A Rat Model of Pressure Overload Induced Moderate Remodeling and Systolic Dysfunction as Opposed to Overt Systolic Heart Failure

Summary

We describe the creation of a rat model of pressure overload induced moderate remodeling and early systolic dysfunction where signal transduction pathways involved in the initiation of the remodeling process are activated. This animal model will aid in identifying molecular targets for applying early therapeutic anti-remodeling strategies for heart failure.

Abstract

In response to an injury, such as myocardial infarction, prolonged hypertension or a cardiotoxic agent, the heart initially adapts through the activation of signal transduction pathways, to counteract, in the short-term, for the cardiac myocyte loss and or the increase in wall stress. However, prolonged activation of these pathways becomes detrimental leading to the initiation and propagation of cardiac remodeling leading to changes in left ventricular geometry and increases in left ventricular volumes; a phenotype seen in patients with systolic heart failure (HF). Here, we describe the creation of a rat model of pressure overload induced moderate remodeling and early systolic dysfunction (MOD) by ascending aortic banding (AAB) via a vascular clip with an internal area of 2 mm². The surgery is performed in 200 g Sprague-Dawley rats. The MOD HF phenotype develops at 8-12 weeks after AAB and is characterized noninvasively by means of echocardiography. Previous work suggests the activation of signal transduction pathways and altered gene expression and post-translational modification of proteins in the MOD HF phenotype that mimic those seen in human systolic HF; therefore, making the MOD HF phenotype a suitable model for translational research to identify and test potential therapeutic anti-remodeling targets in HF. The advantages of the MOD HF phenotype compared to the overt systolic HF phenotype is that it allows for the identification of molecular targets involved in the early remodeling process and the early application of therapeutic interventions. The limitation of the MOD HF phenotype is that it may not mimic the spectrum of diseases leading to systolic HF in human. Moreover, it is a challenging phenotype to create, as the AAB surgery is associated with high mortality and failure rates with only 20% of operated rats developing the desired HF phenotype.

Introduction

Heart failure (HF) is a prevalent disease and is associated with high morbidity and mortality¹. Rodent pressure-overload (PO) models of HF, produced by ascending or transverse aortic banding, are commonly used to explore molecular mechanisms leading to HF and to test potential novel therapeutic targets in HF. They also mimic changes seen in human HF secondary to prolonged systemic hypertension or severe aortic stenosis. Following PO, the left ventricular (LV) wall gradually increases in thickness, a process known as concentric LV hypertrophy (LVH), to compensate and adapt for the increase in LV wall stress. However, this is associated with the activation of a number of maladaptive signaling pathways, which lead to derangements in calcium cycling and homeostasis, metabolic and extracellular matrix remodeling and changes in gene expression as well as enhanced apoptosis and autophagy^2,3,4,5,6. These molecular changes constitute the trigger for the initiation and propagation of myocardial remodeling and transition into a decompensated HF phenotype.

Despite the use of inbred rodent strains and standardization of clip size and surgical technique, there is tremendous phenotypic variability in LV chamber structure and function in aortic banding models^7,8,9. The phenotypic variability encountered after PO in rat, Sprague-Dawley strain, is described elsewhere^10,11. Of those, two HF phenotypes are encountered with evidence of myocardial remodeling and activation of signal transduction pathways leading to a state of heightened oxidative stress. This is associated with metabolic remodeling, altered gene expression and changes in posttranslational modification of proteins, altogether playing a role in the remodeling process^10,12. The first is a phenotype of moderate remodeling and early systolic dysfunction (MOD) and the second is a phenotype of overt systolic HF (HFrEF).

The PO model of HF is advantageous over the myocardial infarction (MI) model of HF because the PO-induced circumferential and meridional wall stresses are homogeneously distributed across all segments of the myocardium. However, both models suffer from variability in the severity of PO^10,11 and in infarct size^13,14 along with intense inflammation and scarring at the infarct site¹⁵ as well as adhesion to the chest wall and surrounding tissues, which are observed in the MI model of HF. Moreover, the rat PO induced HF model is challenging to create as it is associated with high mortality and failure rates¹⁰, with only 20% of the operated rats developing the MOD HF phenotype¹⁰.

