In defining success, Disney did something different. Fred Lee worked at Disney after a distinguished career in hospital senior leadership, consulting, and training. Shortly after his arrival at Disney, he recalls noting Disney’s satisfaction scores were often in the 75% range, much lower than the 90+% satisfaction scores to which he was accustomed at healthcare organizations. He quickly discovered there was more to the story. In contrast to the common practice of lumping, the top couple of choices on a 5-point Likert scale as “satisfied” and hailing teams with 90+% satisfaction scores, anything less than the top score at Disney was counted as a failure.1 This aspirational message on striving for excellence and combating the Lake Woebegon effect, in which “all the children are above average,”2 has appeal in thinking differently about how to define success in medical care. But when it comes to left ventricular assist devices (LVADs), would we rather have the renowned cardiac surgeon Dr. Walt Dembitsky at the helm or Walt Disney?
Advanced heart failure is a mortal disease; 1-year survival lags many feared metastatic malignancies. Mechanical circulatory support has transformed care for this population. With 74.7% of LVAD patients alive and free from disabling stroke or reoperation to replace or remove a malfunctioning device at 2 years in a recent cohort,3 LVADs are firmly established as mainstream care for advanced heart failure. Despite the dramatic improvements in mortality, morbid adverse events remain a prominent reminder of the work remaining to understand the pathophysiology and mitigate adverse events. Gastrointestinal bleeding, stroke, driveline infection, and thromboembolic complications remain distressingly common. The rate of hospitalization following LVAD is also high, and impacts patient quality of life and cost of care beyond the high upfront device expense. These adverse events certainly blunt patient enthusiasm for such a major surgery, and may lead many physicians to shy away from recommending evaluation for implant.
With these tradeoffs in patient outcomes, defining LVAD success will be difficult. From a provider perspective, we have observed disagreements within our center’s professionals over whether implanting an LVAD in a particular patient have been a “success” or a “mistake.” We have also witnessed instances where the medical team would adjudicate a “success,” but the patient or caregiver’s expectations are not met and the patient wishes they had chosen another path of care. Conversely, patients with recurrent complications can rave about how glad they are to be alive and able to ambulate without oppressive dyspnea. Nuances in patient perspectives will challenge efforts to craft a unifying definition of success.
Anwer et al.4 propose a quantitative measure of LVAD implant success, defined as alive or transplanted at 2 years, two or less readmissions in the first year, no major adverse events in the first year, and a New York Heart Association class of ≤II at 6 months. Applying this definition to a single-center cohort of 278 patients, they found that 81 (29%) met this definition of success. This is an important contribution to the discussion of LVAD success and echoes Disney’s thinking—anything less than perfect is failure. One might ask why two readmissions are still perfect and if we are to follow the Disney model, why accept any readmissions? This component of the proposed measure is arbitrary. Regardless of whether we accept this measure, complications will occur even with perfectly delivered care. For some patients, failure to achieve some aspects of a measure of perfect LVAD care may be preferable to the death and disability associated with advanced heart failure. This is particularly troublesome if aggressive application of this definition by regulatory bodies or payors limited access to LVAD care among patients deemed at high risk of an imperfect outcome. As an alternative, the proposed “success” measure might be more helpful in setting the stage for patients and families to understand the likelihood of a “perfect” result. We might frame the risks with a 10% chance of a perfect outcome, 50% chance of avoiding the feared complications such as stroke or GI bleeding, but a 90% chance of experiencing one or more adverse events, if hospital admissions are included. Extensive data on rates of infection, stroke, bleeding, and hospitalization are available and should be presented to candidates, but information on aggregate risk of any event may not be part of usual disclosure. Such a shift in presenting risks might improve communication and expectations.
On a more granular level, outstanding questions remain, and could be considered for future iterations of such efforts. First and foremost is greater insight into the tradeoffs faced by patients as it relates to life with and without an LVAD. For patients who have experienced life with advanced heart failure and then complications of LVAD, which was more significant? Heart failure symptoms and hospitalization or post-LVAD GI bleeding and a driveline infection? This understanding could identify LVAD complications that from a patient’s perspective negate the longevity and symptom benefit provided by LVAD. In a similar way, by weighting different adverse events, we can acknowledge that for many patients, frequent hospitalization is not equivalent to death or disabling stroke. The patient’s voice needs more weight to inform discussion. Second, although we strive for perpetually lower rates of adverse events, should experiencing a known complication be considered a failure, or should we reserve this label for something else? Clinical trials typically make this distinction by designating unexpected adverse events in a separate category. Finally, we should consider degrees of success—perhaps “highly successful” and “moderately successful” to help guide preimplant discussions about anticipated outcomes.
Most people would probably consider receiving an LVAD the antithesis of the customer service of visiting a Disney theme park, but creating a summary measure that accurately reflects success for individuals will be difficult and future efforts will need to more completely incorporate the patient perspective. Nevertheless, collective efforts by the community to reduce the inevitable exposure to the “Adventureland” that follows an LVAD are worthy and ongoing work to define success is essential.
1. Lee F. If Disney ran your hospital: 9 1/2 things you would do differently. 2004.Bozeman, MT, Second River Healthcare Press.
3. Mehra MR, Uriel N, Naka Y, et al; MOMENTUM 3 Investigators; the MOMENTUM 3 Investigators: A fully magnetically levitated left ventricular assist device - final report. N Engl J Med 2019.380: 1618–1627.
4. Anwer LA, Poddi S, Tchantchaleishvili V, et al. Left ventricular assist devices; how do we define success? ASAIO J 2019.65: 430–435.