- Email: [email protected]
Outcomes of Severely Obese Patients Supported by a Centrifugal-Flow Left Ventricular Assist Device
Outcomes of HVAD Patients with Severe Obesity
Journal Pre-proof
Outcomes of Severely Obese Patients Supported by a Centrifugal-Flow Left Ventricular Assist Device Michael S. Kiernan MD, MS , Samer S. Najjar MD , Amanda R. Vest MBBS, MPH , Emma J. Birks MBBS, PhD, BSc, FRCP , Nir Uriel MD , Gregory A. Ewald MD , Katrin Leadley MD , Chetan B. Patel MD PII: DOI: Reference:
S1071-9164(19)30502-0 https://doi.org/10.1016/j.cardfail.2019.10.013 YJCAF 4439
To appear in:
Journal of Cardiac Failure
Received date: Revised date: Accepted date:
8 May 2019 16 October 2019 29 October 2019
Please cite this article as: Michael S. Kiernan MD, MS , Samer S. Najjar MD , Amanda R. Vest MBBS, MPH , Emma J. Birks MBBS, PhD, BSc, FRCP , Nir Uriel MD , Gregory A. Ewald MD , Katrin Leadley MD , Chetan B. Patel MD , Outcomes of Severely Obese Patients Supported by a Centrifugal-Flow Left Ventricular Assist Device, Journal of Cardiac Failure (2019), doi: https://doi.org/10.1016/j.cardfail.2019.10.013
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.
HIGHLIGHTS
CF-LVAD recipients, regardless of severe obesity, had improvements in quality of life
Body mass index > 35 kg/m2 was not associated with reduced survival
Severe obesity is associated with increased risk of driveline infection
Functional capacity gains were more limited for the severely obese, while improvements in quality of life were similar to patients without severe obesity
Awareness of increased adverse event risk associated with severe obesity is necessary to better inform shared decision making
2 Kiernan HVAD BMI Title:
Outcomes of Severely Obese Patients Supported by a Centrifugal-Flow Left Ventricular Assist Device
Authors:
Michael S Kiernan1, MD, MS, Samer S Najjar2, MD, Amanda R Vest1, MBBS, MPH, Emma J Birks3, MBBS, PhD, BSc, FRCP, Nir Uriel4, MD, Gregory A Ewald5, MD, Katrin Leadley6, MD, Chetan B Patel7, MD
Author affiliations: 1
Tufts Medical Center, Boston, MA; 2MedStar Washington Hospital Center, Washington,
DC; 3University of Louisville, Louisville, KY; 4University of Chicago, Chicago, IL; 5
Washington University/Barnes-Jewish Hospital, St. Louis, MO; 6Endotronix Inc. Lisle, IL;
7
Duke University, Durham, NC
Short Title:
Outcomes of HVAD Patients with Severe Obesity
Address for correspondence: Michael S. Kiernan, MD, MSc Department of Cardiology, Tufts Medical Center 800 Washington Street, Box 5931 Boston, MA 02111 Telephone:
617-636-8068
Fax:
617-636-6030
E-mail:
[email protected]
3 Kiernan HVAD BMI Relationships with Industry MSK is a consultant for Medtronic and has received travel grant from Abbott. SSN receives research grant support from and is on the advisory board for Medtronic. EJB receives research grant support from Medtronic and Abbott. NU is a consultant for Medtronic, Abbott, and XDx Inc. GAE is a consultant for Medtronic. KL was an employee of Medtronic. CBP receives conference fee support from Medtronic and Abbott. None of the other authors have a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose.
4 Kiernan HVAD BMI Abstract: Background: Ventricular assist devices provide improved outcomes for advanced heart failure patients, but their benefit in the severely obese is not well documented. Methods: Patients enrolled in the HeartWare ADVANCE trial (n=382) were divided into two body mass index (BMI) groups. Patients with severe obesity (> 35 kg/m2) were compared to a control group with BMI ≤ 35 kg/m2. The association of BMI with survival was tested using Kaplan-Meier analysis and major adverse events were compared. Results: At implantation, 48 (13%) of patients were severely obese. There was no difference in survival through two years of support between severely obese patients and the control group. Severely obese patients were at higher risk of driveline infection (p=0.01) and acute renal dysfunction (p=0.002).
Both groups experienced similar improvements in quality of life.
Functional capacity improved in both severely obese and control patients, although severely obese patients had smaller improvements than controls in their six-minute walk scores. Conclusions: Despite an increased risk of adverse events, severe obesity was not associated with reduced survival or quality of life. A better understanding of the risks and benefits of LVAD therapy in obese patients will help in the shared decision-making of the patient selection process.
