Descriptive assessment of cardiac changes in cats with feline infectious peritonitis Authors: Petra Černá, Evan Ross, Lance Visser, Riley Ellis-Reis, McKenna Willis, Jennifer Hawley, Michael R Lappin Published: 2026 ABSTRACT AbstractBackgroundMyocarditis and myocardial injury associated with feline infectious peritonitis (FIP) recently are being recognized.Hypothesis/ObjectivesProspectively document the frequency of cardiac changes in cats with FIP using echocardiography and cardiac troponin I (cTnI) concentrations.AnimalsTwenty cats with naturally occurring FIP without previous history of heart disease.MethodsCats diagnosed with FIP were tested for common concurrent infections associated with myocarditis by testing serum for feline immunodeficiency virus (FIV) antibodies, feline leukemia virus antigen (FeLV), Dirofilaria immitis antigen, Bartonella spp. immunoglobulin G (IgG), and Toxoplasma gondii IgG and immunoglobulin M (IgM) as well as by testing blood for nucleic acids of Bartonella spp. and coronavirus (SARS-COV-2). Each cat also was tested for serum cTn1 concentration and underwent an echocardiographic examination. After 12 weeks of treatment with antiviral drugs, serum cTnI concentrations were reassessed.ResultsAll echocardiographic measurements were normalized to patient body weight. The left ventricular posterior wall from right parasternal long axis and short axis was thickened in 55% (11/20) and 25% (5/20) of cats, respectively. Pleural effusion was present in 40% (8/20) of cats and pericardial effusion in 25% (5/20) of cats, but a cardiac cause was not identified for these effusions. Median cTnI at initial evaluation was 0.37 ng/ml (interquartile range [IQR], 0.20-0.83; upper reference interval, 0.20 ng/ml). Repeat cTnI was performed after 12 weeks of antiviral treatment in 6 cats that initially had increased cTnI (median initial cTnI, 0.75 ng/mL) and, in all 6 cats, cTnI normalized to < 0.20 ng/mL.Conclusions and clinical importanceCats with FIP can present with increases in cTnI and ventricular wall thickening, findings suggestive of myocarditis or myocardial injury. Introduction As more cats with feline infectious peritonitis (FIP) are being treated with new antiviral drugs, new disease processes that may be associated with FIP are being recognized.1–6 A 9-month-old cat diagnosed with FIP developed left sided congestive heart failure secondary to hypertrophic cardiomyopathy (HCM) phenotype 4 days after initial diagnosis.7 In a study evaluating the efficacy of the nucleoside analog GS-441524 for the treatment of FIP, 2 of 31 cats being treated died of cardiac related causes.3 Another case report described fibrinous epicarditis caused by FIP in a 3-year-old cat with inflammatory infiltrate observed in the epicardium.8 Myocardial histopathology confirmed myocarditis and myocardial necrosis associated with the presence of S gene-mutated feline coronavirus (FCoV; M1058L biotype).9 A study comparing myocardial transcription of cytokines and remodeling enzymes in cats with HCM and cats without cardiac disease including cats with FIP reported the highest myocardial marker transcriptions in cats with multicentric lymphoma, FIP, and HCM, followed by other diseases likely associated with a systemic inflammatory state.10 A different study evaluating cytokine production in tissue samples from cats that died of FIP showed that pyrogenic cytokines interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-a) were upregulated in cardiomyocytes and although the heart did not have FIP lesions in any of these cases, upregulation of these cytokines suggested that cardiomyocytes likely played a role in the systemic inflammation caused by FCoV.11 A recent study showed suspicion of myocarditis in 4 cats that presented with different cardiac manifestations including systolic dysfunction, ventricular ectopy, and wall thickening. These 4 cats completely recovered during treatment with GS-441524 in combination with cardiovascular treatments and remained stable during one year of follow-up.12 Feline coronavirus shares similarities with another coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and myocardial injury in patients with SARS-CoV-2 also has been documented.13–15 Humans with SARS-CoV-2 associated myocarditis often have echocardiographic evidence of systolic dysfunction