Physiology and Pathophysiology

Published on 13/02/2015 by admin

Filed under Cardiothoracic Surgery

Last modified 13/02/2015

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 4088 times

Chapter 1 Physiology and Pathophysiology

In this chapter the physiologic principles relating to conditions and treatments encountered in the cardiothoracic intensive care unit (ICU) are reviewed. Most of the topics relate to the cardiovascular system, but some coverage of the respiratory and renal systems is also included. The physiology of acid/base homeostasis and water and electrolyte balance is covered in Chapters 31 and Chapter 32, respectively.


Resting Membrane Potential

Cardiac muscle cells (myocytes), like all cells in the body, have a charge separation across their cell membranes. For resting myocytes in diastole, the charge is about −90 mV, with the inside of the membrane being negatively charged and the outside positively charged.

Ions diffuse across a cell membrane on the basis of their concentration gradient, the transmembrane electrical gradient, and the permeability of the membrane to the ion. For sodium and potassium, this concentration gradient is established by the sodium-potassium adenosine triphosphate membrane pump (Na+/K+ ATPase), which is a membrane-bound protein that uses ATP to actively transport sodium out of the cell and potassium into the cell. Potassium, which is at a high concentration inside the cell (140 mmol/l) and at a low concentration outside the cell (4 mmol/l), has a concentration gradient that favors diffusion out of the cell, but this is countered by an electrical gradient that favors diffusion into the cell. The equilibrium potential of an ion is the membrane potential at which the concentration and electrical gradients are equal and opposite, and no net diffusion occurs. The equilibrium potential (EQ) is quantified by the Nernst equation, which for potassium is:

(1-1) image

where [K+]i is the potassium concentration inside the cell and [K+]o is the potassium concentration outside the cell. The EQ for potassium is −94 mV. The EQ for sodium ([Na+]i = 15 mmol/L, [Na+]o = 150 mmol/L) is + 61 mV. Cells maintain a transmembrane concentration gradient for a number of ions. The EQs for individual ions, along with the permeability of the membrane to the ions, determine the resting membrane potential.

In the resting cardiac myocyte, permeability to potassium is many times greater than that to sodium or chloride. Thus, potassium is the major determinant of the resting membrane potential, and the EQ for potassium (−94 mV) is close to the resting membrane potential (−90 mV). The small difference between the resting membrane potential and the EQ for potassium is due to the diffusion of sodium into the cell, which draws the membrane toward its EQ (+61 mV), although, because of the low permeability to sodium, this contribution is small (+8 mV). The Na+/K+ ATPase also contributes a small amount (−4 mV) to the resting potential because it transports three sodium ions outward in exchange for two potassium ions inward, resulting in the steady loss of positive charge from the cell.

Cardiac Action Potential

An increase in the membrane potential (i.e., the inside becomes less negative) is called depolarization; a decrease in the membrane potential (i.e., the inside becomes more negative) is called repolarization. Action potentials are cycles of depolarization and repolarization and are the result of minute ion fluxes across the membrane due to changes in the permeability of the membrane to different ions. Alterations in membrane ion permeability are achieved by the opening and closing of transmembrane ion channels. Many of the channels are voltage gated, which means that they open when the membrane potential reaches a certain value.

Two basic types of cardiac action potential are recognized: fast action potentials (Fig. 1-1), which occur predominantly in atrial and ventricular myocytes, and slow action potentials (Fig. 1-2), which occur in pacemaker cells of the sinoatrial (SA) and atrioventricular (AV) nodes.

Action Potential in Cardiac Muscle Cells

Cardiac myocytes are electrically coupled to surrounding myocytes by gap junctions, allowing action potentials to spread through the heart in a coordinated manner. When an activation wavefront (wave of depolarization) approaches a myocyte, the membrane potential, initially at resting potential, becomes less negative. At a potential of about −70 mV (the “threshold”), fast sodium channels open, resulting in the rapid influx of sodium ions into the cell (depolarizing current), causing a sudden increase in the membrane potential to about +30 mV. This represents phase 0, or the depolarization phase, of the action potential. Phase 1 represents a slight repolarization in which the fast sodium channels close and potassium diffuses out of the cell. Phase 2 is the plateau phase of the action potential during which calcium channels open, allowing an inward (depolarizing) calcium current to balance the outward (repolarizing) potassium current. Both sets of channels slowly close during the plateau phase, and the inward and outward currents remain in approximate balance. Phase 3 is the repolarization phase of the action potential, in which potassium permeability increases, and the outward going potassium current repolarizes the membrane. Phase 4 commences with the restoration of the resting potential and finishes with the next activation wave front.

