15: Dyspnea

Heart disease, lung disease, and anemia are the most common causes of dyspnea. Neuromuscular disease and anxiety are less common causes. The simplest approach to the differential diagnosis is to consider the anatomical components of each of these systems. This allows us to develop a fairly comprehensive differential diagnosis of dyspnea.

Differential Diagnosis of Dyspnea

1. Heart

   1. Endocardium: Valvular heart disease (ie, aortic stenosis, aortic regurgitation, mitral regurgitation, and mitral stenosis)

   2. Conduction system

      1. Bradycardia (sick sinus syndrome, atrioventricular block)

      2. Tachycardia

         • (1) Atrial fibrillation and other supraventricular tachycardias

         • (2) Ventricular tachycardia

   3. Myocardium: Heart failure (HF)

      1. Systolic failure (coronary artery disease [CAD], hypertension, alcohol abuse)

      2. Diastolic failure (hypertension, aortic stenosis, hypertrophic cardiomyopathy)

   4. Coronary arteries (ischemia)

   5. Pericardium (tamponade, constrictive pericarditis)

2. Lung

   1. Alveoli

      1. Pulmonary edema (HF or acute respiratory distress syndrome)

      2. Pneumonia

   2. Airways

      1. Suprathoracic airways (ie, laryngeal edema)

      2. Intrathoracic airways

         • (1) Asthma

         • (2) Chronic obstructive pulmonary disease (COPD) (see Chapter 33, Wheezing & Stridor)

   3. Blood vessels

      1. Pulmonary emboli

      2. Primary pulmonary hypertension

   4. Pleural

      1. Pneumothorax

      2. Pleural effusions

         • (1) Transudative

           • (a) HF

           • (b) Cirrhosis

           • (c) Nephrotic syndrome

           • (d) Pulmonary embolism (PE)

         • (2) Exudative

           • (a) Tuberculosis

           • (b) Cancer

           • (c) Parapneumonic effusions

           • (d) Connective tissue diseases

           • (e) PE

   5. Interstitium

      1. Edema

      2. Inflammatory

         • (1) Organic exposures (eg, hay, cotton, grain)

         • (2) Mineral exposures (eg, asbestos, silicon, coal)

         • (3) Idiopathic diseases (eg, sarcoidosis, scleroderma, systemic lupus erythematosus, granulomatosis with polyangiitis [formerly Wegener granulomatosis])

      3. Infectious

3. Anemia

The extensive differential diagnosis for dyspnea necessitates a careful and detailed history, physical exam and review of basic laboratory examinations including chest film, ECG, and hematocrit. The history should detail the time course of the complaint, its severity, associated symptoms, and the patient’s past medical history. The physical exam should include vital signs, a detailed cardiac and pulmonary exam, and a search for signs suggestive of anemia (conjunctival pallor or palmar crease pallor). This process often suggests the diagnosis. However, when the diagnosis is not straightforward, certain pivotal findings can narrow the differential diagnosis and focus the diagnostic search (Figure 15-1). One such pivotal clue is fever. Fever is typically seen in pneumonia but could also be seen in asthma or COPD with concomitant infection. Less common causes of fever and dyspnea include valvular heart disease due to endocarditis, pulmonary emboli, acute respiratory distress syndrome, or interstitial lung disease. Chest pain (covered extensively in Chapter 9) is another pivotal clue in patients with dyspnea. Chest pain may be pleuritic or nonpleuritic and acute or chronic/recurrent. Each of these features can help focus the differential diagnosis (see Figure 15-1, Table 15-1). In brief, common causes of dyspnea and pleuritic chest pain are pneumonia, PE, pneumothorax, asthma, and COPD. On the other hand, many diseases may cause nonpleuritic chest pain (including those diseases already mentioned that may cause pleuritic chest pain). In these patients, the acuity of the chest pain can help narrow the differential diagnosis. Common causes of dyspnea associated with acute chest pain include myocardial infarction (MI), aortic dissection, PE, pneumothorax, arrhythmias (causing angina), and pneumonia. Common causes of dyspnea associated with chronic/recurrent chest pain include angina (caused by CAD, severe anemia, or aortic stenosis), asthma or COPD (which are often associated with chest tightness), and recurrent intermittent arrhythmias.

In patients with any of any of the aforementioned pivotal clues (fever, chest pain), a search for risk factors, associated symptoms and signs of those diseases can help rank the differential diagnosis. Table 15-2 lists the highly suggestive risk factors as well as associated symptoms and signs of common diseases causing dyspnea.

When these pivotal clues are absent, it is appropriate to evaluate the most common causes of dyspnea, namely HF, pneumonia, asthma, COPD, and pulmonary emboli. Once again, looking for their respective risk factors, associated symptoms and specific signs can help determine their likelihood. Features that suggest HF include a history of MI, CAD risk factors, long-standing uncontrolled hypertension, or alcohol abuse. Furthermore, an S₃ gallop or jugular venous distention (JVD) are fingerprints for HF. Fever and cough raise the possibility of pneumonia and a significant smoking history (≥ 20 pack years) raises the possibility of COPD. Wheezing—defined as a multi-pitched sound on exhalation—suggests COPD or asthma. Finally, PE can be obvious or subtle and should be considered in patients with risk factors such as recent immobilization, surgery, a history of cancer, or use of estrogen, or suggestive signs (eg, unilateral leg swelling).

Some patients remain difficult to diagnose. For such patients a more comprehensive review of risk factors, associated symptoms and signs (Table 15-2) as well as a careful review of their chest film findings may be helpful (Table 15-3). A normal chest radiograph makes pneumonia, interstitial lung disease, and acute respiratory distress syndrome unlikely and rules out pneumothorax. Focal infiltrate(s) suggest pneumonia and can also be seen with COPD or asthma due to atelectasis or superimposed infection. Diffuse infiltrates or edema may be seen in a variety of settings including any cause of HF, acute respiratory distress syndrome, and certain pneumonias. The presence of a pleural effusion is a critical clue that should be evaluated.

Other diagnostic testing that is often necessary includes echocardiography, pulmonary function tests, B-type natriuretic peptide (BNP), and CT angiography (CTA). Echocardiography can reveal unsuspected HF or valvular heart disease. Pulmonary function tests can help determine whether the patient has obstructive, restrictive, or vascular lung disease (Table 15-4). Figure 15-1 summarizes a diagnostic approach to patients with dyspnea.

Always quantify the increase in dyspnea from baseline. Significant changes suggest serious disease and warrant thorough evaluations.

Although the differential diagnosis of dyspnea is broad, the patient’s risk factors, symptoms and signs point to a cardiac etiology and immediately focus the differential diagnosis. His past history of MI is a clear risk factor for HF. Furthermore, the JVD, S₃ gallop, and peripheral edema are all very suggestive of HF making this the leading hypothesis. His physical exam also reveals a heart murmur raising several alternative diagnoses (ie, mitral regurgitation, aortic stenosis, or aortic regurgitation). This particular murmur is most consistent with mitral regurgitation. Mr. C’s irregular pulse also raises the possibility of atrial fibrillation. Finally, cardiac ischemia presenting as dyspnea rather than pain is a must not miss possibility. Table 15-5 lists the differential diagnosis.

Pursue highly specific positive physical findings (in this case the S₃ gallop and JVD); they should help drive the diagnostic search.

Leading Hypothesis: HF

Textbook Presentation

Patients typically have fatigue, dyspnea on exertion, orthopnea, paroxysmal nocturnal dyspnea, and edema. Often, there is an antecedent history of either MI or poorly controlled hypertension.

Disease Highlights

1. HF refers to any cardiac pathology that impairs left ventricular filling or ejection, which may arise from diseases of the pericardium, myocardium, or valves. The remainder of this discussion will focus on myocardial causes of HF. Valvular heart disease is discussed separately.

2. Affects 5.8 million patients in the United States and accounts for 1 million hospitalizations and 53,000 deaths annually. Each year, HF is diagnosed in 670,000 patients.

3. Pathophysiologic classification: HF may occur in patients with impaired emptying (and an ejection fraction ≤ 40%) or impaired filling (with a preserved ejection fraction ≥ 50%). The distinction is important because both the etiologies and treatments of these 2 groups are different. HF may also be classified based on whether the primary process affects the left ventricle (LV) or the right ventricle (RV).

   1. Heart failure with reduced ejection fraction (HFrEF)

      1. Previously called systolic HF or systolic dysfunction, HFrEF accounts for approximately 50% of cases of HF.

      2. HFrEF develops when systolic dysfunction impairs LV emptying.

      3. CAD accounts for 66% of all cases of HFrEF.

      4. Other common causes include long-standing hypertension and alcohol abuse.

      5. Less common causes include viral cardiomyopathy, postpartum cardiomyopathy, drug toxicity (ie, adriamycin), and idiopathic cardiomyopathy.