and other changes and a high (27%) mortality rate.13,14 At this time, no evidence exists that myocarditis is directly caused by SARS-CoV-2 in humans, but it is suspected that myocardial inflammation is secondary to a generalized inflammatory reaction induced by cytokines.13,14 Myocarditis also has been documented in 8 cats and 3 dogs with SARS-CoV-2 in the United Kingdom.16 The primary aim was to determine the frequency of cardiac changes in cats with FIP including increases in cardiac biomarkers (Troponin I) and changes in echocardiographic parameters (including increased ventricular wall thickness, atrial enlargement, and pericardial effusion). Our secondary aim was to re-evaluate patients with FIP that exhibited cardiac changes or evidence of myocardial injury after treatment with antiviral drugs for FIP. Materials and methods Study design Our prospective clinical study included 20 cats with an antemortem diagnosis of FIP randomly selected without previous history of heart disease or any cardiac changes. These cats were enrolled in another clinical trial for FIP. Caregivers of cats diagnosed with FIP were offered a cardiac evaluation and cats for which clients consented were enrolled in the study and evaluated for possible myocarditis. Each cat was systemically evaluated with a CBC, serum biochemical profile, and urinalysis. Diagnosis of FIP was made if patients had clinical signs consistent with FIP (eg, body cavity effusion, lethargy, hyporexia, pyrexia), had laboratory findings consistent with FIP (eg, hyperglobulinemia, low albumin:globulin ratio), with or without positive quantitative reverse transcriptase (qRT)-PCR on body cavity effusion or fine needle aspirates of organs (performed by Colorado State diagnostic laboratory as previously described).17 Cats that did not have positive qRT-PCR for FCoV had to have clinical signs consistent with FIP as well as response to antiviral treatment as recommended by the Advisory Board for Cat Diseases (ABCD).18 Echocardiograms were performed using the same ultrasound unit (Philips EPIQ 7C, Philips Healthcare, Andover, MA, USA) equipped with multiple cardiac transducers by a cardiology residents (ER and RE) under supervision of a board-certified cardiologist (LV). If necessary, cats were sedated using butorphanol 0.25 mg/kg IV. Choice of probe carrier frequency and techniques to optimize two-dimensional image quality were at the discretion of the operator (ER and RE). Data were captured digitally for measurements at an off-cart workstation using analysis software (TomTec Imaging Systems GmbH, Chicago, IL). The same imaging protocol was used for each echocardiographic examination by each operator. Care was taken to optimize spectral Doppler recordings and the cursor was aligned as parallel as possible to blood flow. Sweep speed (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} $\ge$\end{document}100 mm/s), baseline, scale, and gain all were manually adjusted to optimize velocity spectra. All echocardiographic measurements were performed by the individual who acquired the images. All measurements were performed from three consecutive cycles, and averaged with the highest quality images or spectra selected for measurement. Maximum left ventricular (LV) wall thickness also was recorded for both the interventricular septum (IVS) and LV posterior wall (LVPW). Imaging planes were acquired based on standard imaging planes for cats.19 Two dimensional (2D) images were obtained for all cats in the study because optimal lineup for M-mode imaging was challenging in many juvenile cats20. The IVS was measured as the distance from the blood-tissue interface to the blood-tissue interface.19–21 For the LVPW, the distance was measured from the blood–tissue interface to the tissue–pericardial interface.19–21 Interventricular septum and LVPW measurements were obtained from both the right parasternal long-axis 4-chamber view (Lx) and right parasternal short-axis view (Sx).19–21 Maximum end-diastolic measurements of the IVS, LVPW, and LV dimension (LVDd) were obtained at end diastole.19–21 Maximum left atrial wall diameter (LADmax) was measured from the Lx view in the cranial-caudal dimension at end-systole, just before mitral valve opening.19–21 The short axis left atrial to aorta diameter (LA:Ao) was obtained from the Sx view at the base timed