The MOD is an attractive HF phenotype and constitutes an evolution of the traditionally created HFrEF phenotype as it allows for early targeting of signal transduction pathways that play a role in myocardial remodeling, especially when it pertains to perturbations in mitochondrial dynamics and function, myocardial metabolism, calcium cycling and extracellular matrix remodeling. These pathophysiological processes are highly evident in the MOD HF phenotype¹¹. In this manuscript, we describe how to create the MOD and HFrEF phenotypes and we address pitfalls while performing the ascending aortic banding (AAB) procedure. We also elaborate on how to best characterize by echocardiography the two HF phenotypes, MOD and HFrEF, and how to differentiate them from other phenotypes that fail to develop severe PO or that develop severe PO and concentric remodeling but without significant eccentric remodeling.

Protocol

All methods and procedures described here have been approved by the Institutional Animal Care and Use Committee (IACUC) of Tulane University School of Medicine.

1. Tools and instruments for AAB model creation

1. Obtain disinfectants, such as 70% isopropyl alcohol and povidone-iodine.
2. Obtain ketamine and xylazine for anesthesia and buprenorphine for analgesia.
3. Obtain a heating pad and heavy absorbency disposable underpad with the dimensions of 18 inches x 30 inches.
4. Obtain a 100% cotton twine roll, a tape and a hair clipper.
5. Obtain a 20 cm x 25 cm plastic board, thickness range between 3-5 mm.
6. Obtain a Z-LITE fiber optic illuminator.
7. Obtain a mechanical ventilator for small animals (e.g., SAR-830/AP).
8. Obtain 2-0 and 3-0 Vicryl taper sutures and nylon 3-0 monofilament suture, sterile gauze pads and sterile extra large cotton tips and sterile gloves.
9. Obtain 16 G angiocath for intubation.
10. Purchase the following surgical tools.
    1. Obtain a Weck stainless steel Hemoclip ligation and stainless-steel ligating clips.
    2. Obtain hardened fine iris scissors.
    3. Obtain Adson forceps.
    4. Obtain two curved Graefe forceps.
    5. Obtain a Halsted-Mosquito Hemostats-straight forceps.
    6. Obtain a Mayo-Hegar needle holder.
    7. Obtain an Alm chest retractor with blunt teeth.
11. Utilize and obtain an autoclave and a bead sterilizer.

2. Ascending aortic banding surgical procedure

1. Anesthetize the animal with an intraperitoneal injection of a mix of 75-100 mg/kg Ketamine and 10 mg/kg Xylazine.
   NOTE: Allow a few minutes for the animal to be completely sedated and flaccid. If the anesthetic dose is not sufficient and the animal is still moving in the cage, re-inject the animal with the same anesthetic dose after allowing enough time, around 5-10 minutes between subsequent injections. Most animals require 1-2 injections to achieve deep sedation and anesthesia.
2. Shave the hair on the surgical site located at the right lateral thoracic area under the right armpit.
3. Stabilize the animal by gently taping all four limbs to the plastic board. Then perform endotracheal intubation with a 16 G angiocath. After the animal is successfully intubated, initiate mechanical ventilation with tidal volumes of 2 mL at 50 cycles/min and FiO₂ of 21%. Look for the symmetrical rise in chest wall with each breath.
4. Turn the animal slowly to lie on its left lateral side, and then bend the tail in a U-shape manner and stabilize it by gently taping it to the plastic board. Then go ahead and disinfect the shaved area with topical application of povidone-iodine. 
5. Infiltrate the skin at the incision site with 50/50 mix by volume of 1-2% Lidocaine/0.25-0.5 % Bupivacaine as preemptive analgesia before making the incision.
6. Perform a right horizontal skin incision, 1-2 centimeters long, in the right axillary area 1 cm below the right armpit. Then, dissect the thoracic muscular layer until reaching the thoracic rib cage. Make a 1 cm thoracotomy between the 2^nd and 3^rd rib cage.
   1. While dissecting the muscular layer of the chest, be careful and avoid injury of the right axillary artery, which runs underneath the right armpit.
      NOTE: Thoracotomy performed between the 1^st and the 2^nd rib carries the risk of banding the right brachiocephalic artery instead of the ascending aorta. Thoracotomy between the 3^rd and the fourth rib makes it hard to visualize and band the ascending aorta, as the operator will be looking at the right atrium.
      NOTE: Avoid extending the thoracotomy too medially towards the sternum to avoid dissecting and injuring the right internal mammary artery.
7. Dissect the two lobes of the thymus gland gently and push them apart on the side. Then identify the ascending aorta and isolate it from the superior vena cava by blunt dissection via a curved Graefe forceps.
   NOTE: Significant manipulation of the thymus gland will render it swollen and makes it hard to visualize the ascending aorta.
   1. Dissect the superior vena cava from the aorta with extra caution to avoid injury or rupture of the superior vena cava, which is fatal. This may be the trickiest part of the procedure and is expected to happen from time to time even in most experienced hands, but often with beginners and learners.
8. Lift gently the ascending aorta with a curved Graefe forceps and place the vascular clip around the ascending aorta.
   1. Adjust the vascular hemoclip ligation tool via a plastic pre-cut 7" piece to obtain a vascular clip of the desired internal area of 1.5 mm² or 2 mm², depending on which HF model is desired.
9. Suture the thorax via a Vicryl 2-0 monofilament suture. Then suture the muscular layer of the chest via a 3-0 Vicryl taper suture. Then suture the skin incision via a Nylon 3-0 monofilament suture.
10. Administer a combination of the following drugs after completion of the surgery for 48-72 hrs to serve as analgesia in the post-operative period: 1) Buprenorphine 0.01-0.05 mg/kg subcutaneously every 8-12h, 2) Meloxicam 2 mg/kg subcutaneously every 12h, and 3) Morphine 2.5 mg/kg subcutaneously every 2-4h as needed for severe pain.
    NOTE: Leave the animal to recover on a heating pad under regular monitoring. Once the animal shows signs of recovery from anesthesia (able to breath spontaneously - without evidence of gasping or use of accessory muscles for more than two minutes - and has good reflexes, red and warm extremities), extubate the animal and return it to the cage.

