Chief symptom: Shortness of breath for 3 days
History: A 60-year-old white man with a medical history of congestive heart failure (CHF) secondary to ischemic heart disease, chronic obstructive pulmonary disease (COPD), and hypertension was brought into the emergency department with 3 days of worsening shortness of breath and bilateral lower extremity swelling. He reported paroxysmal nocturnal dyspnea and 3-pillow orthopnea for 48 hours. He reported no chest pain, palpitations, cough, sputum production, fever, or diaphoresis. He was hospitalized 4 weeks ago with similar symptoms and diagnosed with acute exacerbation of CHF. At the time of this admission, he was unable to provide the name or dose of his current medications. Review of his electronic medical records revealed that he had been discharged 3 weeks earlier on furosemide, 20 mg once a day; metoprolol tartrate, 25 mg twice a day; aspirin, 81 mg once a day; and an albuterol inhaler, 2 puffs every 6 hours as needed.
The patient did not adhere to his home medication regimen due to financial issues. He was unemployed and had recently lost his insurance coverage. He was unaware of the dietary requirements for managing his condition. He reported no other pertinent medical or surgical history. His father had a heart attack at the age of 66. The patient had a 40 pack-year history of smoking but reported no alcohol use, intravenous drug abuse, or prescription narcotic abuse. He had no known drug allergies.
Physical examination: Vital signs were as follows: blood pressure 130/80 mm Hg, heart rate 104 beats/min, temperature 98.7°F, respiratory rate 24 breaths/min, and oxygen saturation (Spo2) 88% on ambient air that improved to 92% with 2 L oxygen per minute by nasal cannula. The patient appeared in mild respiratory distress; however, there was no use of accessory muscles. On cardiac examination, there was regular rhythm, tachycardia, normal S1 and S2 heart sounds with audible S3 gallop, and visible jugular venous distention (JVD) of 6 cm above the sternal angle. His pulmonary exam revealed bibasilar crackles without wheezing. His abdominal exam revealed palpation of the liver 2 finger-breadths below the costal margin of the ribs with no appreciable tenderness. He had 3+ lower-extremity pitting edema bilaterally up to his knees. The rest of the examination was unremarkable.
Labs: B-type natriuretic peptide (BNP) level was 11,000 pg/mL (5,000 pg/mL on last admission); blood urea nitrogen (BUN) level was 45 mg/dL (baseline, 20 mg/dL); and creatinine (Cr) level was 2.1 mg/dL (baseline, 1.2 mg/dL). Liver function tests revealed elevated aspartate transaminase level (AST) of 60 U/L (baseline, 30 U/L) and alanine aminotransferase (ALT) level of 65 U/L (baseline, 35 U/L). Complete blood count (CBC) and comprehensive metabolic panel (CMP) were otherwise within normal limits.
Imaging: Electrocardiogram (ECG) revealed sinus tachycardia at 104 beats/min, normal intervals, and nonspecific ST- and T-wave changes. Chest X-ray revealed a markedly enlarged cardiac silhouette, increased pulmonary vascular congestion, and bilateral pleural effusions. Transthoracic echocardiography (TTE) showed severe systolic dysfunction with an ejection fraction (EF) of 30%, unchanged from the previous admission 4 weeks earlier.
The patient was admitted to the ICU with the diagnosis of congestive heart failure. He was managed with a single dose of IV furosemide, 40 mg, followed by oral furosemide, 40 mg every 12 hours, which improved his symptoms over the next 24 hours. Staff decided to discharge the patient based on resolution of his presenting symptoms. At the time of discharge, his vital signs were as follows: blood pressure 128/80 mm Hg, heart rate 90 beats/min, temperature 98.6 °F, respiratory rate 18 breaths/min, and Spo2 95% on ambient air. His physical exam revealed the continued presence of congestion, which included positive JVD, trace pedal edema, and an S3 heart sound. He was discharged on oral furosemide, 40 mg/d, with continuation of all his prior discharge medications. He was advised to follow up with his primary care physician in 2 to 3 weeks. Ten days later, he returned to the hospital with shortness of breath at rest and hypoxia (Spo2 85%).