      6. Most patients with HFrEF (impaired emptying) also have diastolic dysfunction (impaired filling).

   2. Heart failure with preserved ejection fraction (HFpEF)

      1. Previously referred to as diastolic HF, HFpEF accounts for approximately 50% of all HF cases.

      2. HFpEF develops when an increase in myocardial muscle mass (thickness), infiltration, or fibrosis decreases LV compliance.

         • (1) Decreased LV compliance impairs LV filling.

         • (2) Note that although LV filling is compromised, contractility is preserved and the ejection fraction is normal.

      3. The most common cause of HFpEF is hypertension. Less common causes include aortic stenosis, hypertrophic cardiomyopathy, and infiltrative cardiomyopathies (eg, hemochromatosis, amyloidosis).

   3. The mortality in patients with HFrEF and HFpEF are similar.

   4. Patients with ejection fraction of 41–49% are classified as HFpEF, borderline. Their treatment and outcomes appear similar to patients with HFpEF.

   5. Right- versus left-sided HF

      1. HF may involve the LV, the RV, or both.

      2. Common causes of LV failure include CAD, hypertension, and alcoholic cardiomyopathy.

      3. Common causes of RV failure include severe pulmonary disease (especially COPD) and advanced LV failure.

      4. Peripheral edema, JVD, and fatigue may be seen in LV or RV failure. Pulmonary edema may also be seen in LV failure but not isolated RV failure.

   6. Progression

      1. HF triggers maladaptive neurohormonal changes including increased activation of the renin-angiotensin-aldosterone system and the sympathetic nervous system.

      2. These neurohormonal responses promote sodium retention, increase afterload, and contribute to edema and progressive HF.

      3. Therapies that interrupt these responses reduce mortality (see below).

4. Classifications of HF

   1. New York Heart Association functional classification is descriptively useful but is limited by the ability of patients to move from 1 class to another with therapy.

      1. Class I: Asymptomatic (ie, symptoms only at levels of exertion that would make healthy patients dyspneic)

      2. Class II: Slight limitation of physical activity (eg, climbing stairs)

      3. Class III: Marked limitation of physical activity (eg, walking on flat surface)

      4. Class IV: Symptoms at rest or with any physical activity

   2. American College of Cardiology Foundation/American Heart Association (ACCF/AHA) stages of HF were developed to facilitate identifying stage-specific therapies.

      1. Stage A: Patients at high risk for HF without structural heart disease or symptoms (eg, patients with hypertension or CAD but normal LV function).

      2. Stage B: Patients with structural heart disease (eg, LV hypertrophy or decreased ejection fraction) without signs or symptoms of HF.

      3. Stage C: Structural heart disease and prior or current symptoms.

      4. Stage D: Refractory HF symptoms despite therapy.

5. Complications of HF

   1. Electrical: Heart block, ventricular tachycardia, atrial fibrillation, sudden death

   2. Pulmonary edema

   3. Stroke and thromboembolism

      1. 2–4% annual incidence

      2. Risk increases if atrial fibrillation coexists

   4. Mitral regurgitation (LV dilatation may lead to sufficient dilatation of the mitral annulus that it causes secondary mitral regurgitation [see below])

   5. Death

      1. Symptomatic mild to moderate HF: 20–30% per year

      2. Symptomatic severe HF: up to 50% per year

      3. Mechanism of death

         • (1) Sudden in 50% (secondary to ventricular tachycardia or asystole)

         • (2) Progressive HF in 50%

Evidence-Based Diagnosis

1. The history should assess risk factors for HF, including hypertension, CAD, alcohol abuse, illicit drug use, and adriamycin exposure.

2. Common symptoms include dyspnea on exertion, orthopnea and paroxysmal nocturnal dyspnea (Table 15-6).

3. Physical exam

   1. Clinical signs and symptoms are affected by

      1. Patient’s current volume status

      2. Chronicity. In chronic HF, signs and symptoms are frequently absent despite marked impairment of LV function and marked volume overload.

   2. S₃ gallop

      1. An S₃ gallop occurs when a large volume of blood rushes from the left atrium (LA) into the LV at the start of diastole (just after S₂).

      2. Virtually pathognomonic of volume overload and occurs most commonly in patients with decompensated HF.

   3. S₄ gallop

      1. Occurs when the LA contracts and sends blood into the LV (just before S₁).

      2. An S₄ gallop may be heard in some normal patients and in many patients with LV hypertrophy due to hypertension or other causes.

      3. S₄ is not specific for HF.

   4. JVD

      1. Defined as > 3 cm of elevation above the sternal angle (Figure 15-2).

      2. Highly specific for HF (> 95%); may occur in RV or LV failure.

   5. Table 15-6 summarizes the sensitivities, specificities, and LRs of clinical findings for HF in patients with dyspnea.

      1. Classic signs and symptoms (orthopnea, paroxysmal nocturnal dyspnea, crackles, gallops and edema) are not sensitive for HF and their absence does not rule out HF. Indeed, even in severe chronic HF (mean ejection fraction 18%, pulmonary capillary wedge pressure [PCWP] > 22 mm Hg), 42% of patients did not have crackles, increased JVP, or edema.

         Signs of HF are commonly absent even in advanced HF.

      2. However, certain findings are highly specific and significantly increase the likelihood of HF when present. An S₃ (but not an S₄) and JVD strongly suggest HF.

      3. Other classic symptoms, like orthopnea, paroxysmal nocturnal dyspnea, and crackles, are not specific for HF.

4. Chest radiography

   1. Cardiomegaly is the most sensitive finding on chest film (74% sensitive, 78% specific), and its absence modestly decreases the likelihood of HF (LR– 0.33).

   2. Pulmonary venous congestion and interstitial edema are highly specific (96–97%) and when present strongly suggest HF (LR+ 12).

   3. Pleural effusions are seen in 26% of patients with HF.

      1. The effusions are usually small to moderate in size and unilateral or bilateral.

      2. These effusions are transudative.

      3. When due to HF, pleural effusions are usually accompanied by cardiomegaly, pulmonary vascular redistribution, or edema.

      4. The absence of these findings or the presence of massive pleural effusions suggests some other etiology and warrants further evaluation.

   4. Table 15-7 summarizes the accuracy of the chest radiograph in the diagnosis of HF.

5. ECG, though not diagnostic for HF, can be suggestive if signs of prior MI or LV hypertrophy are present.

6. BNP

   1. Secreted by LV or RV in response to increased volume, pressure, or both.

   2. May be elevated in systolic or diastolic HF

   3. Levels increase proportionately to the degree of HF

   4. Low BNP levels decrease the likelihood of HF in patients with dyspnea.

   5. High levels of BNP increase the likelihood of HF but are still not entirely specific.

   6. The accuracy of the BNP is summarized in Table 15-8.

   7. BNP is also elevated in many patients with pulmonary embolus (average 702 ng/L, and 1876 ng/L in patients with central PE).

   8. The ACCF/AHA concluded that elevated BNP levels lend weight to a diagnosis of HF but should not be used in isolation to confirm or exclude HF.

7. Two-dimensional echocardiogram is the test of choice to diagnose HF and is recommended for all patients with known or suspected HF.

   1. Systolic and diastolic function can be evaluated.

   2. Regional systolic dysfunction suggests an ischemic etiology.

   3. Valve function can be assessed.

   4. Bedside hand-carried ultrasound accurately identifies patients with ejection fraction < 40% when performed by internal medicine residents with limited training (sensitivity, 94%; specificity, 94%; LR+, 15.7; LR– 0.06).

8. Radionuclide tests can quantify ejection fraction but cannot assess LV wall thickness or valvular abnormalities.

9. Cardiac MR, although more costly and difficult to perform than echocardiography, is another option for the evaluation of HF.

   1. Cardiac MR can accurately measure ejection fraction and LV volume as well as assess myocardial perfusion, viability, and fibrosis.

   2. Appropriate in the initial evaluation of patients with new or suspected HF.

   3. Cardiac MR combined with magnetic resonance angiography and gadolinium enhancement can detect underlying CAD and ischemia (sensitivity 100%, specificity 96% in patients in sinus rhythm).

10. HF and COPD

    1. HF is frequently present but unsuspected in patients in whom COPD is diagnosed.

    2. Diagnosis is more difficult in these patients.

       1. Studies report unsuspected HF in ≈25% of patients with COPD. These patients had fewer pack years of tobacco use than patients without HF (9.6 vs 22.7).

       2. Pleural fluid, pulmonary revascularization, and edema were uncommon even in the subgroup with HF (9.1%) but when present strongly suggested HF (LR+ 9.1).

       3. BNP is less sensitive in patients with COPD (sensitivity, 35%; specificity, 90%; LR+, 3.5; LR–, 0.72).

          Clinicians should have a low threshold for checking an echocardiogram in patients with COPD and dyspnea.