3. Echocardiography

1. Sedate the animal with intraperitoneal injection of 80-100 mg/kg ketamine. Ensure adequate sedation for proper acquisition of good quality echo images.
   NOTE: The use of isoflurane as an anesthetic is discouraged for its cardiodepressor effect, especially in the setting of severe pressure overload and might give a false impression of LV dilatation and systolic dysfunction that resolves once animal is off anesthetic.
   1. Be cautious and administer half or even one third of the dose of ketamine in animals that look dyspneic and tachypneic with suspicion that they have developed the HFrEF phenotype.
2. Shave the hair of the chest, anteriorly, in the completely sedated animal.
3. Lay the animal on its back and stabilize it to the plastic board.
4. Acquire 2D parasternal long axis and 2D parasternal short axis view clips at the level of the papillary muscle. Also, obtain M-mode images from the short parasternal axis view at the level of the papillary muscle to measure LV septal and posterior wall thickness in diastole as well as LV end-diastolic and end-systolic diameter.
   1. Acquire images or clips at a heart rate of 370 - 420 beats per minute to ensure proper assessment of LV size and function. Acquisition of images at lower heart rates will lead to a false impression of depressed LV function and LV dilatation.
      NOTE: Acquisition of foreshortened 2D long parasternal axis view images/clips lead to false measurements. For quality control purposes, make sure that the LV apex and the aorto-mitral angle are visualized within the same plane cut.
   2. Acquire 2D short parasternal axis view images/clips at the level of the mid papillary muscle. This will serve as a reference to obtain reliable serial and subsequent LV measurements while following the animals over time throughout the study period.
5. Obtain M-mode images in long parasternal axis view at the level of the aortic valve to assess the relative aortic to left atrium (LA) diameter at end systole.
   NOTE: Animals with the MOD and HFrEF phenotypes should show evidence of LA dilatation with LA/Ao ratio being ≥1.25 and <1.5 in MOD HF phenotype and ≥1.5 in the HFrEF phenotype¹⁰.

