from 10-20 ml/min and severe renal impairment as less than 10 ml/min. Certain drugs will have other cut-offs. For example, enoxaparin's licensed dose is reduced at creatinine clearance less than 30 ml/min and the glitazones have cut-offs that vary from drug to drug. Note that the BNF uses absolute GFR (i.e. not adjusted for body surface area).
• eGFR is not presently validated in pregnancy, gross œdema, patients with muscle wasting or malnourishment and amputees. It may also be affected by food intake.
• eGFR falls with age so many healthy people aged >75 years will have an eGFR of 50 - 60 ml/min/1.73m  . For further information please see www.merseyrenalunits.nhs.uk
• This estimation should not be used in children <18 years, in acute renal failure, pregnancy, oedematous states, muscle wasting states, amputees, malnourished patients. In children less than 18, specialist units in the UK are recommended to use the Schwartz equation.
• Racial variation has been well validated in 'African Americans' in USA and is believed to extend to Black Afro-Caribbeans in UK. The calculation above assumes this or Caucasian race.  eGFR calculations have not been validated in other ethnic groups (eg Indo Asians) or in subjects of mixed race. eGFR performs less well in Chinese patients
• eGFR may be affected by food intake.
• You are expected to check that the results of calculations are reasonable and appropriate for the clinical context. For example very lean frail woman would have GFR overestimated by these equations. Above 89 ml/min/1.73 m² the eGFR has a wide variance (30% or more) and should not be regarded as being as accurate as more direct clinical chemistry measurements (typical variances <10%).

Further information: www.ganfyd.org

© C Ryall, Royal Liverpool Broadgreen University NHS Trust