Cardiac biomarker concentrations in cats with elevated pancreatic lipase concentrations

Abstract

Background

Cardiac biomarker concentrations are elevated in dogs with pancreatitis, but it is unknown if this is also the case for cats.

Methods

The serum feline pancreatic lipase immunoreactivity (fPLI) of serum samples from 60 cats was quantified using the Spec fPL assay. Serum N‐terminal pro‐B‐type natriuretic peptide (NT‐proBNP) and high‐sensitivity cardiac troponin I (HST) concentrations were also measured using commercial assays.

Results

Five of 20 cats with fPLI concentrations of 8.8 mg/L or more had above reference interval NT‐proBNP concentrations. Serum NT‐proBNP concentrations were not correlated with fPLI concentrations (p = 0.31) and were not different in each diagnostic category for fPLI results (p = 0.24). Serum HST concentrations were positively correlated with fPLI concentrations (r = 0.278, 95% confidence interval: 0.02‒0.50, p = 0.03). However, they were not significantly different in each diagnostic category for fPLI results (p = 0.16).

Limitations

Unidentified co‐variates could contribute to the association between fPLI and cardiac biomarker concentrations.

Conclusions

Cats with elevated fPLI concentrations may have elevated cardiac biomarker concentrations.

INTRODUCTION

Pancreatitis is a common cause of gastrointestinal signs in cats. ¹ Pancreatitis can result in systemic complications such as systemic inflammatory response syndrome, disseminated intravascular coagulation, thromboemboli and multiorgan dysfunction. ² , ³ These complications are suspected to influence morbidity and mortality. Cardiovascular complications of pancreatitis are described in both humans and dogs, but less information exists for cats. ⁴ , ⁵ One study reported a positive correlation between lipase activity and cardiac troponin I concentrations in 19 cats. ⁶ However, the reason for concurrent measurement of these analytes was not reported in the abstract, and this may have led to a selection bias.

Cardiac biomarkers are also being used with increased frequency in the diagnostic work‐up of cats with suspected cardiomyopathies. To effectively utilise a cardiac biomarker, it is important to recognise potential populations that could have false positive test results. Dogs with pancreatitis can have elevated N‐terminal pro‐B‐type natriuretic peptide (NT‐proBNP) and troponin I concentrations without overt cardiovascular compromise. ⁵ It is unknown whether this is also true for cats. Thus, the aims of this study were to evaluate cardiac biomarker concentrations in cats with various feline pancreatic lipase immunoreactivity (fPLI) concentrations and to evaluate potential correlations between serum fPLI and cardiac biomarker concentrations.

MATERIALS AND METHODS

Study overview

The Institutional Animal Care and Use Committee at Michigan State University approved an exemption for the study due to the use of residual samples. These samples were originally submitted to the Texas A&M Gastrointestinal Laboratory by the animal's primary veterinarian for the measurement of serum fPLI concentration and were handled and stored according to each practice's standard operating procedures. Residual serum samples were selected on a convenience basis from cats with fPLI concentrations across the dynamic range of the Spec fPLI assay. If more than one sample was available per cat, only the initial sample was used. These samples were then submitted for quantification of NT‐proBNP and high‐sensitivity cardiac troponin I (HST).

Biomarkers

Pancreatic lipase immunoreactivity was measured using the Spec fPL assay (Texas A&M Gastrointestinal Laboratory). ⁷ , ⁸ It has a sensitivity and specificity of 79% and 82% (at a cut‐off of ≥5.4 µg/L), respectively, for a diagnosis of pancreatitis. ⁹ Our study used the following diagnostic cut‐offs: concentrations of 4.4 µg/L or less were considered within the reference interval (RI), ⁷ concentrations of 4.5–8.7 µg/L were considered equivocal for the diagnosis of pancreatitis and concentrations of 8.8 µg/L or more were considered consistent with a diagnosis of pancreatitis. ⁷ These cut‐off values were recently proposed and have been used in other studies. ⁷ , ¹⁰

Serum NT‐proBNP concentrations were measured using an analytical validated assay (Cardiopet proBNP test kit, IDEXX Laboratories). ¹¹ NT‐proBNP is a biomarker of myocardial stretch. ¹² Elevated NT‐proBNP concentrations are useful in detecting occult heart disease, and concentrations may differ in cats with severe versus mild to moderate cardiac disease. ¹³ , ¹⁴ , ¹⁵

Serum HST concentrations were measured using a commercial assay (high‐sensitivity troponin I assay, Siemens Healthineers HST). At this time, a feline‐specific RI has not been reported. Cardiac troponin I is a biomarker of myocardial ischemia.