to aortic valve closure.19–21 Maximum right ventricular outflow tract velocity (VmaxPV) and left ventricular outflow tract velocity (VmaxAV) were obtained using continuous wave Doppler from the Sx view at the base (optimized for the right ventricular [RV] outflow tract) and the left apical 5 chamber view, respectively. Color flow doppler was utilized to identify areas of turbulence, dynamic obstructions to blood flow, and valve regurgitations. Systolic anterior motion of the mitral valve or chordae tendineae was identified as anterior motion of the mitral valve and chordae toward the IVS in systole, identified on either 2D or M-Mode examination. The presence of cavitary effusions was noted and described subjectively.19 Given the wide variety in ages and sizes of cats utilized in the study with naturally occurring FIP, standard reference intervals (RI) for cats were not applicable.20 Instead, RI derived from allometric scaling of healthy cats were utilized.20 For allometric scaling, the 2D measurement (in mm) was divided by the patient’s body weight raised to the corresponding scaling exponent. The allometric scaling exponents used are as follows: maximum long axis left atrial dimension (MaxLAD) = 0.177, interventricular septal thickness in diastole, long-axis view (IVSd,Lx) = 0.158, left ventricular internal diameter in diastole, long-axis view (LVDd,Lx) = 0.202, left ventricular posterior wall thickness in diastole, long-axis view (LVPWd,Lx) = 0.278, interventricular septal thickness in diastole, long-axis view (IVSd,Sx) = 0.158, left ventricular internal diameter in systole, short-axis view (LVDd,Sx) = 0.247, and left ventricular posterior wall thickness in diastole, short-axis view (LVPWd,Sx) = 0.212.20 Each measurement was defined as being enlarged or thickened if the allometrically scaled echocardiographic measurement exceeded the upper 97.5% prediction interval of the derived proportionality constant.20 The measurement was decreased or thin if it was below the 2.5% prediction interval of the proportionality constant.20 An La:Ao >1.6 was defined as increased.19,21 All cats had blood collected at the initial visit for CBC and serum biochemistry. Surplus serum was centrifuged and frozen at −80 °C until analysis. High sensitivity cTnI concentration was measured to assess evidence of ongoing myocardial injury. Although the RI for serum cTnI concentration has not been established for healthy cats, a concentration of >0.2 ng/mL generally is considered abnormal.22 Cats with HCM usually have concentrations <0.06 ng/mL20 or <1.0 ng/mL, whereas higher concentrations generally are associated with ongoing myocardial injury rather than primary cardiomyopathy.23 Each cat was screened for most common infectious diseases previously shown to be associated with myocarditis in cats, including feline leukemia virus (FeLV) antigen, feline immunodeficiency virus (FIV) antibodies, D. immitis antigen using Idexx SNAP Feline Triple Test, Bartonella spp. (antibodies and PCR on blood), Toxoplasma gondii (IgG and IgM antibodies), and SARS-COV-2 (PCR on whole blood) performed by Colorado State Diagnostic Laboratory as previously described.24–27 For cats that had increased cTnl concentration on initial testing (>0.2 ng/mL),22 repeat cTnI concentration was performed after 12 weeks of antiviral treatment for FIP. The study followed ethical guidelines and was approved by the Colorado State University Institutional Animal Care and Use Committee (Protocol number 3965, approval date December 2, 2022). Statistical analysis Data were recorded in Excel (Excel, Microsoft, Redmond, Washington, USA). Variables were assessed for normality by visual assessment of histograms and use of the Shapiro–Wilks test. Descriptive statistics were used to report the data. Data were reported as median (range) and interquartile range (IQR). For echocardiography, descriptive statistics were performed utilizing commercially available software (MedCalc Statistical Software, Ostend, Belgium). Mean and SD are presented for normally distributed data; median and IQR are presented for non-normally distributed data. Categorical data are presented as proportions exceeding upper RI for both the average of three measurements and if a single measurement was increased. Proportions are reported with 95% confidence intervals (CI) calculated