Korean Journal of
Veterinary Service

Various pathophysiological mechanisms can disrupt acid-base homeostasis in veterinary patients with severe illnesses. While several studies have examined acid-base disorders in humans with conditions such as liver, kidney, and pulmonary diseases and tumors (Bruno and Valenti, 2012; Swenson, 2016; Mazzeo and Maimone, 2018), similar research in cats is limited. Acid-base disorders are crucial for their diagnostic, therapeutic, and prognostic implications. The bicarbonate concentration in cats is especially inversely proportional to mortality (Hayes et al., 2011; Stillion and Fletcher, 2012; Kohen et al., 2018). To effectively manage these disorders, various analytical methods have been developed, including traditional and physicochemical approaches. The physicochemical approach, incorporating the Stewart (strong ion model, quantitative) and Fencl-Stewart (semi-quantitative) methods, is particularly effective for identifying underlying causes. Our previous study demonstrated that these physicochemical methods offer superior diagnostic value over traditional approaches in cats (Chun and Yu, 2021). Despite the widespread use of traditional methods in veterinary practice, Stewart and Fencl-Stewart methods have rarely been applied in veterinary clinics, especially in cats. Therefore, this study aimed to analyze metabolic acid-base disorders in cats with various underlying diseases including hepatobiliary, urinary, carcer, gastrointestinal tract, respiratory, cardiovascular, and neurologic disorders, and elucidate the pathophysiology.

Ethics statement

The study retrospectively reviewed the medical records of felines at Gyeongsang National University Animal Medical Center (GAMC) from May 2018 to August 2020, with Institutional Animal Care and Use Committee (IACUC) approval.

Population data collection

The population data were consistent with our previous study (Chun and Yu, 2021). Data collected included case history, signalment, physical exams, diagnostic imaging, and blood analysis results, which covered venous blood acid-base, electrolyte, lactate, serum biochemistry profiles, and clinical diagnosis.

Classification of underlying disorders in cats

Feline disorders were classified into seven categories: cardiovascular, hepatobiliary, gastrointestinal, neurologic, urinary tract, respiratory disorders, and cancer, with unclassified cases excluded from categorized disorder analysis. Control data were obtained from healthy client-owned cats, confirmed through history, physical exams, hematocrit, and total solid concentration based on Hopper et al. (2014). Blood analysis results from 18 healthy cats provided reference values.

Blood sample collection, acid-base analyses, and statistical analysis

Blood samples were drawn from the jugular or cephalic veins using a syringe with a 23-gauge needle or a 24-gauge intravenous catheter. Samples were transferred into appropriate microtubes with anticoagulants: Ethylenediaminetetraacetic acid (EDTA) for complete blood count, lithium heparin for serum biochemistry, or blood gas analysis. Blood gas analysis utilized lithium heparin microtubes or arterial blood collection syringes. Analyses for acid-base, lactate, electrolyte, and biochemistry values were conducted immediately using a Benchtop blood gas analyzer (pHOx Ultra Blood Gas Analyzer; Nova Biomedical) and a chemistry analyzer (IDEXX Catalyst One; IDEXX Laboratories). Complete blood counts were performed with a hematology analyzer (ProCyte Dx; IDEXX Laboratories).

Supplementary Table S1 (Siggaard-Andersen, 1977; Constable, 2000; Boniatti et al., 2009) provides the calculated formula for acid-base analysis. Three methods were employed to investigate metabolic acid-base disorders. Reference ranges from 18 healthy cats defined acidosis as pH <7.35 and alkalosis as pH >7.44, used for classification with traditional methods. Supplementary Table S2 lists the diagnostic criteria for the traditional acid-base analysis method for cats. Acid-base disorder values were defined as ±2 standard deviations from the mean comparison value. A P-value <0.05 was considered significant. Statistical analyses were performed using IBM SPSS version 25.0 and GraphPad Prism 7. After normality tests, variables such as pH, PCO2, HCO₃⁻, SID, SIG, A_TOT, free water effect, chloride, phosphorus, albumin, lactate effects, electrolyte levels, lactate, anion gap, and BE were compared between healthy and disorder groups using independent t-tests or Mann-Whitney U tests. The average of the acid-base-related parameters of each disease group was compared with that of the healthy group to identify indicators showing statistically significant differences.

