Abstract
Objectives
The assessment of homocysteine status in diseased cats has indicated high plasma concentrations in chronic kidney disease and yielded conflicting results with respect to cardiovascular disorders. Previous investigations in small populations of normal cats revealed greater-than-expected variability in plasma homocysteine concentration. The purpose of this study was to determine biological determinants and the reference interval (RI) of plasma homocysteine concentration in the feline species, under strict pre-analytical conditions.
Methods
In this prospective observational study, privately owned healthy adult cats underwent a complete physical examination, urinalysis and blood testing, in order to rule out any signs of disease. Plasma homocysteine concentration was measured using high-performance liquid chromatography–tandem mass spectrometry.
Results
Of 151 cats recruited, 30 cats were not included owing to abnormal physical examination or fractious behaviour, and 30 cats were excluded based on abnormalities on blood work or urinalysis. Plasma homocysteine concentrations >28 µmol/l were associated with a dietary protein content >9.3 g/100 kcal metabolisable energy. The RI for plasma homocysteine concentration was determined to be 6.2–52.3 µmol/l.
Conclusions and relevance
Normal values for plasma homocysteine concentration in cats have a wide RI, suggesting high inter-individual variability. Whether some healthy cats exhibit impaired homocysteine metabolism remains to be elucidated.
Introduction
The assessment of homocysteine status in diseased cats has indicated high plasma concentrations in chronic kidney disease and yielded conflicting results with respect to cardiovascular disorders.1–3 In humans, renal impairment is a major cause of homocysteine accumulation. 4 Hyperhomocysteinaemia has long been considered a causal independent risk factor for several cardiovascular diseases. 5 Yet, controversy arose recently regarding its significance. 6 Homocysteine is a sulfur-containing amino acid derived from the demethylation of methionine. 7 The accumulation of homocysteine may be caused by alterations in its metabolic pathways: re-methylation to methionine, which is mainly dependent on cobalamin (vitamin B12) and folate (vitamin B9), or renal trans-sulfuration to cystathionine. 8 In people, determinants of plasma homocysteine concentration (HCY) include genetic disorders, physiological and environmental factors, acquired clinical conditions and drugs. The most frequent diseases leading to significant acquired hyperhomocysteinaemia are cobalamin or folate deficiencies and renal insufficiency. Biological determinants of HCY are age, sex and lifestyle (including smoking, coffee and alcohol intake, and exercise). 4 Consideration of these various factors is mandatory for the establishment and interpretation of reference intervals (RIs) for the diagnostic screening of hyperhomocysteinaemia. Inadequate blood collection and handling procedures also may result in spurious elevations of HCY. 4 Thus, strict pre-analytical conditions are recommended for the measurement of HCY in people. Notably, immediate blood centrifugation and separation of plasma from cells, or storing blood samples on ice in case of delayed centrifugation, are advocated. 9
Species-specific determinants and the RI of HCY still need to be established in cats. Previous investigations in small populations of normal cats revealed a greater-than-expected variability in HCY.2,3,10 Cats are obligate carnivores that have a high protein requirement. 11 They exhibit several peculiarities in protein and one-carbon metabolism. 12 Re-methylation of homocysteine to methionine is of particular importance for the synthesis of S-adenosylmethionine, the main methyl donor in physiological methylation reactions. Data reported in cats with cobalamin deficiency suggested enhanced capacity of homocysteine re-methylation to methionine via the betaine–homocysteine methyltransferase pathway. 13 To be of clinical value, a prerequisite for assessing the status of homocysteine in disease states is to determine variables affecting, and normal values of, HCY in the feline species. Additionally, experimental procedures and homocysteine measurements have to be conducted with stringent monitoring of the pre-analytical settings.
The aims of the present study were to assess, in plasma from privately owned healthy adult cats, firstly the effects of epidemiological and environmental factors upon HCY, and secondly a 95% RI for HCY, under strict pre-analytical conditions.
Materials and methods
The study was approved by the institutional animal care and use committee (see ‘Ethical approval’). Informed written consent was obtained from cat owners before participation in the study.