Treatment

1. Prevention: Hypertension therapy decreases the incidence of HF by 30–50%.

2. Evaluation

   1. Initial history and physical exam should assess functional capacity and volume status (weight, vital signs, lung exam, JVD, S₃ gallop and edema).

   2. Routine laboratory tests recommended by the ACCF/AHA

      1. CBC

      2. Urinalysis, blood urea nitrogen (BUN), creatinine

      3. Electrolytes

      4. Lipid panel, glycohemoglobin, liver function tests

      5. Thyroid-stimulating hormone

      6. Chest radiograph

      7. ECG

      8. Echocardiogram

   3. Evaluation of CAD

      1. HF develops secondary to ischemia in approximately two-thirds of patients with HFrEF.

      2. Identifying underlying CAD allows clinicians to optimize medical therapy and identify which patients can benefit from revascularization.

      3. CAD should be suspected in patients with chest pain, CAD risk factors, ischemic ECG findings, or regional wall motion abnormalities on noninvasive imaging.

      4. Stress test or angiography can be used depending on the pretest probability of CAD.

3. Treatment

   1. Certain treatments benefit all patients with HF whereas others have greater proven efficacy in patients with HFrEF.

   2. All HF patients

      1. Sodium restriction is recommended for all HF patients. A high sodium diet is associated with a marked increase in acute decompensated HF (an absolute increased risk 31–34%, NNH ≈ 3), hospitalizations, and all-cause mortality.

      2. Diuretics (loop or thiazides)

         • (1) Mainstay of therapy to treat edema and pulmonary congestion (should be used in combination with salt restriction)

         • (2) The clinical assessment of volume status is critical. Increasing weight, edema, JVD, pulmonary edema, or an S₃ gallop suggests patients are volume overloaded.

         • (3) However, multiple studies demonstrate that despite severe chronic HF and marked volume overload (measured by PCWP) patients may not have signs of HF.

         • (4) Therefore, dyspneic HF patients should undergo aggressive diuresis while monitoring renal function.

         • (5) LV filling pressures can also be indirectly estimated using hand carried ultrasound evaluation of the inferior vena cava (IVC).

           • (a) Hand carried devices can measure IVC diameter and collapsibility with inspiration.

           • (b) HF (right or left) is associated with an increase in IVC diameter and a decrease in the normal collapsibility with inspiration.

           • (c) IVC diameter > 2.0 cm suggests elevated PCWP ≥ 17 mm Hg (75% sensitive, 83% specific, LR+ 4.4, LR– 0.3).

           • (d) IVC collapsibility < 45% suggests elevated PCWP ≥ 17 mm Hg (83% sensitive, 71% specific, LR+ 2.9, LR– 0.24).

           • (e) IVC measurements may prove useful as an adjunct to determine the need for further diuresis. Larger IVC diameters and less collapsibility on discharge predicted the subsequent need for readmission.

         • (6) Input and output, daily weights, lung and cardiac exam, electrolytes, BUN and creatinine should be monitored daily in hospitalized patients.

      3. Control of hypertension

      4. CAD revascularization: Coronary artery bypass surgery can improve cardiovascular outcomes in select patients with CAD and HF. Recommendations include the following:

         • (1) Coronary artery bypass surgery should be considered in patients who have HF (HFrEF or HFpEF) with angina and left main stenosis or left main equivalent disease.

         • (2) Coronary artery bypass surgery is also recommended for select patients with HFrEF and multivessel or proximal LAD disease

      5. Influenza and pneumococcal vaccination

      6. Nonsteroidal antiinflammatory drugs and thiazolidinediones increase fluid retention and have been associated with worsening and precipitating HF and should be avoided.

   3. Patients with HFrEF

      1. Therapy with angiotensin-converting enzyme (ACE) inhibitors, beta-blockers, aldosterone antagonists, and hydralazine combined with nitrates have been shown to reduce morbidity and mortality in patients with HFrEF.

      2. ACE inhibitors

         • (1) Indicated in patients with HFrEF or patients with a prior MI

         • (2) Angiotensin receptor blockers (ARBs) may be used in place of ACE inhibitors when a troublesome cough develops in patients taking ACE inhibitors.

         • (3) ARBs may cause angioedema in patients who had angioedema while taking ACE inhibitors.

      3. Beta-blockers

         • (1) Beta-blockers reduce morbidity and mortality in all stages of HF, including severe HF (ejection fraction < 25%)

         • (2) Indicated in patients with HFrEF or patients with a prior MI. Beta-blockers with proven efficacy include carvedilol, sustained-release metoprolol and bisoprolol.

         • (3) Initiate therapy at low doses, when patients are euvolemic and not requiring inotropes.

         • (4) Beta-blockers can precipitate fatal asthma and should be avoided in patients with severe or decompensated reactive airway disease. Selective beta-1-agonists can be used with caution in patients with controlled reactive airway disease.

      4. Aldosterone antagonist (eg, spironolactone)

         • (1) Reduces mortality in patients with class II–IV HF and ejection fraction ≤ 35% (< 40% with prior MI)

         • (2) Should be avoided in patients with a glomerular filtration rate ≤ 30 mL/min, K+ ≥ 5.0 mEq/dL, or patients who cannot have their serum potassium adequately monitored

      5. Statins should be used in patients with HFrEF and prior MI.

      6. Hydralazine and nitrates, in addition to ACE inhibitors and beta-blockers, have been demonstrated to reduce mortality in black patients with class III or IV HF. They may also be useful in patients who are unable to tolerate ACE inhibitors/ARBs.

      7. Digoxin

         • (1) Reduces hospitalizations but not mortality

         • (2) Low serum concentrations (0.5–0.8 mg/dL) are as effective as higher concentrations.

         • (3) ACCF/AHA guidelines recommend digoxin only in symptomatic patients with HFrEF.

         • (4) Digoxin may increase mortality in women and is not advised for women by some authorities.

      8. Cardiac resynchronization therapy: Some patients with HF have prolonged QRS intervals, associated with prolonged and dyssynchronous depolarization. This nonuniform depolarization results in poorly organized contraction and contributes to LV dysfunction. In addition, it contributes to mitral regurgitation and LV remodeling.

         • (1) In cardiac resynchronization, wires are implanted in the atria and both ventricles to allow precise and coordinated depolarization of the atria and left and right ventricles.

         • (2) Cardiac resynchronization therapy improves ejection fraction, quality of life, and functional status, and it reduces hospitalizations and mortality in select patients.

         • (3) Indications are complex and detailed but include patients with symptomatic HF despite optimal medical therapy, an ejection fraction ≤ 35% and a QRS ≥ 0.15 s (and in certain patients ≥ 0.12s).

      9. Implantable cardiac defibrillator

         • (1) A substantial proportion of patients with HF experience sudden death (30% in dilated cardiomyopathy), presumably secondary to ventricular tachycardia and ventricular fibrillation.

         • (2) Implantable cardiac defibrillators are recommended in the following select HF patients:

           • (a) Patients who have survived cardiac arrest, ventricular fibrillation, or hemodynamically destabilizing ventricular tachycardia

           • (b) Symptomatic HF patients (NYHA class II–III) with ischemic or nonischemic HFrEF with ejection fraction ≤ 35%, at least 40 days post MI while taking appropriate therapy

           • (c) Asymptomatic patients with ischemic HF and an ejection fraction ≤ 30%, at least 40 days post MI while taking appropriate medical therapy

           • (d) Placement could be considered in HF patients with unexplained syncope.

      10. Heart transplantation is an option for a few patients with severe HF refractory to intensive medical therapy.

   4. Patients with HFpEF

      1. Systolic and diastolic hypertension should be controlled.

      2. Diuretics can be used to treat pulmonary congestion or edema.

      3. Digoxin has no proven benefit.

      4. Control ventricular rate for patients with atrial fibrillation.

      5. The effectiveness of ACE inhibitors, beta-blockers, or ARBs is less well established. Recent studies suggest ARBs decrease hospitalizations in patients with HFpEF.

   5. See ACCF/AHA guidelines for the treatment of patients with refractory HF and cardiogenic shock.

Alternative Diagnosis: Chronic Mitral Regurgitation

Textbook Presentation

Patients with mitral regurgitation may be identified by hearing a holosystolic murmur at the apex in an asymptomatic patient or during an evaluation of shortness of breath, dyspnea on exertion, edema, orthopnea, and fatigue. Alternatively, it may be discovered during the evaluation of patients with atrial fibrillation.

Disease Highlights

1. Trivial asymptomatic mitral regurgitation is commonly discovered on echocardiogram. The remainder of the discussion will focus on patients with more significant regurgitation.

2. Etiologies: Mitral regurgitation develops secondary to damaged mitral leaflets (primary) or a dilated mitral annulus (secondary).