Q: Was our patient diagnosed correctly?
A: Yes, from a clinical perspective. However, from a documentation perspective, the primary and secondary diagnoses should have been documented as “acute on chronic systolic heart failure” and “acute hypoxic respiratory failure” instead of “congestive heart failure” alone. This improper documentation would have resulted in inappropriate coding of the diagnosis-related group (DRG), leading to lower geometric mean length of stay (GMLOS) and lower reimbursement to the hospital.
Q: Was our patient managed and discharged appropriately?
A: No. He should have been on IV diuretics until optimization of his fluid status. While symptomatically improved, he was discharged inappropriately early from the hospital while still having signs of congestion. Additionally, he was not discharged on guideline-directed medical therapy (GDMT). GDMT for heart failure includes an angiotensin-converting enzyme (ACE) inhibitor and a beta-blocker. Beta-blockers shown to be effective in heart failure include bisoprolol, carvedilol, or metoprolol succinate (1-3). Given the chronic nature of his heart failure with low EF, the addition of an aldosterone antagonist is also indicated (4). However, given his underlying renal insufficiency and his current social circumstances with inadequate outpatient care, withholding aldosterone antagonist therapy until the patient has consistent outpatient follow-up where electrolyte levels can be monitored closely could be appropriate. Since the patient had ischemic heart disease and is at very high risk for recurrent events, a high-intensity statin should also have been included in his discharge regimen (5, 6).
Q: Were care transitions handled appropriately at discharge?
A: No. Poor coordination of the transition and poor communication with the patient must be considered as contributing factors to his readmission. An effort to ensure access to follow-up with a primary care physician should have been made prior to discharge. A patient care coordinator should have been in contact with the patient to encourage and facilitate outpatient care adherence. Patient education, vital for outpatient care adherence and outpatient success, was missing at the time of discharge. Education regarding diagnosis and disease management, including medication, lifestyle changes, self-care measures, and outpatient care expectations, are central to a patient's success. The patient might also have benefited from another day of hospitalization, during which transitional measures, medication optimization, and reconciliation could have been completed.
The main goal of initial inpatient treatment for acute CHF exacerbation is reducing CHF symptoms, with secondary aims of improving quality of life and slowing progression of cardiac failure to decrease mortality.
In the majority of cases, a CHF exacerbation has multiple causes. According to a large multicenter study, common factors implicated in worsening of CHF status included non-adherence to dietary restriction; infectious processes, notably pulmonary infections; and inappropriate reductions in CHF therapy (7). Many of these factors are modifiable and avoidable, meaning that readmissions could be averted.
CHF exacerbation is a clinical diagnosis based on at least 1 of the following symptoms: exertional dyspnea, fatigue, paroxysmal nocturnal dyspnea, orthopnea, cough, early satiety, weight gain, and increasing abdominal girth. Physical exam findings may consist of an S3 gallop, JVD, bibasilar crackles, abdominal distention, hepatomegaly, and peripheral edema. The initial workup should include CBC, CMP, BNP, 12-lead ECG, and a chest X-ray. While there is no consensus on the utility of obtaining serial BNP measurements, measurement of BNP or N-terminal pro-B-type natriuretic peptide may be useful in the diagnosis of acutely decompensated CHF, especially in the setting of clinical uncertainty. A TTE should be completed if there is no documentation of a previous study. A TTE may be repeated when there is a change in clinical status. Decompensation of CHF is not an indication for a repeat echocardiogram.
According to 2013 guidelines from the American College of Cardiology Foundation/American Heart Association, all patients who are volume overloaded and classified as New York Heart Association (NYHA) class II-IV should be on IV loop diuretics during hospitalization (1). There is insufficient evidence to recommend bolus doses over continuous infusion of loop diuretic therapy (8). However, if patients are already receiving loop diuretic therapy, the initial IV dose should equal or exceed their oral daily dose and be given as either intermittent boluses or continuous infusion. Urine output and signs and symptoms of congestion should be assessed serially. The diuretic dose should be titrated for relief of symptoms, to reduce volume excess, and to avoid hypotension (9). High doses should be avoided as they are linked with an increase in mortality. During CHF exacerbation, oxygen supplementation should also be provided for hypoxemia.