Animals
Cats were prospectively recruited via the students and staff of the institution, between June 2016 and January 2018. Free health checks comprising a complete physical examination, blood testing and urinalysis were offered to all owners of apparently healthy cats >1 year of age. The minimum number of cats to be included in the study was set at 120, in accordance with the American Society for Veterinary Clinical Pathology (ASVCP) 2011 guidelines. 14 To assess the effect of age on HCY, cats were divided into four mutually exclusive categories based on the joint American Association of Feline Practitioners and American Animal Hospital Association feline life stage guidelines: junior (1–2 years), prime (3–6 years), mature (7–10 years), and senior and geriatric (>11 years). 15 The enrolment was conducted with the aim of including 30 cats in each group.
Cats examined for health checks were not included in the study if they exhibited fractious behaviour impeding blood collection, if there was a history or clinical signs indicative of any pathological condition, or any drug treatment during the month before sample collection (besides external and internal parasite preventives). Given that obesity predisposes to several inflammatory and metabolic disorders, cats presenting with a body condition score (BCS) of 5/5 were not included in the calculation of the 95% RI but were enrolled in the study of the biological determinants of HCY. 16 Periodontal disease was not viewed as a clinically significant finding. The auscultation of a low-intensity (grade 1–3/6) systolic heart murmur, in the absence of gallop rhythm or arrhythmias, was not considered a non-inclusion criterion owing to the inability to classify such murmurs as pathological or innocent without echocardiography. 17
After inclusion in the study, cats were excluded if their blood chemistry profile (including total protein, albumin, transaminase, alkaline phosphatase, glucose, urea, creatinine, symmetric dimethylarginine [SDMA], cobalamin, folate and free thyroxine [T4]), haematology (along with blood smear examination) or urinalysis revealed any clinically significant abnormalities. More specifically, cats with suspected kidney disorders were excluded on the basis of the following criteria: urine specific gravity <1.035, active urine sediment, urine protein: creatinine ratio >0.4, plasma creatinine >18 mg/dl or SDMA >14 µg/dl. Vitamin B12 and B9 deficiencies were defined as cobalamin and folate concentrations <400 ng/l and <3 µg/l, respectively.
Procedures
Owners were asked about the presence of any signs of disease (including gastrointestinal signs, polyuria/polydipsia, neurological disorders) and ongoing drug treatments. Information on the cats’ environment (outdoor access, tobacco smoke exposure) and dietary history were also obtained. Occasional treats (representing <10% of daily energy consumption) were not taken into account. The energy value of the diets was obtained from the manufacturer’s information on macronutrients and the dietary protein content was expressed as g/100 kcal metabolisable energy (ME). The presence of supplemental methionine was based on the manufacturer’s declaration.
All cats underwent a complete physical examination by a board-certified internist. Blood sampling was performed between 8:30 am and 12:30 pm. All cats underwent a 12 h fast before blood collection. Heparinised blood samples were centrifuged (5 mins at 3130 g) within 20 mins and plasma supernatant was harvested within 30 mins of blood collection. Complete blood counts (ProCyte DX Analyzer; IDEXX Laboratories) and basic biochemistry profiles (RX Daytona; Randox Laboratories) were performed within 2 h of blood collection. For free T4 (Immunotech; Beckman Coulter), cobalamin and folate (SimulTRAC-SNB; MP Biomedical) radioimmunoassays, 450 µl of heparinised plasma was refrigerated and the measurements were conducted within 3 days. Residual heparinised plasma was frozen at −80°C in 250 µl aliquots within 4 h of blood collection and was stored in a biological resource centre (see ‘Author note’) until SDMA and homocysteine measurements were performed. Plasma SDMA concentration was measured as described previously. 18
Urine was collected by ultrasound-guided cystocentesis. Urinalysis, performed within 30 mins of urine collection, consisted of urine specific gravity measurement (manual refractometer), a dipstick test with an automated reader (VetLab UA Analyzer; IDEXX Laboratories), sediment microscopic examination and urine protein:creatinine ratio measurement (RX Daytona; Randox Laboratories).
