Readmissions After Ventricular Assist Device: Etiologies, Patterns, and Days Out of Hospital

Transcription

1 Readmissions After Ventricular Assist Device: Etiologies, Patterns, and Days Out of Hospital Stephen J. Forest, MD, Ricardo Bello, MD, PhD, Patricia Friedmann, MS, Danielle Casazza, RN, Cecilia Nucci, RN, Jooyoung J. Shin, MD, David D Alessandro, MD, Gerin Stevens, MD, and Daniel J. Goldstein, MD Department of Cardiothoracic Surgery and Division of Cardiology, Montefiore Medical Center, Albert Einstein College of Medicine, New York, New York Background. Scarce literature exists describing the patterns of readmission after continuous flow left ventricular assist device (CF-LVAD) implantation. These carry significant cost and quality of life implications. We sought to describe the etiology and pattern of readmission among patients receiving CF-LVADs. Methods. Frequency, reason, urgency, and duration of readmission as well as freedom from readmission were examined in a retrospective review of our institutional experience. As an indirect means of quality of life, the ratio of days out of hospital (OOH)/days alive with device was calculated. Results. From 2006 to 2011, 71 adult patients implanted with a CF device were included. Indication for device implantation was bridge to transplant (n 19), potential bridge to transplant (n 25), or destination therapy (n 27). Length of support averaged 359 days. Total support time was 69.7 patient years. One hundred fifty-five readmissions accounted for a total of 1,659 hospital days. Fifty-six patients were readmitted during the study period. Median time to first readmission was 48 days (range 2 to 663 days). Median length of stay was 5 days. The single most common etiology for readmission was gastrointestinal bleeding accounting for 14% of readmissions. Readmissions were urgent (87%), elective (10%), or life-threatening (3%). Patients on the average enjoyed 92% of their time OOH. Conclusions. Patients undergoing CF-LVAD support are often readmitted within 6 months of discharge. Readmissions tend to be of short duration and the most common reason is for gastrointestinal bleeding. Importantly, following discharge after implant procedure, 51 patients spent at least 90% of days OOH. (Ann Thorac Surg 2013;95: ) 2013 by The Society of Thoracic Surgeons The clinical application of contemporary continuous flow left ventricular assist devices (LVAD) has led to dramatic improvements in early and mid-term outcomes for patients with advanced heart failure undergoing institution of mechanical support as bridge to transplantation or destination therapy [1, 2]. Similarly, benefits in quality of life and functional capacity have been recognized. While details regarding index hospitalization for LVAD implantation have been described in multiple trials [1, 2] and institutional series [3], scarce attention has been paid to the issue of readmission after successful LVAD therapy. With comprehensive healthcare reform firmly on the horizon in the US, reducing readmissions is considered essential to achieving medical care efficiencies and reducing cost. Hence, in an effort to gain an understanding of the problem of rehospitalization after a costly intervention like LVAD therapy, we examined the etiologies and patterns of hospital readmission after successful contemporary continuous flow LVAD implantation at our institution. In particular, we investigated frequency, reason, urgency, and duration of readmissions. Moreover, as an indirect measure of quality of life, the ratio of days out of hospital to days alive with device was calculated. Material and Methods Between June 13, 2006 and December 1, 2011, 92 patients underwent implantation of a continuous flow LVAD at Montefiore Medical Center. This study was approved by our Institutional Review Board. Data were prospectively acquired and maintained in our Institutional Review Board approved VAD database. A waiver of the need to obtain consent from unidentified patients was approved. Timing of readmissions, duration of hospitalization, and total length of device support were evaluated through February 1, Patient follow-up was considered complete when the patient was transplanted, there was myocardial recovery with device explantation, or the patient expired. A patient was otherwise on continued mechanical support. Factors associated with readmission were examined. A univariate analysis was conducted Accepted for publication Dec 21, Address correspondence to Dr Goldstein, Department of Cardiothoracic Surgery, 3400 Bainbridge Ave, MAP 5, Bronx, NY 10467; sforest@montefiore.org. Dr Goldstein discloses financial relationships with HeartWare, Terumo, and Thoratec Inc by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc

