The elevated serum urea : creatinine ratio in canine babesiosis in South Africa is not of renal origin

INTRODUCTION The early diagnosis of acute renal failure in canine babesiosis is crucial. Azotaemia is an important biochemical finding in acute renal failure. The serum concentrations of urea and/or creatinine are commonly used as indicators of azotaemia in canine babesiosis. Serum urea is commonly elevated in canine babesiosis in a manner disproportionate to serum creatinine. In a recent retrospective study of approximately 400 Babesia canis cases, the mean and median serum urea concentrations were approximately double the normal laboratory serum concentrations, whilst mean and median serum creatinine concentrations fell within normal laboratory values. Although it was not evaluated, this phenomenon would have caused an increased serum urea:creatinine ratio. The normal serum urea:creatinine ratio reported in dogs is 10–15 (when calculated using mg/d as the unit of measurement). In human medicine a serum urea:creatinine ratio in an azotaemic patient of ≥ 20 indicates prerenal azotaemia, whereas a serum urea:creatinine ratio of <20 in an azotaemic patient indicates intrinsic renal disease. This would indicate that serum urea is more likely to increase due to prerenal factors than serum creatinine, whereas both parameters are equally likely to increase due to renal disease. It has been speculated that this phenomenon in canine babesiosis may be caused by a falsely raised serum urea, a form of laboratory error inherent in various test methods, or by prerenal azotaemia, or by serum substances interfering with the analysis of serum creatinine. Some of this speculation may have its origins in the increased serum urea:creatinine ratios found in cases of intestinal haemorrhage, excessive protein loading or increased protein catabolism, where hyperureagenesis is common. It is hypothesised that ammonia loading occurs in canine babesiosis as a result of haemolysis, blood transfusions and gastrointestinal haemorrhage. This could lead to a nonrenal related elevation in serum urea concentrations via hyperureagenesis and could cause the increased serum urea: creatinine ratio apparent in canine babesiosis. In humans reduced cardiac output has been associated with a raised urea:creatinine ratio and this could play a role in babesiosis as there is evidence of myocardial injury in this disease. Although minimal intrinsic renal impairment without overt renal failure has been shown to be common in canine babesiosis, renal failure is a reported complication of canine babesiosis infection. It occurs in less than 3% of South African canine babesiosis infections and appears to be more common in the haemoconcentrated form of the disease. When it does occur it is usually fatal. Typically canine babesiosis cases seen at the Onderstepoort Veterinary Academic Hospital (OVAH) with renal failure are severely azotaemic, with both serum urea and creatinine raised. In addition, cases are frequently oliguric or anuric, with any residual urine being darkly pigmented. Cystatin-C is a cysteine protease inhibitor. It is constantly produced by all nucleated cells, it is freely filtered by the glomerulus, it is metabolised by the luminal tubular epithelium, and is without significant unmetabolised tubular reabsorption. As in human medicine, cystatin-C has been proposed as a more sensitive marker of reduced glomerular


INTRODUCTION
The early diagnosis of acute renal failure in canine babesiosis is crucial.Azotaemia is an important biochemical finding in acute renal failure 12 .The serum concentrations of urea and/or creatinine are commonly used as indicators of azotaemia in canine babesiosis.Serum urea is commonly elevated in canine babesiosis in a manner disproportionate to serum creatinine 9 .In a recent retrospective study of approximately 400 Babesia canis cases, the mean and median serum urea concentrations were approximately double the normal laboratory serum concentrations, whilst mean and median serum creatinine concentrations fell within normal laboratory values 9 .Although it was not evaluated, this phenomenon would have caused an increased serum urea:creatinine ratio.
The normal serum urea:creatinine ratio reported in dogs is 10-15 (when calculated using mg/d as the unit of measurement) 28 .In human medicine a serum urea:creatinine ratio in an azotaemic patient of ≥ 20 indicates prerenal azotaemia, whereas a serum urea:creatinine ratio of <20 in an azotaemic patient indicates intrinsic renal disease 6 .This would indicate that serum urea is more likely to increase due to prerenal factors than serum creatinine, whereas both parameters are equally likely to increase due to renal disease.It has been speculated that this phenomenon in canine babesiosis may be caused by a falsely raised serum urea, a form of laboratory error inherent in various test methods 34 , or by prerenal azotaemia, or by serum substances interfering with the analysis of serum creatinine 4,5,9,15,39 .Some of this speculation may have its origins in the increased serum urea:creatinine ratios found in cases of intestinal haemorrhage, excessive protein loading or increased protein catabolism, where hyperureagenesis is common 32,37 .It is hypothesised that ammonia loading occurs in canine babesiosis as a result of haemolysis, blood transfusions and gastrointestinal haemorrhage.This could lead to a nonrenal related elevation in serum urea concentrations via hyperureagenesis and could cause the increased serum urea: creatinine ratio apparent in canine babesiosis 9,30,32 .In humans reduced cardiac output has been associated with a raised urea:creatinine ratio and this could play a role in babesiosis as there is evidence of myocardial injury in this disease 3,10,11,23 .
