Plasma Creatinine for Testing Renal Function: Avoid Misinterpretation
WSAVA 2002 Congress
H.P. Lefebvre (1), DMV, PhD, Dipl. ECVPT; J.P. Braun (1), DMV, PhD; A.D.J Watson (2), BVsc, PhD, Dipl. ECVPT
(1) UMR 181 Experimental Physiopathology and Toxicology, National Veterinary School of Toulouse
(2) Veterinary Medicine, Sydney University

Plasma (or serum) creatinine (Pl-Cr) concentration is mainly measured in dogs and cats to aid diagnosis of renal dysfunction. Most of the data available have been obtained in dogs. The main assumption for interpretation of Pl-Cr is that Pl-Cr increases when renal function decreases, because Pl-Cr is freely filtered by the glomerulus, and not reabsorbed by the renal tubules, except perhaps under conditions of partial urine outflow impairment. The aim of this lecture is to outline the main factors that should be taken into account to avoid misinterpretation of Pl-Cr.

1. Check the conditions before sampling

The animal should be fasted-Food intake and quality of food may influence the Pl-Cr value. There was in dogs for example, no change after feeding raw meat, but Pl-Cr concentrations increased by a mean of 45% one hour after ingestion of cooked meat, and remained increased for 12 hours(9).Prolonged heating of the meat results in the progressive degradation of creatine to Cr. In fasting animals, Pl-Cr concentration is stable 24 hours (10).

The animal should be normally hydrated-Any increase in Pl-Cr can be interpreted as a decrease in renal function only if the hydration state remains constant between the two consecutive samples (see later).

Exercise has limited effect on Pl-Cr-This factor is often mentioned in human literature, but seems to have limited effect in dogs. Pl-Cr increased in Greyhounds immediately after a sprint of 400 m; the increase persisted for about 60 minutes, but remained within the reference range (6). Pl-Cr did not change in sled dogs after a long distance race (3). In untrained Beagles, no major change was observed in Pl-Cr value during or after a 1 hour running exercise(1).

No study has reported a major effect of the time of day of sampling on Pl-Cr value.

2. Sample in optimal conditions

 Sampling site-The sampling site has a minor effect on Pl-Cr; the concentration is slightly lower (up to 15 µmol/L) in the cephalic vein than in the jugular.

 Anticoagulants-Plasma and serum Cr values are very similar.

 Storage conditions-In total blood, Cr is stable up to 24-36 hours at room temperature and up to 4 days at 4°C. In plasma and serum, Cr is stable at 20°C for 3 days. In frozen plasma or serum, the stability is excellent: after 8 months at -20°C, the decrease in Cr concentration is only 6.5 µmol/L (5).

3. Choose the enzymatic assay methods in preference

Two assay methods are currently used. Jaffé's method is the less expensive but not specific and many interfering substances leading have been reported to cause overestimation of Cr concentration. Cephalosporins may increase Pl-Cr up to 50% with Jaffé's method. The enzymatic method is more specific, although an interference by bilirubin can decrease the Pl-Cr value obtained (this occurs also with Jaffé's method). Hemolysis seems not to interfere with Cr assay. Another important point is between-day assay variation, which may induce misinterpretation of Pl-Cr value. Good quality control is therefore mandatory to detect such variations. When serial testing is likely, another solution is to deep-freeze an aliquot of the sample, to be thawed for Cr assay, simultaneously and under the same conditions with the next sample (5).

4. Be wary of interpretations based only on published reference intervals

This is probably the most critical step for the final interpretation of Pl-Cr, i.e., absence or presence of hypercreatininemia.

What is the "true" upper limit? Practically, if the observed Cr value is higher than a threshold value (the upper limit of the reference interval), it is considered as abnormal and if other suggestive clinical and biological findings are present (such as inappropriate urine specific gravity), it may contribute to the diagnosis of renal dysfunction. The reference intervals that are available differ greatly according to the source (various textbooks of veterinary medicine), the analytical method (Jaffé's vs enzymatic), and the sampled population (which is generally poorly documented). Most reference intervals overlap but the upper limit may differ considerably (from 105 µmol/L to about 250 µmol/L in dogs, for example)(5). It is generally considered that Pl-Cr "increases," i.e., becomes higher than the upper limit of the reference interval, when >75% of kidney function has been lost. In practice, however, the observed value is considered abnormal mostly according to the extent of the increase over the upper limit of the reference interval. For example, when the upper limit is 120 µmol/L, a 140 µmol/L value might not be considered as important whereas a 250 µmol/L will be.

The reference interval may changed according to physiological variables, especially age and muscle mass. Controversial results have been obtained about the effect of age on Pl-Cr value: age has been reported variously to increase, decrease or have no effect on Pl-Cr value in adult dogs. In puppies, Cr value increased between 2 weeks and 6 months and then remained stable. Large muscle mass is reported in many textbooks to increase Cr production and therefore Pl-Cr, but no correlation was found between body weight and Pl-Cr value (8). Gender differences in muscle mass do not appear to affect Pl-Cr(2).

