Guidelines for assessing maternal cardiovascular physiology during pregnancy and postpartum

Abstract

Maternal mortality rates are at an all-time high across the world and are set to increase in subsequent years. Cardiovascular disease is the leading cause of death during pregnancy and postpartum, especially in the United States. Therefore, understanding the physiological changes in the cardiovascular system during normal pregnancy is necessary to understand disease-related pathology. Significant systemic and cardiovascular physiological changes occur during pregnancy that are essential for supporting the maternal-fetal dyad. The physiological impact of pregnancy on the cardiovascular system has been examined in both experimental animal models and in humans. However, there is a continued need in this field of study to provide increased rigor and reproducibility. Therefore, these guidelines aim to provide information regarding best practices and recommendations to accurately and rigorously measure cardiovascular physiology during normal and cardiovascular disease-complicated pregnancies in human and animal models.

INTRODUCTION

Pregnancy is a unique physiological stress test that many women will experience within their lifetime. Despite this, there is little understanding of how the cardiovascular system adapts during uncomplicated, healthy pregnancies, with greater emphasis placed on understanding and reducing the pregnancy-associated cardiovascular disease burden. With maternal mortality rates across the world at an all-time high and likely to increase further in subsequent years, having a complete understanding of these cardiovascular adaptations is imperative. To adequately advance the field and to effectively aid in reducing maternal mortality rates and designing rational interventions, a greater understanding is needed concerning the plethora of cardiovascular physiological changes that occur during both normal and cardiovascular disease-complicated pregnancies.

Normal, uncomplicated pregnancies are characterized by several hormonal, systemic, and cardiovascular adaptations necessary to support maternal-fetal circulation and, ultimately, a healthy pregnancy. Understanding of these critical cardiovascular adaptations has been hindered by the lack of experimental consistency among studies and, in some cases, a lack of rigor in reporting key biological variables that impact pregnancy, postpartum, and lactation. Here, we review the best practices and recommendations to accurately and rigorously measure cardiovascular physiology during normal pregnancies and those complicated by cardiovascular disease in human and animal models. We discuss key physiological changes in the cardiovascular system that have been documented in both human and animal models and preferred experimental techniques used to examine these changes in vivo and in vitro, i.e., “pregnancy in a dish.” We highlight key pregnancy-associated variables that should be considered during experimental design, such as gravidity, parity, species used, and selection of time points or periods to intervene during pregnancy. In addition, we discuss the impact of several modifiable stressors of increasing importance, such as the impact of changes in diet, physical activity, and environmental exposures. We finally discuss the advantages and disadvantages of current models of pregnancy-associated cardiovascular diseases, such as hypertensive disorders of pregnancy, myocardial infarction, and peripartum cardiomyopathy, and provide recommendations for their respective uses. After increasing rigor and reproducibility, a secondary goal of these guidelines is to provide a framework for investigators to understand preferred methodologies and necessary considerations for completing meaningful assessments of maternal cardiovascular physiology.

Although this article focuses on assessing maternal cardiovascular physiology during pregnancy and postpartum, investigators with a combined interest in developmental programming should refer to our paired article “Guidelines for In Vivo Models of Developmental Programming of Cardiovascular Disease Risk.”

OVERVIEW OF MATERNAL CARDIOVASCULAR PHYSIOLOGY DURING PREGNANCY

Significant changes in systemic and cardiovascular physiology occur during pregnancy and postpartum, facilitating the physiological demands of a healthy pregnancy, labor and delivery, and postpartum recovery. In particular, there is a ∼48% expansion in plasma volume and a ∼30–40% increase in cardiac output (through increased stroke volume in rodents) above nonpregnant levels to facilitate the increasing metabolic demands of the growing fetus and a new organ, the placenta (1–3). Significant changes in sympathetic and parasympathetic balance during pregnancy contribute to reductions in heart rate variability during early pregnancy (4). A reduction in total peripheral resistance (∼35–40%) coupled with an increase in vascular compliance (1) prevents an increase in arterial pressure, which shows a transient reduction in midgestation in healthy pregnant humans, sheep, and rodents (5–8). Additional changes in maternal physiology include changes in maternal metabolism, necessary to divert nutrients to the growing fetus (2) (highlighted in Fig. 1). Our intent in this section is to briefly introduce the main maternal cardiac and vascular changes that occur during pregnancy. The reader is directed to comprehensive reviews that discuss in detail these adaptations (2, 6, 9, 10).

Figure 1.

Systemic and cardiovascular adaptations during pregnancy. A schematic depicting the key changes in both systemic and cardiovascular physiology that occur in pregnant women and animals. These adaptations must occur to meet the circulatory demands of the mother and fetus. Several hemodynamic changes occur including changes in cardiac output (CO), blood pressure (BP), and total peripheral resistance (TPR). In addition, there is significant remodeling of the ventricles, atria, aorta, and the vasculature, with the latter including remodeling of and increased blood flow to the uterine arteries. In addition, the placenta develops and provides vital nutrients necessary to sustain fetal growth and development, assisted by changes in maternal systemic metabolism. Arrows depict key systemic and cardiovascular changes common among all of the depicted mammalian species. Images were created with a licensed version of BioRender.com.

Cardiac Adaptations

The increased workload placed on the heart during pregnancy results in the physiological growth of the heart, i.e., an increase in heart size and mass and an increase in left ventricular chamber dimensions. The increase in heart size during pregnancy represents physiological hypertrophy that is thought to be reversible and occurs independent of changes associated with pathological hypertrophy, such as fibrosis, reduced cardiac function, and decreased angiogenesis (11, 12). However, this physiological hypertrophy is associated with temporal changes in different cardiac cell compartments and significant changes in cardiac glucose, lipid, and ketone body metabolism (13–15). Many maternal cardiovascular responses to stress are also diminished. For instance, the maternal chronotropic response to hypoxia is markedly reduced (16). This highlights that pregnancy not only affects basal but also stimulates cardiovascular function in the mother (16).

Many preclinical pregnancy studies have used commonly used experimental techniques to assess changes in systemic physiology and metabolism, such as the assessment of circulating factors and metabolites in the blood, through glucose tolerance tests, insulin tolerance tests, and evaluation of key metabolic organs. Studies primarily focused on the cardiac adaptations during a healthy pregnancy comprise standard documentation of mouse cardiac physiology (11–13, 17–19). In recent years, studies have also used omics approaches for interrogation of maternal systemic and cardiovascular adaptations to normal pregnancy (13); however, many have focused mostly on pathological consequences.

Vascular Adaptations

The maternal vascular system undergoes significant adaptations during pregnancy (9). Vascular adaptations occur at the cellular level, determining the function of the main components of the vascular wall and, consequently, facilitating changes in vascular tone and reactivity to circulating and local factors. During pregnancy, endothelial cells increase the production and bioavailability of vasodilatory products such as nitric oxide, accommodate large increases in fluid volumes, and form gap junctions in the uterine circulation that enhance vasodilatory responses (20–22). One of the primary changes in vascular smooth muscle cell function during pregnancy is a change in myogenic tone, primarily in uterine arteries (23–25). In pregnant women, the myogenic tone of myometrial arteries increases in late pregnancy (26), whereas uterine arteries from pregnant mice have reduced myogenic tone compared with nonpregnant mice (27). Perivascular adipose tissue also changes during pregnancy and this change is vascular bed specific. Adipose tissue surrounding rodent mesenteric arteries becomes anticontractile (28), whereas uterine perivascular adipose tissue has antidilatory effects that are mediated by distinct mechanisms in pregnant versus nonpregnant rats (29, 30).

The most pronounced changes in the maternal vascular system during pregnancy occur in the uteroplacental circulation, where hemodynamic adaptations continue a pattern established during the menstrual cycle (31). This pattern involves a reduction in uterine artery impedance and increased uterine artery blood flow, which reaches a >10-fold increase at term compared with prepregnancy levels (32). Extensive growth of the uterine circulation is a hallmark of a healthy pregnancy that facilitates large changes in uterine blood flow and the establishment of blood flow to a newly developed organ, the placenta. In humans and rodents, the placenta is hemochorial, and this anatomical architecture provides a limited barrier to the delivery of nutrients from the mother to the fetus (33). To support an increase in uterine and placental blood flow, uterine arteries, and veins experience outward expansive remodeling and an increase in length (8, 32). In species with hemochorial placentation, remodeling of the uterine spiral arteries, which are nonbranching end arteries that supply the outer endometrial layer (or decidua in pregnancy), contributes to a reduction in downstream resistance, secondary to the placentation process, and acceleration of the arterial blood in upstream vessels (i.e., ascending uterine arteries, arcuate, and radial arteries). This hemodynamic adaptation increases shear stress on the vascular wall and shear stress-induced production of nitric oxide from the vascular endothelium (10). Remodeling and increased dilatory responses of the uterine arteries, as well as local changes in cell signaling and altered responsiveness to paracrine and humoral factors, further contribute to pregnancy-induced increases in uterine blood flow (8, 34–40). The time course of vascular remodeling differs between small/preplacental uterine vessels and upstream large uterine vessels. Changes in large vessels start early in pregnancy (41) and may occur independently of the presence of the placenta (42). Pregnancy-associated adaptations of the uterine circulation differ among mammals, and this variation is due to differences in structure and anatomy of the uterine circulation, type of placentation [i.e., hemochorial in humans vs. epitheliochorial in ruminants or endotheliochorial in dogs (43)], uterine size, and number of offspring (44).

In addition to uterine blood flow, splanchnic (e.g., stomach, spleen, liver, intestines, and pancreas), renal, pulmonary, and skin blood flow increases during pregnancy (45). In the cerebral circulation, receptor, and transporter activity changes, and the upper and lower limits of the cerebral blood flow autoregulatory curve extend during pregnancy, maintaining brain homeostasis and protecting the mother’s brain from blood pressure fluctuations (46). Despite these alterations, the changes in cerebral blood flow during pregnancy are minimal (46). The fetoplacental unit secretes signaling molecules into the maternal circulation that can act upon cardiac and vascular cells to induce functional and structural changes. Thus, fetoplacental signals play a key role in the expansion and adaptability of the maternal cardiovascular system during pregnancy [for detailed discussion, see Osol et al. (9)].