Although minimal intrinsic renal impairment without overt renal failure has been shown to be common in canine babesiosis 24 , renal failure is a reported complication of canine babesiosis infection 16,40 .It occurs in less than 3% of South African canine babesiosis infections and appears to be more common in the haemoconcentrated form of the disease 16 .When it does occur it is usually fatal 24,40 .Typically canine babesiosis cases seen at the Onderstepoort Veterinary Academic Hospital (OVAH) with renal failure are severely azotaemic, with both serum urea and creatinine raised.In addition, cases are frequently oliguric or anuric, with any residual urine being darkly pigmented 16 .
Cystatin-C is a cysteine protease inhibitor 1,19 .It is constantly produced by all nucleated cells, it is freely filtered by the glomerulus, it is metabolised by the luminal tubular epithelium, and is without significant unmetabolised tubular reabsorption 33 .As in human medicine 13,20,33  filtration rate than creatinine in dogs, and thus an earlier indicator of renal disease 1,2,19 .Serum haemoglobin, bilirubin, and triglycerides do not interfere with the cystatin-C assay 22 .The human test for cystatin-C has been validated for canine use 1 .
In a recent study in humans in which cystatin-C was compared with creatinine in falciparum malaria, where renal failure is common in both children and adults 7,14 , it was shown that renal disease appears to have been underestimated when using creatinine as a screening test 13 .This may have been due to intrinsic sensitivity properties of the analyte or due to laboratory perturbations specific to the measurement of creatinine in the presence of haemolysis 12,34,39 .Serum creatinine usually only becomes elevated once two-thirds of renal function is lost.Furthermore the by-products of haemolysis, namely bilirubin and haemoglobin, have been reported to interfere with the spectrophotometric analysis of creatinine.Because there appear to be disease mechanisms in common between canine babesiosis and human falciparum malaria 8 , it is possible that serum creatinine also underestimates renal disease in canine babesiosis.
The objective of this study was to report the presence of an increased serum urea:creatinine ratio in canine babesiosis and to determine whether the elevated ratio originates from elevated serum urea or depressed serum creatinine concentrations or both, using serum cystatin-C concentrations as a reference.

MATERIALS AND METHODS
This was a prospective study involving 25 dogs with babesiosis treated at the OVAH (Group 1), and 13 normal reference dogs (Group 2).Dogs were included in the babesia-infected group if they were positive for babesia parasites on thin blood smear and negative for macroscopic serum lipaemia and had a packed cell volume of ≤0.2 / .Group 2 consisted of clinically healthy dogs determined to be free of canine babesiosis by thin blood smear examination.
A serum sample was collected from the jugular vein of each dog into 3 m serum vacuum tubes.All the dogs in the study also had a peripheral blood smear, PCV (by whole blood centrifugation) and urine specific gravity (collected by cystocentesis) performed.Samples were collected prior to any treatment.Serum chemistry was determined by the OVAH Clinical Pathology Laboratory using the following standard methods 4,19,38 .
Serum cystatin-C concentration was measured using the particle-enhanced turbidimetric immunoassay (Diagnostech, Honeydew Dako Cytomation, Cystatin-C PET kit, Denmark), designed for the determination of human cystatin-C 19 .Normal values were set at <1.7 mg/ according to the group of reference dogs (group 2, n =13) used in this study.Serum urea concentrations were measured using the Technicon Method 38 which is a modification of the enzymatic method of Talke and Schubert for the RA-1000 analyser (Bayer (Pty) Ltd Isando, SA).Normal values were 3.6-8.9mmol/ as established in the OVAH Clinical Pathology Laboratory.Serum creatinine concentrations were measured using the Technicon Method 35 which is a kinetic modification of the Jaffé alkaline picrate reaction for the RA-1000 analyser (Bayer (Pty) Ltd Isando, SA).Normal values were <133 µmol/ as established in the OVAH Clinical pathology Laboratory.
Serum urea, creatinine, cystatin-C and serum urea:creatinine ratios were measured or calculated for each dog.Urea and creatinine were converted to mg/d for the calculation of the serum urea: creatinine ratio.Serum urea, creatinine, cystatin-C and serum urea:creatinine ratios were compared between the groups using the Wilcoxon rank-sum test for difference in medians.Two-tailed tests were used and the significance level was set at α = 0.05.A statistical software package was used for data analysis (NCSS 2004, NCSS, Kaysville, Utah, USA).