Another approach for interpreting variations in Pl-Cr is to repeat measurements over time, at intervals appropriate to the expected rate of development or improvement of disease, and to compare successive values to identify any trends. This is better than the "reference interval" approach, but requires conditions of sampling and assays to be standardised. Another difficulty is to know the critical difference required between consecutive values to conclude that Pl-Cr concentration is increasing or decreasing. This difference depends on both analytical and intra-individual variability. In one study in healthy dogs, the critical difference for Pl-Cr was estimated as 35 µmol/L (4). That means that it is not possible to conclude Pl-Cr is increasing if the difference between the second and the first samples is less than +35 µmol/L. Unfortunately, a critical difference has not been defined in diseased conditions.

The relationship between Pl-Cr and glomerular filtration rate (GFR) is curvilinear (2), which means that Pl-Cr does not increase much during the early stages of renal insufficiency while GFR is falling, but that Pl-Cr may change considerably while GFR remains at very low levels. Thus, GFR may fall from 3 to 1.5 mL/kg/min, whereas Pl-Cr remains under the upper limit of the reference interval. On the other hand, Pl-Cr in a dog with end-stage renal insufficiency may fall from 1000 to 500 µmol/L (i.e., a 50%-decrease) following sustained therapy, but the corresponding GFR value might improve only from 0.2 to 0.3 mL/kg/min (normal value about 3 mL/kg/min), indicating a recovery of only 3% of the normal renal function, that is a poor response to treatment. Such opposite wide variations should be kept in mind when interpreting successive Pl-Cr values.

5. Always remember that Pl-Cr depends not only on renal elimination, but also on production and distribution

Recent data indicate that Pl-Cr should be interpreted according to pharmacokinetic considerations (7, 10). Pl-Cr is actually an hybrid variable and its interpretation requires assumptions about its input rate into plasma, and its clearance and volume of distribution. Any increase in Pl-Cr is mostly considered as an indirect marker of decreased clearance, suggesting decreased renal function. This is true only if production rate and volume of distribution are unaltered. But in dogs with mild subclinical experimental renal failure, the daily production rate of Cr was decreased by about 25% compared to healthy conditions (10). Such a decrease is only partially explained by a decrease in body weight, and could also result from decrease in synthesis of creatine, the precursor of Cr. Whatever the cause, it is evident that the decrease in production rate reduces the effect of renal dysfunction on Pl-Cr. It explains why Pl-Cr is relatively insensitive in early stages of renal dysfunction. The volume of distribution of Cr is about 600 mL/kg, which corresponds to the volume of body water. It is clear also that dehydration may reduce this volume and increase the Pl-Cr concentration. For this reason, if Pl-Cr has to be interpreted as a marker of GFR, it can be done only if the animal is normally hydrated. On the other hand, Pl-Cr may also represent a useful marker of the hydration state in other diseases. When volume of distribution and production rate remain constant, any change in GFR will induce a proportional inverse change in Pl-Cr.

6. Plasma exogenous creatinine clearance test (PECCT)

This test, recently proposed (10), offers a practical way to detect renal insufficiency in a dog in which a renal disease is suspected, based for example on marginal azotemia and/or unexplained polyuria. Cr is injected (dose level 80 mg/kg) in the hospitalised and fasted animal, by bolus i.v. administration. 1-mL (or less) of blood is sampled just before (basal) and at 5 min, 1, 2, 6 and 10 h. The plasma clearance of Cr is then calculated by divided the dose injected by the area under the Pl-Cr vs time curve. If the clearance is lower than 2 mL/kg/min, follow-up examination is recommended. Another option is the dynamic test which requires just two blood samples: one before creatinine injection and one at a fixed time afterwards. If the difference in Pl-Cr between the two exceeds a given threshold value (e.g., 637 µmol/L at 2 hours), the animal may be considered renal-impaired.

In conclusion, Pl-Cr remains one of the most important clinical chemistry determinations, but its interpretation is still a challenge, especially for clinical cases with borderline values. The conditions of sampling should be standardised. Repeated measurements may help considerably by increasing sensitivity. The volume of distribution and the production rate of Cr should be taken into account before any interpretation. The PECCT seems to be a promising additional tool for discrimating between patients with and without renal insufficiency.


1.  Chanoit GP et al. J Vet Med A, 2002, in press

2.  Finco DR et al. J Vet Pharmacol Ther, 1995, 18:418-421.

3.  Hinchcliff KW et al. J Am Vet Med Assoc, 1993, 202:401-405.

4.  Jensen AL, Aaes H. Res Vet Sci, 1993, 54:10-14.

5.  Lefebvre HP, et al. Rev Med Vet, 1998, 149: 1-8.

6.  Rose RJ, Bloomberg MS. Res Vet Sci, 1989, 47, 212-218.

7.  Toutain PL et al. Rev Med Vet, 2000, 151:643-648.

8.  van Den Brom WE, Biewenga WJ. Res Vet Sci, 1981, 30:152-157.

9.  Watson ADJ et al. Am J Vet Res,1981, 42: 1878-1880.

10. Watson ADJ et al. J Vet Intern Med, 2002, 16:22-33.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Hervé P. Lefebvre, DMV, PhD, Dipl. ECVPT
UMR 181 Experimental Physiopathology and Toxicology
National Veterinary School of Toulouse

A.D. J Watson, BVsc, PhD, Dipl. ECVPT
Veterinary Medicine, Sydney University

J. P. Braun, DMV, PhD
UMR 181 Experimental Physiopathology and Toxicology
National Veterinary School of Toulouse

MAIN : : Plasma Creatinine
Powered By VIN