Frailty and Ventricular Assist Devices: Implications for Advanced Practice Registered Nurses

Frailty in advanced heart failure (HF) is a growing and prevalent comorbid condition, occurring in approximately 50% to 65% of advanced HF patients.1–3 Frailty predisposes individuals to nearly a 5-fold increased risk of postoperative complications when undergoing cardiac surgical treatment modalities such as implantation of a durable left ventricular assist device (dLVAD) for advanced HF.4 This increased risk includes longer lengths of stay and increased 30-day readmissions.5 Although there is inconsistency in the conceptual and operational definitions of frailty, the most widely used definition focuses on the concept of frailty as a clinical syndrome with at least 3 of 5 applicable phenotype domains present: weight loss, weakness, self-reported exhaustion, low physical activity, or slow walking speed.6–8 The most common conceptual definition describes an association between this clinical syndrome and decreased reserve, increased vulnerability, and eventually poorer outcomes.6,7 Although the Fried frailty phenotype criteria may be the most commonly used frailty measure, there are more than 60 frailty measures to choose from when assessing frailty in any given population.9 A concerning issue is there have been no reported frailty measures shown to be reliable or valid when used specifically in the dLVAD population, making frailty assessment difficult in this complex population.This column presents a brief overview of current literature highlighting risks of frailty for individuals receiving a dLVAD, the most common frailty measures used in advanced HF and dLVAD therapy, and the implications to consider for advanced practice registered nurses (APRNs) caring for patients with dLVADs in the critical care setting.Recent literature summarized frailty in the HF and dLVAD populations in the United States, describing significant variability in frailty definitions in these patients. Types of frailty measures also varied in this review but most commonly included at least some modified form of the Fried frailty phenotype criteria or the deficit index, which establishes frailty on the basis of accumulation of comorbidities and deficit in activities of daily living.1,2 Increased mortality was found to be associated with frailty, regardless of which frailty measure was used. Frailty-associated outcomes in the review were similarly demonstrated through increased 30-day readmissions, longer lengths of stay, increased costs following dLVAD implantation, and worse depression scores.10 Despite the availability of various frailty measures used in the HF and dLVAD populations to date, there still is no criterion standard frailty measure recommended, demonstrating the ongoing need to further evaluate which valid frailty measure or measures should be used to most accurately assess frailty in these patients. Determining frailty appropriately allows for establishing strategies and interventions to mitigate postoperative outcomes.There are a host of reliable and valid frailty measures available for use in assessing frailty in any given population, however none have been studied to determine validity in the advanced HF or dLVAD populations. When using the Fried frailty phenotype criteria, individuals are considered frail if they have at least 3 of the 5 following domains present: weight loss of more than 10 lb in the last year, certain answers to 2 questions related to exhaustion and occurrence of physical activity, 5-meter walk test longer than 6 seconds, or handgrip strength less than 25% of the patient’s weight, measured in kilograms.6 The complexity of advanced HF includes individuals experiencing cardiac cachexia leading to weight loss, lacking energy due to advancing HF, or experiencing the inability to walk even small distances due to persistent heart failure symptoms, which can interfere with accurate interpretation of whether or not these symptoms are directly related to worsening HF or to frailty. Joseph and colleagues2 used the original Fried frailty phenotype criteria along with a modified version of the phenotype to assess the association between frailty in advanced HF with mortality and length of stay. They found no associations when using the full 5-domain Fried frailty phenotype criteria measure, however when they adapted the frailty assessment to include only 3 criteria (inactivity, hand grip strength, and exhaustion), they found associations with increased mortality and increased time to extubation, suggesting that measures may need to be adjusted to increase accuracy of frailty determination specific to advanced HF and dLVAD.2 Researchers have recommended that frailty assessment include some form of physical reserve assessment,7 albeit recognizing there may be limitations to these measures on the basis of severity of HF at presentation and the patient’s ability to ambulate. When the 5 domains were evaluated individually, sensitivity and specificity were noted to be higher than 80%, except for the domain of weight loss, which was 8.3% and 97.4%, respectively. Positive predictive value based upon the individual 5 items ranged between 12.5% and 52.5% but increased when combining hand-grip strength and gait speed together as a dual measure (87.5%).11A study evaluating the Interagency Registry for Mechanically Assisted Circulatory Support described use of a 5-meter walk test individually as a measure of frailty before and/or after dLVAD implantation.12 The 5-meter walk test is a timed test for an individual in which they walk 5 meters (or 16 feet); this testing occurs in a series of 3 walks, using the fastest walk test for frailty determination for a 5-meter walk test of greater than 6 seconds. More than 40% of individuals in the study were “too sick” to complete the 5-meter walk test, and although there were no concrete numbers to determine frailty in those who were too sick, this may suggest correlation between severity of illness and the existence of additional comorbid conditions such as frailty. There are rising concerns among clinicians working with advanced heart failure patients, as well as researchers, surrounding the lack of accurate measures available for frailty assessment in individuals too sick to walk.12 The 5-meter walk test has been used in patients undergoing a variety of cardiac surgical procedures, with satisfactory test-retest reliability (intraclass correlation coefficient [ICC], 0.86-0.91). This measure also had satisfactory interrater reliability (ICC, 0.88- 0.96). No reliability data on the 5-meter walk test exist in the dLVAD population, however the 5-meter walk test has been adopted for use in a variety of registries for walking assessment speed in patients without cardiopulmonary symptoms.9The deficit index, modified by Dunlay and colleagues1 to use with the dLVAD population, takes into account the accumulation of deficit