Heart Failure Readmissions

Opinion Statement

Heart failure readmissions (HFR) represent a personal burden for patients and a large financial burden for the healthcare system. As such, strategies to decrease HFR are avidly sought and studied. There are many reasons for HFR which challenges programs aimed to reduce the frequency of HFR. Large pharmacological and device trials often incorporate hospital admission as an endpoint and many programs have been developed in varied settings to address HFR. Some of the most successful programs use a multidisciplinary team approach, intensive patient education and system commitment. Many risk factors for HFR have been identified although prediction tools are limited. The reduction of HFR should incorporate a multidisciplinary approach with 1) evidenced-based physician-guided medical and device therapy, 2) institutional programs for effective care transitions, 3) strategies aimed to improve disease management, and 4) engage patients in self-care.

Introduction

HF accounts for 6.5 million hospitalizations per year and is directly or indirectly responsible for 60 to 70% of admissions over the age of 65.(1, 2) Approximately 30% of all patients admitted for HF are readmitted within 30 days.(3) Among all-cause admissions, 27% of Medicare patients are rehospitalized with 37% of these patients presenting for HF at the second admission.(4, 5) The number of total hospitalizations predicts mortality in patients with HF.(6)

Of all HFR, 48% are related to HF with 38% of diagnoses are noncardiac.(7, 8) The major causes of non-HFR include myocardial infarctions, atrial fibrillation, COPD, acute infections (with pneumonia being the most common), and iatrogenic causes.(9) (Table 1) Many readmissions are also linked to medication non-adherence, lack of follow-up after discharge, low socioeconomic status, and poor English skills. (5, 9) As expected, comorbidities, severity of initial presentation, and increased mortality are associated with readmissions, although length of stay of index admission has not been associated.(6, 10-12) Socioeconomic status is likely also an important factor in readmissions indicated by those with Medicaid having higher readmission rates than those with commercial insurance.(13)

Table 1.

Factors Associated with Readmissions(5, 9)

Index Admission Factors	Acute coronary syndrome
	Arrhythmia
	Uncontrolled HTN
	Elevated BNP
	Elevated troponin I
	Dyspnea at rest
	Cognitive impairment
	Medicaid insurance
	Low socioeconomic status
	Poor English skills
Hospital Course	Lack of low sodium diet during stay
	Development of pneumonia
	Decline in renal function
	HF as a new diagnosis
	Intensification of beta-blockade
	Discontinuation of digoxin
	Discharge medications (CCB, NSAIDs, glipizide, gabapentin)
	Poor handoff (e.g. poor communication with nursing home)
Post-discharge factors	Lack of follow-up within 30 days
	Lack of transportation to appointments Alcohol abuse
	Medication noncompliance
	Non-adherence to low sodium diet

The cost of first-time HF admissions is estimated at $28 billion per year.(1) This cost increases significantly for all-cause readmissions to a calculated $44 billion per year.(4) Unplanned HF readmissions are estimated to cost $17.4 billion in Medicare funds alone.(3) An estimated 75% of all readmissions can be avoided, resulting in estimated saving of $12 billion per year.(14) To improve costs and reduce avoidable readmissions, the US Congress passed the Patient Protection and Affordable Care Act, which penalizes hospitals for readmissions for all causes within 30 days of discharge.(15, 16) The financial penalty for readmissions aims to incentivize both insurers and hospitals to develop better strategies to reduce repeat hospitalizations.

Table 1. Factors Associated with Readmissions(5, 9)

Risk Prediction for Rehospitalization

Predicting patients at the highest risk for rehospitalization may allow for concentration of resources. Hospital based transitional care programs which target all patients may be overwhelming for hospitals with limited resources. Readmission risk prediction tools, in general, have poor predictability, are not generalizable and are not matched with specific interventions to prevent readmissions.(17) Older adults with geriatric syndromes including muscle weakness and acute delirium may be at the highest risk for admissions.(18-20) Comorbidities which exacerbate HF, lead to poor self-management and complicate medical therapy are also important in assessing risk.(21) Heart failure patients are often admitted for other causes with respiratory diseases being the common.(22)

Many risk prediction models for rehospitalization have been developed based on both administrative and medical records data. Predictive ability has only been modest with average c-index reported around 0.60.(21) Prediction models are likely best used in a similar patient population to which the model was tested. Variables incorporated in the various models are many and include medical, psychosocial, economic, and system level variables. It is clear that rehospitalizations occur for many different reasons including for patients with unpreventable advancing illness. Prediction of rehospitalization requires further study and changes in policy have out-paced our ability to identify patients at the highest risk for readmission and determination of the appropriate interventions.(17)