   1. Primary mitral regurgitation

      1. Etiologies: mitral valve prolapse, rheumatic heart disease, and endocarditis.

      2. Although most patients with mitral valve prolapse never require valve replacement, it is the most common cause of mitral regurgitation and the need for valve replacement/repair.

   2. Secondary mitral regurgitation

      1. HF: LV dilatation leads to mitral annular dilatation and mitral regurgitation.

      2. Ischemic mitral regurgitation: Leaflet tethering shortens the mitral apparatus, resulting in mitral regurgitation.

3. Pathophysiology

   1. Compensated mitral regurgitation: Mitral regurgitation leads to LA dilatation, and compensatory LV dilatation. If systolic function is maintained, ejection fraction remains normal to high and LV end-systolic volume remains low because mitral regurgitation reduces LV afterload.

   2. Decompensated mitral regurgitation: Systolic function may fail leading to increased LV end systolic volume, decreased stroke volume, and decreased ejection fraction. This may be irreversible.

   3. LA dilatation may lead to atrial fibrillation.

4. Disease progression is slow. Average delay from diagnosis to symptoms is 16 years. However, in patients with severe mitral regurgitation, the annual mortality is 5%.

5. Complications include dyspnea, pulmonary edema, atrial fibrillation, and sudden death.

Evidence-Based Diagnosis

1. Physical exam: The typical murmur is a blowing, holosystolic murmur heard at the apex that radiates to the axilla. S₂ may be inaudible.

   1. Grade 3 or louder systolic murmur

      1. 85% sensitive, 81% specific for moderate to severe mitral regurgitation

      2. LR+, 4.5; LR–, 0.19

   2. S₃ gallop may be heard due to increased flow across the mitral valve.

2. ECG may demonstrate LA enlargement or LV hypertrophy, neither sensitive nor specific for the diagnosis.

3. Chest radiograph may demonstrate LA or LV enlargement, neither sensitive nor specific for the diagnosis.

4. Echocardiography is the test of choice to diagnose and quantify mitral regurgitation and is recommended in all patients with suspected mitral regurgitation. Transesophageal echocardiography provides more precise details on valve anatomy and may help determine whether valve repair (versus replacement) is an option.

Treatment

1. Serial echocardiography

   1. Serial echocardiography is important to detect signs of LV dysfunction, which may occur despite the absence of symptoms.

   2. Echocardiography is recommended annually or semiannually in patients with moderate to severe mitral regurgitation and after a change in signs or symptoms in patients with any degree of mitral regurgitation.

   3. Serial echocardiography is not recommended for asymptomatic patients with mild mitral regurgitation with normal LV size and function.

2. Valve repair versus replacement

   1. Valve repair is superior to valve replacement (when technically feasible).

   2. Valve repair is associated with substantially decreased operative mortality (2% vs 6%), a lower rate of endocarditis, and is associated with a significantly better ejection fraction. Importantly, valve repair does not require subsequent anticoagulation.

3. ACCF/AHA guidelines for valve repair are summarized below.

   1. Mitral valve repair is reserved for patients with severe mitral regurgitation and any of the following:

      1. Symptoms

      2. Atrial fibrillation

      3. Pulmonary hypertension

      4. Likelihood of success > 90%

      5. Mild to moderate LV dysfunction: (ejection fraction 30 – 60%, LV end-systolic diameter 40–55 mm)

   2. Decisions in patients with severe LV dysfunction (ejection fraction < 30%, LV end-systolic diameter > 55 mm) and severe mitral regurgitation are complex.

      1. Mitral valve repair or replacement increases afterload (by preventing ejection of blood from the LV to the LA) and may worsen HF.

      2. Consultation is recommended. HF therapy should be optimized.

   3. Isolated mitral valve surgery is not indicated in patients with mild to moderate mitral regurgitation.

4. Medical therapy of mitral regurgitation

   1. Chronic primary asymptomatic mitral regurgitation with preserved LV function: No medical therapy has been demonstrated to be useful.

   2. Secondary mitral regurgitation: Optimize HF regimen. ACE inhibitors, beta-blockers (particularly carvedilol), diuretics, and digoxin can be useful. Biventricular pacing reduces the severity of mitral regurgitation.

   3. Mitral regurgitation with hypertension: Treat hypertension with ACE inhibitors, diuretics, or beta-blockers.

   4. Treat underlying ischemia.

   5. Endocarditis prophylaxis is recommended for patients following mitral valve replacement with a mechanical or bioprosthetic device and for repairs utilizing a bioprosthetic annuloplasty ring. It is also recommended for patients with a history of endocarditis.

Alternative Diagnosis: Chronic Aortic Regurgitation

Textbook Presentation

Patients with chronic aortic regurgitation typically complain of progressive dyspnea on exertion or the sensation of a pounding heart. Alternatively, the patient may be asymptomatic, and the diagnosis may be suspected when a careful examiner detects an early diastolic murmur.

Disease Highlights

1. Secondary to damaged aortic leaflets or dilated aortic root

2. Etiologies

   1. Valvular abnormalities: Rheumatic carditis, bacterial endocarditis, collagen vascular disease, calcific degeneration, fenfluramine and phentermine

   2. Aortic root dilatation: Hypertension, ascending aortic aneurysm, Marfan syndrome, aortic dissection, syphilitic aortitis

   3. Bicuspid aortic valve disease

      1. Most common form of congenital heart disease affecting 1% of population, and transmitted in an autosomal dominant pattern.

      2. Affects the aortic valve (AV) and the aorta.

      3. Aortic regurgitation may occur due to valve changes or dilation of the proximal aorta.

      4. Dilation of the proximal ascending aorta is due to associated changes in the aortic media independent of the AV function.

      5. Complications include aneurysms, aortic regurgitation, and dissection.

3. Pathophysiology

   1. Regurgitation results in LV remodeling and eccentric and concentric LV hypertrophy to maintain wall stress. LV end-diastolic volume increases to augment the stroke volume so that forward flow remains in the normal range despite the regurgitant volume. During the compensated stage the ejection fraction is normal.

   2. The increasing preload and afterload may eventually result in LV systolic dysfunction, and the LV end-systolic volume increases and ejection fraction decreases. LV end-diastolic pressure increases and pulmonary congestion and dyspnea result. Exertional angina may also occur.

   3. Significant LV dysfunction can become irreversible. Valve replacement should be performed before irreversible LV dysfunction and HF develop (see below).

   4. Progression to symptoms or LV dysfunction in patients with normal LV function develops in 4% of patients per year. Sudden death occurred in 0.2% per year.

Evidence-Based Diagnosis

1. The pulse pressure (systolic–diastolic BP) is often wide in aortic regurgitation due to 2 processes. First, the large stroke volume increases the systolic BP, and second, the regurgitation of blood back into the LV rapidly lowers the diastolic BP.

   1. The wide pulse pressure causes many of the classic physical findings, such as bounding pulses and head bobbing.

   2. Wide pulse pressures (typically defined as systolic BP – diastolic BP ≥ 50% of systolic BP) are not specific for aortic regurgitation. Other causes include anemia, fever, pregnancy, large arteriovenous fistula, cirrhosis, thyrotoxicosis, and patent ductus arteriosa. Elderly patients with systolic hypertension commonly have a widened pulse pressure.

2. Auscultation

   1. May demonstrate an early decrescendo diastolic murmur following S₂. Best heard at the left sternal border.

      1. Auscultation is more sensitive for moderate to severe aortic regurgitation.

      2. Sensitivity is 0–64% among students and residents.

      3. Sensitivity is 80–95% among experienced cardiologists.

      4. Another study reported that the diastolic murmur of mild to moderate aortic regurgitation was rarely detected by attending noncardiologists (sensitivity 4% mild aortic regurgitation, 14% moderate aortic regurgitation).

      5. However, the finding of a diastolic murmur is highly specific (98%).

         A diastolic murmur is always abnormal and warrants evaluation (echocardiography).

   2. A systolic murmur suggesting aortic stenosis may be heard.

      1. Regurgitation results in increasing end-diastolic volumes.

      2. Stroke volumes increase to maintain forward flow.

      3. The increased cardiac output may exceed the capacity of even a normal aortic valve to accommodate flow, resulting in a high flow systolic murmur across the aortic valve. One study reported that 51% of patients with mild to moderate aortic regurgitation had a systolic murmur (86% in moderate aortic regurgitation and 50% in mild aortic regurgitation).

         Although a diastolic murmur strongly suggests aortic regurgitation, systolic murmurs are often the only murmur heard in patients with aortic regurgitation.

   3. Austin Flint murmur

      1. Aortic regurgitant streams may impact the mitral valve leaflets during diastole resulting in functional mitral stenosis and a late diastolic murmur over the apex.