Postpartum Cardiovascular Adaptations

Little is known about the temporal changes of the maternal vascular system during the immediate or long-term postpartum period following a healthy pregnancy. Mean arterial pressure is reduced, and arterial compliance is increased one year postpartum compared with prepregnancy values in women who had a healthy pregnancy, suggesting a long-lasting beneficial effect of pregnancy on the female vascular system (47). Pregnancy-induced remodeling of the uterine arteries is partially reversed during the postpartum period, whereas remodeling of arteries in nonvascular beds, such as the ascending aorta, is fully reversed during postpartum (47).

BEST PRACTICES FOR STUDYING MATERNAL CARDIOVASCULAR PHYSIOLOGY DURING PREGNANCY AND POSTPARTUM

An examination of recent American Journal of Physiology articles in the past 10 years has shown that there appears to be a significant discrepancy regarding reporting key biological and pregnancy-related variables within animal studies. For example, between published articles, there seems to be a lack of consistency between the specific time points examined during pregnancy and postpartum, alongside a lack of information or consistency regarding essential study information, such as the starting age and particular strain of animals used (e.g., C57BL/6 mice; 6 N or 6 J), and pregnancy-related information such as litter sizes, number of pregnancies, gestational length, lactation status, and hormonal status. In addition, sometimes, it is unclear whether the primary focus of the studies is maternal, fetal, or both. In some cases, the key cardiovascular adaptations are not always fully assessed. These considerations will be discussed in further detail in the following sections.

Studying Cardiovascular Physiology of Pregnancy in Human Subjects

The rapid and profound physiological changes of pregnancy significantly affect the conduct and interpretation of standard cardiovascular assessments. Adaptations are progressive but occur at different time lines and trajectories during and following pregnancy (i.e., blood pressure vs. heart rate) (2). Thus, it is important to standardize assessments based on gestational age and time since childbirth when conducting cardiovascular assessments during the peripartum period. The selection of the appropriate control group will be dependent on the primary research question, and whenever possible control or match groups for key population characteristics that are known to influence cardiovascular health [e.g., age, body mass index (BMI), parity, gestational age]. For example, if the primary study aim was to examine the impact of preeclampsia on cardiovascular health, the control group would consist of pregnant women not diagnosed with preeclampsia who were matched primarily by gestational age and, as much as possible, age, parity, and BMI. In contrast, if the primary aim was to examine the impact of pregnancy on cardiovascular health, the control group would be matched by key population characteristics (e.g., age, BMI, parity).

Key considerations for cardiovascular testing during pregnancy.

• 1) Determine (and report) weeks gestation, gravidity, and parity (2).
• 2) Screen for a prior or current history of pregnancy complications known to impact cardiovascular health (e.g., gestational diabetes, gestational hypertension, preeclampsia; emerging risk factors include preterm delivery, intrauterine growth restriction) (48).
• 3) Understand the normal and abnormal physiological adaptations to pregnancy so that medical referral can be made, as appropriate, if observed (e.g., undiagnosed hypertension, hyperglycemia) (49).
• 4) Follow up with the mother postchildbirth to identify pregnancy complications that may have developed later in/after pregnancy, as there is some evidence that cardiovascular dysfunction can be present before diagnosis of complications (e.g., preeclampsia) (50, 51).
• 5) Normal pretest standardization for cardiovascular testing in nonpregnant populations should be followed in pregnant and postpartum populations (e.g., fasted >6 h, avoid exercise >24 h, no intake of caffeine, vitamin C, alcohol, and smoking for >12 h) (52). However, because of pregnancy-induced changes in metabolism and glucose stores, lengthy fasted protocols may not be advisable in some pregnant populations (e.g., insulin-treated gestational diabetes mellitus). When fasted protocols cannot be conducted, standardized pretest nutritional intake is recommended (e.g., a standard snack rationed by grams of macronutrients per kilogram body weight at a prespecified interval before the test). It is, however, recognized that this is not always feasible.
• 6) The time of day that the vascular assessment was conducted should be standardized and reported to account for diurnal changes in cardiovascular function (52).
• 7) As pregnancy progresses, the uterus may compress the aorta and vena cava in the supine position leading to transient alterations in venous return, cardiac output, blood pressure, and heart rate (49). The supine position should be avoided in favor of a left lateral tilt in the recumbent position, or a semirecumbent position when measurements are being made.
• 8) When longitudinal cardiovascular assessments are conducted, maternal posture should be maintained across all assessments from preconception to postpartum.
• 9) When feasible, blood sampling to assess sex hormones (estrogen, progesterone, and testosterone) (49), glucose, lipid, and insulin is recommended given their known influence on cardiovascular health and function.

Although many physiological adaptations reverse quickly after childbirth, others remain transiently (or permanently) altered. Some of the cardiovascular adaptations to pregnancy may persist past the first-year postpartum (e.g., increased cardiac output, aortic diameter) thus postpartum “controls” should not be considered the same as individuals who have never been pregnant. Breastfeeding may also impact cardiovascular health (53). Thus, data on the length and type of breastfeeding (e.g., exclusive vs. mixed) should be recorded.

Modeling Cardiovascular Physiology of Pregnancy in Experimental Animals

There are many mammalian species that can be used to model human pregnancy, each with important strengths and weaknesses (see Table 1). Genetically modified mice are a powerful tool that can be used to understand molecular mechanisms underlying maternal cardiovascular health and disease as a result of their similarity to humans, short gestation period, and the emergence of genomics (54). In mice (55), rats (56), rabbits (57), and guinea pigs (58), the maternal cardiovascular adaptations to pregnancy can be studied noninvasively, for instance focusing on measurements of maternal cardiac geometry and function using echocardiography, or calculating pulsatility index (PI, the measure of blood flow velocity in a vessel during the cardiac cycle), as a marker of uteroplacental vascular resistance via Doppler blood flow velocimetry. In addition, although used less frequently, cardiac magnetic resonance imaging has been used to assess cardiac function in monkey pregnancy studies (59). In the smaller species, echocardiography is usually performed under reversible anesthesia and, therefore, can be done serially. However, care must be taken because anesthesia can only be induced a fixed number of times in any one animal, and the anesthetic itself may depress maternal cardiovascular function (60). Radiotelemetry in rodents circumvents some of these problems, but cardiovascular measurements obtained using this method are currently limited to changes in resting blood pressure and heart rate (61). A few studies have attempted chronically instrumented preparations in rats surgically prepared with arterial and venous catheters and a femoral Transonic flow probe (see Table 1). Following postsurgical recovery, drugs can be administered to study not only basal but also stimulated cardiovascular function (62–64). In mice, rats, rabbits, and guinea pigs, tissues can also be isolated at terminal endpoints. One can study molecular mechanisms in the placenta (65, 66) and cardiovascular function in the pregnant and postpartum animal via isolated Langendorff preparations (67, 68), and perfusion or wire myography (69, 70). However, some rodent species present important weaknesses, which need careful consideration. Rats are a better model for studying placentation due to the presence of hemochorial placentation, observed in both humans and rats (71). However, in rats and mice, cardiovascular maturation continues past birth, becoming complete by the second week of postnatal life (72). Rodents also give birth to litters, so there may be differences in the maternal metabolic adaptation to pregnancy. Smaller species will also have low blood volume, limiting the number of blood samples that could be taken serially, for instance, with advancing gestation. In contrast, sheep and humans share similar tempos of prenatal cardiovascular development (72) and some breeds of sheep, like Welsh Mountain, give birth primarily to singleton lambs of similar weight to term human babies (7). Most importantly, the sheep permits surgical instrumentation of the mother and fetus for long-term in vivo cardiovascular recording and serial blood sampling for endocrinology and daily monitoring of blood gases and metabolic status (7, 73–76). Recently, there has been an important development for research in sheep with the introduction of CamDAS technology (shown in Fig. 2). CamDAS is a wireless data acquisition system that permits the simultaneous recording of multiple pressure and blood flow signals from mother and fetus in vivo over long periods of gestation in free-moving animals (7, 73–76). The systems not only improve animal welfare and, thereby, the physiological quality of the recording, but they also permit the continuous measurement of cardiovascular function in the mother and the fetus beat-to-beat in different environments for weeks to months. For instance, by combining CamDAS technology with the creation of hypoxic isolators, it has been possible to investigate longitudinally the effect of hypoxic pregnancy on the maternal cardiovascular adaptation to pregnancy and the fetal physiology (7, 73–76). Clearly, there are gross anatomical differences between the human hemochorial and the ovine cotyledonary placenta (77). However, there are also important similarities. Both sheep and humans have placental countercurrent flow of maternal and fetal blood within the placental villous tree, comparable transplacental oxygen gradients and oxygen consumption rates, as well as similar nutrient transporter expression (7). At the molecular level, the induction of oxidative and endoplasmic reticulum (ER) stress and the activation of the unfolded protein response (UPR) are highly conserved pathways across species (78). This is also the case for rodents, in which activation of the placental UPR^ER has been demonstrated in hypoxic pregnancy (79). Llamas, which share similar advantages to sheep and are amenable to similar experimentation, have been used to model human adaptation to high-altitude pregnancy (80). Nonhuman primates like macaques and baboons are clearly the most similar to humans. However, clear disadvantages of the larger and more sentient species, including sheep, llamas, and nonhuman primates, are ethical concerns and high-maintenance costs. For this reason, study of maternal cardiovascular function in nonhuman primate species have been mostly noninvasive in multiuser colonies that are kept very long term (81). Additional examples of examination of maternal cardiovascular physiology are listed in Table 1 (82–98).

Table 1.