Representative Results

Characterization of the HF phenotypes, that develop 8-12 weeks following AAB, could be easily performed via echocardiography. Representative M-mode images of Sham, Week 3 post-AAB, MOD and HFrEF phenotypes are presented in Figure 1A. Figure 1B and Figure 1C are showing the vascular clip size for the creation of the MOD HF phenotype and HFrEF phenotype, respectively. The LV end-diastolic (LVEDV) and end-systolic (LVESV) volumes could be calculated by using the formulae of the area length method: V=5/6×A×L, where V is the volume in ml; A is the cross sectional area of the LV cavity in cm², obtained from the short parasternal axis view at the level of the mid-papillary muscle in diastole (Ad) and in systole (As); and L is the length of the LV cavity in cm, measured from the long parasternal axis view as the distance from the endocardial LV apex to the mitral-aortic junction in diastole (Ld) and in systole (Ls). Representative 2D long parasternal axis and short parasternal axis echocardiography images, with illustration on how to measure Ld, Ls, Ad and As, in Sham and MOD HF phenotype are presented in Figure 2. The LVEDV in the MOD HF phenotype usually ranges between 600 - 700 µL, with very few animals having LVEDV greater than 700 µL and up to 1000 µL; whilst, the LVESV in the MOD phenotype ranges between 120 - 160 µL (Table 1). From the 2D short parasternal axis view echocardiography images presented in Figure 2, one could appreciate the degree of LVH in the MOD phenotype compared to the sham. Representative pressure-volume loop tracings of the Sham, Week 3 post-AAB, MOD and HFrEF phenotypes are presented in Figure 3. The LV maximum pressure is at least 200 mmHg, even at week 3 post-AAB, and increases further at week 8 post-AAB due to the mismatch between the growth of the animal and aorta and the fixed created stenosis in the ascending aorta. Note that the animals at week 3 post-AAB are fully compensated with shift of the LVEDV and LVESV to the left compared to the sham. With progressive eccentric hypertrophy and remodeling, there is a shift in LVEDV and LVESV to the right in the MOD and HFrEF phenotypes compared to week 3 post-AAB. One could appreciate also the significant increase in LVESV in the MOD phenotype and the profound increase in LVESV in the HFrEF phenotype, which reflects the significant and profound decreases in stroke volume and LVEF in the MOD and HFrEF phenotypes, respectively, compared to week 3 post-AAB. Moreover, one could appreciate the significant increase in LVEF at week 3 post-AAB and the significant decrease in LVEF in the HFrEF phenotype compared to the sham.

The rat PO induced HF model is associated with high mortality and failure rates. Only about 20% of the rats that undergo AAB, with a vascular clip of 2 mm² in internal diameter, will transition to develop the MOD HF phenotype. Representative M-mode images of the failed phenotypes are presented in Figure 4. Figure 4A is showing representative M-mode images of animals that did not develop LVH at week 8 post-AAB, and had completely lost the PO with complete regression of LVH (sham-like) or had variable degree of LVH and PO at week 8 post-AAB causing a mild-moderate LVH phenotype. The second failed phenotype group is presented in Figure 4B showing representative M-mode images of animals with severe PO (LV maximum pressure >200 mmHg) and severe LVH who remained compensated with no evidence of eccentric remodeling, concentric remodeling (CR) group, or with a mild (MILD group) eccentric remodeling. Echocardiography and hemodynamic data of the sham, failed, and successful/desired phenotypes are presented in Figure 5 and Table 1. Note the progressive increases in heart weight and LV weight as the animals transition from a compensated phenotype to a more eccentric and remodeled phenotype. Also, there is an exponential increase in LVESV and decrease in LVEF as the animals transition from a compensated concentric remodeling to a decompensated eccentrically remodeled phenotype. Of particular interest is that both the MOD and HFrEF HF phenotypes have a similar degree of myocardial stiffness as measured by the stiffness-coefficient β of the end-diastolic pressure volume relationship (EDPVR (mmHg/µL)) compared to all the other phenotypes, whereas there is a gradual decrease in LV efficiency as the animals transition to a more eccentrically remodeled phenotype. LV efficiency is calculated from the end-systolic pressure volume relationship (ESPVR) divided by the arterial elastance (EA). Despite that there is no significant statistical difference in ESPVR and ESPVR/EA between the MOD and HFrEF phenotypes and the sham group, this is falsely the case as the MOD and HFrEF phenotypes have a significantly higher LV end-systolic pressure compared to the sham, making the ESPVR slope falsely steeper with shift in V0 to the right compared to the sham. Moreover, when the MOD and HFrEF phenotypes are compared to the compensated and concentrically remodeled phenotypes, which have the same degree of PO, then one could appreciate the significant and progressive increase in LVESV and drop in ESPVR and ESPVR/EA with progressive eccentric remodeling, as observed in the MOD and HFrEF phenotypes compared to the CR and MILD phenotypes (Figure 5 and Table 1).