RESULTS
A summary of the descriptive statistics of the serum chemistry data is provided in Table 1.Serum urea was above the reference range in 17/25 anaemic dogs.Serum urea was significantly elevated in the anaemic babesiosis group compared with the reference group (P < 0.001).Serum creatinine was not elevated in any of the 25 babesiosis dogs.Serum creatinine did not differ significantly between the reference group and the babesiosis group (P = 0.27).Serum cystatin-C was not elevated (>1.7 mg/ ) in any of the babesiosis dogs.Serum cystatin-C levels did not differ significantly between the reference group and the babesiosis group (P = 0.45).The serum urea:creatinine ratio was elevated in 23/25 babesiosis and 3/13 reference dogs.The urea:creatinine ratio was significantly elevated in the babesiosis group compared with the reference group (P < 0.001).Free serum haemoglobin was >1.6 g/ in 8/25 babesiosis dogs.Free serum haemoglobin was significantly elevated in the anaemic babesiosis group compared with the reference group (P = 0.002).

DISCUSSION
This study confirmed that serum urea is often disproportionately increased relative to serum creatinine and serum cystatin-C in anaemic babesiosis dogs.It is therefore not surprising that the serum urea:creatinine ratio was found to be significantly 0038-2809 Tydskr.S.Afr.vet.Ver.( 2006) 77(4): 175-178 The elevated serum urea concentrations encountered therefore represent either a form of laboratory error in the measurement of serum urea or serum creatinine, a form of pre-renal azotaemia, or a form of hyperureagenesis from substrate loading.The most likely cause of laboratory error would be photometric interference by free serum haemoglobin causing a positive bias on the measurement of serum urea concentrations and a negative bias on serum creatinine concentrations 9,15,34,38 .Free serum haemoglobin was significantly elevated in the anaemic babesiosis group compared with the control group in this study.Serum urea was measured by an enzymatic UV method 38 .This method has its absorbance spectra at wavelength 340 nm.This falls within the 300-500 nm range of high intrinsic absorbance that free serum haemoglobin displays 4,5 .Therefore it may be that a proportion of the elevated concentrations of serum urea encountered in the anaemic babesiosis group is a result of laboratory error in which free serum haemoglobin causes a positive bias on its measurement.Retrospectively calculated urea:creatinine ratios (unpublished data) in a study where comparative concentrations of free serum haemoglobin to those encountered in canine babesiosis were infused into dogs to measure the effect on the kidney, were however normal 25 .This indicates that free serum haemoglobin alone is not likely to be the cause of the elevated serum urea encountered in this study.Other potential causes of prerenal azotaemia in humans and dogs include dehydration, hypotensive shock, cardiac disease and rhabdomyolysis 3,11,17,28 .
There are several possible pathophysio-logical explanations for the disproportionate increase in urea in canine babesia infections.Malaria studies have shown that total body water may be depleted in this disease 27 .It has been postulated that a similar event may occur in canine babesiosis 36 .Serum urea may increase before serum creatinine in an acute hypovolaemic state 6 .Hypotensive shock is thought to be a possible cause of renal failure in severe human falciparum malaria patients 21,31,41 .Hypotension has been reported in both complicated and severe uncomplicated canine babesiosis dogs 18 .In a human study on causes of an increased urea:creatinine ratio, chronic congestive cardiac disease was found to be major contributor 3 .There is some evidence for cardiac pathology in canine babesiosis 10,23 .The reporting of two cases of rhabdomyolysis and a significant increase in creatine kinase activity in South African canine babesiosis has led to speculation that increased muscle catabolism may account for the disproportionate increase in serum urea concentrations compared to serum creatinine concentrations observed 17,40 .Other possible sources of substrate for hyperureagenesis may include erythrocyte components released during haemolysis and blood from gastric ulceration 30,32 .A gastrointestinal (GIT) form of canine babesiosis has been described in South Africa 28 Pancreatitis has been shown to be a complication following babesiosis in a few dogs 30 .According to Maegraith 26 , the erythrocyte count may fall by 1 × 10 12 / of blood/day of babesia infection.This represents a daily loss of approximately 12.5 % of the dog's RBC.
The normal percentage loss of RBC per day is 0.9 %.This means that during severe haemolysis 14 times the normal daily haemoglobin load on the body can occur.
Concentrations of both plasma ammonia and serum urea may also increase during a severe catabolic state due to the deamination process involved in using tissue proteins as an energy source.A catabolic state is likely in anaemic canine babesiosis dogs because of systemic inflammation, decreased appetite, tissue hypoxia, metabolic acidosis, pyrexia and the increased work of breathing.Substantial catabolism of endogenous proteins was found in critically ill dogs in a study by Michel 29 .Hyperureagenesis therefore remains a potential cause of the elevated serum urea:creatinine ratio seen in canine babesiosis.
