Cardiac extracellular volume fraction in cats with preclinical hypertrophic cardiomyopathy

Abstract Background Cardiac magnetic resonance imaging (CMR) allows for detection of fibrosis in hypertrophic cardiomyopathy (HCM) by quantification of the extracellular volume fraction (ECV). Hypothesis/Objectives To quantify native T1 mapping and ECV in cats. We hypothesize that native T1 mapping and ECV will be significantly increased in HCM cats compared with healthy cats. Animals Seventeen healthy and 12 preclinical HCM, age‐matched, client‐owned cats. Methods Prospective observational study. Tests performed included indirect blood pressure, CBC, biochemical analysis including total thyroid, urinalysis, transthoracic echocardiogram, and CMR. Cats were considered healthy if all tests were within normal limits and a diagnosis of HCM was determined by the presence of left ventricular concentric hypertrophy ≥6 mm on echocardiography. Results There were statistically significant differences in LV mass (healthy = 5.87 g, HCM = 10.3 g, P < .0001), native T1 mapping (healthy = 1122 ms, HCM = 1209 ms, P = .004), and ECV (healthy = 26.0%, HCM = 32.6%, P < .0001). Variables of diastolic function including deceleration time of early diastolic transmitral flow (DTE), ratio between peak velocity of early diastolic transmitral flow and peak velocity of late diastolic transmitral flow (E : A), and peak velocity of late diastolic transmitral flow (A wave) were significantly correlated with ECV (DTE; r = 0.73 P = .007, E : A; r = −0.75 P = .004, A wave; r = 0.76 P = .004). Conclusions and Clinical Importance Quantitative assessment of cardiac ECV is feasible and can provide additional information not available using echocardiography.


| INTRODUCTION
Hypertrophic cardiomyopathy (HCM) is the most common heart disease in cats. 1,2 It is characterized by variable patterns and distributions of left ventricular (LV) hypertrophy. 3 Hallmark histopathologic lesions include myofiber disarray, small coronary arteriosclerosis, and interstitial and replacement fibrosis. 4 In humans with HCM, myocardial interstitial changes can be focal or global and the final common end point of irreversible fibrosis has been linked to increased risk of cardiac complications. [5][6][7] Cardiac magnetic resonance imaging (CMR) can be used to noninvasively assess myocardial fibrosis using late gadolinium enhancement (LGE); however, quantitative evaluation is limited. [8][9][10][11] Although LGE has been performed in normal cats and Maine Coon cats with mild to severe HCM, only 1 cat displayed focal LGE and there was no difference in overall myocardial contrast enhancement between normal cats and cats with HCM. 12 This study indicates that LGE can detect myocardial fibrosis but is only useful for detection of focal fibrosis, where a discrete area of fibrosis is surrounded by normal myocardium. As such, in order to detect fibrosis using LGE, normal myocardium is necessary to establish a standard of reference. In cases where the pattern of fibrosis is diffuse and limited or no normal myocardium is available for reference, LGE cannot be utilized.
One method to overcome the limitation of LGE to detect diffuse fibrosis is quantitative myocardial mapping. T1 mapping measures the longitudinal or spin-lattice relaxation time, which is determined by how rapidly protons reequilibrate their spins after an excitation radiofrequency pulse. All tissues have inherent T1 relaxation times that are based on a composite of their cellular and interstitial components. 8 The 2 most important biological determinants of an increase in native T1 values are interstitial edema secondary to infarction with associated cellular destruction and increased interstitial space from fibrosis. 13 Native T1 values are a composite signal of myocytes and extracellular volume (ECV), whereas contrast-enhanced T1 mapping can specifically calculate the ECV fraction. Gadolinium-based contrast agents are distributed throughout the extracellular space and shorten T1 relaxation times of the myocardium proportional to the local concentration for gadolinium. 14 Areas of fibrosis and scar will therefore exhibit shorter T1 relaxation times, after contrast administration. 15 Quantification of the ECV can be determined according to An increased ECV is due to excessive collagen deposition and is highly correlated with histological measures of collagen and fibrosis. [16][17][18] The use of CMR to evaluate HCM in people has determined that native T1 values are prolonged and ECV increased in HCM. 19,20 T1 mapping and ECV assessment have the advantage of assessing fibrosis noninvasively and can detect diffuse fibrosis more accurately than LGE. The objective of this study was to quantify myocardial fibrosis, namely T1 mapping and ECV, in healthy and preclinical HCM cats using CMR and their association with echocardiographic variables.