Modeling human pregnancy in different animal species: comparative advantages and disadvantages in the study of maternal cardiovascular physiology

	Pros and Cons		Possible Measurable Outcomes
Species	Main Advantages	Main Disadvantages	Echo/Doppler	Serial Blood Sampling	In Vivo Chronically Instrumented Preparation	Langendorff Preparation	Perfusion/ Wire Myography	Molecular	Stereology/Histology
Mouse	Genetic manipulation (PMID: 35510543) Short gestation; amenable to transgenerational studies Hemomonochorial placenta (ISBN:978-3-540-78796-9)	Litter bearing Born immature; does not mirror the tempo of human development (PMID: 35534925). Small; limited blood volume	✓ Under anesthesia PMID: 30293577	× Not reported	Limited PMIDs: 27927648, 24935937, 22014504, 20100997, 11242593	Limited PMIDs: 24705556, 25051449	✓ PMIDs: 35510543, 27181166, 26110512	✓ PMIDs: 35510543, 30293577, 25051449, 26110512	✓ PMID: 26110512
Rat	Larger than mouse Short gestation; amenable to transgenerational studies Hemomonochorial placenta (ISBN:978-3-540-78796-9)	Litter bearing Born immature; does not mirror the tempo of human development (PMID: 35534925) Small; limited blood volume	✓ Under anesthesia PMID: 37506535	× Not reported	Limited PMIDs: 23856654, 33788974, 34757774	✓ PMIDs: 33788974, 30354714	✓ PMIDs: 15326063, 37615097, 34694887, 34634151	✓ PMIDs: 33788974, 30248337, 22289909, 34694887, 34634151	✓ PMIDs: 30248337, 22289909
Guinea pig	Mirrors tempo of human development (PMIDs: 29633280, 35534925) Midrange gestation; amenable to transgenerational studies Hemomonochorial placenta ISBN:978-3-540-78796-9	Litter bearing Small; limited blood volume	✓ Under anesthesia PMID: 29633280	× Not reported	Limited PMID: 29847165	✓ PMID: 36487188	✓ PMID: 27739590	✓ PMID: 29633280	✓ PMID: 29633280
Rabbit	Short gestation; amenable to transgenerational studies Hemomonochorial placenta ISBN:978-3-540-78796-9	Litter bearing Born immature; does not mirror the tempo of human development PMID: 35534925 Small; limited blood volume	✓ Under anesthesia PMID: 35927671	× Not reported	× Not reported	✓ PMID: 12271152	✓ PMID: 26806799	✓ PMID: 35927671	✓ PMID: 35927671
Sheep	Mainly singleton-twin bearing Mirrors tempo of human development PMID: 35534925 Large; high blood volume	Cotyledonary placenta ISBN:978-3-540-78796-9 Expensive	✓ Under anesthesia PMID: 35534925	✓ PMIDs: 35534925, 31039691, 26926316, 32862711, 34694887	✓ PMIDs: 35534925, 31039691, 26926316, 32862711, 34694887	✓ PMID: 26806799	✓ PMID: 34694887	✓ PMIDs: 35534925, 34694887	✓ PMIDs: 35534925, 34694887
Llama	Mainly singleton-twin bearing Large; high blood volume	Epitheliochorial placenta ISBN:978-3-540-78796-9 Expensive	✓ Under anesthesia PMID: 11262476	✓ PMIDs: 12855051, 18006479	✓ PMIDs: 12855051, 18006479	× Not reported	✓ PMIDs: 12855051, 18006479	✓ PMIDs: 12855051, 18006479	✓ PMIDs: 12855051, 18006479
Nonhuman primate	Mainly singleton-twin bearing Mirrors tempo of human development PMID:35534925 Large; high blood volume Hemomonochorial placenta ISBN:978-3-540-78796-9	Expensive	✓ Under anesthesia PMID: 21447636	✓ PMID: 9546793	× PMID: 8633635	× Not reported	✓ PMID: 11262476	✓ PMIDs: 21447636, 9546793	✓ PMIDs: 21447636, 9546793

Figure 2.

The CamDAS system used for recording maternal and fetal cardiovascular function during hypoxic pregnancy. A: a specially designed nitrogen-generating system supplies compressed air and nitrogen to bespoke isobaric hypoxic chambers housed at The Barcroft Centre, University of Cambridge. Each chamber is equipped with an electronic servo-controlled humidity cool steam injection system to return the appropriate humidity to the inspirate (i). Ambient partial pressures of oxygen and carbon dioxide, humidity, and temperature within each chamber were monitored via sensors (ii). For experimental procedures, each chamber had a double transfer port (iii) to internalize material and a manually operated sliding panel (iv) to bring the ewe into a position, where daily sampling of blood could be achieved through glove compartments (v). Each chamber incorporated a drinking bowl on continuous water supply and a rotating food compartment (vi) for determining food intake. A sealed transfer isolation cart could be attached to a side exit (vii) to couple chambers together for cleaning. Waste could be disposed via a sealable pipe (viii). B: some ewes can be instrumented with the CamDAS system during surgery under general anesthesia. Following postsurgical recovery (usually 5 days), the system can monitor continuous longitudinal changes in maternal and fetal cardiovascular function, for example: maternal and fetal arterial blood pressure, maternal and fetal heart rate and maternal and fetal femoral blood flow, uterine blood flow, umbilical blood flow, and fetal carotid blood flow. The wireless CamDAS system is contained in two parts in a custom-made sheep jacket: the data acquisition box (ix) on one side and a box containing the pressure transducers (x) on the other side. Cables (xi) provide connection between the two boxes and to two lithium battery packs. Measurements made using the CamDAS system are transmitted wirelessly via Bluetooth technology (xiii) to a laptop on the outside (xii), on which it is possible to continuously measure and record the maternal and fetal cardiovascular function during the experimental period. CC-BY: reproduced with permission (7, 73–75).

Studying Cardiovascular Physiology of Pregnancy in a Dish: In Vitro Studies

Pregnancy is a multifaceted physiological event affecting numerous organ systems. Studying its molecular pathways in the cardiovascular system in humans is challenging because of required invasive techniques, leading researchers to often use animal models. However, because these models do not fully capture human molecular physiology, consideration of in vitro models for examining pregnancy-related effects should be made in addition to in vivo models. Human pluripotent stem cells, derived from embryonic stem cells or sources like dermal fibroblasts or peripheral blood mononuclear cells (PBMCs), serve as a primary resource. These cells can be studied in homogeneous cultures (99), cocultures (100, 101), or engineered tissues (102, 103).

The systemic nature of pregnancy necessitates the incorporation of various associated factors into these models, including circulating profiles like hormones (104), metabolic substrates (105–107), and hemodynamic changes such as heart rate alterations (108) and plasma volume shifts (109). Integrating these factors can be intricate, especially when combining multiple elements. These parameters also change as pregnancy progresses and should be modeled in relation to the stage of pregnancy to be studied.

In vitro heart rate elevation can be achieved via chronic adrenergic stimulation (110, 111) or electrical pacing (112–114). Electrical pacing offers a sustained response but can be detrimental to cardiomyocytes over time. In contrast, the efficacy of adrenergic stimulation may diminish because of receptor desensitization. While blood pressure sees a minor rise during normal pregnancy (108), it can be modeled in vitro (i.e., as increased preload) by applying cyclic mechanical stretching (115, 116). Modeling increased cardiac afterload in monolayer cultures is problematic due to cell culture substrate rigidity. Engineered cardiac tissues provide a viable method for simulating preload and afterload, as evidenced in earlier studies (102, 117). To model increased afterload, one can embed the tissue within a framework of rigid pillars that maintain the tissue’s dimensions at its natural resting state. Conversely, preload simulation can be achieved by attaching the tissue to a pliable strip that exerts a pulling force, thereby extending the tissue beyond its resting length.

Metabolic substrate and hormone changes during pregnancy are well documented (105–107), allowing culture media adjustments accordingly. Incorporating these hormones at appropriate concentrations is possible, but supplementing culture medium with serum from pregnant women provides a comprehensive approach.

Depending on the research focus, integrating multiple aspects is ideal to best represent pregnancy in vitro. Both two-dimensional and three-dimensional cultures can accommodate these factors, but cocultures or engineered tissues mimicking the heart and vasculature currently offer the most accurate representation (see Fig. 3).

Figure 3.

Models to mimic aspects of pregnancy in vitro. Stem cells can be differentiated into cardiomyocytes and other cell types that can be studied in homogeneous cultures or in coculture systems (e.g., tissue engineering). Techniques for heart rate modulation, blood pressure modeling, and culture media adjustments are also depicted, highlighting the comprehensive approach to simulate pregnancy conditions. PBMC, peripheral blood mononuclear cell.

RELEVANT MODELS OF COMMON PREGNANCY-ASSOCIATED DISEASES AND RISK FACTORS

Hypertensive Disorders of Pregnancy

Hypertensive disorders of pregnancy (HDP) are leading causes of maternal and fetal mortality and morbidity, affecting 116 per 100,000 women of reproductive age globally (118). The most prevalent of these disorders are preeclampsia and related complications [i.e., eclampsia and hemolysis, elevated liver enzymes, and low platelet count (HELLP) syndrome], gestational hypertension, chronic hypertension, and preeclampsia superimposed on chronic hypertension (119–121).

Preventative and therapeutic interventions to treat HDP and its cardiovascular sequelae are limited; thus, there is an urgent need for research on the mechanisms of HDP and therapeutics. Because of ethical aspects and restrictions in pregnant persons’ inclusion in clinical trials, the scientific community has turned their efforts to the development and use of experimental animal models. In vivo models range from large animals such as nonhuman primates, baboons, rhesus monkeys, dogs, and sheep to small animals such as rats, guinea pigs, and mice (122). Induction of HDP in an animal model is experimentally accomplished by disruption of maternal cardiovascular adaptations and pregnancy homeostasis. The most used experimental approaches include 1) surgical reduction of uteroplacental blood flow or placental ischemia (123–133), 2) in vivo knock out or overexpression of genes using transgenic animals (82, 134–145), 3) exogenous administration of pharmacological agonists or antagonists of key molecular pathways (146–162), and 4) environmental stressors such as prenatal high-cholesterol diet (163–168) or hypoxia (7) (see Table 2). Rodents with preexisting hypertension or other preexisting comorbidities superimposed by signs of preeclampsia during pregnancy have also been used to understand the early development of the syndrome (see Table 2) (169–175).