Systolic versus Preserved Systolic Heart Failure

Preserved ejection fraction HF (HFpEF) is distinct from systolic heart failure with distinctive demographics and treatment strategies. As many as 20-60% of patients with HF are estimated to have HFpEF, and studies comparing systolic versus HFpEF demonstrate similar readmission rates.(23-25) HFpEF is associated with advanced age, female gender, and hypertension and guideline based therapy recommendations are significantly less robust for patients with HFpEF.(1) Heart failure trials predominately study systolic failure therefore treatments for HFpEF which reduce hospitalizations and improve outcomes are less clear.(1)

Strategies to Reduce Readmission

Strategies to reduce readmissions fall into 3 broad categories:

1. Optimization of pharmacologic and device therapies

2. Monitoring strategies to recognize early decompensation

3. Patient behavioral modification and lifestyle change to eliminate behaviors which exacerbate HF

Pharmacologic Therapies

The pathophysiology of chronic HF includes reflexive and excessive neurohumoral activation in response to a decrease in effective circulatory volume. Both the sympathetic nervous system (SNS) and the renin-angiotensin-aldosterone system (RAAS) contribute to volume retention, arrhythmias, and cardiac remodeling. Medical therapy for HF is focused on decreasing neurohumoral activation while controlling intravascular volume through diuresis. Medications that block RAAS and SNS have a marked benefit on stabilizing and improving heart failure. Major studies evaluating medical therapies routinely measure HF hospitalizations, not specifically readmissions. However, appropriate medication management is the best way to control decompensation. Successful medication management includes both the assurance that the patient takes the medications appropriately and monitoring for adverse effects.

• Angiotensin-converting enzyme (ACE) inhibitors/Angiotensin receptor blockers (ARBs)

  • ○Drugs which block RAAS, ACE inhibitors and ARBs, are at the cornerstone of HF therapy. These drugs result in vasodilatation of peripheral vasculature, reversal of cardiac remodeling, and limitation of sodium retention. Numerous studies using ACE inhibitors in chronic HF with reduced systolic function demonstrate a clear mortality benefit and reduction of hospitalizations, although rehospitalizations have not been directly studied.(1, 26) In HFpEF treatment with the ARB candesarten showed a decrease in hospitalizations although not a mortality benefit.(27)
• Aldosterone antagonists

  • ○The final common pathway of RAAS activation leads to increased aldosterone synthesis. Aldosterone antagonists serve to further inhibit an increase in effective circulatory volume. For symptomatic patients with severe left ventricular dysfunction, aldosterone antagonists not only assert a mortality benefit but also decrease hospitalizations.(28) Caution is advised with initiation of aldosterone antagonists especially for older adults, since there is an increase in hyperkalemia-related admissions.(29) A recent study of aldosterone antagonism in HFpEF showed no significant improvement in clinical symptoms or exercise tolerance.(30) Further randomized controlled trials are required prior to the development of strong guideline recommendations for the chronic maintenance of HFpEF. The recently completed TOPCAT trial investigates the effect of aldosterone antagonists on a composite endpoint of cardiovascular mortality, aborted cardiac arrest or hospitalization, will illuminate the utility of this drug class for HFpEF patients. (31)
• Beta-receptor blockers (beta-blockers)

  • ○Beta-blockers reduce SNS-mediated activation of the heart and reduce cardiac demand. When used in the chronic maintenance of HF, beta-blockers demonstrate improvement in symptoms, mortality, and hospital admission rates.(1) Long-acting beta-blockers, such as metoprolol and carvedilol demonstrate increased efficacy as compared to short-acting analogues with respect to these endpoints.(32) Uptitration of beta-blockers post hospitalization does not predict readmissions as long as uptitration is gradual and according to medication guidelines.(33)
• Loop diuretics:

  • ○Loop diuretics potentiate diuresis by inhibiting sodium reabsorption in the nephron loop of Henle. Although there is no clear mortality benefit with the use of loop diuretics, diuretics provide symptomatic relief. With clinician oversight, patients can be taught to titrate their own diuretics on a day to day basis. These flexible diuretic regimens give the patient control to titrate their loop diuretics based on weight and symptoms and can be an important strategy in controlling volume status. One study suggests that 30-day readmissions could be reduced using flexible diuretic regimens.(34)
• Digoxin

  • ○ The cardiac glycosides, digoxin, improves cardiac inotropy by increasing intracellular calcium concentrations. Patients with reduced systolic function maintained on digoxin demonstrate decreased rates of hospitalizations as well as symptomatic relief although no mortality benefit.(35)

Controversial Therapies

• Nesiritide:

  • ○ Nesiritide, a recombinant form of brain natriuretic peptide which disrupts increased RAAS activation and relaxes vascular smooth muscle is used for acute decompensated HF. There was hope that this medication would improve outcomes. In a large clinical trial nesiritide did not reduce mortality or rehospitalization rates compared to placebo and is therefore not recommended for routine inpatient use.(36)
• Ultrafiltration:

  • ○ Ultrafiltration is an alternative to loop diuretics in decreasing effective circulatory volume. Moreover, the deleterious effects of loop diuretics, namely electrolyte abnormalities, are absent with ultrafiltration. Studies defining the role of ultrafiltration in acute exacerbation of HF are mixed. In one study, ultrafiltration resulted in greater volume loss during index hospitalization and decreased rehospitalization rates when compared to standard diuretic therapy.(37) However, results trended towards worsening renal function. A subsequent trial demonstrated acute worsening of renal function with ultrafiltration (versus diuretic protocols) with no measurable improvement in weight loss or rehospitalization rates. Long-term renal function was no different between the ultrafiltration and diuretic groups.(38)
• Brain Nateuretic Peptide (BNP) Guided Therapy

  • ○ The TIME-CHF trial compared BNP vs. symptom-guided intensification of medical therapy. The results of the trial indicated that BNP guided therapy decreased HF but not all-cause hospitalizations when compared with symptom-guided therapy. However, patients over the age of 75 did not benefit from this approach.(39) BNP-guided medical intensification may increase the risk of adverse drug events. In a meta-analysis of several studies assessing natriuretic guided HF management, HF hospitalizations were reduced.(40)

Device Therapies

Heart failure hospitalizations occur despite optimization of medical therapy. Patients often fail to recognize or report symptoms of early decompensation. Technologies aimed at reliably measuring intravascular volume to identify early decompensation are gaining favor. Chronically elevated filling pressures increase patient risk for HF events.(41) Invasive strategies for volume removal have also been employed, although with varied results and no clear consensus guidelines.

• Invasive Monitors:

  • ○ Invasive monitors have become an emerging strategy aimed at identifying early decompensation and therefore the potential to avoid readmissions. Monitoring requires that timely action be taken based on the collected patient data. Right ventricular pressure sensors provide measurement of right ventricular systolic pressure, a surrogate of pulmonary artery diastolic pressure. A nonrandomized trial in patients with NYHA II-III heart failure demonstrated an appreciable reduction in HF-related hospitalizations, but this relationship was not validated in subsequent trials.(42, 43) Similarly, a left atrial pressure sensor resulted in uptitration of ACE inhibitors and beta-blockers, and decrease in loop diuretics; functional class improved.(44) However, randomized trials are needed to study effects on hospitalization.
  • Implantable pulmonary artery pressure sensors provide real-time information on hemodynamics thus allowing more rapid outpatient management changes. These devices are less invasive than right ventricular or left atrial pacers in that they can be placed during a right heart catheterization. They have been associated with fewer HF-related hospitalizations, with one randomized trial demonstrating a thirty-seven percent reduction in HF admissions over a six months period.(45) It is unclear, however, whether these results can be extrapolated to the routine use of these devices as the patients enrolled in this trial were very carefully monitored using industry resources.
  • Cardiac resynchronization therapy (CRT) results in improved cardiac output and reverse remodeling. These devices are associated with improved functional class and exercise capacity. In addition, when combined with implantable cardiac defibrillators (ICD), there is a mortality and hospitalization benefit. It is important to note, however, that guideline recommendations for such devices only include patients with systolic dysfunction and a widened QRS complex.(46)
  • Another strategy to prevent hospitalizations directly measures electrical impedance as a surrogate for volume overload using CRT/ICD devices. Increased pulmonary vascular congestion results in improved electrical current conduction thereby reducing intrathoracic electrical impedance. Studies using impedance are currently limited, with preliminary results showing increased sensitivity in assessing clinically significant weight gain. Whether these devices decrease hospitalization rates is unknown. (47)

Diet and Lifestyle

Diet and lifestyle changes are an important part of the management of HF. A low sodium diet, restricted fluid intake, in some circumstances, and self-monitoring of weight and signs and symptoms are strongly encouraged. However, testing the effect of these self-management strategies on outcomes is difficult.

• The adoption of a low-sodium diet of approximately two grams per day or less, can control volume retention and is recommended in the chronic maintenance of HF. The evidence for this guideline, however, rests on case studies and expert opinion.(1) One recent study raises doubts about the benefit of more stringent sodium restriction, citing an increase in rehospitalization rates. This inverse relationship is postulated to be due to a reciprocal increase in neurohormonal activation in response to low sodium intake.(48) Recommendation for controlled sodium intake is standard of care.