Once acute symptoms of exacerbation have resolved, patients should be placed on both an ACE inhibitor and a beta-blocker (1). ACE inhibitors (1, 10), beta-blockers (1-3, 11, 12), and aldosterone antagonists (4) have been shown to reduce both mortality and hospitalizations. ACE inhibitors as a class have shown mortality benefit, and there is no single “preferred” ACE inhibitor. In contrast, among the beta-blockers, the only 3 shown to provide a mortality benefit are bisoprolol, carvedilol, and metoprolol succinate. Hydralazine and nitrates were shown to be particularly beneficial, as early as 1987, in reduction of mortality in younger white veterans with a lower ejection fraction and those with a history of hypertension (13). In African-Americans with NYHA class III-IV HF who are persistently symptomatic, hydralazine and nitrates have shown to reduce mortality as well (14, 15). Aldosterone antagonists have been shown to reduce mortality in NYHA class II-IV CHF patients with an EF less than or equal to 35%. They also reduce mortality in post-acute myocardial infarction (MI) patients with an EF less than 40% or diabetes mellitus with clinical findings of CHF (16).
Many angiotensin-receptor blockers (ARBs) have been evaluated in patients with heart failure. Valsartan has been shown to reduce mortality, although studies have shown mixed results regarding other ARBs (17-20). Another medication used frequently in heart failure is digoxin. Patients with heart failure who have been taking digoxin for a long time have an increased risk of a recurrent CHF event if digoxin is withdrawn (21). In women and elderly patients, digoxin should be used cautiously, as there are studies that have shown digoxin to increase mortality in this population (1).
Statins reduce hospitalizations in patients with a recent or remote history of MI. Statins are shown to have a strong, independent association with improved survival of patients with heart failure (22). In the absence of coronary artery disease and prior MI, there is no added benefit from statins if used primarily for CHF management, especially in the elderly population (5, 6).
Multiple steps should be considered prior to discharging a patient who has been admitted for a CHF exacerbation (23, 24). Determining when acute symptoms have resolved is crucial in evaluating a patient's readiness for discharge. Physicians, while addressing exacerbating factors, must also address a patient's activity level, dietary habits, and weight monitoring. Patients should also be counseled on smoking cessation. A thorough review of changes in medications must be completed with the patient, who should demonstrate a full understanding prior to discharge. Physicians must describe warning signs and symptoms of an exacerbation and provide clear instructions on how to respond.
Multidisciplinary teams must be used to ensure that transition-of-care measures are completed, which include but are not limited to social support, access to medications, and appointment confirmations with primary care 7 to 10 days following discharge. Some patient factors that are predictors of readmission are medication non-adherence; smoking history; male sex; prior admission within 6 months; current length of stay more than 7 days; creatinine level above 2.5 mg/dL at discharge; BNP of 1,000 pg/mL or greater; associated pulmonary, liver, or kidney disease; heart rate above 90 beats/min at discharge; and socioeconomic factors. Some factors in care that are predictors of readmission include inadequate discharge instructions, inadequate diuretic dosage, too early discharge, and failure to connect with ambulatory care (25-28).
Proper documentation is essential to facilitate continuity of care, communication between clinicians, and appropriate billing and reimbursements. For heart failure, it is important to specify the primary diagnosis as acute and/or chronic and as systolic and/or diastolic CHF. Specific documentation of the severity of a patient's condition along with comorbidities is crucial for appropriate DRG classification and GMLOS (Table) (29, 30).
Additionally, CMS's Value-Based Purchasing (VBP) initiative requires documentation of core measures for patients with a diagnosis of CHF: evaluation of left ventricular systolic dysfunction with EF less than 40% and initiation of an ACE inhibitor or ARB (29, 30). In the case of our patient, a VBP program would have lowered reimbursements to the hospital because the patient had a readmission within 30 days and incomplete documentation of core measures.
In conclusion, this case demonstrates the potential pitfalls of managing a patient with heart failure: failures of management, documentation, and transitions of care leading to readmission.