Selection of cats analyzed in this study

During the study from May 2018 to August 2020, 327 venous blood-gas analyses were conducted on 75 cats. Of these, 69 met the study criteria, allowing for the comparison of three analysis methods. The cats included 29 neutered males, 4 intact males, 24 spayed females, 8 intact females, and 4 with unspecified sex, with a median age of 6 years. Predominantly, Domestic Shorthairs were represented (42%), followed by mixed breeds and several purebreds like Russian Blue and Persian. Eleven cats lacked breed information. Diagnoses were recorded for 66 cats. A comparison group of healthy cats included 8 neutered males, 2 intact males, and 8 spayed females, with a mean age of 4 years. Domestic Shorthairs were again predominant (77.8%). The pH reference interval was 7.35∼7.44, and various calculated parameters for the Fencl-Stewart method were included in the analysis. Supplementary Table S3 provides the signalments of the cats with diseases and healthy cats.

Classification of disease category in 69 cats analyzed in this study

The 69 cats with various underlying diseases were classified into the following disease categories: hepatobiliary system (n=6/69 cats), urinary system (n=18/69 cats), gastrointestinal system (n=11/69 cats), cancer (n=20/69 cats), respiratory system (n=7/69 cats), cardiovascular system (n=11/69 cats), and neurologic disease (n=5/69 cats). Some cats could be diagnosed with more than one different disease because of multi-organ disorders.

Comparison of acid-base status between diseased and healthy groups

A parameter showing significant differences in the mean comparison of each disease category and healthy group is described in the following paragraph. A comparison graph between each disease group and the healthy group using the Stewart (SID_a, A_TOT, SIG) and Fencl-Stewart (free water, Cl⁻, phosphate, albumin, lactate, UA effects) methods is shown in Fig. 1∼7. The hepatobiliary disease category showed statistically significant differences in mean pH (P<0.001), BE_ECF (P<0.001), HCO₃⁻ (P=0.001), Cl⁻ (P=0.040), SID_a (P=0.002), SIG (P=0.040), Cl⁻ effect (P=0.018), albumin effect (P=0.040), and UA effect (P=0.009) (Fig. 1). The urinary system disease category showed a difference in mean pH (P=0.001), BE_ECF (P=0.002), HCO₃⁻ (P=0.012), and SID_a (P=0.003) (Fig. 2). The cancer category showed a difference in mean pH (P<0.001), BE (P=0.017), and SID_a (P=0.034) (Fig. 3). The gastrointestinal system disease category showed differences in mean pH (P=0.007), BE (P=0.001), HCO₃⁻ (P=0.001), K⁺ (P=0.050), Ca⁺⁺ (P=0.010), A_TOT (P=0.048), and albumin effect (P=0.015) (Fig. 4). The respiratory disease category showed a difference in pH only (Fig. 5). The cardiovascular system and neurologic disease categories did not show differences in the means of any values (Fig. 6, 7).

Figure 1. Comparison graph of acid-base Stewart (SID_a, A_TOT, SIG) and Fencl-Stewart (Free water, Cl⁻, phosphate, albumin, lactate, UA effects) method between hepatobiliary system disorder cats and healthy cats.
A_TOT, total quantity of weak acids; SID, strong ion difference; SIG, strong ion gap; UA effect, unmeasured anion effect.

Figure 2. Comparison graph of acid-base Stewart (SID_a, A_TOT, SIG) and Fencl-Stewart (Free water, Cl⁻, phosphate, albumin, lactate, UA effects) method between urinary system disorder cats and healthy cats.
A_TOT, total quantity of weak acids; SID, strong ion difference; SIG, strong ion gap; UA effect, unmeasured anion effect.

Figure 3. Comparison graph of acid-base Stewart (SID_a, A_TOT, SIG) and Fencl-Stewart (Free water, Cl⁻, phosphate, albumin, lactate, UA effects) method between cancer cats and healthy cats.
A_TOT, total quantity of weak acids; SID, strong ion difference; SIG, strong ion gap; UA effect, unmeasured anion effect.

Figure 4. Comparison graph of Stewart (SID_a, A_TOT, SIG) and Fencl-Stewart (Free water, Cl⁻, phosphate, albumin, lactate, UA effects) method between gastrointestinal tract disorder cats and healthy cats.
A_TOT, total quantity of weak acids; SID, strong ion difference; SIG, strong ion gap; UA effect, unmeasured anion effect.

Figure 5. Comparison graph of acid-base Stewart (SID_a, A_TOT, SIG) and Fencl-Stewart (Free water, Cl⁻, phosphate, albumin, lactate, UA effects) method between respiratory system disorder cats and healthy cats.
A_TOT, total quantity of weak acids; SID, strong ion difference; SIG, strong ion gap; UA effect, unmeasured anion effect.