During phase 2 and the early part of phase 3 the membrane is completely resistant to further depolarization (the absolute refractory period). During the later part of phase 3, the membrane can undergo depolarization only in response to a supranormal stimulus (the relative refractory period).

A cardiac action potential is transmitted deep within the myocyte by invaginations of the cell membrane known as T tubules. The entry of calcium into the cell during an action potential leads to the release of calcium from the sarcoplasmic reticulum into the cytoplasm of the myocyte, precipitating muscular contraction.

Cardiac Conducting System

The heartbeat is initiated and then conducted through the heart by specialized myocytes collectively known as the pacemaker-conduction system. Under normal circumstances the heartbeat originates from the SA node, which is a collection of modified myocardial cells found at the junction of the superior vena cava and the right atrium. The blood supply to the SA node comes from branches of the right coronary artery. From the SA node, a wave of activation spreads through the atrial muscle to the AV node. The AV node is a collection of myocardial cells at the AV junction on the posterior septal wall of the right atrium, adjacent to the origin of the coronary sinus. The blood supply to the AV node comes from branches of the posterior descending coronary artery. The AV node has two important functions. First, it delays conduction of the impulse to the ventricles; the transit time of the AV node is about 130 ms. This delay allows sufficient time for the atria to contract before the ventricles are activated. Second, the AV node has a long refractory period, which prevents rapid atrial rates (such as occur with atrial flutter) from being transmitted to the ventricles.

From the AV node, the cardiac impulse passes very rapidly over the ventricles. Initially the impulse passes through the His bundle, which divides into the left and right bundle branches. The right bundle branch crosses the right ventricular cavity via the moderator band to the free wall papillary muscle. The left bundle branch almost immediately divides into the anterosuperior fascicle (running to the anterolateral papillary muscle) and the posteroinferior fascicle (running to the posteromedial papillary muscle). From the ventricular septum, the electric impulse passes over the surface of the ventricles via the complex Purkinje system. Rapid conduction within the ventricles allows the ventricular muscle mass to contract almost simultaneously during systole. The pacemaker-conduction system ensures that electric activation of the atria and ventricles is finely coordinated, resulting in effective pumping action by the heart.

Components of the conducting system can become disrupted by disease (e.g., damage to the His bundle by an aortic root abscess) or as a consequence of cardiac surgery. The blood supply to the SA node may be interrupted during a superior septal approach to the atria (commonly used during surgery involving both the mitral and the tricuspid valves). Damage to the AV node may occur as the result of a misplaced suture during tricuspid valve surgery.

Action Potentials within Pacemaker Cells

Cells of the SA and AV nodes have slow action potentials that undergo spontaneous depolarization and are therefore known as pacemaker cells. Under normal circumstances, the rate of spontaneous depolarization of the AV node cells is less than that of the SA node. Thus, the SA node is the dominant pacemaker, and it inhibits the slower pacemakers. This process is known as overdrive suppression.

The action potential within pacemaker cells of the SA and AV nodes is different from that within normal myocytes. Pacemaker cells have a higher resting membrane potential (−60 mV), lack fast sodium channels, and undergo slow, stable depolarization during phase 4 to threshold. This spontaneous depolarization occurs as a consequence of decreased potassium diffusion out of the cell and increased diffusion of calcium and sodium into the cell. Once the membrane reaches threshold (−40 mV), calcium channels open, resulting in a “slow” phase 0 depolarization. Repolarization (phase 3) occurs as a consequence of reduced diffusion of calcium and increased diffusion of potassium. The slope of phase 4 determines the speed with which the membrane reaches threshold and another action potential is initiated. Therefore, the slope of phase 4 determines heart rate.

In addition to the SA and AV nodes, spontaneous depolarization can also occur within myocytes of the conducting system (latent pacemakers). These pacemakers are typically slower than the atrial ones. In the presence of complete AV block, slow ventricular escape rhythms originating from these ventricular pacemakers usually occur. In other conditions (e.g., epicardial pacing or after myocardial infarction) nonspecialized cardiac myocytes can take over the pacemaker function of the heart.