2 Ann Thorac Surg FOREST ET AL 2013;95: READMISSION AFTER CF-LVAD IMPLANTATION 1277 using the 2 test and Fisher exact test for categoric variables, and the Wilcoxon rank sum test for continuous variables. A logistic regression analysis using backwards elimination was conducted to assess multivariate predictors of readmission. A p value less than 0.05 was considered statistically significant. Results Patients and Devices The final study group consisted of 71 patients (77%). Seventeen patients who died prior to discharge on index hospitalization, two pediatric patients, one patient who was transplanted during the same hospitalization as device implantation, and one patient who was electively managed at an outside facility for geographical reasons were not included in the readmission analysis. Indications at device implantation included bridge to transplantation in 19 (27%) patients, potential bridge to transplantation in 25 (35%), and destination therapy in 27 (38%). Mean age of recipients was years and 23% were female. Etiology of cardiomyopathy was ischemic in 22 (31%) patients, non-ischemic in 28 (39%) and other in 21 (30%) - 13 acute MI, 5 post-cardiotomy, 2 post partum, 1 device exchange. Median Interagency Registry for Mechanically Assisted Circulatory Support category at time of implantation was 2. Three different devices were implanted in our patients. The HeartMate II (Thoratec Inc, Pleasanton, CA) was implanted in 58 patients, the Heartware HVAD (Heartware Inc, Framingham, MA) in 9 patients, and the Ventrassist VAD (Ventracor Inc, no longer exists) in 4 patients. No patient was lost to follow-up. Readmissions The median postoperative length of stay for index hospitalization was 25 days (range 10 to 113 days). Mean length of mechanical support was days with a total support time of 69.7 patient years. Thirty-eight readmissions were for transplantation (5 were aborted) and these were excluded from analysis. After index discharge, 56 patients (79%) were readmitted at least once to the hospital after a mean of days out of hospital. Distribution of readmissions is depicted in Figure 1. Time to first readmission ranged from 2 to 663 days, with 25 readmitted within 30 days of discharge from index hospitalization. The median length of stay for the first readmission was 5 days. There were a total of 155 readmissions for a total of 1,659 hospital days (Fig 2). Freedom from readmission appears in Figure 3. The number of readmissions per year of LVAD support was Readmissions were categorized as urgent in 87% of the cases, elective in 10%, and life threatening in only 3% of cases. The single most common etiology for readmission was gastrointestinal bleeding, accounting for 14% of readmissions. Elective readmissions included 8 for desensitization, 3 for elective noncardiac surgery, 3 for chemoradiation, and 1 for preoperative transplantation evaluation. Etiology of readmission was stratified by type of VAD, centrifugal or axial, and is shown in Figure 4. At the time of study close out, 46% of patients had undergone transplantation, 35% were on ongoing support, 4% had undergone successful LVAD explantation after documentation of complete ventricular recovery and 14% had expired. Causes of death included sepsis (5), intracranial bleed (2), respiratory failure (2), and unknown cause for 1. Patients who had at least 1 readmission were more likely to have prolonged device support, vs days, and longer cardiopulmonary bypass time during implantation of the device ( vs minutes, p 0.02). Readmitted patients were older than those who were not readmitted (57 12 vs years, p 0.08) (Table 1). In this cohort of 71 patients, we found no significant predictors for 30-day readmission (Table 2). There were no multivariate predictors for any readmission or 30-day readmission in this analysis. Fig 1. Distribution of readmissions: 155 readmissions in total; 56 patients (79%) were admitted at least 1 time.

3 1278 FOREST ET AL Ann Thorac Surg READMISSION AFTER CF-LVAD IMPLANTATION 2013;95: Fig 2. Total number of readmission days. Total days of readmission 1,659 days. When excluding elective admissions to the hospital, LVAD recipients enjoyed an average of days out of hospital out of days alive with device. Stated differently, on the average, patients spent 92% of their time out of hospital after discharge from index hospitalization (range 41% to 100%). Total out of hospital time was 21,464 out of 23,034 days (93%). Comment Nowhere is the issue of readmission more under the spotlight than in the management of heart failure, a disease entity that affects 1% of the western world and accounts for greater than 1,000,000 hospitalizations and 6,500,000 hospital days per year [4]. Marked improvements in readmission rates for heart failure have been attained with the use of multidisciplinary congestive heart failure outpatient clinics [5], dedicated nurse educators [6], and telephone support after hospital discharge [7]. Left ventricular assist device therapy has become standard for the management of advanced heart failure patients as a bridge to transplantation or as permanent therapy [1, 2]. While a growing body of evidence strongly suggests laudatory outcomes after implantation of second and third generation continuous flow LVADs, the complex issue of hospital readmission has not been Fig 3. Freedom from readmission with time from readmission after initial hospitalization. Median time to first readmission is 51 days.