using the Wilson score method. Results Study group Twenty cats with FIP were recruited between November 2022 and September 2024 (Supplementary Table S1). Of these, the majority were domestic shorthairs (15/20; 75%) or longhairs (1/20; 5%) and only 4/20 (20%) cats were pedigred breeds: Maine Coon (1 cat), Sphynx,1 Bengal,1 and Siberian.1 There were 14/20 (70%) neutered males, 4/20 (20%) spayed females, and 2/20 (10%) intact females. The majority (13/20; 65%) of cats were <2 years of age at diagnosis; median age was 0.9 years (IQR, 0.5-4.9) and the oldest cat enrolled in the study was 9.6 years old. Median weight was 3.3 kg (IQR, 2.1-3.5) and median body condition score was 3.0/9.0 (IQR, 3.0-4.0). Some cats had received medications before the enrollment: antibiotics (amoxicillin-clavulanate or doxycycline), appetite stimulants (mirtazapine), and anti-emetics (maropitant). Clinical findings Most cats (12/20; 60%) were diagnosed with effusive FIP. Non-effusive FIP was diagnosed in 8/20 (40%) cats, 5/20 (25%) cats had ocular involvement, and 3/20 (15%) cats had neurologic involvement. Out of all cats, 13/20 (65%) had FIP confirmed by qRT-PCR, whereas 7/20 (35%) cats had presumptive FIP, with 5 (25%) cats PCR negative and in 2 (10%) cats no PCR was performed. Only 6/20 (30%) cats presented with heart murmurs that ranged between grade I/VI and IV/VI. Sixteen of 20 (80%) had blood pressure measured by Doppler according to the American College of Veterinary Internal Medicine (ACVIM) consensus statement on systemic hypertension in dogs and cats.28 The median systolic blood pressure (BP) was 97 mmHg (IQR, 82-151). According to the ACVIM consensus statement guidelines for the identification, evaluation, and management of systemic hypertension in dogs and cats, a normotensive cat is considered to have BP <140 mmHg.29 Infectious disease testing All cats were negative for FeLV antigen, FIV antibody, and D. immitis antigen. All cats were negative for SARS-CoV-2 nucleic acids based on whole blood PCR and negative for toxoplasma based on serology. None of the 20 cats were positive for Bartonella spp. based on whole blood PCR. Of the 18 cats tested for Bartonella spp antibodies, 12 (67%) were negative, 5 (28%) cats were seropositive at 1:64, and 1 (5%) cat was seropositive at 1:128. Troponin measurements The cTnl concentration was increased (>0.2 ng/mL) in 11/20 (55%; 95% CI, 34.2-74.2) cats and the median concentration at initial evaluation was 0.37 ng/mL (IQR, 0.20-0.83; upper RI, 0.20 ng/mL). Evaluation cTnI was performed after 12 weeks of antiviral treatment in 6 cats that initially had increased cTnI concentration (median initial cTnI, 0.75 ng/mL) and in all 6 cats, cTnI concentrations normalized to <0.20 ng/mL. Echocardiography results Twenty cats had echocardiograms performed. The results for each echocardiographic variable (following allometric scaling, where applicable) are provided in Table 1; echocardiographic data for each cat before allometric scaling also are provided in Supplementary Table S2. Three cats (15%) had increased LA:Ao whereas LADmax was not increased in any cat. It was more common for the LVPW to be thickened compared to the IVS. For the average of three measurements, the LVPWd,Lx was increased in 6 (30%) cats and LVPWd,Sx was increased in 4 (20%) of cats. For maximum LV wall thickness, the LVPWd,Lx was increased in 11 (55%) cats and LVPWd,Sx was increased in 5 (25%) cats. Average IVSd was not increased in any cat, and 1 cat had maximum IVSd,Sx that was increased (Figure 1). Two (10%) cats had decreased LVIDd Lx and 1 cat had decreased LVIDd,Sx. Right parasternal short-axis view at the level of the papillary muscle with an increased interventricular septal thickness and a moderate amount of pleural effusion. This figure demonstrates moderate thickening of the interventricular septum (measuring 6.2 mm) in diastole (IVSd,Sx). The right ventricular lumen (RV), left ventricular lumen (LV), and left ventricular free wall (LVFW) are labeled as well. For cats in this study, it was more common for the LVFW to be thickened (55% and 25% of cats had an increased maximal LVFW in long axis and short axis, respectively). This cat (shown here) had an IVS measuring thickened in the short axis view. Table 1Descriptive statistics of echocardiographic variables and proportions of wall