      2. Sensitivity varies from 0% to 100%.

3. Doppler echocardiography is the test of choice and should be performed in all patients with a diastolic murmur and in those patients with a dilated aortic root.

4. Patients with biscuspid valves in whom the aortic root is not adequately visualized with transthoracic echocardiography should undergo additional imaging to evaluate the aortic root (eg, transesophageal echocardiography or cardiac MR).

5. Exercise testing can help assess LV function during stress.

Treatment

1. Serial echocardiography is important to detect signs of LV dysfunction, which may occur in patients without symptoms. It should be performed 3 months after the initial study to ensure stability and then periodically and whenever there is a change of symptoms.

2. AV replacement

   1. LV dysfunction and death are more common in symptomatic patients and those with depressed ejection fraction or an increased LV systolic volume.

      1. Mortality in symptomatic patients without surgery is 10–20% per year.

      2. Outcomes are worse in medically treated asymptomatic patients with a LV end-systolic diameter > 50 mm. Symptoms, LV dysfunction, and death develop at 19% per year, compared with 6% per year with a LV end-systolic diameter 40–50 mm and none with a LV end-systolic diameter < 40 mm.

   2. AV replacement is recommended for select groups with severe aortic regurgitation.

      1. All symptomatic patients

      2. Asymptomatic patients with an ejection fraction ≤ 50%

      3. Asymptomatic patients with a LV end-diastolic diameter > 75 mm or a LV end-systolic diameter > 55 mm

      4. When aortic regurgitation is secondary to dilatation of the aortic root, valve repair is recommended in patients with aortic regurgitation of any severity associated with an aortic root > 4.5–5 cm.

3. Replacement valves may be either mechanical or bioprosthetic (eg, porcine valves).

   1. Mechanical valves are more durable and are often chosen for young patients to minimize the need for subsequent AV replacement. However, patients with mechanical valves require lifelong anticoagulation.

   2. Bioprosthetic valves are used more often in older patients (> 70 years) with shorter life expectancies and patients at greater risk for bleeding while receiving anticoagulation therapy.

   3. Endocarditis prophylaxis is recommended for patients following AV replacement with a mechanical or bioprosthetic device and for repairs utilizing a bioprosthetic material. It is also recommended for patients with a history of endocarditis.

4. Afterload reduction

   1. Should not be substituted for AV replacement in patients with an indication for valve replacement

   2. Indications

      1. Severe aortic regurgitation

         • (1) Symptomatic patients or those with LV dysfunction as short-term preoperative therapy to improve their hemodynamic function.

         • (2) Symptomatic patients or those with LV dysfunction who are not surgical candidates

         • (3) For asymptomatic patients with LV dilatation but normal systolic function, the ACCF/AHA concluded that vasodilators “may be considered.” The evidence is inconclusive.

      2. Recommended for patients with any degree of aortic regurgitation and hypertension

   3. Not indicated in asymptomatic, normotensive patients with normal systolic function and mild to moderate aortic regurgitation

5. Beta-blockers are relatively contraindicated. Prolonged diastole increases regurgitation and accelerates progression.

Alternative Diagnosis: Aortic Stenosis

See Chapter 31, Syncope.

Alternative Diagnosis: Atrial Fibrillation

Textbook Presentation

Classically, patients with atrial fibrillation seek medical care for palpitations. The abrupt onset often prompts patients to be seen emergently. Patients may also complain of shortness of breath and dyspnea on exertion. Occasionally, atrial fibrillation is detected during a routine office visit when an irregularly irregular pulse is discovered and evaluated.

Disease Highlights

1. Atrial fibrillation is the most common clinical arrhythmia; its incidence increases with age (3.8% of patients ≥ 60 years old to 9% in those ≥ 80 years old).

2. May be episodic or persistent

3. Secondary to multiple wavelets of excitation that meander around the atria

4. Etiologies

   1. Most common etiologies are hypertension, CAD, and HF.

   2. Acute coronary syndrome: In 2–5% of patients presenting to the emergency department with new-onset atrial fibrillation, it is secondary to an acute MI.

   3. Other etiologies include alcoholic heart disease, valvular heart disease, cor pulmonale, thyrotoxicosis, and PE.

5. Complications

   1. Stroke: Stasis promotes thrombus formation within the atria. Subsequent embolization results in stroke and other systemic emboli.

      1. Atrial fibrillation accounts for one-sixth of all strokes.

      2. Stroke is more common in patients with atrial fibrillation who have other clinical risk factors:

         • (1) Valvular heart disease

         • (2) Prior transient ischemic attack or stroke

         • (3) Increasing age

         • (4) Hypertension

         • (5) Diabetes

         • (6) HF

         • (7) Sex (women affected 1.5–3.0 times more than men)

      3. The annual stroke rate in atrial fibrillation patients not receiving anticoagulation is 4.1% per year. However, there is substantial variation depending on the presence or absence of these other risk factors. For the subgroup of patients with a prior transient ischemic attack or stroke, the annual stroke rate increases to 13% per year.

   2. Worsening HF due to loss of atrial kick; especially important in patients with stiff LV (ie, diastolic dysfunction)

Evidence-Based Diagnosis

1. Easily recognized on ECG (Figure 15-3)

2. Episodic atrial fibrillation can be detected with Holter monitoring or event recorders.

Treatment

1. Evaluation

   1. ECG can document atrial fibrillation, as well as suggest underlying etiologies (ischemia or right heart strain in PE.)

   2. Baseline echocardiogram to assess LV function and stroke risk

   3. Thyroid function tests, electrolytes, BUN and creatinine are recommended.

   4. Consider evaluation for other etiologies (eg, MI, PE).

2. Rhythm control versus rate control

   1. Cardioversion should be performed immediately in unstable patients (with ischemia, hypotension, or HF).

   2. In stable patients, 2 options exist: rhythm control or rate control.

      1. Rhythm control attempts to restore normal sinus rhythm using cardioversion and antiarrhythmic agents.

      2. Rate control allows persistent atrial fibrillation. The ventricular response is controlled with atrioventricular nodal blocking agents (eg, beta-blockers, diltiazem, verapamil, or digoxin). Anticoagulation is used to prevent stroke.

      3. Studies show that rhythm control and rate control results in similar mortality and stroke rates, even in patients with underlying HF.

      4. Long-term anticoagulation: The American College of Chest Physicians (ACCP) recommends continuing long-term antithrombotic therapy in patients managed with a rate or rhythm control strategy based on their risk factors regardless of the appearance of persistent normal sinus rhythm (see below).

      5. Rate control is the recommended strategy in most patients. (Patients with their first episode of atrial fibrillation or with symptoms or exercise intolerance may choose rhythm control.)

         • (1) A resting HR of < 110 bpm is recommended.

         • (2) Uses beta-blockers, diltiazem, verapamil, or digoxin

         • (3) Beta-blockers, diltiazem, and verapamil should be avoided in patients with decompensated HF.

         • (4) Digoxin

           • (a) Less effective at controlling ventricular response during activity and in paroxysmal atrial fibrillation

           • (b) Should not be used as the sole agent for patients with paroxysmal atrial fibrillation

           • (c) Useful in patients with HFrEF

           • (d) Second-line drug

         • (5) Combination therapy

           • (a) Verapamil and beta-blockers should not be used concurrently in the same patient due to a high frequency of complications (bradycardia or HF).

           • (b) Digoxin and beta-blockers or digoxin and diltiazem or verapamil may be used concurrently to achieve HR control.

         • (6) AV ablation can be used to achieve rate control when pharmacologic therapy is unsuccessful or not tolerated.

         • (7) Amiodarone can be used for HR control if other measures fail.

      6. Rhythm control therapy

         • (1) Cardioversion options

           • (a) Flecainide, dofetilide, propafenone, ibutilide, amiodarone, or direct current cardioversion can be used to restore normal sinus rhythm.

           • (b) Electrical cardioversion is contraindicated in patients with hypokalemia or digoxin toxicity.

         • (2) The probability of conversion to normal sinus rhythm decreases the longer the atrial fibrillation lasts.

         • (3) The maintenance of sinus rhythm is complex and beyond the scope of this text.

         • (4) Cardiology consultation is recommended.

         • (5) Anticoagulation therapy for cardioversion.

           • (a) Anticoagulation is often used preceding and following cardioversion (whether electrical or chemical) to decrease the risk that preexistent or new thrombi form or embolize.

           • (b) Cardioversion should not be delayed in unstable patients. If the atrial fibrillation was < 48 hours in duration they should undergo cardioversion without delay. If the duration of atrial fibrillation was ≥ 48 hours or unknown, patients should undergo cardioversion while concurrently receiving an unfractionated heparin bolus and maintenance therapy to achieve an activated partial thromboplastin time of 1.5–2 times control until warfarin increases the INR to 2.0-3.0.