Plasma HCY measurement
Measurements of HCY were performed within 6 months of blood collection. A high-performance liquid chromatography–tandem mass spectrometry method (Api3000; Applied Biosystems/MDS Sciex) was used, as described for human plasma samples. 19 For biological validation of HCY measurement in feline plasma samples, a matrix effect was excluded with a recovery value test (see the supplementary material). 20
Statistical analysis
Statistical analysis was performed using open-source software R (R Core Team, V3.3.3). Continuous data were assessed graphically and with the Shapiro–Wilk’s test for normality. Non-normally distributed data are expressed as median (interquartile range [IQR]). The effects of physiological variables (breed, age, sex, body weight [BW], BCS) and environmental conditions (outdoor access, tobacco smoke exposure, dietary protein content) on HCY were examined with a Spearman correlation test for continuous variables and with a Mann–Whitney U-test (two independent groups) or a Kruskal–Wallis test (more than two independent groups) for categorical data. The impact of dietary protein content on HCY was further studied with a scatter plot and values were divided into two clusters by hierarchical cluster analysis. Significant differences were assumed for a type 1 error risk of 0.05.
The determination of the 95% RI for HCY was conducted according to the ASVCP 2011 guidelines. 14 Data were represented graphically using a histogram. The 95% RI was determined with Reference Value Advisor (V2.1), a freeware set of macroinstructions for Microsoft Excel (2016). 21 A non-parametric method was used to calculate the central 95% RI, bounded by the 2.5th and 97.5th percentiles. The 90% confidence intervals (CIs) around the upper and the lower limits of the 95% RI were determined using a bootstrap method.
Results
Study population
One hundred and fifty-one cats were examined. The process of recruitment of healthy cats is described in Figure 1. Thirty cats did not complete the study (Table 1). Of the 121 cats included, 30 cats were excluded owing to at least one significant biological abnormality (Table 1). The final study population consisted of 91 cats (Table 2).

Flow diagram describing study population enrolment
Non-inclusion and exclusion criteria applied in the recruitment of healthy cats
Description of biological characteristics of the study population
DSH = domestic shorthair; DLH = domestic longhair; IQR = interquartile range; BCS = body condition score; ME = metabolisable energy
Biological determinants of HCY in cats
All biological variables were documented in all cats, except for dietary protein content which was known in only 81 cats (Table 2). The effects of breed and tobacco smoke exposure were not included in the statistical analysis owing to the low number of purebred cats and individuals exposed to second-hand smoke. All individuals with an HCY >28 µmol/l were fed diets in which the protein content was >9.3 g/100 kcal ME (Figure 2). An HCY <28 µmol/l was associated with a wide range of dietary protein content (6.2–12 g/100 kcal ME). Two cats from the same household were mainly fed with a methionine-supplemented diet for urinary acidification that had a dietary protein content of 10.9 g/kcal ME: their HCY was 29.5 µmol/l and 36.9 µmol/l, respectively.

Relationship between dietary protein content and plasma homocysteine concentration in 81 healthy adult cats. Triangles and circles constitute two clusters established by a hierarchical cluster analysis, with a cut-off for plasma homocysteine concentration set at 28 µmol/l (dashed horizontal line). ME = metabolisable energy
RI of HCY in healthy cats
Seven obese cats were excluded from the calculation of the 95% RI. Distribution of HCY in the 84 healthy cats is represented in Figure 3. The median (IQR) HCY in the final population was 13.9 (10.5–20.5) µmol/l. The 95% RI was 6.2–52.3 µmol/l. The 90% CI for the lower and upper limits of the 95% RI were 5.3–7.0 µmol/l and 42.0–54.3 µmol/l, respectively.

Distribution of plasma homocysteine concentration in 84 healthy adult cats. The black curve is the fitted distribution of plasma homocysteine concentration. The vertical solid grey lines represent the limits of the 95% reference interval. The dashed grey lines represent the 90% confidence intervals for the lower and upper limits of the 95% reference interval
Discussion
This is the first study to establish biological determinants of, and a 95% RI for, HCY. We assessed a large population of privately owned healthy adult cats, and HCY measurements were performed under strict pre-analytical conditions. Our data are of clinical relevance because previous investigations in smaller populations revealed a greater-than-expected variability in HCY.2,3,10 A study, such as ours, was a prerequisite for assessing the status of homocysteine in disease states.