Table 2.

Common pregnancy-associated diseases and relevant surgical, diet-induced, pharmacologically induced, genetically induced, and spontaneously induced animal models

Pregnancy-Associated Disease or Risk Factor	Approach	Model	Species	Hemodynamics	Maternal Vascular Function	Maternal Cardiac Function	Postpartum Cardiovascular Complications	Key Selected References (PMID)
Preeclampsia	Surgical	Uterine artery ligation	Primates	↑BP	?	?	?	17377512
		RUPP (uterine and abdominal aortic occlusion)	Rats Mice	↑BP ↑Uterine artery resistance	Endothelial dysfunction Glomerular endotheliosis ↑ Vascular oxidative stress	↓ LV EF Cardiac hypertrophy	Cerebral edema Neuroinflammation ↓ LV EF ↓ Stroke volume BP ↑ or normalized Vascular dysfunction	25298513 10642326 31669926 29588233 28202440 34727994 27658290 31266978
		Selective (s)RUPP (ovarian and uterine occlusion)	Rats	↑BP ↑Uterine artery resistance	↑vascular oxidative stress	?	?	31266978
	Diet induced	High-cholesterol diet	Rats	↑BP	Endothelial dysfunction Cerebrovascular dysfunction	?	Cerebrovascular dysfunction	32856035 36689585 35137612 35615682
	Genetic/Transgenic	Angiotensinogen/renin	Rats Mice	↑BP	Vascular dysfunction Glomerular endotheliosis	↑Cardiomyocyte size	↑LV Mass ↓Global longitudinal strain ↑Cardiomyocyte size ↑Cardiac output	26394667 31786987 27927648
		STOX1	Mice	↑BP	?	Cardiac hypertrophy	BP and albuminuria normalized	23357179 26758611
		C1q−/−	Mice	↑BP ↑stroke volume (postpartum) ↑end diastolic volume (postpartum) ↑end systolic volume (postpartum)	Vascular dysfunction	?	↓BP after delivery ↑BP late postpartum Mild vascular dysfunction, ↑LV mass ↑myocardial cell hypertrophy and fibrosis	32374620 29669946
		COMT−/−	Mice	↑BP	↑Vascular dysfunction	?	?	22392899 18469803
	Pharmacological/Infusion	AVP	Rats Mice	↑BP	Glomerular endotheliosis	?	?	30282823 25001273
		TLR agonists	Rats Mice	↑BP	Vascular dysfunction ↑Vascular oxidative stress	?	?	29437897 26873968 37352412 25853857 19779466
		TNF-α	Rats	↑BP	Impaired BBB	?	?	15928030 26400187
		sFlt-1	Rats Mice	↑BP	Vascular dysfunction	?	No long-term effects on BP and vascular function	17403433 19915174 20026766 20567176
		Nω-nitro-l-arginine methyl ester	Rats Mice	↑BP	Vascular dysfunction	?	Hypertension resolved Impaired vascular reactivity	36260752 25240245
		Chronic AT1-AA	Rats	↑BP	Renal vascular dysfunction	?	↑ LV Mass ↑ Cardiac collagen deposition Myocardial hypertrophy	21600854 24979132
	Spontaneous	Dahl Salt-sensitive (SS)	Rats	↑BP ↑uterine artery resistance	↑Cerebral blood flow autoregulation ↑Myogenic tone ↑Blood brain permeability	?	?	25904684 30612494 33275518
		BPH/5	Mice	↑BP ↑Uterine artery resistance	Decidual artery remodeling		BP recovery after delivery	11882569 16957025
Diabetes	Diabetic pregnancy	Diabetic-NOD	Mice	↓Mean uterine artery velocity ↓MAP mid to late pregnancy	Endothelial dysfunction	Early LV dilation No ↑ in LV CO	?	17827401 22014504 23636813
	Gestational diabetes	Db/+	Mice	↑Uterine artery resistance	Endothelial dysfunction	?	↑Long-term endothelial dysfunction	21266665 20554929
Peripartum Cardiomyopathy	Genetic/transgenic	Cardiomyocyte STAT3−/− and Cardiomyocyte STAT3−/− with caAKT	Mice	No change in HR	Increased Vasohibin, Endothelial apoptosis and dysfunction, Vascular insufficiency Capillary dropout	Systolic heart failure (after 1–2 pregnancies) ↓ FS% ↑ HW ↑ LVEDD, LVESV Energetic deficits fibrosis	Systolic heart failure and cardiomyopathy (after 1–2 pregnancies)	17289576 28201733 24448315
		Cardiomyocyte PGC1-α−/−	Mice	↓HR	↓endothelial gene expression Capillary dropout, ↓microvascular density, ↓VEGF	Systolic heart failure (after 1–2 pregnancies) ↓FS% ↑HW:TL ↑LVEDD, LVESD, ↑E/E′	Dilated cardiomyopathy (after 1–2 pregnancies)	22596155
		Titin (TTN) mutation/ TTNtv knockin	Mice	?	?	TTNtv mutation similar that seen in women with PPCM, ↑fibrosis ↑LVEDD ↓ FS%	?	24558114 19406126
		Cardiomyocyte Ndufs4−/−	Mice	?	?	↑HW:TL ↓FS% ↑LV dimension	Cardiomyopathy (after 1–6 pregnancies)	23931755
		Cardiomyocyte Perk−/−	Mice	No change in BP	Capillary dropout, ↓VEGFa	Systolic heart failure (after 1–2 pregnancies) ↓FS% ↑HW ↑LVEDD, Oxidative stress and protein aggregates, apoptosis	Cardiomyopathy and ↑mortality (after 1–3 pregnancies)	34548576
		Cardiac NPR1−/−	Mice	No change CO	?	↓FS, EF, ↑LVEDV, LVESV, ↑fibrosis, myocyte size, HW:TL (during lactation period).	Cardiomyopathy and ↑mortality, fibrosis, myocyte size, HW:TL (after 1–5 pregnancies)	31665900
	Pharmacological	Notch1 inhibition (with LY-411575)	Mice	?	Capillary drop out, ↓VEGFa	↓FS, EF ↑LVEDD, LVEDV, LVESD, LVESV ↑HW:TL and myocyte size ↑fetal gene expression fibrosis	↓FS, EF ↑LVEDD, LVEDV, LVESD, LVESV ↑HW:TL and myocyte size ↑fetal gene expression ↑fibrosis	32529705
		ErbB1/ErbB2 inhibitor (Lapatinib)	Rats and mice	No change in HR	?	↑Mortality, ↑HW:TL ↓ FS, EF ↑LVEDD, No fibrosis No apoptosis	Cardiomyopathy	21186272
Environmental exposures		Nano-TIO2 (Dose range: 9–12 mg/m3 for 4–6 h/day; number of maternal exposure days varies within studies).	Rat	↑Uterine artery vasoconstriction ↓Placental blood flow	Uterine artery endothelial dysfunction Impaired placental vascular reactivity	↓Mitral value deceleration time ↑E/A ratio	↓Female pup weight ↑Oxidative stress in Dam and Offspring	35642938 35066857 36950144 35260159 37181648 30734150 31215478
		Ozone (Dose: 1 ppm O3 for 4 h on GD10.5)	Mice	?	?	↓SV ↓LV EF Altered transcriptomics	?	37695302
		Ozone (Dose: 0.4–0.8 ppm O3 for 4 h/day on GD5 and 6)	Rat	↑Uterine artery resistance	?	?	?	29269335
		Diesel Exhaust (Dose: 0.3 mg/m3 for 6 h/day, 5 d/wk, from GD0.5 to 17.5).	Mice	?	↑Increase embryo reabsorption ↑Placental hemorrhage	?	12-wk-old offspring: increased susceptibility to pressure-overload-induced heart failure	24533117
		Electronic cigarette (Dose: aerosol mass or chemical concentration not reported – exposure condition was 3 h/d, 5 day/wk, from GD5–20).	Rat	↓Uterine artery and fetal blood flow	?	?	↓Offspring weight ↓Crown-rump length	30653941

BBB, Blood brain barrier; BP, blood pressure; CO, cardiac output; EF, ejection fraction; FS, fractional shortening; GD, gestational day; GFR, glomerular filtration rate; HW, heart weight; LV, left ventricular; LVEDD, LV end-diastolic dimension; LVEDV, LV end-diastolic volume; LVESD, LV end-systolic dimension; LVESV, LV end-systolic volume; MAP, mean arterial pressure; PGC-1α, peroxisome proliferator-activated receptor-γ coactivator 1-α; PPCM, peripartum cardiomyopathy; RUPP, reduced uterine perfusion pressure; TL, tibia length; TLR, Toll-like receptor.

When deciding which HDP animal model is appropriate to address a research question, the following concepts should be considered:

• 1) Comparative physiological variation between humans and the experimental species of interest (176). In addition to variations in placental morphology and function (77, 177), the maternal cardiovascular adaptations and the maternal immune and hormonal changes during pregnancy and postpartum should be contemplated in a comparative fashion.
• 2) Heterogeneity of clinical presentation and potential clinical subtypes of HDP (178). For example, although an animal model may not recapitulate the heterogeneity in the diagnostic onset of preeclampsia (i.e., early vs. late-onset preeclampsia), it is still valuable in studying other cardiovascular features of the syndrome, such as maternal endothelial dysfunction, increased maternal blood pressure, impaired maternal cerebrovascular function, or maternal cardiac systolic and diastolic function.
• 3) Opportunity to study later-life maternal and offspring health. Animal models provide preclinical evidence for mechanistic pathways to later-life cardiovascular health. Because of shorter lengths of gestation and life spans, experimental models afford feasibility for long-term follow-ups of the mother and the offspring after a pregnancy with HDP.
• 4) Other considerations: gestational age and duration, timing of treatment or induction of HDP, time points of postpartum assessment for early versus late postpartum, and the relation of these reproductive stages to animal’s life span and expected reproductive transitions.