4 Ann Thorac Surg FOREST ET AL 2013;95: READMISSION AFTER CF-LVAD IMPLANTATION 1279 Fig 4. Etiology of readmission by type of ventricular assist device. Other hemorrhage (3) included hematuria, hemarthrosis, and post-catheterization hemorrhage. Elective surgery (3) included carotid endarterectomy, excision of salivary gland, and herniorrhaphy. Other readmissions included mechanical fall (3), chemoradiation (3), altered mental status, anaphylaxis, diabetic ketoacidosis, encephalopathy, headache, hypokalemia, myocardial infarction, neck mass, osteoarthritis-knee pain, pancreatitis, peripherally inserted central venous catheter line malfunction, preoperative evaluation, rash, transient shortness of breath, and urinary retention. (ARF acute renal failure; GI gastrointestinal; N/V/D/Abd pain NOS nausea/vomiting/diarrhea/abdominal pain not otherwise specified; TIA transient ischemic attack; VAD ventricular assist device.) addressed in detail in the literature. Given the current healthcare environment focus on cost reductions, it is critical to begin to gather data on rehospitalizations as it relates to LVAD therapy, a costly intervention that has gained widespread acceptance for patients with advanced heart failure. Table 1. Univariate Analysis for Any Readmission: 56 Patients Were Admitted at Least 1 Time and 15 Patients Were Free From Readmission Table 2. Univariate Analysis for Readmission Within 30 Days: 25 Patients Were Admitted Within 30 Days and 46 Patients Were Not Variable Not Readmitted (n 15) Readmitted (n 56) p Value Variable Not Readmitted Within 30 days (n 46) Readmitted Within 30 days (n 25) p Value Age Male (13) 87% (42) 75% 0.49 Diagnosis of ischemic (8) 53% (27) 48% 0.72 Cardiomyopathy Indication for device BTT (5) 33% (14) 25% 0.70 DT (6) 40% (21) 38% PBTT (4) 27% (21) 38% Diabetes (7) 47% (23) 41% 0.70 Dialysis (0) 0% (2) 4% 1.00 IABP support (7) 47% (15) 27% 0.21 PVD (2) 13% (4) 7% 0.60 COPD (0) 0% (7) 13% 0.33 CVD (0) 0% (8) 14% 0.19 Ejection fraction Perfusion time (minutes) Index LOS (days) Discharged to home (7) 47% (29) 52% 0.72 Days of device support Type of device Axial (11) 73% (47) 84% 0.45 Centrifugal (4) 27% (9) 16% BTT bridge to transplantation; COPD chronic obstructive pulmonary disease; CVD cerebrovascular disease; DT destination therapy; IABP intraaortic balloon pump support; LOS Length of stay; pbtt possible bridge to transplant; PVD peripheral vascular disease. Age Male (34) 74% (21) 84% 0.33 Diagnosis of Ischemic (22) 48% (13) 528% 0.74 Cardiomyopathy Indication for Device BTT (11) 26% (7) 28% 0.73 DT (19) 41% (8) 32% PBTT (15) 33% (10) 40% Diabetes (21) 46% (9) 36% 0.43 Dialysis (0) 0% (2) 4% 0.12 IABP support (17) 37% (5) 25% 0.14 PVD (5) 11% (1) 4% 0.41 COPD (4) 9% (3) 12% 0.69 CVD (4) 9% (4) 16% 0.44 Ejection fraction Perfusion Time (min) Index LOS (days) Discharged to home (23) 50% (13) 52% 0.87 Days of device Support Type of device Axial (37) 80% (21) 84% 1.00 Centrifugal (9) 20% (4) 16% BTT bridge to transplantation; COPD chronic obstructive pulmonary disease; CVD cerebrovascular disease; DT destination therapy; IABP intraaortic balloon pump support; LOS length of stay; pbtt possible bridge to transplant; PVD peripheral vascular disease.