Relationship between the Action Potential and the Electrocardiogram

The electrocardiogram (ECG) is the surface recording of the electrical activity of the heart. (Its interpretation is described in Chapter 8.) The relationship between the action potential and the ECG is shown in Figure 1-1. The P wave corresponds to depolarization of the atria during late ventricular diastole. The PR interval is the time from the onset of atrial activation to the onset of ventricular activation—a significant portion of which is taken up by the delay through the AV node. The QRS complex corresponds to ventricular depolarization. The ST segment spans the time when the ventricular myocytes are in phase 2 (plateau) of their action potentials, and the T wave corresponds to ventricular repolarization.

Control of Heart Rate and Cardiac Conduction

The electric activity in the heart is controlled by the autonomic nervous system and circulating epinephrine. Parasympathetic stimulation via the vagus nerve causes the release of acetylcholine that binds to muscarinic receptors on cells within the SA and AV nodes, leading to an increase in potassium permeability within these cells. Increased potassium permeability hyperpolarizes the cell membrane (meaning the membrane potential becomes more negative) and reduces the slope of phase 4 of the action potential (see Fig. 1-2), thus reducing heart rate and prolonging conduction through the AV node (↑PR interval). Intense vagal stimulation (e.g., during laryngoscopy) can lead to asystole (SA block) or complete heart block (AV block).

Sympathetic stimulation causes the release of norepinephrine, which activates β1 receptors on the cellular membrane. This leads to decreased potassium permeability and increased calcium and sodium permeability, which reduces the extent of repolarization and increases the slope of phase 4 in pacemaker cells (see Fig. 1-2), thus increasing heart rate and shortening the PR interval. Sympathetic nervous system activation also causes increased excitability throughout the entire conducting system.

A completely denervated heart has a resting rate of about 100 beats/min, this being the intrinsic rate of discharge of the SA node. The normal resting heart rate is 60 to 70 beats/min, indicating that parasympathetic tone dominates in the normal heart at rest. Abnormalities of impulse generation and conduction are discussed in Chapter 21.

Cardiac Cycle

The cardiac cycle is divided into ventricular systole (contraction and ejection) and ventricular diastole (relaxation and filling) (Fig. 1-3).


Diastole commences with the closure of the aortic and pulmonary valves. Intraventricular pressure falls but there is very little increase in ventricular volume (isovolumetric relaxation). Once ventricular pressure falls below atrial pressure, the mitral and tricuspid valves open and ventricular filling begins. Initially, the pressure gradient between the atria and the ventricles is high and ventricular filling is rapid (the phase of rapid early filling). Under normal circumstances about 70% of ventricular filling occurs during this phase. As diastole progresses, ventricular pressure rises and the rate of filling slows (the phase of diastasis). The final 25% of filling during ventricular diastole results from atrial contraction (the phase of atrial systole). When the pressure in the ventricles rises above the pressure in the atria the mitral and tricuspid valves close and diastole is complete. Isovolumetric relaxation and the first part of rapid early filling are active, energy-requiring processes.

In various disease states diastolic filling is abnormal. For instance, with mitral stenosis a high proportion of ventricular filling occurs late in diastole. In this circumstance, shortening of diastole due to tachycardia or loss of atrial systole due to the development of atrial fibrillation can cause marked hemodynamic compromise. A similar situation exists when active relaxation is prolonged (e.g., due to myocardial ischemia or left ventricular hypertrophy). Conversely, in some circumstances (e.g., restrictive cardiomyopathy) a greater proportion of diastolic filling occurs early in diastole. In this circumstance, cardiac output may be improved with modest tachycardia. Diastolic dysfunction is discussed in Chapter 20.

Pressure-volume Loops

A useful way of evaluating cardiac function experimentally is by plotting ventricular pressure against ventricular volume throughout the cardiac cycle (Fig. 1-4). Families of pressure-volume loops can be generated under different physiologic conditions. Stroke volume (SV) is the difference between the end-diastolic volume (EDV; see Fig. 1-4, position b) and the end-systolic volume (ESV; see Fig. 1-4, position d). Ejection fraction (EF) is the proportion of the end-diastolic volume that is ejected during systole:

(1-2) image

The area bound by the pressure volume-loop gives myocardial work. Characteristic changes in the pressure-volume loop are seen with alterations in the loading conditions or contractile function of the ventricle and with disease (Figs. 1-4, 1-5, 1-6, and 1-7).