Gestational Diabetes and Diabetic Pregnancies

Pregestational and gestational diabetes are both characterized by elevated blood glucose, which can lead to both maternal and fetal complications (179). The incidence of women with preexisting diabetes is ∼2%, and the global incidence of gestational diabetes is 14% (179, 180). Women with gestational diabetes exhibit subclinical left ventricular structural changes, and those with insulin-dependent diabetic pregnancy display early restrictive filling patterns (181, 182). In addition, pregnant diabetic subjects exhibit endothelial dysfunction and altered uterine-placental vasculature (183, 184). To study the impact of maternal diabetes, animal models are frequently used (185). Animal models of diabetic pregnancy have hyperglycemia at the onset of or starting in early pregnancy, whereas models of gestational diabetes should have hyperglycemia onset later in pregnancy that ideally reverses with parturition (186). Blood glucose levels vary with food intake, thus fasting glucose levels are routinely compared. Blood glucose concentrations are monitored with a glucometer validated for the species or by biochemical assay (187). For most rodents, fasting blood glucose approximates 100 mg/dL. In rodents, mild hyperglycemia is ∼120–200 mg/dL and severe hyperglycemia over 250 mg/dL, although the cutoffs for mild and severe are subjective (185, 188).

Diabetic animal models include surgical pancreatectomy and viral approaches, but chemically induced and genetic models are the most common (see Table 2) (84, 186, 189–192). Chemical treatment with streptozotocin (STZ) or alloxan damages β cells, and animal age, dose, and timing relative to mating will determine the degree of hyperglycemia (193). The effects of STZ and alloxan are mostly irreversible making their use a more appropriate model of diabetic pregnancy rather than gestational diabetes. Using an insulin receptor antagonist to block insulin action in pregnant mice is reversible, making this model more representative of gestational diabetes (194).

Genetic rodent models of diabetic pregnancy include nonobese diabetic (NOD), Ins2 Akita, New Zealand Obese mice, specific strains of C57BL/6 mice, and Zucker diabetic fatty, GK, BB, and Tet29 rats (185, 186). NOD mice undergo spontaneous β cell destruction between 12 and 30 wk of age resulting in severe hyperglycemia (195). Because NOD mice become diabetic at different ages, this model has the advantage of providing age-matched nondiabetic controls. Heterozygous Ins2 Akita possess a mutated Ins2 gene and are a diabetic pregnancy model with a range of glucose levels allowing evaluation of both mild and severe maternal hyperglycemia (196).

Few rodent genetic models mimic human gestational diabetes. Db/+ mice, heterozygous for a leptin receptor mutation, do not initially become diabetic. They develop moderate glucose intolerance during pregnancy that resolves after parturition, making them a good model for gestational diabetes (185). A less well-characterized gestational diabetic model, C57BL/6N mice, also acquires glucose intolerance during pregnancy (186).

Larger animal models such as nonhuman primates, sheep, pigs, and dogs have also been used as with surgically and chemically induced diabetes; however, high cost and ethical concerns limit studies (193) (see Table 1). Other models of diabetes during pregnancy, such as C57BL/6J mice, can be achieved through dietary manipulation, such as high-fat diet (186).

Myocardial Ischemia in Pregnancy

The drastic changes in the cardiovascular system during pregnancy are usually well tolerated in healthy individuals. However, some healthy individuals can develop severe adverse cardiac events in late pregnancy (197) such as myocardial infarction (MI) (198, 199) which has increased recently because of rising maternal age, smoking, chronic hypertension, obesity, and diabetes (200–203). Pregnancy-associated MI accounts for over 20% of maternal cardiac deaths (199). The relative risk of MI during late pregnancy is approximately three- to fourfold higher than the rates in age-matched nonpregnant individuals (203). MI during late pregnancy and the peripartum period carries a markedly worse prognosis than in nonpregnant people for unclear reasons (202, 204). Eghbali’s (205, 206) laboratory successfully modeled this observation in rodents using both in vivo and ex vivo models of ischemia-reperfusion injury. In an in vivo model, nonpregnant and late-pregnant rats underwent occlusion of the left anterior descending coronary artery for 45 min, followed by 3 h reperfusion (205). At the end of the reperfusion, Evans blue dye was injected to assure both experimental groups were exposed to the same ischemic risk. This in vivo model showed myocardial infarct size in late pregnant animals was approximately fourfold greater than in nonpregnant animals (205). To examine the translational potential of gene therapy or microRNA delivery, it is necessary to extend the reperfusion duration to 24 h, allowing sufficient time for regulation of the expression of target genes.

In the ex vivo Langendorff perfusion, isolated mouse hearts were subjected to a 20-min ischemic event followed by 40 min of reperfusion (205, 206). This model allows for the real-time measurement of hemodynamic parameters at the baseline, during ischemia, and at the reperfusion. Although, the hemodynamic parameters were similar between nonpregnant and late pregnant before ischemia as assessed by the rate pressure product, an indicator of the oxygen requirements of the heart and measured as the left ventricular developed pressure × heart rate, the functional recovery of the heart was extremely poor in late pregnant mice upon ischemia compared with nonpregnant mice (205). Remarkably, there was partial recovery in all hemodynamic parameters observed in hearts in mice 1 day postpartum, and even more improvements were noted in hearts 7 days postpartum (205). As the first few minutes of reperfusion is very critical for any intervention, this ex vivo model is also very effective for measurements of calcium retention capacity and signaling pathway analysis, in which the duration of reperfusion should be reduced to 5–10 min (205, 206).

Hypoxia Models

Gestational hypoxia arising from anemia, smoking, sleep apnea, and residence at high altitude (HA > 8,250 ft) plays an important role in the etiology of pregnancy complications, including fetal growth restriction (FGR) and preeclampsia (207, 208), and exerts long-term impacts on offspring health (76, 209). Pregnant women living at HA are exposed to chronic hypobaric hypoxia. In Leadville, Colorado (3,100 ft), for instance, the partial pressure of inspired oxygen is ∼33% lower than at sea level (100 mmHg vs. 150 mmHg) (210). HA pregnancy therefore provides a natural in vivo model for understanding the mechanisms by which hypoxia influences placental function and hypoxia-related disorders of pregnancy. HA reduces birth weight (211–213), increases the incidence of FGR, and hypertensive disorders of pregnancy (214–216) as well as maternal susceptibility to cardiovascular dysfunction during and long after pregnancy (217). It has also been shown to reduce the normal rise in uterine artery blood flow (216, 218, 219) and cardiac output (220), and blunt myometrial artery relaxation with exposure to specific vasodilatory agents (221, 222); these effects are reminiscent of the impaired vasodilation and remodeling of uterine vessels, lesser uterine artery blood flow, and placental insufficiency often observed in preeclampsia and FGR (223–227). Placental morphology, function, and transcriptional profiles have also been reported to be affected by HA (228–231). Advantages of the human model include the ability to study the effects of hypoxia without the confounding effect of underlying disease and the enhanced potential for translational impact. Differences between ancestry groups with respect to the magnitude by which high altitude reduces fetal growth (212, 216, 232) are also valuable for determining the role of genetics for hypoxia-related disorders of pregnancy. One limitation is the inability to serially sample the placenta and umbilical or uterine vessels across gestation. This prohibits access to the samples required for directly establishing the impact of gestational hypoxia on transplacental oxygenation or nutrient exchange during pregnancy and determining whether such effects precede or follow the onset of fetal growth restriction or maternal vascular dysfunction.

Various animal models have been developed to understand the pathophysiology of human pregnancy complications associated with hypoxia and to explore the potential intervention of these disorders. Commonly used animal models of gestational hypoxia include rodents and sheep, achieved by exposing the animals to normobaric hypoxia ranging from 10% to 14% O₂ at approximately sea level or to hypobaric hypoxia at high altitude during different states of gestation (7, 62, 76, 233–240). The chick embryo is also frequently used to study impacts of chronic hypoxia on fetal growth and the developing cardiovascular system independent of maternal and placental effects (241). These models have successfully recapitulated many features of preeclampsia and fetal growth restriction and demonstrated desirable efficacy of trial drugs (7, 79, 235, 242, 243). To more precisely mimic human conditions, factors such as placenta type, gestational duration, and metabolism rate should be taken into consideration when selecting an animal model. For example, the guinea pig is preferred if placentation is studied as it shares with humans the hemochorial placental type (77, 244). Knockout (KO) mice are useful tools to link the function of a gene and its pathway to the pathophysiology of pregnancy complications associated with gestational hypoxia (144, 145). Moreover, the sheep model is the choice if fetal instrumentation and physiology recording are required (76, 207).

Peripartum Cardiomyopathy

Peripartum cardiomyopathy (PPCM) is an idiopathic systolic dysfunction that occurs in the last trimester of pregnancy and during the first 6 to 12 mo postpartum. Although rare, the incidence of PPCM is increasing, currently impacting around 1:1,000 to 1:3,000 births, each year in the United States (245). Several risk factors and comorbidities contribute to the development of PPCM, such as advanced maternal age, high blood pressure, multiple gestations, and lactation (245). Significant racial disparities exist, with black and African-American women being disproportionately impacted by PPCM.

The etiology of PPCM is poorly understood. Currently, mutations in ∼21 genes have been linked with PPCM, with 15% of PPCM arising from mutations in several sarcomeric proteins, including titin, myosin heavy chain 6, and myosin heavy chain 7 (176). However, it is important to note that mouse models harboring these mutations have either not been examined in the context of pregnancy or do not recapitulate the PPCM phenotype in mice. This also means that around 85% of PPCM originates from nongenetic causes. Some studies suggest increased risk of PPCM following viral infection and exaggerated immune responses. However, much of what has been uncovered regarding the underlying mechanisms contributing to PPCM indicates key roles for hormonal changes (i.e., prolactin and its cleavage) and angiogenic imbalance (i.e., antiangiogenic form of prolactin and increased circulating levels of sFlt1) (246, 247).