thickening/chamber in cats with FIP.Echocardiographic variableMean (SD)/median (IQR)Upper reference intervalThickened/dilated (average value) with 95% CIThickened/dilated (single maximum value) with 95% CI Heart rate (bpm) 213.4 (± 30.0) bpm MaxLAD (mm/kg0.177) 10.60 (± 1.58)13.650%0% LA:Ao 1.40 (± 0.18)1.615% (5.2-36.0)15% (5.2-36.0) IVSd,Lx (mm/kg0.158) 3.33 (0.75)4.930%0% LVIDd,Lx (mm/kg0.202) 10.03 (± 1.64)13.930%0% LVFWd,Lx (mm/kg0.177) 3.43 (± 0.44)3.5130% (14.5-51.9)55% (34.2-74.2) IVSd,Sx (mm/kg0.158) 3.57 (± 0.54)5.010%5% (0.9-23.6) LVIDd,Sx (mm/kg0.247) 10.00 (± 1.54)12.820%0% LVFWd, Sx (mm/kg0.212) 3.45 (0.76)3.9720% (8.1-41.6)25% (11.2-46.9) VmaxPV (m/s) 0.98 (± 0.26) VmaxAV (m/s) 1.21 (± 0.40)Abbreviations: AV = aortic valve; IVSd,Lx = interventricular septal thickness in diastole, long-axis view; IVSd,Sx = interventricular septal thickness in diastole, short-axis view; IVSs,Lx = interventricular septal thickness in systole, long-axis view; IVSs,Sx = interventricular septal thickness in systole, short-axis view; LA:Ao = left atrium to aortic root ratio in short axis; LVFWd,Lx = left ventricular free wall thickness in diastole, long-axis view; LVFWs,Lx = left ventricular free wall thickness in systole, long-axis view; LVFWd,Sx = left ventricular free wall thickness in diastole, short-axis view; LVIDd,Lx = left ventricular internal diameter in diastole, long-axis view; LVIDs,Lx = left ventricular internal diameter in systole, long-axis view; LVIDd,Sx = left ventricular internal diameter in diastole, short-axis view; LVIDs,Sx = left ventricular internal diameter in systole, short-axis view; MaxLAD = maximum long axis left atrial dimension; PV = pulmonary valve.Data summarized as mean (SD) for normally distributed data and median (IQR) for non-normally distributed data. The upper reference interval for allometrically scaled data is based on the upper reference limit (97.5) of the proportionality constant derived from healthy control cats. The percent of cases with enlarged or thickened echocardiographic values are also presented along with 95% CI. Descriptive statistics of echocardiographic variables and proportions of wall thickening/chamber in cats with FIP. Abbreviations: AV = aortic valve; IVSd,Lx = interventricular septal thickness in diastole, long-axis view; IVSd,Sx = interventricular septal thickness in diastole, short-axis view; IVSs,Lx = interventricular septal thickness in systole, long-axis view; IVSs,Sx = interventricular septal thickness in systole, short-axis view; LA:Ao = left atrium to aortic root ratio in short axis; LVFWd,Lx = left ventricular free wall thickness in diastole, long-axis view; LVFWs,Lx = left ventricular free wall thickness in systole, long-axis view; LVFWd,Sx = left ventricular free wall thickness in diastole, short-axis view; LVIDd,Lx = left ventricular internal diameter in diastole, long-axis view; LVIDs,Lx = left ventricular internal diameter in systole, long-axis view; LVIDd,Sx = left ventricular internal diameter in diastole, short-axis view; LVIDs,Sx = left ventricular internal diameter in systole, short-axis view; MaxLAD = maximum long axis left atrial dimension; PV = pulmonary valve. Data summarized as mean (SD) for normally distributed data and median (IQR) for non-normally distributed data. The upper reference interval for allometrically scaled data is based on the upper reference limit (97.5) of the proportionality constant derived from healthy control cats. The percent of cases with enlarged or thickened echocardiographic values are also presented along with 95% CI. Pleural effusion was present in 9 cats, and was characterized as moderate volume in 4 cats (Figure 2) and mild in the remaining 5 cats. All cats with mild effusion had echocardiograms performed after thoracocentesis. Pericardial effusion was present in 5 cats, and was moderate in volume in one cat and minimal in the remainder; 2 cats had both pleural and pericardial effusion. A primary cardiac cause for the effusions was not identified in any cat. Four cats had mild mid-left ventricular obstruction, and 2 cats had systolic anterior chordal motion of the mitral valve. The LVOT velocity did not exceed 2.1 m/s in any cat. Trace tricuspid regurgitation was seen in 6 cats, trace mitral regurgitation in 2 cats, trace aortic insufficiency in 1 cat, and dynamic right