| MATERIALS AND METHODS
The study protocol was reviewed and approved by the Institutional Animal Care and Use Committee (Protocol #17281) at the University of Illinois at Urban-Champaign.

| Cats, clinical examinations, and group assignment
Twenty-nine client-owned cats were prospectively studied and recruited over a 2-year period and emphasis was placed on recruiting older normal cats. Tests performed included physical examination, indirect blood pressure by Doppler method, CBC, biochemical analysis including total thyroid, urinalysis, transthoracic echocardiogram, and CMR with contrast. Cats were considered healthy if all diagnostic tests were within normal limits and a diagnosis of preclinical HCM was determined by the presence of either focal or generalized LV concentric hypertrophy ≥6 mm on echocardiography. 21 All cats were preclinical and asymptomatic at the time of evaluation. None of the healthy cats or preclinical HCM cats were receiving medications other than topical heartworm and flea prevention. for off-line analysis. Each study was analyzed by the same observer (R.C. Fries) at the end of the recruitment period in random order. All studies were labeled by random identification number only and each measurement was repeated 5 times and the mean values were used for statistical analysis. Echocardiographic studies were performed before CMR on the same day.

| Echocardiography
Assessment of LV size and function was performed using standard right parasternal short-axis and long-axis views, and left apical parasternal long-axis views. 22 Two-dimensional variables measured included LV internal dimensions at end-diastole (LVIDd) and endsystole (LVIDs), LV free-wall thickness at end-diastole (LVFWd) and end-systole (LVFWs) and interventricular septal thickness at enddiastole (IVSd) and end-systole (IVSs). The LV fractional shortening was calculated using the following formula: Assessment of left atrial (LA) size was performed from standard right parasternal long-axis and short-axis views. Variables measured included LA diameter (LA SAX ) and aortic diameter (Ao) measured from a right parasternal short-axis view in early diastole timed to the earliest frame in which the closed aortic value cusps could be visualized. The ratio between LA SAX to Ao (LA : Ao) was calculated. Additionally, the LA septal-to-free wall dimension maximum (LAD Max ) and minimum (LAD Min ) were measured from the right parasternal long-axis 4-chamber view. The LAD Max and LAD Min were measured midchamber approximately parallel to the mitral annuls at end LV systole immediately before mitral valve opening (LAD Max ) and end LV diastole immediately after mitral valve closure (LAD Min ). The LA fractional shortening was calculated using the following formula: LA