5 1280 FOREST ET AL Ann Thorac Surg READMISSION AFTER CF-LVAD IMPLANTATION 2013;95: In the seminal paper describing the outcomes of a randomized trial of pulsatile versus continuous flow devices as destination therapy, Slaughter and colleagues [1] documented a 94% rehospitalization rate for the 133 recipients of the HeartMate II LVAD (event rate 2.64 per patient-year). This was significantly lower than the 4.25 patient-year event rate for recipients of the pulsatile HeartMate XVE device. The rehospitalization rate in our institutional experience was 79%, excluding those instances where the patient was readmitted when a suitable donor became available for transplantation. Our readmission rate of 2.46 per patientyear of LVAD support is comparable with that reported by Slaughter and colleagues. In a group of 133 patients undergoing HeartMate II LVAD support as a bridge to transplant, Miller and colleagues [2] described a median of 60 days out of hospital before transplantation, readmission, or death (range 0 to 418 days). Fifty-four discharged patients required rehospitalization for complications with a median duration of 4 days [2], similar to that seen in the present study. Neither of these seminal trials provided any details regarding the etiology and nature of these hospitalizations. In a systematic review of LVADs as destination therapy in the United Kingdom aimed at assessing cost-effectiveness of this therapy, readmissions had an event rate of 3 per patient at a total cost of 16,174 [8]. Again, no detail regarding the patterns of readmission, causes, urgency or length of stay is provided. Our institutional experience demonstrated that early readmission after successful implantation and discharge is nearly universal within 6 months of the implant procedure. First readmissions tended to be of short duration (median 5 days) and patients were readmitted on the average 2.2 times. The large proportion (41%) of these readmissions is for complications, most often related to bleeding or infectious processes. The recent recognition of acquired von Willebrand syndrome as a possible mediator of gastrointestinal bleeding in recipients of continuous flow devices [9] may lead to management strategies that will reduce mucosal bleeding and need for readmission for transfusion therapies. Similarly, the observation that driveline infections may be drastically reduced by maintaining the velour covering completely within the subcutaneous tissues [10] may lead to a reduced need for readmission for intravenous antibiotics and wound care. Additional preventative measures to reduce readmission are somewhat difficult to assess from our statistical analysis. The power of the study is limited. Even though perfusion time is a predictor of readmission at any time during device support, it is not a significant predictor within the first 30 days after discharge. It is difficult to ascertain why longer perfusion time could be a predictor, but it would seem more likely to be relevant in the more acute postoperative setting which it was not, as displayed in the 30-day readmission analysis. Longer duration of support obviously renders a patient more likely to be readmitted and was confirmed in this analysis. It is encouraging to confirm previous reports of the reliability of continuous flow LVADs in that lifethreatening complications related to device malfunction or failure were very rare. Perhaps most importantly, patients enjoyed 91 out of 100 days of days alive with the device out of hospital, most of which were at home. This finding underscores the positive impact that these technologies have on quality of life of the patient with advanced heart failure. While we recognize our limited statistical power, very limited data exists in the literature focusing on this important issue facing recipients of these miniaturized VAD therapies. Readmission after LVAD is recognized as a critical and growing problem as LVAD technology gains a wider penetration and greater focus is paid to curbing rehospitalizations. As clinical experience continues to grow and the technology matures, we anticipate growing numbers of patients will be supported by LVAD for several years. Guidelines for prevention of infectious [11] and bleeding complications will evolve and new challenges, including gastrointestinal bleeding and de novo native aortic valve insufficiency [12], will be effectively managed as to provide LVAD recipients with optimal quality of life while maximizing time out of hospital. Limitations The retrospective nature of this study, derived from a tertiary, inner city, academic medical center, limits its generalizability to other institutions with different geographic and patient populations. The power of this study is limited and an analysis including a larger cohort of patients could help identify additional univariate and potential multivariate predictors of readmission. Addendum Since submission of our manuscript, an article by Hasin and colleagues [13] describes the readmissions for 115 axial flow LVAD recipients. There were 224 readmissions in 83 patients. The overall readmission rate was per patient-year of follow-up with gastrointestinal bleeding as the leading cause of rehospitalization, a finding similar to ours. Unlike our experience, the authors were able to describe geography, hemoglobin and brain natriuretic peptide levels as predictors of reduced readmissions. References 1. Slaughter MS, Rogers JG, Milano CA, et al. Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med 2009;361: Miller LW, Pagani FD, Russell SD, et al. Use of a continuousflow device in patients awaiting heart transplantation. N Engl J Med 2007;357: Williams ML, Trivedi JR, McCants KC, et al. Heart transplant vs left ventricular assist device in heart transplant-eligible patients. Ann Thorac Surg 2011;91: Hunt SA, Abraham WT, Chin MH, et al focused update incorporated into the ACC/AHA 2005 Guidelines for the Diagnosis and Management of Heart Failure in Adults: a report of the American College of Cardiology Foundation/ American Heart Association Task Force on Practice Guide-