ventricular outflow tract obstruction in 2 cats. Right parasternal long-axis 4-chamber view including a moderate amount of pleural effusion. The right atrium (RA), right ventricle (RV), left atrium (LA), and left ventricle (LV) are labeled. Treatment and follow up All cats were treated with antiviral medications. Fourteen of 20 (70%) cats were treated with molnupiravir (median dose, 14.3 mg/kg PO q12h; IQR, 12.2-17.4) for 12 weeks and the remaining 6/20 (30%) were treated with the nucleoside analog GS-441524 (doses not available); 6 cats were treated with GS-441525 (4 with unlicensed product before compounded GS-441525 was available in the United States and 2 with prescribed compounded product) for 12 weeks.30,31 At the end of 12 weeks of treatment, 17 cats (85%) were alive and 3 cats died or were euthanized 3-6 days after starting treatment. The cats that died had increased cTnI concentrations but based on necropsy findings, myocarditis did not appear to be the cause of death. Necropsy examination was performed in all 3 cats and they had mildly increased heart weight: 0.48%, 0.56%, and 0.82% of body weight (normal, 0.3%-0.45%). The BCS of these cats was 2 out of 9 in one cat and 3 out of 9 in two cats. Two cats had no clinically relevant histologic lesions in the heart. The third cat had patchy perivascular edema with small numbers of neutrophils and mononuclear cells and had positive immunoreactivity to FCoV on immunohistochemistry (IHC) within infiltrating mononuclear cells throughout the subpleural and adjacent pulmonary parenchyma. Two of the 3 cats that died were positive on qRT-PCR for FCoV before death and all had consistent findings on necropsy and were positive IHC for FCoV. Discussion Based on data collected in humans with SARS-CoV-2,13,14 documented myocarditis in cats with FIP,12 and newly developing treatments for FIP that have prolonged survival of these patients,1,12,32 further investigation of possible development of myocardial injury in cats with FIP is warranted. The goal of our study was to evaluate echocardiographic and cardiac biomarker changes in cats diagnosed with FIP without known cardiac disease to better understand the prevalence of cardiac changes associated with this disease. Myocarditis is characterized by the presence of inflammation in response to physical, chemical, and infectious agents. Cats with infectious myocarditis caused by systemic diseases, such as protozoa (Toxoplasma gondii, Hepatozoon species, Sarcocytsis felis), viruses (FIV), bacteria (Bartonella species) and, in some cases, opportunistic fungi have been described.24,33–36 Clinical and laboratory findings associated with Toxoplasma gondii and Bartonella spp. associated myocarditis can be confused with those of non-effusive FIP.25,37 In particular, T. gondii, Bartonella spp., and non-effusive FIP commonly share the findings of fever and hyperglobulinemia.38 Because bartonellosis and toxoplasmosis are readily treated with antibiotics, it was important to exclude these agents from the differential diagnosis before making a suspect diagnosis of coronavirus associated myocarditis. All cats were negative for retroviruses and none of the cats included in the study had toxoplasmosis or bartonellosis, although the diagnosis of Bartonella infection can be challenging. Culture of Bartonella is the gold standard for diagnosis, and a combination of pre-enrichment culture and PCR increases sensitivity.39 Sensitivity also can be increased by multiple blood cultures or PCR performed on several biological samples (blood, lymph node, oral swab, tissues),40 which was not done in our study and is a limitation. Some of the cats in our study had positive serology, but serology is more useful for exclusion than for confirmation of the disease, because of its high negative predictive value but low positive predictive value.41 Because of the high prevalence of infection in healthy cats in endemic areas, a positive culture or PCR is not confirmatory of disease in cats with compatible signs, and other compatible diagnoses must be ruled out. All cats had FIP that was confirmed by positive qRT-PCR for FCoV or had clinical signs consistent with FIP as well as response to antiviral treatment as recommended by the ABCD.18 Moreover, all of the cats that were seropositive for Bartonella in the study were