Figure 1.7 Pressure-volume loops showing aortic regurgitation and stenosis.

(Redrawn from Jackson JM, Thomas SJ, Lowenstein E: Anesthetic management of patients with valvular heart disease. Semin Anesth 1:239, 1982.)

Determinants of Cardiac Output

Cardiac output is the product of stroke volume and heart rate. Stroke volume is determined by preload, afterload, and contractility. Cardiac output may be divided by body surface area to obtain the cardiac index. The normal value for cardiac output in awake normotensive subjects is 1.9 to 3.5 L/min/m2 (see Chapter 8).


The functional contractile unit of the myocyte is the sarcomere, which is composed of overlapping thick and thin filaments. The thick filaments contain the protein myosin; the thin filaments contain the proteins actin, tropomyosin, and the troponin complex. Activation of the troponin complex by calcium leads to binding between actin and myosin and contraction of the sarcomere. The force of contraction is partly dependent on the degree of overlap of the thick and thin filaments. In the resting state the sarcomere is 1.8 to 2.0 μm long. Maximum overlap of the filaments occurs at a sarcomere length of about 2.3 μm—that is, when the sarcomere is prestretched above its resting length. This property of the sarcomere underlies the Starling law of the heart, which states that the degree of fiber stretch at end-diastole (preload) determines the force of contraction. In the intact heart, this is represented by the relationship between end-diastolic volume and stroke volume and can be displayed as a ventricular function curve (Fig. 1-8). Over the physiologic range, the relationship is relatively linear; thus, the ejection fraction, which is the slope of the ventricular function curve (SV/EDV), is relatively preload independent. In the ICU, left ventricular end-diastolic volume (or its surrogate, end-diastolic area) may be estimated by echocardiography. The effect of increasing preload on the pressure-volume loop is shown in Figure 1-4.

Left Ventricular Compliance

Because end-diastolic pressure is related to end-diastolic volume through the passive pressure-volume relationship (see Fig. 1-6), end-diastolic pressure is used as a surrogate for end-diastolic volume as a measure of preload. Clinically, left ventricular end-diastolic pressure is usually inferred from the pulmonary artery wedge pressure (PAWP; Chapter 8), which is obtained by means of a pulmonary artery catheter. Unfortunately, the relationship between end-diastolic volume and end-diastolic pressure (ventricular compliance) is not linear. At high ventricular volumes, a small increase in end-diastolic volume is associated with a big increase in end-diastolic pressure (see Fig. 1-6). Furthermore, left ventricular compliance may be altered by disease. For example, compliance is decreased in aortic stenosis and increased in aortic regurgitation (see Fig. 1-7). In some situations, an increase in filling pressure may actually be associated with a reduction in preload (e.g., pericardial tamponade). Thus, the estimation of preload from the PAWP may be misleading (see Chapter 8).1,2

Ventricular Interactions

The term preload generally refers to left ventricular end-diastolic fiber stretch because it is easier to model mathematically and also because left ventricular preload determines systemic stroke volume. However, when intravenous fluid is administered, right ventricular preload is augmented. Because the left and right ventricles are in series (series interdependence), a change in the loading conditions (preload or afterload) of one ventricle is, within a few cardiac cycles, transmitted to the other ventricle. Thus, increased right ventricular preload leads to an increase in right ventricular output, thereby increasing left ventricular preload.

If the left and right ventricles have normal systolic and diastolic function, and if tricuspid valve function is normal, right ventricular end-diastolic pressure (central venous pressure; CVP) may be used as a surrogate for left ventricular preload. However, when using CVP to estimate preload, the following points should be borne in mind:


Figure 1.9 Output of the left and right ventricles at different atrial filling pressures.

(Redrawn from Guyton AC, Hall JE: Textbook of Medical Physiology, ed 10. Fig. 9.10, p. 104. Philadelphia, WB Saunders, 2000.)