Figure 1: Representative heart failure phenotypes at week 8 following ascending aortic banding. (A) Representative M-mode images of sham animals, animals three weeks following ascending aortic banding (AAB) and eight weeks following AAB. Figure 1A has been modified from Chaanine et al., American Journal of Physiology-Heart and Circulatory physiology, 2016. (B) Vascular clip size for the creation of severe left ventricular hypertrophy (LVH) with moderate eccentric remodeling (MOD). (C) Vascular clip size for the creation of severe LVH with overt systolic heart failure (HFrEF). Figures 1B and 1C has been obtained and modified from Chaanine et al., Methods in Molecular Biology, 2018. Please click here to view a larger version of this figure.

Figure 2: Calculation of left ventricular volumes by echocardiography using the area-length method. Representative 2D long parasternal and 2D short parasternal axis view echocardiography images to measure left ventricular (LV) cavity length in diastole (Ld) and in systole (Ls) and LV cavity cross sectional area in diastole (Ad) and in systole (As) in order to calculate LV volumes at end of diastole and systole. Please click here to view a larger version of this figure.

Figure 3: Pressure-volume loop tracings were obtained via a 1.9 F rat pressure-volume catheter using the open chest and left ventricular apical puncture approach. Representative pressure-volume loop tracings in Sham, week 3 following AAB, MOD and HFrEF phenotypes at week 8 following AAB. Figure has been modified from Chaanine et al., Circulation: Heart Failure, 2013. Please click here to view a larger version of this figure.

Figure 4: Encountered phenotypes at week 8 following AAB with failure to develop the desired heart failure phenotype(s). (A) Representative M-mode images of animals that lost the pressure overload (PO) and did not develop LVH (Sham-like) and those with variable PO and LVH (mild-moderate LVH) phenotypes. (B) Representative M-mode images of animals that developed severe PO, LVH and concentric remodeling (CR), but without (CR) or with mild (MILD) eccentric remodeling phenotypes. Figure 4B has been modified from Chaanine et al., Journal of American Heart Association, 2017. Please click here to view a larger version of this figure.

Figure 5: Echocardiography and pressure-volume loop parameters in the different phenotypes. Data are presented as individual values (dots) with median (horizontal line) in the different phenotypes at week 8 post-AAB. Statistical analysis results of the presented data in the different phenotypes are shown in table 1. LVESV: left ventricular end-systolic volume, LVEF: left ventricular ejection fraction, EDPVR: end-diastolic pressure volume relationship, ESPVR: end-systolic pressure volume relationship, EA: arterial elastance. Please click here to view a larger version of this figure.