positive on qRT-PCR for FCoV. None of the cats was treated with antibiotics and it is therefore possible to make a suspect diagnosis of coronavirus associated myocarditis because they responded to antiviral treatment and their cTnI concentrations normalized with antiviral treatment in all the cats that were retested. Bartonella henselae has been reported after coronavirus (COVID-19) infection in humans with lymphadenopathy after close contact with cats,42,43 and more studies are needed to see if cats after recovery from FIP might be more prone to developing bartonellosis. All of the cats in our study that survived were doing well 1.5-3.5 years after treatment and did not develop any signs of bartonellosis after antiviral treatment while being in remission from FIP. Few reported cases have shown a link between myocarditis and FIP, and FCoV shares similarities with another coronavirus, SARS-CoV-2, the etiologic agent for the COVID-19 pandemic. Myocardial injury in patients with SARS-CoV-2 has been documented13–15 and myocarditis has now been associated with SARS-CoV-2 in 8 cats and 3 dogs in the United Kingdom.16 All cats in our study were tested for SARS-CoV-2 and all were negative. Myocarditis caused by SARS-CoV-2 was therefore considered unlikely. Given the young age of most of the cats in our study with naturally occurring FIP, standard RI for cats were not applicable.20 Instead, RI derived from allometric scaling of healthy cats was utilized.20 In our study, 15% had increased LA:Ao whereas LADmax was not increased in any cat. It was more common for the LVPW to be thickened compared with the IVS. For the average of three measurements, LVPWd,Lx was increased in 30% of cats and LVPWd,Sx was increased in 20% of cats. For maximum LV wall thickness, the LVPWd,Lx was increased in 55% of cats and LVPWd,Sx was increased in 25%. Pleural effusion was present in 8 cats and was characterized as moderate in 3 cats and mild in the remaining 5 cats. All cats with mild effusion had echocardiograms performed after thoracocentesis. Pericardial effusion was present in 5 cats. It was moderate in volume in one cat and scant in the remainder, but a primary cardiac cause for the effusions could not be identified in any cat. High sensitivity cTnI was measured to assess evidence of ongoing myocardial injury. In veterinary medicine, increased cTnI serves as the most reliable non-invasive biomarker of myocardial injury, with concentrations > 0.2 ng/mL considered abnormal.22 Increased cTnI can be seen in cats with hypertrophic cardiomyopathy, but increases generally are mild in primary cardiac disease (<1.0 ng/mL).23,44 Mild increases in cTnI also can occur in other conditions including in patients with azotemia, sepsis, and systemic hypertension.44 Higher concentrations of cTnI are more suggestive of ongoing myocardial injury or myocarditis.44,45 Definitive diagnosis of myocarditis remains challenging on echocardiography alone because alterations in myocardial echogenicity are subtle and inherently subjective without histopathologic confirmation.24,25,33,36 Echocardiographic evidence of LVFW or IVS thickening may provide additional supportive evidence of myocardial inflammation.25,26 Confirmation of myocarditis ultimately requires myocardial biopsy and histopathologic evaluation, which were beyond the scope of our study. In our population, 11/20 cats had increases in cTnI. The high frequency of increased cTnI along with normalization of cTnI after treatment in 6, suggest myocardial injury plausibly was present in some cases. Only 1 out of 20 cats was azotemic at diagnosis; this cat also had increased cTnI and it is possible the renal disease contributed to the cTnI increase in this cat.46 Myocarditis in these cats could have been a consequence of viral spread throughout the body or secondary to systemic inflammation. A previous study evaluating myocardial expression of cytokines and remodeling enzymes in cats reported the highest myocardial marker transcriptions in cats with multicentric lymphoma, FIP, and HCM, and inflammatory marker transcription predominated in the myocardium of cats with systemic inflammatory disease.10 Another study evaluating cytokine production in tissue samples from cats that died of FIP showed that pyrogenic cytokines IL-6 and TNF-a were upregulated in cardiomyocytes. This