Although the study was not designed to compare babesia-induced renal failure with babesia without renal failure, over the time during which material was collected for this study, 3 cases of the haemoconcentrating form of babesia infection were seen.These dogs had a mean haematocrit of 0.54 / .One of these dogs had a mildly elevated serum urea (13 mmol/ ) in the presence of a normal serum creatinine (86 µmol/ ) and serum cystatin-C (0.6 mg/ ) concentrations.The urea:creatinine ratio for this case was 37.43 and the dog made a full recovery.The second dog also had a mildly elevated serum urea (14.4 mmol/ ) but in the presence of mildly elevated serum creatinine (171 µmol/ ) and serum cystatin-C (2.6 mg/ ) concentrations.The urea:creatinine ratio for this case was 20.85 and this dog also made a full recovery.The third dog had acute renal failure proven on post mortem examination and histopathology.The serum urea in this dog was markedly elevated (41.8 mmol/ ) in the presence of markedly elevated serum creatinine (608 µmol/ ) and serum cystatin-C (5.7 mg/ ).The urea:creatinine ratio in this dog was 17.02.This small group of dogs reinforces the findings of the larger group namely that urea is not useful in the detection of acute renal failure and that cystatin-C and creatinine are as useful as one another for making this diagnosis.It is also interesting to note that in the case with proven acute renal failure, cystatin-C concentration tracked the change in creatinine concentration.
We conclude that serum urea is often elevated due to non-renal factors in canine babesiosis dogs, which causes an elevated serum urea:creatinine ratio.The cause of the elevated ratio remains undetermined, but may be as a result of laboratory error, pancreatitis, gastric ulceration, hypovolaemia, hypotensive shock, cardiac disease or hyperureagenesis.Therefore, serum urea concentration should not be used as an indicator of reduced glomerular filtration in canine babesiosis.The authors advocate the measurement of serum creatinine and hourly urine production to detect significant renal disease in canine babesiosis dogs.There appears to be no added benefit to be derived from the measurement of serum cystatin-C concentrations in canine babesiosis.

Table 1 : Summary of serum chemistry data in canine babesiosis and human falciparum malaria cases.
P-value for the Wilcoxon-rank sum test for differences in medians between babesiosis and control groups. * Journal of the South African Veterinary Association 71: 232-239 31.Naqvi R, Ahmad E, Akhtar F, Naqvi A, Rizvi A 2003 Outcome in severe acute renal failure associated with malaria.Nephrology, Dialysis, Transplantation 18: 1820-1823 32.Prause L C, Grauer G F 1998 Association of gastrointestinal hemorrhage with increased blood urea nitrogen and BUN/creatinine ratio in dogs: a literature review and retrospective study.Veterinary Clinical Pathology 27: 107-111 33.Randers E, Erlandsen E J 1999 Serum cystatin C as an endogenous marker of the renal function -a review.Clinical Chemistry and Laboratory Medicine 37: 389-395 34.Reyers F, Myburgh E 2000 The effect of sample colour on the results produced by dry-and wet chemistry analysers.Proceedings of the World Small Animal Veterinary Association and FECAVA Congress, Amsterdam, Holland, 507 35.Rossignol B D R, Petit C 1984 Improvement of creatinine measurement on RA-1000.Clinical Biochemistry 17: 203-204 36.Schetters T P, Kleuskens J, Scholtes N, Gorenflot A 1998 Parasite localization and dissemination in the Babesia-infected host.Annals of Tropical Medicine and Parasitology 92: 513-519 37. Simao C, Stone R, Almeida M 2003 Kidney failure associated with Plasmodium falciparum infection.Acta Medica Portuguesa 16: 93-95 38.Tiffany T O, Jansen J M, Burtis C A, Overton J B, Scott C D 1972 Enzymatic kinetic rate and end-point analyses of substrate, by use of a GeMSAEC fast analyzer.Clinical Chemistry 18: 829-840 39.Weber J A, van Zanten A P 1991 Interferences in current methods for measurements of creatinine.Clinical Chemistry 37: 695-700 40.Welzl C, Leisewitz A L, Jacobson L S, Vaughan-Scott T, Myburgh E 2001 Systemic inflammatory response syndrome and multiple-organ damage/dysfunction in complicated canine babesiosis.Journal of the South African Veterinary Association 72: 158-162 41.Zinna S, Vathsala A, Woo K T 1999 A case series of falciparum malaria-induced acute renal failure.Annals of the Academy of Medicine, Singapore 28: 578-582