	Sham (n=5)	Sham-like (n=5)	Mild-mod LVH (n=8)	CR (n=11)	MILD (n=14)	MOD (n=14)	HFrEF (n=5)
Body weight (g)	594 ± 37	466 ± 66	464 ± 22	497 ± 43	530 ± 59	478 ± 39	546 ± 18
HW (mg)	1269 ± 124.5	1328 ± 119	1614 ± 177	1645 ± 191a	1821 ± 169a,b	2106 ± 292a,b,c,d,e	2897 ± 182a,b,c,d,e,f
LVW (mg)	897 ± 94	968 ± 91	1161 ± 144	1222 ± 152a	1372 ± 135a,b	1580 ± 219a,b,c,d,e	1726 ± 82a,b,c,d,e
RVW (mg)	218± 22	218 ± 23	266 ± 24	239 ± 26	249 ± 26	283 ± 42a,b	565 ± 76a,b,c,d,e,f
IVSd (cm)	0.19 ± 0.01	0.21 ± 0.01	0.23 ± 0.01a	0.29 ± 0.01a,b,c	0.28 ± 0.02a,b,c	0.28 ± 0.01a,b,c	0.28 ± 0.02a,b,c
LVPWd (cm)	0.20 ± 0.01	0.21 ± 0.02	0.24 ± 0.01a,b	0.29 ± 0.02a,b,c	0.28 ± 0.02a,b,c	0.28 ± 0.01a,b,c	0.30 ± 0.02a,b,c
LVEDV (μl)	560.5 ± 25.8	570 ± 32	668 ± 143	442 ± 42,c	583 ± 45d	697 ± 129d,e	881.5 ± 55.7a,b,c,d,e,f
LVESV (μl)	105.9 ± 8.9	93 ± 15	111 ± 20	59 ± 7a,b,c	85.3 ± 10.6d	139.7 ± 22.5a,b,c,d,e	319.2 ± 51.5a,b,c,d,e,f
LVEF (%)	81.1 ± 1.2	83.7 ± 2.9	83.1 ± 2.5	86.5 ± 2.2a,c	85.4 ± 1.7a	79.8 ± 1.9b,c,d,e	64.1 ± 3.6a,b,c,d,e,f
LVPmax (mmHg)	121 ± 19	126 ± 23	186 ± 23a,b	218 ± 18a,b	221 ± 22a,b,c	*234 ± 25a,b,c	262 ± 16a,b,c,d,e
EDPVR (mmHg/μl)	0.018 ± 0.005	0.017 ± 0.004	0.041 ± 0.013	0.043 ± 0.017	0.039 ± 0.015	*0.068 ± 0.025a,b,c,d,e	0.079 ± 0.017a,b,c,d,e
ESPVR/EA	1.57 ± 0.67	1.96 ± 0.61	2.63 ± 1.52	3.35 ± 1.23a	2.62 ± 0.55	*1.63 ± 0.41d	0.82 ± 0.24c,d,e
Data are presented as mean ± standard deviation. Statistical analysis was performed using One-way ANOVA. P < 0.05 was considered significant.
aP < 0.05 vs Sham
bP < 0.05 vs Sham-like
cP < 0.05 vs Mild-modearte LVH
dP < 0.05 vs CR
eP < 0.05 vs MILD
fP < 0.05 vs MOD
*n=6
Abbreviations: HW: heart weight, LVW: left ventricular weight, RVW: right ventricular weight, IVSd: septal wall thickness in diastole, LVPWd: left ventricular posterior wall thickness in diastole. LVEDV: left ventricular end-diastolic volume, LVESV: left ventricular end-systolic volume, LVEF: left ventricular ejection fraction, LVPmax: left ventricular maximal pressure, EDPVR: end-diastolic pressure volume relationship, ESPVR: end-systolic pressure volume relationship, EA: arterial elastance.

Table 1: Echocardiography and pressure-volume parameters in Sham, Sham-like, Mild-moderate LVH, CR, MILD, MOD and HFrEF phenotypes.

Discussion

Following PO related to AAB in rat, the LV undergoes concentric remodeling by increasing LV wall thickness, known as concentric LVH, as a compensatory mechanism to counteract for the increase in LV wall stress. Increase in LV wall thickness becomes noticeable during the first week following AAB and reaches its maximum thickness at 2-3 weeks post-AAB. During this time period, activation of maladaptive signal transduction pathways lead to progressive enlargement of the LV with increases in LV volumes, a process known as eccentric hypertrophy or remodeling. It is expected that the HF phenotype in rat develops around 8 weeks following AAB in most of the animals with few of them developing HF at week 12 following AAB. Two HF phenotypes ensue depending on the severity of AAB. The MOD phenotype is obtained via the creation of ascending aortic banding (AAB) with a vascular clip of 2 mm² in internal diameter, whilst, the creation of the HFrEF phenotype requires AAB with a tighter vascular clip of 1.5 mm² in internal diameter. It is important to perform echocardiography at 2-3 weeks following ascending aortic banding to verify the presence of severe concentric LVH. Severe LVH is defined as LV septal and posterior wall thickness ≥1.5 times normal (0.19 cm), and usually ranges between 0.27 - 0.3 cm. Animals that do not develop severe LVH at week 3 following AAB, will be deemed as having unsuccessful AAB and should not be followed thereafter. Those that have developed severe LVH at week 3 following AAB, will undergo echocardiography at week 8 following AAB to assess for the development of the desired HF phenotype. It is not rare to encounter animals that had severe LVH at week 3 following AAB to have regression or resolution of LVH at week 8 following AAB, for reasons that we will address in the latter section of the discussion. Animals with severe LVH and concentric remodeling without or with mild eccentric remodeling at week 8 following AAB, therefore the CR and MILD phenotypes, respectively, are unlikely to develop further eccentric remodeling even if they are followed for an extended month or two. Those that are in between the MILD and MOD phenotype, may develop the MOD HF phenotype if they are followed for one more month.