when determining a patient’s frailty. This measure consists of 31 deficits including 17 comorbid conditions such as hypertension, diabetes, and hemiplegia, as well as 14 questions pertaining to activities of daily living.13 Frailty was determined in this study if the deficit index score was greater than 0.25, which was shown to have association with increased mortality.1 The deficit index, as a modified version of the original frailty index,14 is not validated in the literature for use in the dLVAD population; however, Dunlay and colleagues1 noted that when the deficit index was used in patients before dLVAD implantation, frailty was present in more than 60% of the sample.Mortality in the first year after receiving a dLVAD was higher in individuals determined to be frail by the provider eyeball test, as well as for those patients who were too sick to ambulate.12 The provider eyeball test is a subjective assessment measure used to assess whether the provider feels frailty exists before and/or after implantation. One benefit to the use of this measure is that there is no physical activity required of the patient; however, a major limitation to the measure is the judgment subjectivity of the provider. Many clinicians feel the provider eyeball test is one accurate way to determine frailty despite the lack of reliability and validity for this measure, however there is inconsistency of overall frailty determination in the literature when comparing provider eyeball frailty testing with objective measures such as the 5-meter walk test or the Fried frailty phenotype criteria.12 Due to this inconsistency described in the literature and what is seen in clinical practice, many clinicians question the singular utility of the provider eyeball test in accurately determining frailty by this simple subjective measure, when compared with objective measures.One additional option for frailty assessment many dLVAD centers are using is the hand-grip strength test, where hand grip strength of less than 25% has been associated with frailty, postimplantation mortality, and higher postoperative complication rates.13 The hand grip tool, known as a dynamometer, measures forearm musculature and may reflect upper body strength. A recent review of the utility of handgrip strength testing in the dLVAD population found handgrip strength testing to be simple and reliable and found correlations between improved handgrip strength with improved 6-minute walk tests.15 Test-retest reliability was found to be excellent in previous literature between the Jamar (0.98) and Smedley (0.92) dynamometers. Poor validity was noted with both dynamometers at 2 different time points.16 Previous correlation coefficients have been found as R2 = 0.80 for males and R2 = 0.75 for females.17Currently, none of the 4 previously mentioned measures have been found reliable or valid in the dLVAD population. The Figure describes the 4 frailty measures in more detail.1,6,7,12,13,15 The Fried frailty phenotype criteria measure is the most common and complex of the 4 measures and approaches frailty from a multilevel physical assessment. One advantage of the Fried frailty phenotype criteria is that this will likely be the most accurate and valid of the 4 measures. By contrast, the deficit index takes into consideration comorbid conditions and ability to perform activities of daily living, providing a different assessment approach for frailty that does not require the patient to be able to walk. The provider eyeball assessment test is quick and easy to use when clinicians are trying to identify possible frailty. However, this measure is subjective and depends on the clinician’s understanding of what frailty conceptually is and is not. Lastly, handgrip strength testing is emerging as a potential frailty measure, as it does not require the patient to be able to walk. One limitation to handgrip strength testing is the inability to accurately complete the study if the upper arm strength of the patient is decreased from noncardiac-related comorbid conditions. Clinicians need a quicker and more accurate way to determine frailty in order to begin strategizing how to lessen the effects of frailty following dLVAD implantation.Determining frailty before dLVAD implantation may be helpful in postimplantation management, as pre-dLVAD frailty is associated with higher 1-year mortality, longer lengths of stay, and increased 30-day readmissions.5 Knowing the risk of frailty, APRNs caring for dLVAD patients in the immediate postimplantation period need to focus on early initiation and monitoring of effects of physical and occupational therapies. Adequate nutrition is also important during both the preimplantation and the postimplantation phases. Success of dLVAD therapy depends on the early initiation of physical and occupational therapy, adequate nutrition, and strengthening.18Consideration of rehabilitation options following implantation is also imperative to aid in decreasing lengths of stay and the risk of 30-day readmissions. Many rehabilitation options exist, including inpatient rehabilitation, skilled nursing, and outpatient physical or occupational therapy; however, it has been posited that patient frailty may improve with placement of a dLVAD.19 One study described improved frailty in nearly half of the participants within 3 to 6 months following the placement of a dLVAD, including improvements in quality-of-life scores in this same time frame. Ambulation measured in feet when using the 6-minute walk test and albumin levels reflecting nutrition have been shown to significantly improve after dLVAD placement as well.20 Frailty has also been shown to improve or completely reverse after a dLVAD implantation that was followed by cardiac transplantation.21 The Table includes guidance for diagnosis and management of frailty-related issues occurring after implantation of a dLVAD that the APRN should be prepared to address.Regardless of which frailty assessment measurement is used, the determination of frailty before dLVAD placement can aid in planning of interventions that can increase the likelihood of better outcomes for patients receiving dLVADs. Frequently, patients arrive at dLVAD implanting centers in deteriorated condition in which there is insufficient time for the patient to undergo rehabilitation preimplementation, known as prehab. Monitoring for exacerbation of the coexisting comorbid conditions is just one way to limit the effects of frailty after surgery. It is imperative to involve therapy teams early in post-dLVAD implantation recovery to address the loss of muscle mass that has occurred over time from the effects of advancing HF. Further research is needed to determine whether a single measure or a multidimensional measure is the most accurate way to assess for frailty in the dLVAD population. This essential research will lead to optimization of care and outcomes for patients after dLVAD implantation.