Around 12 mouse models exist that recapitulate many of the key characteristics of PPCM (176) (see Table 2) (246–256). Of these mouse models, the cardiomyocyte-specific signal transducer and activator of transcription 3 (STAT3) (251) and cardiomyocyte-specific peroxisome proliferator-activated receptor-γ coactivator 1-α (PGC1α) KO mice (247) are among the most widely used. These models recapitulate the hormonal and angiogenic imbalance seen in women with PPCM and can be rescued using the prolactin secretion inhibitor, bromocriptine. However, despite their use, both STAT3 and PGC1α have not been shown to change in the hearts of patients with PPCM. In addition, the impact of comorbidities and risk factors known to contribute to PPCM in women, such as aging, multiple gestations, and so forth have not been previously considered in these models. At present, none of the existing models fully recapitulate the PPCM phenotype so caution must be taken when selecting a model for use. Moving forward, additional models are required to further clarify underlying mechanisms contributing to PPCM.

Advanced Maternal Age

Defining advanced maternal age and its impact on pregnancy.

Advanced maternal age pertains to pregnancy in women and females aged 35 yr or older and is associated with an increased risk of complications such as preeclampsia, intrauterine growth restriction, preterm birth, stillbirth, and gestational diabetes mellitus (257–261). Understanding the drivers of increased complications is crucial due to delayed childbearing becoming increasingly common place, necessitating effective clinical management and intervention opportunities. Moreover, there is emerging concern about the long-term cardiovascular health of postpartum mothers and offspring born to older mothers (257, 262–265).

Impact of aging on cardiovascular adaptations.

Increased susceptibility to pregnancy-related complications in women of advanced maternal age may, in part, stem from inadequate cardiovascular adaptations during pregnancy. Aging effects encompass reduced vascular compliance, impaired endothelium-dependent function, and heightened activity of the sympathetic nervous system, contributing to increased arterial stiffness and hypertension (257, 266, 267). Evidence from rodent models and in women indicates that advanced maternal age impacts vascular function in pregnancy (257, 268–271).

Methodological recommendations for studying cardiovascular function at advanced maternal age.

For investigating cardiovascular adaptations in advanced maternal age, consider the age of animal models (and the impact that will have on fertility) and age-associated body weight changes. In both human and animal studies, select the appropriate vascular bed for investigation (268–273).

To measure cardiovascular function in advanced maternal age, a variety of commonly used techniques can be used:

• 1) Doppler ultrasound: a noninvasive imaging technique to assess blood flow velocity in maternal and fetal vessels (257). This requires the use of anesthesia (most commonly isoflurane) that can be considered a confounding factor. However, both control and experimental animals should be treated the same way, and care should be used to minimize the impact by paying close attention the animal’s body temperature (often a small space heater is required in addition to a heated animal handling table), heart rate, and breathing rate, and adjusting the anesthesia as required (it is therefore advantageous to use an inhaled anesthetic such as isoflurane, where adjustments can be made to the level of anesthesia).
• 2) Maternal blood pressure measurements: performed noninvasively using a sphygmomanometer or continuously through telemetry devices in animal studies (7, 257, 268, 271, 274). Telemetry devices are the gold standard because they do not require the animal to be restrained, but their use in female mice (especially young mice) can be challenging because of the size of the telemetric transmitter that needs to be implanted; a person highly skilled in such small animal surgery will have greater success. Ideally, surgery should be performed before pregnancy to prevent additional impacts on the pregnancy. Using a sphygmomanometer is quite simple but can cause stress to the animals since they are restrained in acrylic holders, which can add variability to the data, and could impact the pregnancy outcomes. An appropriately sized restraint should be selected, taking into account that the animal might be pregnant, and a larger restraint might be required in later pregnancy. Intra-arterial catheters can also be used for blood pressure measurements in both conscious and anesthetized animals.
• 3) Isolated vascular studies: In animal models and tissue from human pregnancy, small blood vessels can be isolated to study their contractile and relaxation responses, as well as arterial stiffness using wire myographs and pressure arteriographs. Using pressure arteriographs, ex vivo vascular compliance, myogenic tone, and flow-mediated vasodilation can also be assessed. One potential confounding variable in pregnancy research is the lack of capacity to hold preconstricted vascular tone in some vascular beds to assess vasorelaxation. In addition, for the pressure arteriography, selection of vessels without branches is essential.
• 4) 
  Cardiac function:

  • 1) 
    echocardiography offers insights into cardiac structure and function and has been measured in offspring of advanced maternal age dams (263) and can also be measured in the dam. Anesthesia and body temperature, heart and respiratory rate considerations outlined in the Doppler ultrasound section are also relevant for echocardiography.
  • 2) 
    Continuous, noninvasive measurements of cardiac output can be estimated using the bioreactance method. This method involves the provision of alternating high-frequency electrical current that traverses the thoracic cavity, in which four surface electrodes collect information on relative phase shifts of current, which are used to calculate blood flow (see Ref. 275 for additional information on the bioreactance method). This method can also be used to collect information on stroke volume, heart rate, peripheral vascular resistance, and mean arterial pressure (274).
  • 3) 
    Isolated working heart studies: Assess capacity to recover from ischemia-reperfusion injury (257) and offers the ability to study basal as well as simulated cardiac function. Preliminary experiments may need to be performed to determine the optimal period for ischemia, especially in older animals. A previous study in offspring from advanced maternal age dams reduced the global ischemia period to 10 min as none of the hearts were recovering with the standard 20 min ischemia (262).

Understanding the impact of advanced maternal age on pregnancy outcomes, including cardiovascular adaptations, warrants further exploration. Advanced cardiovascular measurement techniques may enhance the understanding and optimize care for older expectant mothers.

CONSIDERATIONS FOR RIGOROUS EXPERIMENTAL DESIGN

Timed Pregnancy Studies

Timed pregnancy studies afford investigators the opportunity to directly examine time points during pregnancy (see next section for time-point selection), which is particularly advantageous if a period of vulnerability has been identified for examination of physiological experimental endpoints or specific timed inventions. For timed murine pregnancy studies, estrus should first be stimulated in female mice using the concepts put forward by Whitten (276), whereby soiled bedding containing male pheromones is placed into the female cage to stimulate estrus. Because the female estrus cycle in rodents is 4–5 days (277), male studs should be paired with female mice until either the presence of a copulatory plug or if time-points postpartum are of greater interest, until one cycle has elapsed. Male mice should be paired with females in the early evening and removed and single housed the next morning and plugs checked early and daily because sometimes these can fall out or dissolve. Once a copulatory plug is identified this signifies gestational day 0.5. A vaginal swab to confirm the presence of sperm in the vagina can be performed as per Caligioni’s (277) and Byers et al.’s (278) study, as an additional confirmation of copulation; however, it is important to note that this is somewhat invasive and can result in stress. Because the presence of a copulatory plug and the use of vaginal swabs only indicates copulation has occurred rather than confirming pregnancy, regular body weight measurements can be obtained as a secondary confirmation of pregnancy. This is advantageous in strains with characteristic weight gain during pregnancy, such as FVB/NJ mice, which gain ∼3–4 g in the first 8 days following the identification of a copulatory plug (279). Although ultrasound is a commonly used methodology to assess the human pregnancy stage and has been used to determine the pregnancy stage in rodents (280), this is less commonly used in the field for pregnancy determination in animal models.

Parity, Gravidity, and Time Points

The use of preclinical animal models to investigate physiological and pathophysiological adaptations during pregnancy and postpartum is well accepted. Yet, several factors should be considered when designing preclinical models for studying healthy pregnancies, pregnancies with complications such as HDP, and postpartum physiology. Key considerations include timing of interventions and end point measurements. With studies focused on the assessment of cardiovascular physiology during an uncomplicated pregnancy, selecting time points for investigation is often impacted by the type of animal and the strain chosen. For example, the commonly used mouse strains, FVB/NJ and C57BL/6, although both mouse strains have different gestational lengths (281) so care must be taken when selecting time points. In addition, time-point selection will likely also be governed by the primary end point of the study and the time frame during pregnancy that it is likely to be changed. If the latter is unknown, investigators should select time points across pregnancy and postpartum.

Key considerations should also be made for studies focused on pregnancy-associated diseases. For example, many experimental models of HDP, such as preeclampsia, include surgical (282), pharmacological (54, 165, 283, 284), dietary (165), environmental (7), or immunological (285–287) approaches (Table 2) that are initiated in mid to late gestation, mimicking midgestational manifestation and diagnosis of preeclampsia. Genetic approaches to replicate HDP begin before pregnancy (134, 288) allowing for investigation of early gestational changes before the development of the hypertensive phenotype (134) (Table 2). Several preclinical models develop hypertension during pregnancy mimicking superimposed preeclampsia (169, 289, 290) where preeclampsia symptoms arise in the presence of preexisting hypertension. Thus, many experimental models are limited to study of late gestational events whereas others allow for study of mechanisms related to early gestational events but often in the presence of superimposed preeclampsia (169, 289, 290). Exceptions include the rodent model of preeclampsia induced via gestational infusion of arginine vasopressin that is associated with alterations in placental morphology in mid gestation (158) and the rodent model of preeclampsia achieved through midgestational treatment with leptin (54).