A further effect that must be considered is the fact that the left and right ventricles have a shared septum (parallel interdependence). Severe right ventricular volume overload (e.g., due to tricuspid regurgitation or right ventricular infarction) leads to leftward displacement of the ventricular septum, impairing left ventricular diastolic function. In this situation, administration of fluid to augment left ventricular preload will cause further leftward displacement of the ventricular septum and worsen left ventricular filling. Similarly, severe left ventricular dilatation can impair right ventricular diastolic filling.


Afterload can be defined as ventricular wall stress during systole. It is determined by the impedance to ejection and ventricular geometry. Wall stress can be calculated by using the Laplace law:

(1-3) image

where r = the radius of the ventricle, ΔP = the pressure gradient across the ventricular wall, and w = wall thickness. Thus, left ventricular afterload is increased by left ventricular dilatation and reduced by left ventricular hypertrophy. The transmural left ventricular pressure gradient, and therefore afterload, is increased by high systemic vascular resistance, high arterial blood pressure, and a noncompliant aorta. Transmural ventricular pressure is reduced by high intrathoracic pressure such as that which occurs with mechanical ventilation and positive end-expiratory pressure (PEEP).

A sudden increase in afterload is associated with an immediate fall in stroke volume. Over the next few beats, stoke volume gradually recovers due to increased diastolic ventricular volume (see Fig. 1-5). However, patients with congestive cardiac failure who are operating on the plateau part of their ventricular function curve (see Fig. 1-8) are unable to increase stroke volume by increasing preload—that is, they have reduced preload reserve. Thus, in patients with congestive cardiac failure, increased afterload (e.g., due to phenylephrine) can cause a precipitous fall in cardiac output. Indeed, afterload reduction is a fundamental principle of the treatment of left ventricular failure.

A sudden fall in afterload is associated with an immediate increase in stroke volume. If venous return is also increased, a sustained increase in stroke volume occurs. This is the mechanism by which cardiac output is increased to supranormal levels in patients with septic shock and also the mechanism by which vasodilating drugs preserve cardiac output in patients with congestive cardiac failure.

Clinically, ventricular wall stress is difficult to measure, and systemic vascular resistance (see later material) derived from a pulmonary artery catheter is commonly used as a surrogate. However, systemic vascular resistance reflects only the nonpulsatile components of afterload and does not take into account ventricular dimensions and the compliance of the arterial tree. This is important clinically. For instance, with severe aortic regurgitation, marked left ventricular dilatation occurs, increasing left ventricular wall stress and afterload. However, measured systemic vascular resistance may be normal or low. Systemic vascular resistance is a particularly unhelpful surrogate of left ventricular afterload in mechanically ventilated cardiac surgery patients who have stiff aortas and dilated ventricles.

Myocardial Dysfunction

Potentially reversible cardiac dysfunction can occur as a consequence of myocardial ischemia, stunning or hibernation, or remodeling.7 Irreversible myocardial dysfunction occurs due to myocyte loss caused by infarction or replacement (e.g., with amyloid or fibrotic tissue).

Ischemia, Stunning, and Hibernation

Myocardial ischemia results from an imbalance between myocardial oxygen supply and demand (as outlined subsequently in Oxygen Supply and Demand in the Coronary Circulation). If oxygen supply is inadequate to replenish the ATP consumed during the repetitive coupling and uncoupling of actin and myosin molecules, systolic and diastolic dysfunction occurs. Brief periods (<10 min) of total oxygen deprivation or longer periods of reduced oxygen delivery produce dysfunction that is potentially reversible with restoration of the blood supply. More prolonged periods of ischemia result in irreversible myocardial infarction.

Relief of ischemia does not always result in an immediate return of contractile function. Myocardial stunning is temporary myocardial dysfunction that persists following the resolution of an ischemic episode (postischemic dysfunction). This dysfunction occurs in the presence of normal, or near normal, coronary blood flow and in the absence of irreversible cellular damage. Return of contractile function occurs over a period of hours to days. Myocardial stunning arises primarily from reperfusion injury. The mechanisms of reperfusion injury and myocardial stunning are incompletely understood, but they involve the generation of oxygen-derived free radicals (e.g., superoxide, peroxynitrite), altered concentration and sensitivity to intracellular calcium, and endothelial dysfunction.3,4 Myocardial stunning is encountered in a number of clinical situations, such as after cardiopulmonary bypass and subsequent to reperfusion therapy for an acute coronary syndrome.