           • (c) Anticoagulation for 3 weeks prior to cardioversion is recommended in stable patients with atrial fibrillation ≥ 48 hours or of unknown duration. For patients with atrial fibrillation < 48 hours, anticoagulation decisions may be based on other patient risk factors, (eg, mitral valve disease, HF, or prior embolism).

           • (d) Low-risk patients (without the aforementioned risk factors) and atrial fibrillation of recent onset (< 48 hours) can undergo cardioversion without delay.

           • (e) Alternatively, a transesophageal echocardiography can be performed to look for thrombi. Patients without thrombi can receive unfractionated heparin and then undergo cardioversion. Patients with thrombi should receive 3 weeks of anticoagulation therapy prior to cardioversion.

           • (f) Patients typically receive a vitamin K antagonist (target INR 2.0–3.0) for 4 weeks following cardioversion. Patients with left atrial thrombi seen on a transesophageal echocardiogram should probably receive anticoagulation therapy longer.

3. Stroke prevention

   1. Antithrombotic therapy with anticoagulants, typically using vitamin K antagonists, or antiplatelet agents (typically aspirin) has been used to prevent strokes in atrial fibrillation patients.

   2. The 2 most widely studied therapies are warfarin and aspirin.

      1. Multiple studies suggest that warfarin is superior to aspirin at preventing stroke with a relative risk reduction of 64% compared to 19% for aspirin. In the absence of contraindications, the benefit of warfarin usually outweighs the risk.

      2. The standard INR target is 2.0–3.0.

      3. The absolute benefit of warfarin increases as the risk of stroke increases.

      4. Warfarin reduces the absolute rate of stroke on average by 2.7% per year, but in patients with prior transient ischemic attack/cerebrovascular accident, the absolute risk reduction is 8.4% per year.

      5. Contraindications to warfarin therapy include recent gastrointestinal or central nervous system hemorrhage, recent trauma or surgery, uncontrolled hypertension, noncompliance, syncope, or alcoholism.

      6. The ACCF/AHA have published guidelines for stroke prevention in persistent or paroxysmal atrial fibrillation (Table 15-9).

   3. Two special groups are worth mentioning.

      1. Although physicians worry about bleeding complications in the elderly, studies show that elderly patients with atrial fibrillation are at high risk for stroke and benefit from anticoagulation if they are carefully selected.

      2. Patients with lone atrial fibrillation (age < 60 years, no heart disease, hypertension or risk factors) are at the lowest risk for stroke (< 1% per year when treated with aspirin).

         • (1) The absolute risk reduction from warfarin is very small and similar in magnitude to the risk of hemorrhage from warfarin.

         • (2) The AHA/ACC/ACCP does not recommend warfarin in these patients.

         • (3) The utility of aspirin is uncertain and the AHA/ACC concluded that the effectiveness of aspirin for this population has not been established.

   4. New oral anticoagulants provide alternatives to vitamin K antagonist.

      1. The new anticoagulants directly inhibit thrombin (dabigatran) or factor Xa (rivaroxaban and apixaban).

      2. Rivaroxaban and dabigatran accumulate in renal impairment.

      3. Compared with warfarin, dosing is easier, anticoagulation does not require monitoring, and there are fewer drug and food interactions.

      4. However, unlike warfarin, antidotes are not available to reverse anticoagulation in hemorrhaging patients.

         • (1) Vitamin K and fresh frozen plasma are ineffective.

         • (2) Preliminary data suggest that prothrombin complex concentrate may reverse anticoagulation due to Xa inhibitors (eg, rivaroxaban).

      5. The precise role of these agents compared with vitamin K antagonist remains an area of active clinical research.

   5. Aspirin combined with clopidogrel has been compared with either aspirin or warfarin monotherapy in the prevention of stroke in patients with atrial fibrillation.

      1. Combination therapy is less effective than warfarin at stroke prevention but associated with the same rate of major bleeding. Annual event rate on combination therapy of 5.6% vs 3.9% on warfarin.

      2. Combination therapy was slightly more effective than aspirin alone at preventing stroke, (0.9% per year absolute risk reduction), but associated with a 0.7% per year absolute increase in the rate of major bleeding.

Alternative Diagnosis: CAD

See Chapter 9, Chest Pain.

Mr. C’s echocardiogram confirms HF and rules out significant valvular heart disease as the primary etiology of his dyspnea. Similarly his ECG does not demonstrate atrial fibrillation. The likely etiology of his HFrEF is ischemia given his history of MI, ECG, and regional wall motion abnormalities on echocardiogram. An angiogram would be recommended by the AHA/ACC guidelines if not already performed.

The angiogram confirms CAD as the cause of Mr. C’s HF.

Mrs. L’s shortness of breath is not only severe but markedly worse than baseline. Both of these features should prompt a thorough investigation. Unfortunately, the clinical information does not suggest a specific diagnosis (Figure 15-1). She has no pivotal clues that can help limit the differential diagnosis (fever, pleuritic chest pain, or other chest pain.) Furthermore, there are no clues that might suggest one of the common causes of dyspnea: No risk factors for HF (CAD, hypertension, or alcohol abuse that commonly cause HF), no history of wheezing or smoking to suggest asthma or COPD, no fever or cough to suggest pneumonia, and no associated symptoms or risk factors to suggest PE (chest pain, cancer, immobilization, prior VTE). A careful exam is vital to look for helpful clues.

Despite a thorough exam, the leading diagnosis is unclear. In such cases, it is particularly important to systematically review the differential diagnosis in order to arrive at the correct diagnosis (Table 15-2). Each item on the list should be reviewed in light of the history and physical to determine whether it remains in the differential and should be explored further, or whether the existing information makes it highly unlikely.

Reviewing Table 15-2, the absence of a murmur makes mitral regurgitation and aortic stenosis unlikely, since the clinical exam is 85–90% sensitive for these conditions. The clinical exam is less sensitive for aortic regurgitation (see above). Therefore, aortic regurgitation remains on the differential diagnosis. An arrhythmia is essentially ruled out by the patient’s normal heart rate during symptoms. HF is not particularly suggested by the history and physical exam, but it cannot be excluded given the low sensitivity of the S₃ gallop and JVD. The patient denies any history of chest pain, but dyspnea is occasionally an anginal equivalent, thus CAD remains a possibility. Pneumonia is highly unlikely given the normal chest film and the lack of fever and cough. Asthma remains a possibility although this is not particularly suggested by the history or physical exam. COPD is effectively ruled out by the trivial smoking history. PE cannot be excluded by the current information and remains on the list, although the presentation is not particularly classic for PE. Since PE is associated with a high mortality, it should be considered a must not miss possibility. A significant pleural effusion and pneumothorax are ruled out by the normal chest radiograph, which also makes interstitial disease unlikely (although not impossible). Anemia is ruled out by the normal CBC. We can now focus on the clinical clues and diagnostic tests for these remaining possible diagnoses (aortic regurgitation, HF, CAD, asthma, and PE). Table 15-10 lists the differential diagnosis.

A methodical approach to the differential diagnosis is vital whenever the leading diagnosis is unclear or when the leading hypothesis cannot be confirmed.

Considering each diagnosis in turn, the patient’s physical exam and echocardiogram exclude HF and aortic regurgitation. The patient’s pretest probability of CAD is quite low given her age, sex, and risk factors (3.2%; see Chapter 9, Chest Pain). In addition, the Framingham data suggest the likelihood of a coronary event in a female patient with these CAD risk factors to be < 1% over the ensuing 8 years. The history and normal pulmonary function tests with methacholine challenge make asthma very unlikely. Although her history sounds atypical for PE, she is taking hormone replacement therapy, a known risk factor for VTE. Given the exclusion of the other diagnoses, PE becomes more probable. You revise your differential diagnosis and make PE both your leading and must not miss diagnosis.

Leading Hypothesis: PE

Textbook Presentation

Classically, patients with PE experience the sudden onset of shortness of breath and severe chest pain that increases with inspiration. Patients may complain of hemoptysis and associated unilateral leg swelling.

Disease Highlights

1. Pathophysiology: Most commonly occurs when a lower extremity venous thrombosis embolizes to the lung. Upper extremity thrombi may also cause PE.

   1. 80% of patients with PE have deep venous thrombosis (DVT).

   2. 48% of patients with DVT have PE (often asymptomatic).

2. Symptoms vary markedly. Massive obstruction may result in RV failure and death, whereas lesser obstruction may be asymptomatic.

3. 3-month mortality is 17.5%

4. Risk factors

   1. A variety of risk factors increase the odd ratio for VTE including a personal history of VTE (2.9), estrogen use (2.3), surgery within the last 4 weeks (2.3), personal history of thrombophilia (2.0), active or metastatic cancer (1.9), immobilization (1.7), age over 50 (1.5).

