Wallach's Interpretation of Diagnostic Tests: Pathways to Arriving at a Clinical Diagnosis (52 page)

   ECG with tachyarrhythmia such as atrial fibrillation/AVNRT may suggest tachycardia-induced cardiomyopathy (typically requires 120–200 bpm for prolonged periods). Heart block, often complete, is seen with cardiac sarcoid or Lyme carditis.

   
Chest radiography
: Useful to differentiate CHF from pulmonary disorders. Cardiomegaly, cephalization of pulmonary vessels, Kerley B lines, effusions, and valvular calcification are associated with CHF. Diagnostic sensitivity of x-ray findings is high (83%), with lower specificity (68%).

   
Coronary angiography/stress imaging
: Coronary angiography is a Class I ACC/AHA recommendation for patients presenting with HF and angina-like symptoms unless the patient is not a candidate for revascularization. It is also recommend in young patients with systolic HF to exclude congenital coronary anomalies. Noninvasive stress imaging to detect ischemia or myocardial viability is reasonable (Class IIa) in HF patients with established CAD.

   
Viability testing
(PET with FDG, MRI, SPECT, Dobutamine stress echocardiogram): Over 80% of ischemic HF demonstrates viability, but only modest benefit seen with revascularization in the only randomized trial to date.

   
Right heart (pulmonary artery) catheterization
: Not recommended (Class III) to be performed routinely in HF patients due to lack of evidence in guiding medical therapy. Class I (recommended) to guide therapy when intracardiac filling pressures cannot be established by clinical measures or for worsening renal function. Also recommended for the consideration of advanced treatment options (Class II a).

   Initial
laboratory evaluation
for patients with HF include a complete blood count to assess for anemia or infection that may mimic, or worsen, HF-associated dyspnea. Blood urea nitrogen and creatinine may help differentiate states of volume overload due to renal disease, or if HF is confirmed, demonstrate cardiorenal syndrome as a consequence of low-output HF. Urinalysis in CHF is notable for mild albuminuria (<1 g/day) with isolated RBCs and WBCs, hyaline, and granular casts (less common). Specific gravity is high (>1.020) as opposed to primary renal disease. Hyponatremia generally indicates HF, although may be reflective of excessive diuresis (along with potassium depletion). Lowered serum sodium is a powerful predictor of HF mortality.

   Liver function tests may be altered due to hepatic congestion and is often accompanied by elevated INR with signs of right heart failure. Primary liver dysfunction is also associated with a dilated cardiomyopathy. Serum albumin and total protein are decreased in HF. The presence of elevated total protein may indicate HF due to infiltrative disease (amyloid, sarcoid).

   ESR may be decreased due to decreased serum fibrinogen.

   
BNP or NT-proBNP
is suggested in evaluation of all patients in whom CHF is suspected when the diagnosis is uncertain and is a class I indication to support the diagnosis of decompensated HF in hospitalized patients. Both have similar concentrations in normals but are significantly elevated in HF (NT-proBNP are fourfold higher). Natriuretic peptides are elevated in both systolic and diastolic heart failure and aid in the diagnosis with similar accuracy for both conditions.

   A number of factors affect natriuretic peptide levels including age, renal function, nutritional state, and gender. Natriuretic peptides are also known to be elevated in sepsis, atrial fibrillation, pulmonary hypertension, valvular heart disease, coronary artery disease, and other conditions that may present with dyspnea. Elevated levels alone should not be used as the sole determinant for the presence of CHF nor exclude the presence of other diseases.

   There are little data directly comparing BNP to NT-proBNP, but both have been show to aid in diagnosis and prognosis of HF, as well as assessing efficacy of HF therapy and predicting readmission rates. An NT-proBNP level of >900 pg/mL has equivalent accuracy of a BNP >100 pg/mL for diagnosis (predictive accuracy of 83%). Values below this have a very high negative predictive value for HF as a cause of dyspnea.