Hibernating myocardium is a state of chronic ischemic dysfunction. Hibernating myocardium was initially thought to be caused by low baseline blood flow. However, it is now recognized that baseline blood flow may be near normal and that the primary problem is reduced vasodilator reserve. Small increases in oxygen demand can then provoke acute ischemia that is typically painless.5 As with stunning, hibernating myocardium is not associated with permanent disruption of cellular integrity. However, there is some loss of the contractile elements, along with disorganization of the cytoskeletal proteins and interstitial inflammation.6 The downregulation of cellular function that occurs with myocardial hibernation serves to reduce myocardial oxygen requirements, helping to minimize cellular damage. Once blood supply has been restored, recovery of normal function is slower than with myocardial stunning, taking some days. If hibernating myocardium is not revascularized, myocardial fibrosis and irreversible dysfunction eventually occur. Hibernating myocardium is a significant cause of congestive cardiac failure in patients with coronary artery disease; the other two causes are infarction and remodeling (see subsequent material).

Collectively, ischemic, hibernating, and stunned myocardium are referred to as viable myocardium. The distinctions among normal, viable, and infarcted myocardium are of tremendous importance in terms of prognosis and treatment options for patients with coronary artery disease, and are discussed in Chapter 5.


Remodeling7 is an alteration in ventricular structure that occurs as part of normal growth or due to a pathologic process such as hypertension, valvular heart disease, myocardial infarction, or a cardiomyopathy. The primary feature of remodeling is hypertrophy. Ventricular hypertrophy is an adaptive response to a change in loading conditions that helps to attenuate ventricular dilatation, reduce wall stress (see Equation 1-3), and stabilize contractile function. Ventricular hypertrophy is initiated by myocardial stretch and various neuroendocrine processes.

Chronic pressure overload such as that which occurs with hypertension and aortic stenosis typically results in concentric hypertrophy, in which ventricular wall thickness is increased out of proportion to the increase in chamber size. Chronic volume overload such as that which occurs with aortic and mitral regurgitation typically results in eccentric hypertrophy, in which wall thickness is increased in proportion to chamber size.

Following a myocardial infarction, ventricular remodeling (hypertrophy, dilation, impaired contractility) can occur in sites adjacent to or remote from the zone of infarction, that is, within normal myocardium. The signal for remodeling within this normal myocardium is complex; it involves alterations in ventricular loading conditions, activation of neuroendocrine pathways (including the sympathetic nervous system, the renin-angiotensin-aldosterone system [RAAS], and natriuretic peptides; see subsequent material), and inflammation within infarcted and noninfarcted myocardium.

Unchecked, ventricular remodeling can eventually result in irreversible cardiac dysfunction due to myocardial fibrosis. However, if appropriate therapy is introduced early enough, remodeling can be interrupted or reversed. If possible, treatment should be directed to the underlying cause: for valvular heart disease, this involves valve repair or replacement; and for hypertension, it involves effective control of blood pressure. Following myocardial infarction, pharmacologic antagonists (angiotensin-converting enzyme [ACE] inhibitors, β blockers, aldosterone antagonists) to the neuroendocrine pathways involved in the remodeling process have proven to be partially effective (see Chapter 19).

Clinical Assessment of Cardiac Function: Stroke Volume and Ejection Fraction

Ejection fraction and stroke volume are two parameters that are commonly measured in the ICU to evaluate cardiac performance. However, stroke volume and ejection fraction do not always change in parallel and are affected differently by changes in loading conditions.

Ejection fraction is the proportion of diastolic volume ejected during ventricular contraction (see Equation 1-2). The normal range of left ventricular ejection fraction is 55% to 75%. In the ICU, ejection fraction is usually estimated by echocardiography.

These two parameters are related by the end-diastolic volume. Thus, a patient with an ejection fraction of 60% and an end-diastolic volume of 90 ml has a stroke volume of 54 ml. Similarly, a patient with an ejection fraction of 30% and an end-diastolic volume of 180 ml also has a stroke volume of 54 ml. Therefore, if ventricular volumes are high, it is possible to have a low ejection fraction but a normal stroke volume (as in chronic stable heart failure) or a low-normal ejection fraction and a high cardiac output (as in septic shock).

Buy Membership for Cardiothoracic Surgery Category to continue reading. Learn more here