   2. Thrombophilia

      1. Antiphospholipid antibodies: Present in 2–8.5% of patients with VTE

      2. Factor V Leiden

         • (1) Most common thrombophilia

         • (2) Mutation in factor V causes resistance to cleavage by activated protein C

         • (3) 11% of patients with DVT

         • (4) Confers a 2.7 × increased risk of VTE

         • (5) Combined with oral birth control pill, mutation increases risk 35 times

      3. Prothrombin gene mutation

      4. Protein C or S deficiency (rare)

         • (1) Protein C and S are naturally occurring anticoagulants

         • (2) Deficiency is associated with hypercoagulability.

         • (3) Synthesis of protein C and S requires vitamin K.

         • (4) Warfarin decreases synthesis of both factors.

         • (5) Assays for protein C and S must be performed while patients are not taking warfarin.

      5. Antithrombin III deficiency (also rare): Assay must be done while patient is not taking heparin.

      6. Hyperhomocysteinemia: 3 × increased risk of VTE

      7. Increased factor VIII: 6 × increased risk

5. PE commonly presents as a “COPD exacerbation.”

   1. Studies have reported that 16% of patients with a COPD exacerbation have pulmonary emboli.

   2. The rate was 25% among patients with an unexplained exacerbation of COPD, compared with 8% in those with an exacerbation of known etiology.

   3. Unexplained exacerbation was defined as patients without parenchymal consolidation on chest radiograph without fever or chills (ie, not obviously due to pneumonia) and patients who lacked the common factors precipitating COPD exacerbations:

      1. Lower respiratory tract infection (increased sputum, purulence, fever, cold or sore throat)

      2. Exposure to noxious irritants

      3. Objective signs of HF

      4. Noncompliance

         Consider pulmonary emboli in patients with a COPD exacerbation that is unexplained.

Evidence-Based Diagnosis of PE

1. Clinical presentation: The diagnosis of pulmonary emboli is complicated because patients may present in a variety of ways and signs and symptoms are neither sensitive nor specific.

   1. Although dyspnea and chest pain are the most common symptoms, neither is sufficiently sensitive to rule out the diagnosis when absent nor sufficiently specific to rule in pulmonary emboli when present.

   2. The accuracy of the signs and symptoms are shown in Table 15-11.

   3. The HR and RR are similar in patients with and without PE and neither confirm nor rule out PE.

      The classic presentation of PE is actually the exception. Patients may have very few symptoms. A high index of suspicion must be maintained for the diagnosis of PE.

2. The chest film, arterial blood gas level, and ECG cannot reliably rule out PE.

   1. Chest film: May reveal focal oligemia (45%), wedge-shaped infiltrate (15%), or pleural effusions (45%) but is normal in 12–50% of patients with PE.

   2. Arterial blood gas measurement: May demonstrate hypoxemia and hypocarbia, but findings are neither sensitive nor specific for PE. PaO₂ < 80 mm Hg: sensitivity 58–74%; LR+ 1.2, LR– 0.8

      Patients with PE may not be hypoxic. Therefore, normal arterial oxygen does not rule out PE. On the other hand, unexplained hypoxia, particularly in the company of a normal chest radiograph, should increase the suspicion of PE.

   3. ECG

      1. Useful to diagnose other conditions (ie, MI).

      2. Certain findings suggest PE but are unusual: S1Q3T3 (19–50% sensitive), transient right bundle-branch block (6–67% sensitive).

      3. T wave inversions in V1–V4 when seen in conjunction with T wave inversions in lead III strongly suggests PE (88% sensitive, 99% specific; LR+ 88, LR− 0.12).

3. Decision rules and D-dimer testing

   1. Decision rules and D-dimer testing are useful in combination but should not be used separately to rule out PE.

      1. Wells, Geneva, and others decision rules have been developed to estimate the pretest probability of PE. (See below) A recent meta-analysis demonstrated that they are not sufficiently accurate to rule in or rule out PE without additional testing (LR+ 1.4–3; LR– 0.24–0.5).

      2. PERC rule

         • (1) The PERC rule is a recently developed clinical decision rule designed to use clinical criteria to rule out PE without other objective testing (D-dimer or CTA)

         • (2) Criteria include age < 50 years, HR < 100 bpm, SaO₂ > 94% on room air, no prior history of VTE, no trauma or surgery in the last 4 weeks, no hemoptysis, no exogenous estrogen, and no unilateral leg swelling.

         • (3) Patients with none of the criteria are classified as PERC–.

         • (4) One large study suggested that PERC– patients are at low risk for PE (1%) and could safely avoid further testing for PE (D-dimer or CTA). However, the diagnostic protocol for PE in this study was not rigorous and may have missed patients with PE.

         • (5) A more recent study found the incidence of PE in PERC– patients to be 5.4% (95% CI, 3.1 – 9.3%).

         • (6) Should not be used as a stand-alone tool to rule out PE.

      3. D-dimers

         • (1) D-dimers are fibrin breakdown products that are often elevated in patients with VTE but are nonspecific and elevated in many other conditions: surgery, trauma, cancer, and end-stage renal disease.

         • (2) Enzyme-linked immunosorbent assay (ELISA) and quantitative rapid ELISA are more sensitive than other assays (95–98%; LR– 0.05–0.11).

         • (3) Need to be used in conjunction with decision rules since not sufficiently sensitive to rule out PE in patients at high risk.

   2. Decision rules combined with D-dimer testing

      1. The precise rule is dependent on the sensitivity of the D-dimer assay used.

         • (1) A negative quantitative D-dimer assay in a patient with a Wells score of ≤ 4 has a failure rate (missed diagnoses) of 0.5%.

         • (2) A qualitative D-dimer assay in a patient with a Wells score of ≤ 2 has a failure rate of 0.9%.

         • (3) Patients with an elevated D-dimer value or a Wells score > cutoff (2 or 4) should be evaluated with a CTA (or a ventilation-perfusion [V/Q] scan if CTA is contraindicated) (Figure 15-4).

4. CTA

   1. Test of choice in moderate- to high-risk patients

   2. May demonstrate filling defects in proximal pulmonary arteries

   3. Makes alternative diagnosis in 25% of patients (lymphadenopathy, tumor, aortic dissection)

   4. Accuracy

      1. A wide range of sensitivities and specificities have been published due to variations in CT methodology (single vs multidetector), standard for diagnosis, duration of follow-up and differing patient populations (sensitivity, 83–90%; specificity, 94–100%).

      2. Follow-up studies of patients with negative CTA have documented subsequent VTE in 1.3–5.3%. Rates are higher in patients with a high pretest probability of disease.

      3. The PIOPED II study reported on the accuracy of multidetector CT using stringent diagnostic criteria.

         • (1) Multidetector CTA was 83% sensitive and 96% specific for PE (LR+ 19.6, LR– 0.18).

         • (2) The negative predictive value was 96% in low-risk patients (Wells score < 2), 89% in intermediate-risk patients (Wells score 2–6) and only 60% in high-risk patients (Wells score > 6).

5. V/Q scan

   1. Radionuclear study used less frequently since advent of CTA

   2. Radio-isotope infused and inhaled. V/Q images are compared yielding the following potential results:

      1. High probability scan

         • (1) Multiple areas of normal ventilation but no perfusion

         • (2) Effectively rules in PE

           • (a) 60% sensitive, 96% specific

           • (b) LR+ 15, LR− 0.4

      2. Normal or near normal perfusion scan effectively rules out PE. (Normal scans are seen in 0–2% of patients with PE.)

      3. Nondiagnostic scan (low or intermediate probability)

         • (1) Matched areas of ventilation and perfusion abnormalities

         • (2) 67% of patients who undergo V/Q testing have this pattern

         • (3) Neither rules in or out PE

6. Angiography

   1. Gold standard

   2. Invasive and rarely used; serious complications occur in 0–3% of patients.

7. Strategy

   1. First use the Wells score to estimate the patients’ probability of disease.

      1. Clinical signs and symptoms of DVT: 3 points

      2. PE is as likely or more likely than alternative diagnoses: 3 points

      3. Heart rate > 100: 1.5 points

      4. Immobilization ≥ 3 days or recent surgery (< 4 weeks): 1.5 points

      5. Prior VTE (DVT or PE): 1.5 points

      6. Hemoptysis: 1 point

      7. Active cancer (treated in last 6 months or palliative): 1 point

   2. A negative D-dimer in conjunction with a low pretest probability of PE effectively rules out PE (Figure 15-4). All other patients should have CTA unless contraindicated.

      1. Positive CTA: effectively confirms the diagnosis of PE

      2. Negative CTA

         • (1) Effectively rules out PE in low- to moderate-risk patients

         • (2) Additional testing may be appropriate in intermediate-risk patients.