study also showed that the heart was not directly affected by FIP lesions in any of these cases, suggesting that the upregulation of these cytokines was likely a consequence of cardiomyocytes playing a role in the systemic inflammation caused by FCoV.11 Evidence however exists that FIP can affect the heart, with one case report describing fibrinous epicarditis caused by FIP in a 3-year-old cat in which inflammatory cell infiltration was observed in the epicardium, composed mainly of macrophages, plasmacytes, and lymphocytes, and IHC was positive for FCoV.8 Another report described a pyogranuloma found in the myocardium of a cat with heart disease, and the histopathologic finding was myocarditis and myocardial necrosis that was associated with the presence of S gene-mutated FCoV (M1058L biotype).9 Our study had some limitations. Importantly, study size was small with only 20 cats and not all cats had the diagnosis of FIP confirmed by PCR or IHC. However, a presumptive diagnosis of FIP can be made based on signalment, clinical signs, and diagnostic investigations, as well as response to antiviral treatment as recommended by the ABCD.18 The cats did not have echocardiograms repeated at the end of the study. Blood pressure measurements were not available in 20% of the cats at the initial visit. Moreover, not all cats had repeated cTnI concentrations, but in 50% of cats where cTnI was measured after 12 weeks of antiviral treatment, cTnI normalized to <0.20 ng/mL. Additional studies measuring cTnI concentrations in cats with FIP before and after antiviral treatment are needed. Not all cats had the diagnosis of FIP confirmed by PCR or IHC. Another limitation is that the cats were only screened for the most common infectious diseases associated with myocarditis, and a few case reports have described other infectious diseases in cats such as Borrelia burgdorferi, Streptococcus canis and Sarcocystis spp. causing myocarditis.47–49 However, these diseases are rare causes of myocarditis in cats. Although serologic testing for T. gondii IgG and IgM was performed, we did not perform T. gondii PCR to detect active infection, and histopathology to visualize tachyzoites. However, all cats had cytology of fine-needle aspirates or fluid performed, and no tachyzoites were visualized. Moreover, none of the 3 cats that died and had necropsy performed had tachyzoites visualized on histopathology. Based on these findings and negative serology, toxoplasmosis therefore was considered unlikely. In regard to echocardiographic data, it was challenging to determine increased ventricular wall thickness in study patients with a wide variety of ages and sizes. The cutoff for increased LV wall thickness generally is considered to be >6 mm, but this measurement applies to normal-sized adult cats.21 For this reason, allometric scaling was utilized to determine wall thickening and chamber enlargement. However, this method has not been utilized broadly for determination of wall thickness in cats, especially in young or growing cats. Thus, false positives are possible, especially in underweight cats. This factor is a potential limitation in any study that utilizes allometric scaling. Additionally, increased LV wall thickness can be caused by disorders other than myocarditis, including primary hypertrophic cardiomyopathy (especially in the older cats in the study), systemic hypertension, hyperthyroidism, other infiltrative diseases, and pseudohypertrophy. For the cats with increased LV wall thickness, it is not definitive that myocarditis was directly related to the presence of FIP. The clinical implication of increased LV wall thickness in our study is unclear, especially because no cats had cardiovascular signs associated with FIP. Finally, it is not definitive that myocarditis would solely manifest with LV wall thickening, and it is likely that some cats with myocarditis would not have thickened LV walls on echocardiography. In conclusion, our study suggests that cats with FIP can present with increases in cTnI and increased ventricular wall thickness, both of which could be suggestive of myocarditis or myocardial injury. Antiviral treatment may decrease ongoing myocarditis in cats with FIP, and additional studies are needed to further evaluate involvement of the myocardium in cats with FIP. Supplementary Material