eGFR

eGFR measurement

• GFR correlates well with risk adverse outcomes, CKD
• More accurate than serum creatinine
• CKD-EPI equation
• Consider re-testing within 7 days
• Use serum creatinine, not GFR in pregnancy
• If using eGFR for drug dosing
• consider body size – if need to correct use formula

SI units

• Urea mmol/L (Normal 2-7)
• Creatinine umol/L (Normal 49-90)
• Convert umol/L to mg/dL (umol/L x 88.4)
• Convert mg/dL to umol/L (mg/dL x 0.0113)
• Normal creatinine clearance for adults 80-120 ml/min
• Abnormal when woman creatinine 1.2 x normal; male 1.4x normal
• Creatinine clearance =
  • {Creatinine (24 hour urine in mg/dL) / Creatinine (serum in mg/dL)}      X   {Volume (24 hour urine in ml) / Time (hours x 60min)

How is GFR measured or estimated?

1. Creatinine (urine and serum):
   1. Filtered but also mildly secreted in PCT (therefore overestimates) 
   2. Endogenously produced
   3. a product of muscle metabolism with near constant production.
2. Inulin (gold standard):
   1. Filtered only
   2. Not made endogenously and must be injected.
3. Urea:
   1. Endogenous product of protein intake. 
   2. It is filtered and reabsorbed therefore not an ideal marker.
4. Cystatin C:
   1. non-glyosylated protein produced by all nucleated cells that is less variable and less affected by age and sex. 
   2. Freely filtered and reabsorbed plus catabolised by tubular epithelial cells. urinary clearance cannot be measured. 
   3. May be a more accurate filtration marker than creatinine and better predictor of adverse events in elderly (mortality, cardiac failure, peripheral arterial disease)

equations estimate GFR from serum creatinine

1. Cockcroft-Gault
   • Developed 1970s from 250 men with a wide range of CrClr.
   • Not adjusted for BSA. Adjusted for age, lean body weight and sex
   • Since unadjusted for BSA, can be used for drug doing
   • Pharmacokinetics used Cockcroft-Gault to determine level of kidney function and drug dosage adjustment. NOTE! The creatinine assays were not standardised, therefore still inconsistent translation into clinical practice
2. MDRD (Modification of Diet in renal Disease)
   • Developed from 1630 patients with CKD. It is more accurate than Cockcroft Gault.
   • It is adjusted for BSA, age, sex and race (if African American) but not weight
   • Less accurate at eGFR >60ml/min/1.73m3
   • Not validated in <18 years old, pregnant and elderly
   • For drug dosing, MDRD needs to be unadjusted for BSA
3. CKD-EPI (CKD Epidemiology Collaboration)
   • is the preferred method in Australia for calculating kidney function in clinical practice and has been shown to have greater accuracy and precision compared to other formulae
   • Accurate as MDRD in the subgroup with eGFR <50ml/min/1.73m3 but also accurate in the subgroup with eGFR>60ml/min/1.73m3
   • Adjusted for age, sex and race
   • Not validated for children <18, pregnant and some ethnic subgroups

Problems associated using eGFR:

1. Using serum creatinine as filtration marker
2. Decreased accuracy at higher eGFR
3. non-steady state conditions for filtration marker when GFR is changing
4. Serum creatinine is affected by
   • production (muscle and diet)
   • PCT secretion and extrarenal (GIT and liver) excretion which causes a wide variation amongst individuals therefore wide-range of abnormal cutoffs
5. As such GFR must decline to approximately 50% of normal before serum creatinine concentration rises above the upper limit of normal. 
6. Always remember drug interactions, some can inhibit creatinine secretion e.g. cimetidine

Clinical situations where eGFR results may be unreliable and/or misleading:

• Acute changes in kidney function (e.g., acute kidney injury)
• People on dialysis
• Recent consumption of cooked meat (consider re-assessment when the individual has fasted or specifically avoided a cooked meat meal within 4 hours of blood sampling)

• Exceptional dietary intake (e.g., vegetarian diet, high protein diet, creatine supplements)
• Extremes of body size
• Diseases of skeletal muscle, paraplegia, or amputees (may overestimate eGFR)
• High muscle mass (may underestimate eGFR)
• Children under the age of 18 years
• Severe liver disease present
• eGFR values above 90 mL/min/1.73m2
• Drugs interacting with creatinine excretion (e.g., fenofibrate, trimethoprim)
• Pregnancy