Statistical analysis

The data were assessed for normality using Shapiro–Wilk testing. The data were reported as medians and interquartile ranges and non‐parametric statistical testing was performed due to data distribution. Kruskal–Wallis testing was used to compare HST and NT‐proBNP concentrations among cats with various serum fPLI concentrations. Spearman's rank correlation coefficients were calculated to evaluate potential associations between fPLI and cardiac biomarker concentrations. Statistical analyses were conducted using commercially available software (R Statistical Software v4.3.0, R Foundation for Statistical Computing), and a p‐value of less than 0.05 was considered significant for all comparisons. Descriptive statistical analyses were also performed.

RESULTS

Animals

Serum was available from 20 cats in each of the diagnostic bins for the fPLI assay. The median age of sampled cats was 12.9 years (interquartile range [IQR]: 5.2 years). Thirty‐three cats were male neutered, and 27 cats were female spayed. Most cats were domestic shorthair cats (n = 42) and five were domestic longhair cats. Less than five each of other breeds were represented. Serum fPLI concentrations were non‐normally distributed (p < 0.001). The median fPLI concentration was 6.7 µg/L (IQR: 10.5 µg/L).

Serum NT‐proBNP concentrations

Serum NT‐proBNP concentrations were non‐normally distributed (p < 0.001). The median NT‐proBNP concentration was 35.5 pmol/L (IQR: 74.0 pmol/L). Five of the 20 cats with fPLI concentrations of 8.8 µg/L or more had elevated (>100 pmol/L) NT‐proBNP concentrations. Ten of 20 cats with fPLI concentrations of 8.8 µg/L or more had NT‐proBNP concentrations greater than those reported in a prior population of healthy cats (>24 pmol/L). ¹⁶

Serum NT‐proBNP concentrations were not correlated with fPLI concentrations (p = 0.31) (Figure 1). Serum NT‐proBNP concentrations were not significantly different between cats with fPLI concentrations in each of the diagnostic bins for the fPLI assay (r = 0.135, 95% confidence interval [CI]: ‒0.13 to 0.38, p = 0.24) (Figure 2). The median NT‐proBNP concentration in cats with a within RI fPLI concentration was 25.0 pmol/L (IQR: 26 pmol/L), the median NT‐proBNP concentration in cats with an equivocal fPLI concentration was 39.0 pmol/L (IQR: 80 pmol/L) and the median NT‐proBNP concentration in cats with an fPLI concentration above 8.8 µg/L was 32.5 pmol/L (IQR: 105 pmol/L).

FIGURE 1.

Correlation between feline pancreatic lipase immunoreactivity (fPLI) concentrations and serum N‐terminal pro‐B‐type natriuretic peptide (NT‐proBNP) concentrations (a) and high‐sensitivity cardiac troponin I (HST) concentrations (b) in 60 cats. The red line represents the fitted correlation, while the grey shading indicates the 95% confidence interval for predictions made by the model. Individual data points are indicated by dots

FIGURE 2.

Serum N‐terminal pro‐B‐type natriuretic peptide (NT‐proBNP) concentrations (a) and high‐sensitivity cardiac troponin I (HST) concentrations (b) in cats with normal, equivocal and elevated feline pancreatic lipase immunoreactivity (fPLI) concentrations. NT‐proBNP concentrations of 800 pmol/L or more and HST concentrations of 800 pg/mL or more were excluded to maintain graphical clarity. Box and whisker plots were generated using Tukey's method. No significant differences between groups were noted

Serum HST concentrations

Serum HST concentrations were non‐normally distributed (p < 0.001). The median HST concentration was 59.9 pg/mL (IQR: 106.9 pg/mL). Serum HST concentrations were weakly positively correlated with fPLI concentrations (r = 0.278, 95% CI: 0.02‒0.50, p = 0.03) (Figure 1). Serum HST concentrations were not significantly different between cats with fPLI concentrations in each of the diagnostic bins for the fPLI assay (p = 0.16) (Figure 2). The median HST concentration in cats with a within RI fPLI was 50.0 pg/mL (IQR: 53.0 pg/mL), the median HST concentration in cats with an equivocal fPLI was 56.7 pg/mL (IQR: 278.2 pg/mL) and the median NT‐proBNP concentration in cats with an fPLI of 8.8 µg/L or more was 84.0 pmol/L (IQR: 149.8 pmol/L).