The PO rat model can be frustrating due to the associated high mortality and failure rates¹⁰, despite use of a standardized vascular clip size and surgical technique, which also adds to the research expense, due to the large number of animals that need to undergo AAB in order to achieve the desired target number (n), and the length of time that the animals need to be followed before they develop the desired HF phenotype. Failure to develop severe LVH is related to either unsuccessful banding or banding of the right brachiocephalic artery instead of the aorta, which is not uncommon. Regression and/or resolution of severe LVH in subsequent follow up assessments is related to aneurysm formation and peri-band aortic remodeling that leads to loss in the severity of PO⁹. It remains unclear why animals with severe LVH and PO develop phenotypic variability in regard to eccentric remodeling despite having the same clip size, sex and strain. It is recommended to visualize the ascending aorta to screen for peri-band aortic remodeling and aneurysmal formation. Animals that develop ascending aortic aneurysm ≥1 cm in diameter should be euthanized, as this will cause dyspnea and distress to the animal due to impingement on surrounding structures. Also, it is recommended to check for turbulent flow across the band by color Doppler, but unfortunately precise estimation of pressure gradient across the band by continuous Doppler is not feasible due to the inability to align the continuous Doppler with the blood flow direction in the ascending aorta.

The MOD is an attractive HF phenotype and constitutes an evolution of the traditionally created HFrEF phenotype as it allows for targeting of signal transduction pathways that play role in myocardial remodeling early on in the disease process, especially when it pertains to perturbations in mitochondrial dynamics and function, myocardial metabolism and calcium cycling and extracellular matrix remodeling and myocardial stiffness; features that are highly evident in the MOD HF phenotype¹¹. Also, the early postoperative mortality (defined as mortality in the first 7 days post-AAB) is lower with the clip size of 2 mm², for the creation of MOD phenotype, than the clip size of 1.5 mm², for the creation of HFrEF phenotype¹⁰, (5% vs 21%, P = 0.009 using Fisher's exact test). However, the success rate between the two clip sizes, for the creation of MOD and HFrEF phenotypes, is not statistically significant¹⁰, (20% vs 13%, P = 0.56 using Fisher's exact test). Moreover, the aortic banding by vascular clip is advantageous over the aortic banding by tightening a nylon suture against a 27 G needle, a technique often used to constrict the transverse aorta in mice, because there is less variation in clip size and less trauma to the aorta compared to the suture technique.

The PO model of HF is advantageous over the myocardial infarction (MI) model of HF because the PO-induced circumferential and meridional wall stress is homogeneously distributed across all segments of the myocardium. However, both models suffer from variability in the severity of PO^10,11 and in infarct size^13,14 along with intense inflammation and scarring at the infarct site¹⁵ as well as adhesion to the chest wall and surrounding tissues observed in the MI model of HF. Moreover, the rat PO induced HF model is challenging to create as it is associated with high mortality and failure rates¹⁰, with only 20% of the operated rats developing the MOD HF phenotype¹⁰. When compared to the spontaneously hypertensive rat (SHR) model, the PO-induced HF model is a better model to study pathways related to myocardial remodeling. The increase in afterload and myocardial wall stress in systole is much higher in the PO-induced HF model than the SHR model. It takes about two years for the SHR to develop systolic HF and the mechanism of systolic HF is not entirely known and is confounded by aging¹⁶. The SHR model and other models of hypertension, such as the DOCA salt model, are more frequently used to investigate mechanisms and therapies related to hypertension and possibly diastolic dysfunction¹⁶.

In conclusion, the MOD HF phenotype is an attractive model to study signal transduction pathways in the context of myocardial remodeling and can be utilized for application and testing of potential therapeutic strategies, before validation of their efficacy in large animal models and in human heart failure.

Materials

Name	Company	Catalog Number	Comments
Adson forceps	F.S.T.	11019-12	surgical tool
Alm chest retractor with blunt teeth	ROBOZ	RS-6510	surgical tool
Graefe forceps, curved	F.S.T.	11152-10	surgical tool
Halsted-Mosquito Hemostats, straight	F.S.T.	13010-12	surgical tool
Hardened fine iris scissors, straight	Fine Science Tools F.S.T.	14090-11	surgical tool
hemoclip traditional-stainless steel ligating clips	Weck	523735	surgical tool
Mayo-Hegar needle holder	F.S.T.	12004-18	surgical tool
mechanical ventilator	CWE inc	SAR-830/AP	mechanical ventilator for small animals
Weck stainless steel Hemoclip ligation	Weck	533140	surgical tool