In rodents, multiparity can influence maternal body composition and inflammation, contributory mechanisms in the etiology of preeclampsia (291). In terms of long-term postpartum health, clinical studies report that multiparity is associated with decreased cardiovascular health scores (292), decreased hippocampal volumes, and worse memory scores in women with five or more deliveries (293). In rodent models of stroke, multiparity results in smaller infarcts and reduced neuroinflammation despite having features of increased metabolic risks (294). It is also challenging to separate multiparity from aging. Advanced maternal age is associated with adapted responses resulting in beneficial effects on maternal blood pressure and vascular function in late gestation in dams that deliver pups at term (268), which is unique to rodents. Vascular function is worsened postpartum in dams with advanced age, in particular those that do not deliver (264). Thus, use of first-time pregnant rodents may reduce biological variability for studies with a focus on pregnancy and postpartum outcomes (165). An additional caveat to the use of preclinical models involves placental structure that can differ based on species, human compared with rat (295) (See Table 1). However, rodents, which have a gestational period of ∼3 wk, are extensively studied and numerous investigators report that maternal renal and systemic hemodynamic changes in the rat mimic the time line in the human (296–298) highlighting that benefit for rodent models.

Pharmacological Properties of Drugs during Pregnancy

Many preexisting conditions demand continuation of maternal pharmacological treatments throughout pregnancy and postpartum. Mothers must also be treated for diseases they acquire during pregnancy or postpartum and maternal vaccination may occur during the perinatal period. In addition, pregnancy-induced complications require the use of maternal therapeutics. Experimental animal models provide a platform for testing drug efficacy, side effects, and mechanisms of action that can aid in safe and efficacious implementation of an intervention.

Pregnancy and the postpartum period are distinct physiological states, involving multisystem adaptations that significantly affect pharmacological properties of drugs (299, 300). Little is known about the interaction between the early postpartum period and drug properties. Furthermore, there is a paucity of research on the effects of maternal therapeutics during pregnancy on long-term maternal health outcomes. The implementation of a maternal treatment during pregnancy or postpartum is a dilemma because while a treatment may reduce maternal symptoms, it can impair offspring development via placental transfer during gestation or milk transfer during lactation (301). On the other hand, this exact same principle of maternal-fetal communication is the basis of the systemic maternal treatments to improve fetal outcomes.

Key considerations for drug properties and their efficacy on pregnancy-associated physiology.

• 1) Increase in the volume of distribution of hydrophilic substrates (302–305). Impact: dosing of hydrophilic drugs.
• 2) Serum albumin is reduced, and compensatory respiratory alkalosis develops. Impact: Drug bioactivity and changes in blood acid/base influence breathing patterns, which can impact exposure conditions for inhalation studies (see Environmental Exposures).
• 3) Increase in glomerular filtration rate. Impact: elimination rates of drugs metabolized by the kidneys and their half-life.
• 4) Gastric mobility is reduced during pregnancy. Impact: bioavailability of orally administered medication and/or alter time for ingestion-related exposures (see Environmental Exposures).
• 5) Drug metabolizing enzymes change during pregnancy and sex steroid concentrations are increased. Impact: drug metabolism.
• 6) Nausea and vomiting are common features during pregnancy. Impact: drug absorption may be reduced. This may not be an issue in preclinical rodent models in initial testing of drug efficacy they lack the emetic reflex (306).

Lactation and Breastfeeding

Lactation and breastfeeding are associated with maternal cardiovascular health benefits (307). Lactation is the period that starts right after delivery and ends once the mother’s body has ceased to produce milk. Breastfeeding or nursing is when a mother feeds her baby using breasts. Infants could also be fed milk that mothers have expressed or pumped and saved in a bottle. Thus, because of the different ways of breastfeeding, the infant defines maternal physiology during this period.

The duration of lactation differs among the species and is dependent on the maturity of the baby at birth. In precocial species, newborns are considerably independent, their behavior and morphology are more mature, and they can survive with limited support (308). Babies unable to move independently and requiring extensive parental care and food provision are born to altricial species (308). Rodents, extensively used in pregnancy and lactation studies, are altricial species: their newborns are ready for life by themselves at 3 wk of age. Thus, rodent dams’ lactate for 3 wk (309). Sheep and nonhuman primates are precocial species. Artificially, sheep are weaned at 1–2 days of age. However, wild sheep can nurse for up to 12 mo (310). Lactation in nonhuman primates lasts 71–551 days, depending on the species. Nonhuman primates that undergo a menstrual cycle experience a prolonged lactation-induced anovulatory period: lactation amenorrhea, which needs to be taken into consideration when designing experimental studies (311). Finally, humans are the altricial species since the newborns are helpless at birth and require much parental care (312). In humans, the American Academy of Pediatrics and the World Health Organization recommend that newborns be exclusively breastfed until 6 mo of age (313).

Several experimental conditions are specific only to the lactation period, significantly impacting the study outcomes. Study groups in women should be defined according to lactation and breastfeeding status: exclusive breastfeeding, mechanical breast pumping, bottle feeding, and any combination of the aforementioned.

Several hormones are released only during lactation and breastfeeding and are known to affect the maternal cardiovascular system. Among them, oxytocin is the most studied. In addition to its role in parturition, lactation, and modulating social behavior, oxytocin lowers blood pressure, heart rate, and obesity (314). Moreover, breastfeeding mothers have higher oxytocin levels in plasma than formula feeders, and it was associated with lower blood pressure and heart rate. Breastfeeding induces immediate and short-lasting (20 min) oxytocin release (315). Oxytocin is released in pulses that become longer as lactation proceeds (315). Thus, experiments should be timed accordingly and under the same conditions for each mother: during breastfeeding, 10 min (as an example) before and after breastfeeding.

Other confounders or study group determinants for studies in women are mode of delivery (spontaneous or C-section); duration of pregnancy (term or preterm); pregnancy complications (gestational diabetes, preeclampsia); nutrition; smoking status; prior history of breastfeeding; psychosocial stress; and medications. All these conditions influence lactation, breastfeeding, and oxytocin production and release, subsequently influencing the cardiovascular system (316). For animal studies, one must remember that if the mother is taken away from her pups for any period, there is always a risk that the dam might not recognize her offspring and refuse to nurse them leading to neonatal mortality (317).

ACCOUNTING FOR ENVIRONMENTAL FACTORS AND MODIFIABLE STRESSORS

Diet and Obesity

Diet choice.

There are several key considerations to make when establishing models of dietary manipulation during pregnancy. In terms of modeling Western-style diets, there are two main approaches 1) using high-fat diets (318) and 2) using high-fat high-simple carbohydrate obesogenic diets (319). Rodents are generally good at regulating food intake in response to a high-fat diet, and therefore often do not gain excess weight when fed ad libitum such a diet (unless the feeding regimen is over an extended period). In contrast, diets rich in simple carbohydrates generally override natural satiety signals. Therefore, feeding of such diets ad libitum leads to increased food intake, weight gain, and increased adiposity and other aspects of metabolic dysfunction.

Standardization of maternal physiology.

There may be interanimal variation in response to diet, for example in nonhuman primates it has been reported that in response to a high-fat diet, females can be divided into responders (i.e., gain excess weight) or nonresponders (i.e., do not gain excess weight) (320). It is therefore important to record maternal weight (and ideally body composition) longitudinally. In addition, longitudinal maternal blood sampling to allow measurement of hormones such as insulin and glucocorticoids and metabolites such as glucose is advisable. This will allow correlation analysis between maternal physiology and cardiovascular phenotyping and ultimately if interventions are tested, allow causal relationships to be identified.

Minimizing external factors in offspring.

When addressing maternal effects on offspring, the mother represents the statistical unit. Therefore, when carrying out studies in a litter-bearing species such as mice, this can be capitalized on, and littermate siblings used to address the effect of offspring sex or diet (e.g., by weaning half of the litter on to a chow diet and the other half on an obesogenic diet). Feeding status (fed or fasting) should also be standardized as should the time of day when experiments are performed (noting that rodents are nocturnal so will feed during the dark phase). Offspring measurements should also be carried out longitudinally, with the timing of phenotyping (e.g., before/after sexual maturity, before effects of altered diet results in changes in body composition, etc.) also considered.

Fitness and Exercise Considerations in Human Pregnancy

Current guidelines recommend engaging in at least 150 min of moderate-intensity physical activity on three or more days of the week during pregnancy to derive clinically meaningful health benefits (321, 322). These include a ∼40% reduction in the risk of developing preeclampsia, gestational hypertension, gestational diabetes, and a reduced risk of depression by nearly 70% (323, 324). An individualized and gradual resumption of physical activity is encouraged following delivery, with appropriate rehabilitation of pelvic floor dysfunction, musculoskeletal pain, and other symptoms that may limit exercise to support physical and mental health (325, 326).

Exercise training is associated with cardiovascular health benefits during pregnancy including a small increase in fitness (↑V̇o_2max ∼2.8 mL/kg/min) and reduced resting heart rate and blood pressure (327). Exercise training is anticipated to improve vascular health (e.g., endothelial function and arterial stiffness) during and following pregnancy, especially in those who have underlying cardiovascular dysfunction (328). Given the impact of physical activity and exercise on cardiovascular health, an objective (i.e., accelerometer) or subjective (i.e., questionnaire) assessment of physical activity is recommended in conjunction with cardiovascular assessments in pregnancy and postpartum.

Emerging data from women have identified an association between cardiorespiratory fitness and HDP (329). Before fitness testing, it is essential to screen for medical conditions where exercise is not advised (e.g., severe preeclampsia). The “Get Active Questionnaire for Pregnancy” was developed for this purpose (https://csep.ca/2021/05/27/get-active-questionnaire-for-pregnancy). Fitness testing to maximal intensities is reserved for experienced researchers who perform concurrent fetal monitoring. Submaximal exercise tests where the peak exercise intensity falls within the recommendations of current guidelines are considered to be low risk in those without contraindications (330), and fetal monitoring is optional.

Key considerations for exercise testing and training in pregnant human subjects.

• 1) Maintain adequate nutrition and hydration before, during, and after exercise.
• 2) A nutritious snack ∼1 h before exercise is recommended to maintain maternal glucose concentrations, especially before longer duration and or higher intensity activities.
• 3) Avoid exercise in excessive heat, especially with high humidity.
• 4) Complete a preexercise assessment of resting heart rate, blood pressure, and glucose so that these values can be interpreted within the context of pregnancy/postpartum physiological adaptations and identify undiagnosed abnormal values.