In our study population, no effects of epidemiological variables on HCY were identified. Besides environmental factors, the main physiological determinants of HCY in people are age and sex. 4 As HCY increases with age, partly as a result of declining renal function and/or B vitamin deficiencies, a specific 95% RI is recommended in the elderly population.4,22–25 Our study was designed to assess a potential effect of age on HCY. The intention was to enrol 30 animals in each age category; unfortunately, only 25 senior and geriatric cats were proposed for inclusion, of which 16 did not complete the study or were excluded after biological testing. Consequently, we were not able to assess the influence of age on HCY.
In people, HCY is higher in men than in women after puberty, but the differential is reduced in older people because HCY increases in women after the menopause.22,23,26,27 These physiological variations would be due not only to the influence of sex hormones, but also to differences in lean body mass between the two sexes. HCY was shown to be positively correlated with fat-free mass and testosterone, but negatively correlated with oestradiol. 28 In healthy adult dogs, differences in HCY were demonstrated in cycling and spayed bitches, but no differences were identified between castrated males and spayed females.29,30 In our feline population, the neutered status of all cats may partly explain the absence of sex-related variations in HCY. A similar study conducted on experimental Turkish Van cats also failed to show any significant effects of age and sex on serum HCY. 31 Breed-specific differences were described in dogs, with Chinese Shar-Peis and Greyhounds being predisposed to higher serum HCY.30,32 Unfortunately, our study lacked a sufficient number of purebred cats to assess the effect of breed on HCY. So far, our data do not advocate partitioning the 95% RI for HCY on the basis of epidemiological determinants.
We assessed the effects of environmental variables on HCY in healthy cats and revealed a possible influence of dietary protein content. In people, the effects of food consumption on HCY may also be largely a result of the protein content of the diet. Meals that are rich in proteins may cause an increase of up to 20% in HCY for at least 8 h. 33 However, dietary protein content may not affect fasting HCY. 34 In our population of healthy cats, a 12 h fast was observed before blood sampling. Yet, all individuals with HCY >28 µmol/l were receiving a diet with protein content >9.3 g/100 kcal ME. Cats are obligate carnivores that have a high protein requirement. 11 According to the European Pet Food Industry Federation, the minimum recommended protein content of feline diets is 6.25 g/100 kcal ME, based on maintenance energy requirements of 100 kcal/kg BW0.67. 35 However, a recent study stated that 5.2 g protein/kg BW would be necessary to maintain lean body mass in adult cats, representing 8.7 g/100 kcal ME assuming an average caloric intake of 60 kcal/kg BW. 36 Thus, all individuals with HCY >28 µmol/l were receiving diets with protein content considerably above the published minimum protein allowance. Yet, not all cats usually fed with a diet with high protein content showed HCY in the upper limits of the 95% RI. If a significant relationship between consumption of a high protein meal and HCY is further confirmed, one possible reason to account for this discrepancy is that the dietary protein content does not equate to the amount of protein ingested before fasting. We were not able to quantify this, because many cats were fed ad libitum and/or shared daily rations with cohabiting cats. Another hypothesis is that some cats may have had a more extended fasting period, as the instruction given to the owners was to withdraw food 12 h before the scheduled time of blood sampling. Additionally, the composition of dietary proteins may have had an impact on the metabolism of homocysteine. We noticed that two cats fed with a methionine-supplemented diet exhibited high HCY, but the diet also had a high protein content. Basic research is needed to assess both quantitatively and qualitatively the influence of dietary protein intake on HCY and its kinetics in the feline species.
In people, several lifestyle determinants of HCY have been identified, such as smoking, coffee or alcohol consumption, and exercise. 4 In our feline population, exercise was assessed on the basis of outdoor access but did not show any effect on HCY. Unfortunately, tobacco smoke exposure was not included in the statistical analysis owing to the low number of cats exposed to second-hand smoke.
Because no effects of epidemiological variables on HCY were identified, the calculation of the 95% RI was based on the entire study population, minus obese cats. In our population, the 95% RI for HCY was much wider and included values higher than those reported in humans. The reference values established in people are <15 µmol/l in adults and <20 µmol/l in seniors, in the absence of folate supplementation. 4 Other studies providing data on HCY in smaller populations of client-owned healthy cats also encountered high and also wide-ranging HCY values.2,3,10 An experimental study, conducted in healthy cats with basal HCYs comprised in our 95% RI, documented impaired homocysteine metabolism in half of the animals based on a methionine loading test. 37 Whether higher values of HCY are normal in the feline species, or whether some apparently healthy cats exhibit alterations in homocysteine metabolism, remains to be elucidated.
Despite its prospective design, this study suffered from several limitations. Owing to the number of cats that did not complete the study or that were excluded from analysis, we did not manage to reach 120 healthy adult cats for the calculation of the 95% RI. The 90% CI for the upper limit of the 95% RI was larger than recommended by the Clinical and Laboratory Standards Institute’s guidelines. 38 As a consequence, the precision of the upper limit of the 95% RI is questionable and its numerical value should be interpreted cautiously. Our study did not involve long-term follow-up of HCY in healthy cats. Repeated sampling in some reference individuals would be valuable to assess the intra-individual biological variation and calculate the index of individuality of HCY in cats, in order to estimate the relevance of a population-based 95% RI, as described previously for other biochemistry analytes in cats. 39 Some subgroups of our study population contained small numbers of cats, precluding the assessment of the effects of breed, age and tobacco smoke exposure on HCY.
Conclusions
This study did not identify any epidemiological determinants of HCY in cats. Yet, a possible influence of dietary protein content was suggested, with an HCY >28 µmol/l being associated with dietary protein content >9.3 g/100 kcal ME. The RI we established for HCY in the feline species was 6.2–52.3 µmol/l. This wide RI reflects a large variability in HCY in cats. If the upper limit of the 95% RI is confirmed by further research, then it should be used to define hyperhomocysteinaemia. These data will have a major clinical impact, underlining the necessity to increase sample sizes to achieve appropriate study power when assessing homocysteine status in disease states.
Supplemental Material
Supplemental Material
Validation of homocysteine concentration measurement in feline plasma samples with a high-performance liquid chromatography–tandem mass spectrometry method
Footnotes
Acknowledgements
We acknowledge IDEXX Laboratories for the free-of-charge measurements of SDMA, and the Center of Animal Biological Resources of Oniris for the storage of all frozen plasma samples. We thank the staff of the Department of Biochemistry and Genetics (University Hospital of Angers) and the staff of Laboniris (Oniris – Nantes Atlantic National College of Veterinary Medicine, Food Science and Engineering) for their kind collaboration. We also want to thank Guillaume Hermouet and Maggy Daunas, as well as the Medical Imaging Unit of Oniris, for their technical assistance. We are grateful to the owners of the 151 cats for their participation in this study.
Author note
The biological resource centre referred to in the ‘Procedures’ section of the study was the Centre de Ressources Biologiques Animales d’Oniris (CRBA); partner of the biobank CaniDNA, which is part of the CRB-Anim infrastructure, ANR-11-INBS-0003, funded by the French National Research Agency in the frame of the ‘Investing for the Future’ programme; Nantes, France; accessed 105 times, last accessed 30 January 2018 (BIORESOURCES).
Preliminary results of this study were presented as a poster communication at the 2018 Spring of Cardiology – Congress of Fundamental and Clinical Research, Montpellier, France, 4–6 April 2018.
Supplementary material
The following file is available online:
Validation of homocysteine concentration measurement in feline plasma samples with a high-performance liquid chromatography–tandem mass spectrometry method.
Conflict of interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Ethical approval
This work involved the use of experimental animals; or involved the use of non-experimental animals (owned or unowned) outside of established internationally recognised high standards (‘best practice’) of individual veterinary clinical patient care. The study therefore had ethical approval from an established committee as stated in the manuscript. The study was approved by the Ethics Committee for Clinical and Epidemiological Veterinary Research of Oniris (approval number CERVO-2016-1-V).
Informed consent
Informed consent (either verbal or written) was obtained from the owner or legal custodian of all animal(s) described in this work for the procedure(s) undertaken. No animals or humans are identifiable within this publication, and therefore additional informed consent for publication was not required.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