         • (3) Additional testing is recommended in high-risk patients (Wells > 6).

         • (4) Options for additional testing include duplex leg ultrasonography, indirect CT of the leg veins CT venography, V/Q scanning, and pulmonary angiography.

           • (a) Since a diagnosis of DVT is taken as a surrogate marker for the diagnosis of PE, duplex leg ultrasonography is the most commonly used test in this situation.

           • (b) Another alternative to rule out DVT is CT venography, which can be performed at the same time as CTA. However, some studies suggest that it is less sensitive than leg ultrasound (sensitivity 60–100%, specificity 93–100%).

           • (c) A combined strategy of CTA and duplex leg ultrasonography has an excellent negative predictive value (≈100%) and is advisable for patients with a high likelihood of PE.

   3. In patients with acute or chronic kidney disease or contrast allergy, V/Q scanning can be considered in place of CTA (see below).

   4. Pulmonary angiography is rarely used.

   5. The approach is illustrated in Figure 15-5.

Evidence-Based Diagnosis of DVT

1. Given the overlap of PE and DVT (and the same therapy), the diagnosis of DVT is often taken as evidence of concomitant PE.

2. Several clinical features modestly increase the likelihood of DVT including malignancy (LR+ 2.7), prior DVT (LR+ 2.3), immobilization (LR+ 2.0), and recent surgery (LR+ 1.8).

3. Signs and symptoms are not very helpful at ruling DVT in or out. Leg swelling is only 32% sensitive, LR+ 1.45, LR– 0.67; leg pain LR+ 1.08, LR– 0.9; Homan sign LR+ 1.4, LR– 0.87.

   The clinical exam for DVT is insensitive. Clinicians must have a low threshold for ordering D-dimer or duplex studies.

4. The diagnostic strategy for DVT integrates the pretest probability of DVT with diagnostic tests of appropriate sensitivity to achieve a satisfactory negative predictive value. In short, as the probability of DVT increases, strategies utilize more sensitive testing strategies to rule out DVT.

   1. The Wells score can help predict risk of DVT.

      1. Patients are assigned 1 point for each of the following: active cancer, immobilization of the lower extremity, recently bedridden (≥ 3 days) or surgery within the last 12 weeks requiring general anesthesia, tenderness along the deep venous system, swelling of the entire leg, calf swelling ≥ 3 cm larger than the unaffected leg, pitting edema of the symptomatic leg, collateral superficial veins, and prior DVT.

      2. 2 points are subtracted if an alternative diagnosis is at least as likely as a DVT.

      3. The prevalence of DVT is 5% for patients with low probability score (≤ 0), 17% for patients with a moderate probability score (1–2), and 53% for patients with a high probability score (≥ 3).

   2. Diagnostic tests

      1. D-dimer

         • (1) D-dimers are very sensitive for VTE but not specific (88–92% sensitive, 45–72% specific).

         • (2) Used in strategies to help rule out DVT when negative.

         • (3) In patients with a low clinical risk, the negative predictive value of a negative D-dimer is 99%.

         • (4) D-dimer never confirms DVT. When elevated, ultrasonography is indicated.

      2. Duplex ultrasonography

         • (1) Ultrasonography may visualize the whole leg or be limited to the proximal leg.

           • (a) Proximal ultrasonography can rule out proximal DVT but will not detect distal DVT (that rarely embolize but can extend and then embolize).

           • (b) 89–96% sensitive for symptomatic proximal DVT, 94–99% specific; LR+, 24; LR–, 0.05

           • (c) Even whole leg ultrasound less sensitive for distal (below the knee) DVT 73–93%

      3. Other options include venography (invasive) and magnetic resonance direct thrombus imaging (accurate but costly).

   3. Diagnostic strategy

      1. In short, DVT is ruled in by a positive ultrasound.

      2. DVT may be ruled out in the following situations:

         • (1) Low probability Wells score: Negative D-dimer test

         • (2) Moderate probability Wells score: Negative highly sensitive D-dimer test

         • (3) High probability Wells score: Negative whole leg ultrasound or negative proximal leg ultrasound and negative D-dimer

         • (4) In patients with a negative proximal leg ultrasound and a positive D-dimer, repeat ultrasound is often recommended to ensure a distal DVT has not extended.

      3. Figure 15-6 illustrates the testing strategy recommended by the ACCP for evaluating patients for their first DVT.

Treatment

1. Oxygen should be administered to patients with hypoxemia.

2. Options include anticoagulation (with a variety of agents) or thrombolytic therapy.

3. Thrombolytic therapy is rarely used but recommended in hemodynamically unstable patients in whom the benefits outweigh the risks. If such patients have contraindications to thrombolysis, surgical embolectomy is recommended.

4. Anticoagulation: Overview

   1. A variety of options for anticoagulation exist, including intravenous unfractionated heparin, subcutaneous low-molecular-weight heparin (LMWH), oral vitamin K antagonists (eg, warfarin), oral thrombin inhibitors (dabigatran), and oral factor Xa inhibitors (rivaroxaban and apixaban). See overview of anticoagulants in section above (Atrial Fibrillation, Prevention of Stroke).

   2. Research is ongoing to determine optimal therapy and the choice of anticoagulants is likely to evolve rapidly. Two recent studies suggest apixaban and rivaroxaban are as effective and safer than vitamin K antagonists in patients with VTE. Further studies are needed.

5. 2012 ACCP Clinical Practice Guidelines for treatment of VTE

   1. Initiation of anticoagulation

      1. Initial treatment with either LMWH or fondaparinux followed by vitamin K antagonist (target INR 2.0–3.0).

      2. Anticoagulation should be continued for a minimum of 5 days and until the INR is > 2.0 for ≥ 24 hours.

   2. Isolated distal DVT (below the knee)

      1. Hold anticoagulation and repeat imaging over 2 weeks to rule out extension.

      2. Initiate anticoagulation (3 months) if patient has severe symptoms, extension of clot (even if still distal), or risk factors for extension (active cancer, recent surgery or transient risk factor), prior VTE, inpatient status, extensive thrombosis).

   3. Acute proximal DVT or PE

      1. Duration of anticoagulation

         • (1) Modified by the bleeding risk and inciting cause of the DVT or PE.

         • (2) Anticoagulation is effective during therapy. After discontinuation, patients with persistent risk factors or idiopathic VTE are at increased risk for recurrence.

         • (3) Surgery provoked: 3 months.

         • (4) Nonsurgical transient risk factor: 3 months

         • (5) Unprovoked:

           • (a) Patient at substantial risk for bleeding: 3 months

           • (b) Patient at low to moderate risk for bleeding: Extended therapy

         • (6) Recurrent DVT/PE: Extended therapy unless the patient is at high risk for bleeding.

         • (7) DVT/PE associated with active cancer: Extended therapy is recommended. LMWH is recommended over vitamin K antagonist.

6. DVT/PE in patients with a contraindication to anticoagulation or in whom anticoagulation therapy has failed: IVC filter is recommended.

7. Symptomatic DVT: Compression stockings are also recommended for 2 years to decrease the likelihood of postthrombotic syndrome.

8. Work-up for thrombophilia: Guidelines for routine testing have not been determined. Consider tests for thrombophilic states in patients without clear precipitant of VTE.

9. Primary prevention of VTE

   1. 25% of venous thromboembolic events are associated with hospitalization. At particularly high risk are patients undergoing hip fracture surgery, hip or knee replacement, and those with spinal cord injury.

   2. Hospitalized medical patients account for 50–75% of hospital-associated VTE.

      1. Anticoagulation prophylaxis has been demonstrated to reduce fatal and nonfatal VTE.

      2. At particularly high risk include patients who are older than 40 years, who are hospitalized for ≥ 3 days with limited mobility, and who have ≥ 1 of the following: Acute infectious disease, class III or IV HF, acute MI, cerebrovascular accident, active cancer, acute respiratory disease, rheumatic disease, body mass index > 30, recent surgery or trauma, thrombophilia, or prior VTE.

   3. The intensity of the prophylaxis depends on the specific clinical situation. Options include compression stockings, pneumatic compression devices (particularly useful in patients who have active bleeding or are at high risk), and anticoagulation. The ACCP updates guidelines frequently.

You review the patient’s Wells score and assign her 3 points since PE is the most likely diagnosis. She also takes hormone replacement therapy, which is a risk factor for PE. Although there is some debate about the use of D-dimer to rule out PE in a patient with a score of 3, you elect to perform a CTA rather than a D-dimer because PE is the most likely diagnosis.

The CTA is highly specific for PE. At this point, PE is ruled in and further confirmation is unnecessary. There is no need for further testing to exclude alternative diagnoses.

Alternative Diagnosis: Asthma

See Chapter 33, Wheezing and Stridor.

Alternative Diagnosis: CAD

See Chapter 9, Chest Pain.