Detailed recommendations and considerations on exercise testing during pregnancy have been previously described by Wowdzia and Davenport (331). Guidelines for cardiorespiratory exercise testing specific to the postpartum period have not been developed. Adult testing guidelines (e.g., American College of Sports Medicine) may be used following an individualized and gradual resumption of moderate-to-vigorous intensity physical activity after childbirth. It is essential to be aware of how the unique considerations of the postpartum period (e.g., type of delivery and healing times, impact of sleep deprivation and breastfeeding, complications persisting from pregnancy or newly developed postpartum) may impact the results and or applicability of the test. Detailed recommendations on appropriate milestones and progression toward moderate-to-vigorous intensity physical activity are outlined by Christopher et al. (332).

Environmental Exposures

Environmental exposures generally fall into three categories, inhalation, ingested, and dermal (see Table 2) (333–343). When designing any exposure experiment it is important to determine the media for the contaminate (e.g., air, liquid, or solid media) and route of exposure (e.g., inhalation, ingested, and dermal) that is most relevant to the environment and biological outcome under investigation. It is always critical to understand and report the “exposure dose.” Dosing related to liquid and solid media often seems more intuitive and more easily assessed with standard laboratory equipment, whereas dosing from inhalation exposures typically requires more specialized instruments and expertise and may be more complex depending on the desired measurement. It is important to emphasize that solely describing the exposure conditions (e.g., the number of hours of exposure) is not reporting the “dose” of the exposure. This error is most often made with inhalation exposures when reporting only the number of puffs or time in the chamber rather than reporting the size and/or concentration of the airborne media (i.e., particulate matter, PM). Depending on the experimental question or conditions, the chemical composition of PM (or gaseous phase) may also be relevant, and will require additional expertise and equipment to obtain, interpret, and determine biological relevance.

Regardless of media or route, there are different approaches that can be used when assessing dose (344). The potential dose, which is the amount of contaminant brought into or on the body, is commonly used for inhalation exposures, and in particular with pregnancy exposures when the objectives require study of progeny after birth. The “potential dose” gives a measure of the amount of contaminate or PM that is brought into the mouth/nose or on the skin. But it is important to realize that this measure does not account for what is absorbed.

The applied dose is the amount of contaminant entering the body at the absorption barrier (e.g., reaching the respiratory tract, gastrointestinal tract, or absorbed by the skin). Since most inhalation exposures (including during pregnancy) are attempting to mimic or replicate a deviant environmental condition (i.e., air pollution, smoking, occupational environment, etc.) reporting the “potential” or “applied dose” is the minimal standard to provide rigor and reproducibility. For inhalation exposures, the “applied dose” (e.g., lung burden) can be estimated from the “potential dose” using several validated mathematical modeling approaches, [such as, multiple path particle dosimetry (MPPD), computational fluid dynamics (CFD), etc.]. Interested readers are directed to the following references for more details on the modeling approaches (345–347).

Finally, measuring the internal dose or biologically effective dose will typically provide better correlation and understanding of toxin-related exposures, but may not always possible or practical to obtain. The “internal dose” is the amount of contaminant that gets past the absorption boundary and into the blood, whereas the “biologically effective dose” is the amount of contaminant that interacts with the internal target tissue or organ. Whether this is possible/practical to measure will depend on the contaminate and/or study context. For example, the “internal dose” is often better suited to single toxin effects. In contrast, complex PM (e.g., cigarette smoke, e-cig aerosol, urban wildland fires, etc.) are comprised of many contaminates/toxins, therefore it is often difficult (if not impossible) to attribute biological outcomes to any single toxin. In the context of pregnancy exposure, the uteroplacental interface is the absorptive barrier that is relevant to the fetal dose, so the study design (e.g., maternal vs. fetal/offspring focus) and contaminate will largely dictate whether it is possible to assess these “doses.”

Regardless of the study design, studies that do not report a “dose” parameter and only reference the exposure condition, do not allow other investigators (or any subsequent reader) to know the real conditions of the treatment applied, and thus provide limited value for translation and/or comparison to other studies in the field. Also, as previously highlighted (see Pharmacological Properties of Drugs during Pregnancy), it will be important to consider and understand how any physiological changes occurring during pregnancy might influence or affect the dosimetry of environment exposures. For example, pregnancy-related increases in maternal tidal volume (299) will alter normal breathing patterns in a manner that can directly influence the location and amount of PM distributed in the lung (348). Likewise, changes in gastrointestinal mobility, blood acid/base status, and metabolic enzymatic activity are altered during pregnancy (299) and also need to be considered in the context of any relevant environment exposure condition (e.g., inhalation, ingestion, and dermal).

CONCLUSIONS

Recent epidemiological studies highlight the heightened contribution of cardiovascular complications to maternal mortality and morbidity worldwide. This evidence has triggered an increase in research efforts on maternal cardiovascular health during pregnancy and postpartum. In the pursuit of new knowledge and further understanding of maternal cardiac and vascular physiology, it is imperative that the scientific community follows best and standardized practices to conduct research studies in pregnant people and experimental models. The American Journal of Physiology-Heart and Circulatory has previously published guideline articles with step-by-step protocols of how to study cardiac and vascular function in nonpregnant humans and experimental models (for reference, see Table 3) (349–355). Pregnancy is a physiological state with unique characteristics of maternal cardiovascular function that dynamically change throughout gestation and postpartum. Hence, our objective was to compliment and extend those previous publications by outlining pregnancy physiology-specific considerations that could lead to appropriate modifications and applications of established experimental protocols. Specifically, we provided theoretical information and practical tips of what factors to consider when studying cardiovascular physiology in pregnant human subjects, experimental animals, or in a dish. Furthermore, we discussed strengths and limitations of various commonly used models of pregnancy-associated diseases and risk factors. Finally, we addressed methodological considerations for the design of rigorous experimental studies in pregnant people and experimental animals. Using these guidelines and underlined best practices, scientists will be empowered to conduct rigorous studies that result in interpretable and translatable outcomes that benefit human maternal health during pregnancy and postpartum.

Table 3.

AJP-Heart and Circulatory references on guidelines for the assessment of cardiac and vascular physiology in nonpregnant experimental models, ex vivo preparations, and cell culture models

Title	PMID Citation	Year
Guidelines for assessment of cardiac electrophysiology and arrhythmias in small animals	36269644	2022
Guidelines for the measurement of vascular function and structure in isolated arteries and veins	33989082	2021
Guidelines for evaluating myocardial cell death	31418596	2019
Guidelines for authors and reviewers on antibody use in physiology studies	29351459	2018
Guidelines for measuring cardiac physiology in mice	29351456	2018
Guidelines for experimental models of myocardial ischemia and infarction	29351451	2018
Statistical considerations in reporting cardiovascular research	30028200	2018

GRANTS

This work was funded by the following sources: National Institutes of Health Grants R01HL163003 (to H.E.C.), R01HL147844 (to H.E.C.), R01HL163818 (to H.E.C.), R01HL143459 (to B.T.A.), P20GM104357 (to B.T.A.), P20GM121334 (to B.T.A.), R01HL159865 (to M.E.), R01HL131182 (to M.E.), R01HL138181 (to C.G.J.), R01HD088590-06 (to C.G.J.), R21HD111908 (to C.G.J.), P20GM103499 (to H.A.L.), F31ES034646-01 (to I.M.O.), T32ES032920 (to I.M.O.), 1R01HL150472-01A (to E.B.P.), R56HL159447-01 (to J.P.W.), R01HL149608 (to L.Z.), R01HL169157 (to L.Z.), R01HL155295 (to L.Z.), P01HD083132 (to L.Z.), R21NS103017 (to L.Z.), R01HL146562 (to S.G.), and R01HL146562-04S1 (to S.G.); Jewish Heritage fund for excellence faculty support grant (to H.E.C.); National Health and Medical Research Council Ideas Grant APP2002129 (to A.S.C.); National Heart Foundation future leader fellowship (to A.S.C.); Christenson Professor in Active Healthy Living (to M.H.D.); Canadian Institutes of Health Research Foundation grant (to S.T.D.); Distinguished University Professor (to S.T.D.); British Heart Foundation (to D.A.G.) and Grant RG/17/12/33167 (to S.E.O.); Medical Research Council UK (to D.A.G.) and Medical Research Council Grant MC_00014/4 (to S.E.O.); Biotechnology and Biological Sciences Research Council (to D.A.G.); Lister Institute (to D.A.G.), Royal Society (to D.A.G.); Dutch Heart Foundation Dekker Grant 2021T017 (to M.F.H.); U.S. Department of Defense (to C.G.J.); American Heart Association Grant 20CSA35320107 (to I.M.O.), and West Virginia University School of Medicine Synergy Grant Award (to I.M.O.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

H.E.C., B.T.A., A.S.C., M.H.D., S.T.D., M.E., D.A.G., M.F.H., C.G.J., H.A.L., I.M.O., S.E.O., E.B.P., J.P.W., L.Z., and S.G. conceived and designed research; H.E.C., D.A.G., M.F.H. prepared figures; H.E.C., B.T.A., A.S.C., M.H.D., S.T.D., M.E., D.A.G., M.F.H., C.G.J., H.A.L., I.M.O., S.E.O., E.B.P., J.P.W., L.Z., and S.G. drafted manuscript; H.E.C., B.T.A., A.S.C., M.H.D., S.T.D., M.E., D.A.G., M.F.H., C.G.J., H.A.L., I.M.O., S.E.O., E.B.P., J.P.W., L.Z., and S.G. edited and revised manuscript; H.E.C., B.T.A., A.S.C., M.H.D., S.T.D., M.E., D.A.G., M.F.H., C.G.J., H.A.L., I.M.O., S.E.O., E.B.P., J.P.W., L.Z., and S.G. approved final version of manuscript.