• Fluid restriction of less than two liters per day in patients with severe hyponatremia (serum sodium < 130) is recommended per expert opinion. This recommendation is further extrapolated to include patients with refractory volume retention despite diuretic therapy.(49) One study of patients with reduced systolic function demonstrated decreased rates of HFR with more stringent volume restriction.(48) Benefits vs. risks of fluid restriction and for which patients requires further study.

• Patients are often instructed to weigh themselves daily, maintain documentation of weight changes, and use a flexible diuretic regimen based on weight changes. There is some indication that this practice can decrease hospitalizations studies are limited.(50)

• In patients with symptomatic systolic dysfunction, participation in a structured exercise program is safe and modestly reduced composite endpoint of cardiovascular mortality and HF hospitalizations.(51)

Supportive Programs for HF Care

Transitional Care

The immediate post-hospitalization time period is the most vulnerable time for a patient to be readmitted. Regardless of disease, transitional care interventions which focus on the post-hospitalized patient decrease rehospitalization rates.(52-55) Many transition interventions use health education and coaching. Rapid follow-up (7 days) for the home-going patients results in fewer hospitalizations and may be the most effective strategy to improve adherence to a treatment plan, self-management and identify impending adverse events.(56, 57) The rapid hospital follow-up appointment continues to be under-utilized.(58) Many of the frailest patients transition to skilled nursing facilities for rehabilitation after hospitalization and the numbers going to an institutional setting is increasing.(59) Although those patients going to SNF have similar re-hospitalization rates as those who are discharged home, the SNF HF patient has significantly higher mortality which may indicate the higher complexity. Programs that span the period from hospitalization through early post-discharge, regardless of discharge destination, are most likely to have the best chance to prevent rehospitalization.

Programs for transitional care have been tested in both general and HF patient populations with reduction of rehospitalizations. Important components of these programs are patient engagement in self-care, recognition of disease exacerbation and a plan for intervention, medication management and reconciliation.

Heart Failure Disease Management Programs

Heart failure disease management programs including the multidisciplinary HF clinic have increased in popularity and decrease rehospitalizations rates. These programs usually incorporate a multidisciplinary team and often manage the most at risk and complex patients with HF. The lack of uniformity between programs makes it difficult to compare and to identify the most important components which decrease rehospitalizations.(60) However, the multidisciplinary approach to care planning can improve outcomes. Evidence indicates that programs which improve outcomes emphasize self-care and patient education.(61-63)

Heart Failure Telemonitoring

In-Home Telemonitoring has mixed evidence as to its ability to reduce rehospitalizations.(64) Many strategies have been employed to monitor outpatients especially in the immediate post-hospitalization period. It is still unclear how to best incorporate these technologies into practice and is an active area of study.

• Nursing telephone follow-up regardless of prior hospitalization has proven to be effective in some studies.(65, 66) Calls usually entail review of signs and symptoms of HF, medication review and review of daily weights. Intervention for symptoms and volume retention are key for stabilization of early decompensation. Nurses may be able to titrate diuretics effectively as demonstrated with other HF medications although there needs to be more study of the effect of this nurse guided practice on rehospitalizations.(50, 67, 68)

• In-home telemonitoring devices which monitor weights, blood pressure and pulse daily and are followed by a nurse, usually at a central monitoring station at a home health care. Monitoring patient data daily allows nurses to intervene on changes in vital signs and weights. This practice requires that nurses have appropriate physician or nurse practitioner involvement so rapid therapeutic changes can be made based on incoming data. Some services (insurance-based) are monitoring patients but have no established relationship with the patient’s medical team and therefore are unsuccessful at intervening in a timely fashion. Even for home health care agencies with established physician participation, real-world challenges exist in successful communication of telemonitoring data and timely response to data.(69)

• Failed nurse-physician communication regarding patients undermines the utility of this technology. Patients can be taught to titrate their own diuretics and this practice with the nurse as a patient coach may be the best strategy for early intervention for the patient receiving home health care. Randomized trials of telemonitoring for HF have had mixed results on hospital admissions and health status.(64, 69-71)

Advanced Heart Failure and the End of Life

HF patients are frequently admitted at the end of life.(72) There is also an increase in resource use including hospice services. Discussing prognosis and setting realistic goals with patients is an important component of care. Decision making regarding the care plan for patients with HF may be an important aspect of reducing readmission rates, although this is difficult to study. Hospital-based care for the decompensated patient is frequently unavoidable in order to control symptoms and regain function.(73) Physiologic decline and worsening disease often means consideration of advanced therapies for HF and complicated by other chronic processes, such as kidney disease, anemia, and malnutrition. The plan of care should be adjusted as HF advances and is important in order to avoid unnecessary procedures and hospitalizations. Close follow-up and ongoing management with a multidisciplinary team is an important for these patients.(74)