6 Ann Thorac Surg FOREST ET AL 2013;95: READMISSION AFTER CF-LVAD IMPLANTATION 1281 lines: developed in collaboration with the International Society for Heart and Lung Transplantation. Circulation 2009;119:e Ducharme A, Doyon O, White M, Rouleau JL, Brophy JM. Impact of care at a multidisciplinary congestive heart failure clinic: a randomized trial. CMAJ 2005;173: Manning S. Bridging the gap between hospital and home: a new model of care for reducing readmission rates in chronic heart failure. J Cardiovasc Nurs 2011;26: Inglis SC, Clark RA, McAlister FA, et al. Structured telephone support or telemonitoring programmes for patients with chronic heart failure. Cochrane Database Syst Rev 2010;8:CD Clegg AJ, Scott DA, Loveman E, Colquitt J, Royle P, Bryant J. Clinical and cost-effectiveness of left ventricular assist devices as destination therapy for people with end-stage heart failure: a systematic review and economic evaluation. Int J Technol Assess Health Care 2007;23: Crow S, Chen D, Milano C, et al. Acquired von Willebrand syndrome in continuous-flow ventricular assist device recipients. Ann Thorac Surg 2010;90: Akhter S, Valeroso T, Pisarski B, et al. Technique of HeartMate II percutaneous lead externalization is associated with incidence of infection. J Heart Lung Transplant 2011;30:S Hannan MM, Husain S, Mattner F, et al. Working formulation for the standardization of definitions of infections in patients using ventricular assist devices. J Heart Lung Transplant 2011;30: Pak SW, Uriel N, Takayama H, e al. Prevalence of de novo aortic insufficiency during long-term support with left ventricular assist devices. J Heart Lung Transplant 2010;29: Hasin T, Marmor Y, Kremers W, e al. Readmissions after implantation of axial flow left ventricular assist device. J Am Coll Cardiol 2013;61: INVITED COMMENTARY The introduction of continuous-flow left ventricular assist devices (LVADs) has had a major impact on the beneficial outcomes and acceptance of this therapy for patients with advanced heart failure. These long-term LVADs are typically implanted for three primary indications: bridge to transplant, bridge to recovery, and destination therapy, in addition to bridge to decision in patients who may become candidates for transplantation after a period of mechanical circulatory support (MCS). The REMATCH trial [1] was a landmark study demonstrating a significant 1-year and 2-year survival benefit of the pulsatile, first generation, HeartMate XVE LVAD versus optimal medical therapy in a cohort of patients with advanced heart failure who were ineligible for transplantation. Current studies with continuous-flow LVADs show 1-year survival rates of approximately 85% [2]. Given this dramatic improvement in patient survival, a current emphasis in the field of long-term MCS is preventing complications associated with these devices and maximizing quality of life. This study from Forest and coauthors comprehensively evaluates their readmissions after LVAD implantation in the current era of MCS [3]. The study population included 71 adult patients who underwent continuous-flow LVAD implantation at their institution with either the Thoratec HeartMate II (n 58), the Heartware HVAD (n 9), or the Ventrassist VAD (n 4). The overall benefit of this therapy was highlighted by their finding that, on average, LVAD patients spent 92% of their time out of the hospital. The majority of this time was at home. However, 56 patients were readmitted during the mean follow-up period (359 days), with the median time to first admission being 48 days. The most common cause of admission after implantation of the LVAD was gastrointestinal bleeding, which accounted for 14% of readmissions. Only 3% of all readmissions were life-threatening. Other reasons for readmission included LVAD-related and non LVAD-related infections, volume overload, arrhythmias, stroke/transient ischemic attack, syncope, and elective desensitization. At the end of the study period, 85% of patients were alive and were receiving ongoing LVAD support (35%), had undergone transplantation (46%), or had undergone device explantation for recovery (4%). The development of new strategies to minimize postimplantation complications and reduce readmissions is critical for the expansion of LVAD technology by optimizing the long-term survival, quality of life, and cost-effectiveness of this therapy. Shahab A. Akhter, MD Department of Surgery University of Chicago Medical Center 5841 S Maryland Ave MC 5040 Chicago, IL sakhter@surgery.bsd.uchicago.edu References 1. Rose EA, Gelijns AC, Moskowitz AJ, et al. Long-term mechanical left ventricular assistance for end-stage heart failure. N Engl J Med 2001;345: Starling RC, Naka Y, Boyle AJ, et al. Results of the post-u.s. food and drug administration-approval study with a continuous flow left ventricular assist device as a bridge to heart transplantation: a prospective study using the INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support). J Am Coll Cardiol 2011;10;57: Forest SJ, Bello R, Friedmann P, et al. Readmissions after ventricular assist device: etiologies, patterns, and days out of hospital. Ann Thorac Surg 2013;95: by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc