GFR in practice: assessment of glomerular filtration rate in dogs (2015)

R Heiene, Oslo, Norway, HP Lefebvre, Toulouse, France and ADJ Watson, Sydney, Australia


Practising veterinarians use information from several sources to detect kidney disease, including elements from the history, physical examination and various diagnostic aids (for example, urine specific gravity and protein content, blood creatinine and urea concentrations, renal imaging...). In most cases, careful consideration of these findings will enable the clinician to determine whether overall kidney function is significantly compromised.

But the decision is not always clear-cut and some difficult cases remain – for example, those with polyuria/polydipsia where other diagnostic possibilities have been excluded, or where there is persistent, unexplained azotaemia.

In such instances, further assessment of renal function might be desired. One approach might be to institute “watchful waiting”, that is, to continue monitoring and to reassess at intervals subsequently.

An alternative could be to determine the dog's glomerular filtration rate (GFR), as GFR is considered to be the single most useful and sensitive indicator of overall renal function. In the past, GFR assessment has been considered impractical in most veterinary clinics, but two recently described plasma clearance methods have improved the situation.

Both these methods require repeated blood sampling over a period of several hours following intravenous administration of a suitable marker. A suitable marker is a substance that is filtered freely by glomeruli, undergoes no renal tubular resorption or secretion, and is neither metabolized nor eliminated by extrarenal means. Both iohexol and creatinine have been shown to be suitable markers in dogs and cats.

Using these techniques, GFR can be estimated by measuring the plasma clearance (Cl) of the marker. Cl is calculated by dividing the intravenous dose administered (D) by the area under the curve (AUC) of plasma concentration plotted against time (i.e., GFR = Cl = D/AUC). The AUC is determined manually or by computer using a 1-compartment or 2-compartment model, or by the trapezoidal method (Heiene and Moe 1998). If using creatinine, first subtract the basal (preinjection) sample value from the concentrations in each of the other timed samples.

Whichever method is chosen, accuracy is important for the dose of marker administered and the exact times samples are collected – errors with either will give false results when calculating GFR. For example, if the 120 minute sample is collected 4 minutes late, record the collection time as 124 minutes, because it is this exact sampling time that is required when calculating AUC and therefore GFR.

Further studies are needed to clarify reference ranges for GFR in dogs, and these may depend on the method chosen and some laboratory factors. However, most published GFR values for dogs are from approximately 2 to 5 ml/min/kg, so values < 1.5 ml/min/kg can be considered abnormal (Heiene and Lefebvre 2007).

Presented here is a summary of the techniques used with iohexol and creatinine. More detailed information on GFR assessment can be found in the accompanying article “Glomerular filtration rate in dogs and cats” (

Plasma iohexol clearance

Materials required

  • Iohexol solution (300 mg iodine/ml).
  • Sterile saline (0.9% NaCl)
  • Intravenous catheter with stopper, and syringes, needles and tubes for blood samples
  • Timer


  1. Fast patient overnight.
  2. Hospitalize patient, continuing to provide water but not food.
  3. Record current body weight (kg).
  4. Taking care to be accurate, administer the iohexol solution as a bolus through the intravenous catheter with stopper, using 0.5 ml/kg for dogs with azotaemia or 1 ml/kg for those without.
  5. Flush the dead space of the catheter with 2 ml of saline, immediately start the timer, then remove the catheter.
  6. Collect blood samples (each at least 1 ml) at 2, 3 and 4 hours after administration of iohexol, using sites different from where iohexol was administered to avoid contamination by residual iohexol at this site. Note the exact time each sample is collected.
  7. Centrifuge samples and harvest plasma to send to the laboratory. Two laboratories currently offering iohexol assays on a commercial basis are:
    1. The Royal Veterinary College, UK, in collaboration with deltaDOT (
    2. Michigan State University (
  8. It is advisable to contact the laboratory beforehand for information on pricing and sample requirements, and to ascertain the laboratory's reference range and the method used to determine the clearance value.


  • Iohexol is stable and samples can be mailed to the laboratory without special precautions.
  • Total time for the procedure is short.


  • Samples have to be sent away for analysis.
  • The cost of iohexol may be a limiting factor for some owners of large dogs.
  • Adverse reactions have been reported (very rare in dogs).
  • A correction needs to be applied to the clearance value calculated with the 3 samples to improve accuracy (Heiene and Moe 1999). The correction formula is:
    Clcorr = - 0.03 +1.06 (Clobs) - 0.07012(Clobs)2
    where Clcorr = corrected value and Clobs is the value determined from the 3 samples.

Plasma creatinine clearance

Materials required

  • Anhydrous creatinine powder (Sigma Aldrich)
  • Sterile saline (0.9% NaCl)
  • Filter (0.2 μm)
  • Intravenous catheter with stopper, and syringes, needles and tubes for blood samples
  • Timer


  1. Fast patient overnight.
  2. Hospitalize patient, continuing to provide water but not food.
  3. Record current body weight (kg).
  4. Collect blood sample to measure basal value (creatinine concentration at time zero).
  5. Prepare creatinine solution (80 mg/ml): dissolve creatinine in saline and sterilize by filtration through a 0.2 μm filter.
  6. Taking care to be accurate, administer the creatinine solution (1 ml/kg) as a bolus through the intravenous catheter with stopper, followed by 2 ml of saline, and immediately start the timer. Remove the catheter.
  7. Collect 1 ml blood samples at 10 minutes and at 1, 2, 6 and 10 hours after administration, noting exact time each sample was collected.Collection sites should be different from the administration site to avoid contamination by residual injected creatinine.
  8. Determine creatinine concentrations in all 6 samples in the same batch of assays.
  9. Calculate the plasma clearance value from the dose administered and the area under the plasma creatinine vs time curve (AUC), after first subtracting the basal value from each of the 5 post-administration values.


  • Assays for creatinine in plasma or serum are readily available in practice laboratories or externally.
  • Required sample volume is small.
  • Time required to perform creatinine assays is short.


  • A suitable creatinine solution is not commercially available and must be prepared specially. This is best done in an appropriate laboratory or pharmacy, rather than in a practice laboratory, to ensure it is sterile and pyrogen-free, and that the concentration is precise (in order to know the exact dose given).
  • Longer hospitalization is required than with iohexol.

Further reading

Heiene R and Lefebvre HP (2007) Assessment of renal function. In “BSAVA Manual of Canine and Feline Nephrology and Urology” 2nd edn, eds J Elliott and GF Grauer, pp 117-125.

Heiene R and Lefebvre HP (2013) Glomerular filtration rate in dogs and cats (

Heiene R and Moe (1999) The relationship between some plasma clearance methods for estimation of glomerular filtration rate in dogs with pyometra. J Vet Intern Med 13:587-596.