| Cardiac magnetic resonance imaging
The morning of the CMR procedure, all cats had 1 to 2 mL of blood drawn for determination of hematocrit, before being placed under general anesthesia. Premedication protocols were determined by a board-certified anesthesiologist (S. Kadotani) based on clinical assessment and cardiovascular status. Anesthesia was induced with alfaxalone 2 mg/kg IV to effect and maintained with isoflurane in 100% oxygen following intubation. Cats were monitored throughout the CMR using ECG, pulse oximetry, end-tidal carbon dioxide, direct arterial blood pressure monitoring, and assessment of anesthetic depth, heart rate, respiratory rate, and temperature with values recorded every 5 minutes. Lactated Ringer's solution was administered at 5 mL/kg/hour during the anesthetic period in healthy cats and 2.5 mL/kg/hour in HCM affected cats.
All cats underwent CMR studies performed on a MAGNETOM Skyra 3T scanner with a 4-channel phased array flex coil (software version syngo MR E11 Siemens Healthcare, Erlangen, Germany).
Scout images were used to identify the long-and short-axis views of the left ventricle as well as the 2-and 4-chamber views of the heart. Thereafter, cine-images of 3 long-axis views (4-chamber, 2-chamber, and 3-chamber view) were acquired using a balanced steady-state precession sequence in combination with parallel imaging and retrospective gating during an expiratory breath-hold. Frequency scout scans were performed before each cine image and optimal frequency was subjectively assessed and determined by the operator (RCF). Frequencies (−200 Hz to +200 Hz) were optimally adjusted before each cine loop based on the frequency scout. 23  Left ventricular segments with LGE were qualitatively evaluated and enhancement was defined as 6 standard deviations above the manually selected normal area (maximally suppressed myocardium on TI scout). Native and postcontrast MOLLI images were processed using commercially available software (cvi 42 , version 5.10, Circle Cardiovascular Imaging Inc, Calgary, Alberta, Canada). Four, manually drawn, regions of interest were used to create individual regional T1 values and ECV in both the basal and mid-ventricular slice. The septum was divided into 2 regions, as was the free wall including the papillary muscles ( Figure 1). Global T1 and ECV values were calculated as the average all regions in both slices.

| RESULTS
Demographic data and results of physical examination are summarized in Table 1. The study sample was comprised of 17 healthy cats and 12 cats with HCM. There were no differences between groups related to complete blood count, biochemistry profile, total thyroid, urinalysis, or demographic data, except for presence of a heart murmur detected more frequently in HCM cats. Two HCM cats had single premature ventricular complexes, which did not require treatment, noted during screening echocardiograms. None of the HCM cats or healthy cats had ectopic heart rhythms during anesthesia for their CMR studies.
Comparison of echocardiographic variables is summarized in Table 2 No significant differences in the distribution of ECV were found.
Inter-and intraobserver measurement variability for echocardiographic and CMR variables are summarized in Table 4.  The role of CMR in veterinary medicine is limited, but provides novel information about cardiac composition that cannot be determined from echocardiography. Although none of the cats in this study were in heart failure, there is reason to believe that diffuse fibrosis plays an important role in disease progression. 27 In people with heart failure and preserved ejection fraction, higher ECV is associated with a higher rate of all-cause mortality and first heart failure hospitalization. 28 Further studies are warranted to determine if ECV can aid in the detection of HCM in cats, in particular for cats with secondary left ventricular hypertrophy and equivocal echocardiographic findings, and if ECV confers prognostic information.
Results of this study also demonstrated that ECV was correlated with multiple measures of diastolic function, LA size, and LA function  echocardiography in HCM, suggesting that diffuse fibrosis plays an important role in the pathophysiology of diastolic dysfunction. 29,32 Furthermore, in humans with heart failure with preserved systolic function, there is a correlation between the amount of collagen type 1 found on endomyocardial biopsy and echocardiographic indices of diastolic dysfunction. 33   ROC results in this study are for detecting HCM based on the echocardiographic criteria, and it remains to be seen if CMR is more sensitive than echocardiography at discriminating between affected and unaffected cats in an equivocal range. Histopathology was not performed in this study, therefore even though ECV has been correlated with histopathology in multiple other species, we cannot definitely conclude that ECV has the same relationship in cats. Finally, correlations in this study do not equate to outcomes and prospective studies are warranted to determine the outcome relationship between ECV and HCM.
In this study sample, cats with HCM had significantly higher ECV values compared with healthy controls and ECV significantly correlated with worsening diastolic function and left atrial size. Quantitative assessment of ECV is feasible in cats and can provide additional information not available using standard imaging techniques. Larger prospective studies are necessary to validate our findings and the effect of ECV on prognosis in cats with HCM requires further study.