Figure 6. Comparison graph of acid-base Stewart (SID_a, A_TOT, SIG) and Fencl-Stewart (Free water, Cl⁻, phosphate, albumin, lactate, UA effects) method between cardiovascular system disorder cats and healthy cats.
A_TOT, total quantity of weak acids; SID, strong ion difference; SIG, strong ion gap; UA effect, unmeasured anion effect.

Figure 7. Comparison graph of Stewart (SID_a, A_TOT, SIG) and Fencl-Stewart (Free water, Cl⁻, phosphate, albumin, lactate, UA effects) method neurologic disorder cats and healthy cats.
A_TOT, total quantity of weak acids; SID, strong ion difference; SIG, strong ion gap; UA effect, unmeasured anion effect.

To determine if there were differences in terms of an abnormal pH according to the disease categories, three or more cases of acidosis or alkalosis were identified in each disease group. In hepatobiliary diseases, urinary system disease, gastrointestinal disease, cancer, respiratory disease, cardiovascular disease, and neurologic system disease, acidosis was identified in 6, 14, 9, 17, 6, 6, and 3 cases, respectively (Table 1). Alkalosis was identified in urinary system disease and cardiovascular system disease. In the following variables of acidosis and alkalosis cats in each category, the null hypothesis in the Mann-Whitney method was rejected or significant mean differences in the t-test were shown (Table 2). When analyzed using the Stewart methods, a significant change in the SID_a was observed only in urinary diseases. The albumin effect was identified as statistically significant when analyzed using the Fencl-Stewart method in cases of hepatobiliary and gastrointestinal diseases. The lactate effect significantly impacted acid-base changes in all disease groups except hepatobiliary disease. In respiratory diseases, while pH changes were observed using traditional methods, no statistical significance was found in other parameters. However, the lactate effect in the Fencl-Stewart method showed a significant difference between healthy and diseased cats, with similar results observed in cats with neurological diseases.

Table 1 . Number of abnormal pH in each disease category*.

	Hepatobiliary	Urinary	Gastrointestinal	Cancer	Respiratory	Cardiovascular	Neurologic
Acidosis	6	14	9	17	6	6	3
Alkalosis	0	3	1	1	1	3	0
pH in the reference interval	0	1	1	2	0	2	2
Total cats	6	18	11	20	7	11	5

*Cases can have more than one disorder..

Table 2 . Statistically significant differences between healthy cats and acidosis cats in each group.

	Hepatobiliary disease	Urinary disease	GI disease	Cancer	Respiratory disease	CV disease	Neurologic disease
Traditional
pH	○	○	○	○	○	○	○
AG	×	○	×	×	×	×	×
BE	○	○	○	○	×	×	○
HCO3−	○	○	○	○	×	×	×
Stewart
SIDa	○	○	×	×	×	×	×
ATOT	×	×	×	○	×	×	×
SIG	○	×	×	×	×	×	×
Fencl-Stewart
Albumin effect	○	×	○	×	×	×	×
Lactate effect	×	○	○	○	○	○	○
UA effect	○	×	×	×	×	○	×
Cl−effect	○	×	×	×	×	○	×

GI, gastrointestinal; CV, cardiovascular; AG, anion gap; BE, base excess; SID_a, apparent strong ion difference; A_TOT, total quantity of weak acids; SIG, strong ion gap; UA effect, unmeasured anion effect..

Acid-base homeostasis is maintained and regulated by three major organs: kidney, liver, and lung. The lung excretes CO₂ through breathing, increasing pH inside the body. Most of the CO₂, volatile acid, produced during the metabolic process is eliminated through the lungs. Many studies have reported acid-base disorders arising from certain diseases of the lungs, such as chronic obstructive pulmonary disease and acute lung damage (Bruno and Valenti, 2012). The liver plays an important role in the production of non-volatile acids that must be excreted daily. Ammonium produced during protein metabolism is converted into urea, and during this process, a hydrogen ion is produced, titrating bicarbonate. Many studies have reported acid-base disorders associated with liver failure, liver cirrhosis, and acute liver damage (Funk et al., 2007; Scheiner et al., 2017; Mazzeo and Maimone, 2018). The kidney excretes ammonium through urine and maintains acid-base homeostasis by engaging in the absorption and reproduction of hydrogen ions and bicarbonate during urinary production (Hamm et al., 2015). Many studies reported acid-base disorders related to kidney diseases, such as chronic kidney disease, acute kidney injury, and urinary tract infection (Elliott et al., 2003; Dhondup and Qian, 2017; Raphael, 2018).

In addition, most serious diseases cause acid-base disorders with various mechanisms. Respiratory acid-base disorders are caused by changes in PCO₂, and not only the respiratory system (lung and upper respiratory tract) but also pain and anxiety can cause respiratory acid-base disorders (Vollmer et al., 2015). Similarly, metabolic acidosis/alkalosis can be caused by problems in various metabolic processes. Acid-base disorders can result from diseases in organs other than the three major organs mentioned above. As mentioned above, many studies have been conducted to identify the causes of acid-base disorders arising from certain diseases and their treatments. However, studies determining the utility of acid-base analysis methods for certain diseases remain insufficient.

In this study, it was compared the acid-base parameters of each disease group with the healthy group to identify indicators showing statistically significant differences between the two groups. Significant mean differences were found in more than one variable for each of the three methods in hepatobiliary and gastrointestinal system disease groups. In addition, no significant differences were identified in the three different independent variables in respiratory, cardiovascular, and neurological disease groups.

According to the analysis of acid-base disorders in 75 cats, the most severe acidosis was found in cancer cats with post-operative low perfusion status, and the most severe alkalosis was identified in a case with iatrogenic alkalosis induced by bicarbonate administration. Those are well-known and common causes of severe acid-base disorders. Metabolic acidosis caused by cancer or inadequate perfusion was reported both in human and veterinary medicine. Previous studies reported the acidosis effect of cancer that affects its exacerbation, metastasis, resistance against chemotherapy, and tumor behavior and the acidosis mechanism often being set as the treatment goal (Pillai et al., 2019).

When cats with acidosis (pH <7.39) in each disorder category were analyzed, more than one significant mean differences for each of the three methods were found in three disease groups: hepatobiliary disease, urinary system disease, and cancer group. This suggests that all three methods are appropriate for interpreting the acid-base status of those organs.

When analyzed using the Stewart methods, a significant change in the SID_a was observed only in urinary diseases. This is thought to result from the nature of these diseases, which readily cause imbalances in several electrolytes, including sodium and chloride. The albumin effect was identified as statistically significant when analyzed using the Fencl-Stewart method in cases of hepatobiliary and gastrointestinal diseases. This is thought to be due to decreased albumin synthesis in liver diseases or changes in albumin concentration as an acute phase protein in systemic inflammatory diseases. In GI diseases, it can be attributed to the common occurrence of protein-losing enteropathy, which generally leads to the loss of plasma proteins, including albumin. Furthermore, the lactate effect was found to have a significant impact on acid-base changes across all disease groups except for hepatobiliary disease. In particular, in respiratory diseases, despite changes in pH observed with the traditional method, no statistical significance was found in any other parameters. However, the lactate effect in the Fencl-Steart method showed a significant difference between healthy and diseased cats. A similar result was observed in the group of cats with neurological diseases. This suggests that measuring changes in lactate concentration could be utilized to diagnose and predict acid-base disorders in most diseases.

This study is retrospective. The location of blood collection and the time between the blood collection and its analysis are not clear. A previous study emphasized the importance of proper sample handling for accurate blood gas analysis and the differences caused by sample collection sites (arterial and venous blood) and methods were studied. Numerous studies argued that the dilution by heparin could affect the pH, PCO₂, and HCO₃⁻ (Haskins, 1977; Hutchison et al., 1983). When blood is exposed to the air, pH may have alkalotic changes due to the reduction of PCO₂ (LeGrys et al., 2015). This limitation can be eliminated by using the prospective study that controls the variables. In addition, this study has a limitation of a small sample size of 69 subjects. Analysis of large samples can yield more accurate RI and allow the analysis of various cats by their specific diseases, thus improving the reliability of research results. In this study, the author only analyzed the difference in mean values between the variables of the healthy group and the disease group, however, using various statistical techniques such as median and regression analysis would analyze the differences in utility of the disease-specific analysis method.

In conclusion, the association between different disorder categories and acid-base analysis result in this study suggests that appropriate analysis methods for different disorder categories can exist. Based on the results of this study, further prospective studies are needed to determine the diagnostic and prognostic value of acid-base analysis in veterinary patients and to find a golden standard analysis method for cats with different disorders.

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2020R1C1C1008675).

The online version contains supplementary material available at https://doi.org/10.7853/kjvs.2024.47.4.261.

No potential conflict of interest relevant to this article was reported.