DISCUSSION

In this study, one in four cats with an elevated fPLI concentration (≥8.8 µg/L) had an above RI NT‐proBNP concentration, and 50% of these cats had NT‐proBNP concentrations greater than those reported in a study of healthy cats. ¹⁶ While it is not definitively known why each veterinarian had submitted a serum sample for fPLI quantification, it is assumed that these cats were clinically suspected to have pancreatitis or gastrointestinal signs. This finding may have clinical significance, as it suggests the potential for elevated NT‐proBNP concentrations in cats with a non‐cardiac primary disorder. This finding is similar to dogs. ⁵ Potential explanations for this include secondary cardiac injury or potentially the effects of fluid resuscitation. ¹⁷ Alternatively, other co‐variates, such as hypertension or unknown comorbid disease, may be responsible. ¹⁸ , ¹⁹ , ²⁰ , ²¹ Unfortunately, case‐specific data were unavailable in this short communication given the study design (use of residual serum), but additional large‐scale investigations originating from the results of this study should consider additional cardiac screening (echocardiogram and Holter monitoring), measurement of blood pressure and evaluation of thyroid status. This will help to further elucidate any potential relationship between pancreatitis and elevated cardiac biomarker concentrations in cats.

We were able to identify a weak positive correlation between fPLI and HST concentrations. This means that cats with higher fPLI concentration had greater cardiac troponin concentrations, which are considered the gold standard for detection of myocardial injury. ²² This mirrors the results of a prior retrospective study of 19 cats, which used a 1,2‐o‐dilauryl‐rac‐glycero‐3‐glutatic acid‐(6'‐methylresorufin) ester (DGGR) lipase assay. ⁶ , ²³ Dogs with pancreatitis have also been shown to develop elevated cardiac troponin I concentrations. ⁵ The median HST concentration was numerically greater in cats with fPLI concentrations suggestive for pancreatitis (8.8 µg/L) than in cats with equivocal (4.5–8.7 µg/L) or normal (≤4.4 µg/L) fPLI concentrations. However, this difference was not statistically significant. Elevated troponin concentrations have been noted in several inflammatory conditions. The data reported in this short communication could be used to generate a power calculation for subsequent large‐scale investigations. Myocardial injury in cats with elevated fPLI concentrations could be due to hypotension, hypoxemia or microthrombi. ²⁴ , ²⁵ , ²⁶

This study had limitations. Given the study design, limited clinical data were available for each cat. This may mean that unknown co‐variates could have influenced the study, and we cannot definitively exclude concurrent cardiac disease in individual cats. We believe that the sample size will help to mitigate the effects of any single cat having unknown cardiac disease on the results of the study. A diagnosis of pancreatitis was based on an assumed clinical suspicion for pancreatitis (primary veterinarian requested fPLI quantification) and the presence of an elevated serum fPLI concentration. Imaging data, which are commonly used as part of a clinical diagnosis of pancreatitis, were unavailable. ¹ This may limit the diagnostic accuracy for pancreatitis; however, the fPLI assay used is highly sensitive and specific. ⁸ , ⁹ , ¹⁰ , ²³ We also used the recently proposed higher fPLI cut‐off values, which should further increase diagnostic specificity. ⁷

To conclude, cats with elevated fPLI concentrations often had NT‐proBNP concentrations greater than previously reported in healthy cats, and we observed a correlation between HST and fPLI concentrations in our cohort of cats.

AUTHOR CONTRIBUTIONS

Acquisition of data, analysis and interpretation of data, drafting the article and revising article for intellectual content: Harry Cridge. Conception and design, analysis and interpretation of data and revising article for intellectual content: Jorg Steiner. Acquisition of data, analysis and interpretation of data and revising article for intellectual content: Barry DeCicco and Stephanie Monterrosso.

CONFLICT OF INTEREST STATEMENT

Jorg Steiner serves as director of the Gastrointestinal Laboratory at Texas A&M University, which offers measurement of serum fPLI and cardiac troponin I quantification on a fee‐for‐service basis. He also acts as a paid consultant and speaker for IDEXX Laboratories, the manufacturer of the Spec fPL assay, which also offers fPLI testing on a fee‐for‐service basis.

ETHICS STATEMENT

The Institutional Animal Care and Use Committee at Michigan State University approved an exemption for this study as only residual serum samples were used.

Cridge H, DeCicco B, Monterroso S, Steiner JM. Cardiac biomarker concentrations in cats with elevated pancreatic lipase concentrations. Vet Rec. 2024;e4599. 10.1002/vetr.4599

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.