Key Words: Left Ventricular Assist Device, Patient Outcomes, Obesity
5 Kiernan HVAD BMI Introduction With advancements in technology leading to improvements in survival as well as quality of life(1,2), left ventricular assist devices (LVADs) are increasingly used to support patients with end-stage heart failure as either a bridge to cardiac transplantation (BTT) or as destination therapy (DT). Concurrent with the epidemic of heart failure(3), is also the problem of obesity which is present in greater than one third of U.S. adults when defined as a body mass index [BMI] > 30 kg/m2. Obesity is a risk factor for the development of heart failure(4) and many obese patients will progress to end-stage disease requiring evaluation for cardiac transplantation and LVAD therapy. Among heart transplant candidates, obesity is associated with increased wait times and may lead to worse outcomes following cardiac transplantation(5,6). Accordingly, society guidelines recommend that patients achieve a BMI ≤ 35 kg/m2 prior to listing for cardiac transplantation (7). In contrast to the heart transplant literature, differences in long-term survival have generally not been seen between obese and non-obese patients undergoing LVAD implantation(8-14). There are mixed findings as to whether obesity affects the rates of certain adverse events such as device infection, but differences in results may be due to varying definitions of obesity utilized as well as the heterogeneity of the populations previously studied. With the expanding use of LVADs(15), appropriate patient selection is important to maintain optimal outcomes. While outcomes of obese patients undergoing implantation of an axial flow LVAD have been reported(8), it remains unclear if these findings are device specific or applicable to all technologies. We performed a retrospective analysis of the HeartWare ADVANCE trial and continued access protocol (CAP) to further explore outcomes of severely obese patients undergoing
6 Kiernan HVAD BMI implantation of a centrifugal flow LVAD as a bridge to cardiac transplantation. The primary aim was to determine the impact of severe obesity on survival, adverse events, health related quality of life (HRQL), and functional capacity in CF-LVAD recipients.
Methods The HeartWare HVAD system was evaluated as a BTT therapy in the United States in a prospective multicenter trial. The study design of the ADVANCE bridge-to-transplant and CAP trials have been previously described(2). A total of 382 patients received an HVAD between August 2008 and November 2012. The National Institute of Health (NIH) classifies obesity as follows: Class I – 30.0 to 34.9 kg/m2; Class II – 35.0 – 39.9 kg/m2; Class III - ≥ 40 kg/m2 (16). Patients from the ADVANCE trial were divided into two groups. The severely obese cohort was comprised of patients with body mass index > 35 kg/m2 at the time of implant and the nonseverely obese (control group) included patients with BMI ≤ 35 kg/m2. Adverse events were defined using Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) definitions(17) and patients were followed to transplant, death, or alive at last follow up. The studies were conducted in compliance with FDA regulations for Good Clinical Practices and approved by the Institutional Review Board designated by each participating study site. Statistical analysis Descriptive statistics were used to evaluate baseline clinical and demographic characteristics as well as LVAD parameters. Results are reported as mean ± standard deviation for continuous variables and as percent (count x 100/sample size) for binary variables. Comparisons between the obese and non-obese patients are made with a two-sample t-test for
7 Kiernan HVAD BMI continuous variables and Fisher’s exact test for categoric variables. Adverse events are reported as events per patient year (number of events) and comparisons between the severely obese and control groups were made with Poisson regression. Kaplan-Meier methodology is used to generate survival estimates and freedom from driveline infection curves for the two groups. Several HRQL and functional capacity measures including Kansas City Cardiomyopathy Questionnaire (KCCQ) Overall Summary Score, EuroQOL-5D (EQ-5D) Overall Health State Score, and six minute walk distance (6MWD) were assessed at baseline, 6, 12, and 24 month visits. A score of zero was imputed if the site indicated the six minute walk test could not be performed. Comparisons of the follow up visits to the baseline measurements were made with a paired t-test while differences between the obese and control groups were assessed with a two sample t-test. Following the same methodology, patient characteristics of BMI and serum albumin levels were also compared. To account for multiple hypothesis testing and thereby reduce the risk of Type I error, mixed effect ANOVA modeling was also used to allow for variability across patients and the fixed effects of BMI and time. A sensitivity analysis was also performed to compare characterisitics as well as outcomes between patients with a BMI > 35 and ≤ 40 kg/m2 versus > 40 kg/m2. All statistical analyses are performed with SAS v. 9.2 software (SAS Institute, Cary, NC).
Results The distribution of baseline body mass index of the 382 patients enrolled in the ADVANCE and CAP trials ranged from 14.89 to 48.09 (Figure 1). A total of 48 patients had BMI > 35 kg/m2 (severely obese) while 334 had BMI ≤ 35 kg/m2 (controls). Baseline characteristics and LVAD parameters of severely obese patients and controls are presented in
8 Kiernan HVAD BMI Table 1. The severely obese group has significantly higher weight and body surface area (BSA) compared to controls. The severely obese group was younger, more commonly diabetic, and had a higher proportion of black patients. Right atrial (RAP) was higher in severely obese patients: 13.4 ± 5.94 mmHg versus 10.6 ± 6.59 mmHg (controls). LVAD parameters were also different between groups. Speed settings were higher in severely obese patients compared to controls, as were measured power and calculated flow. Preoperative hemodynamic data were not available for the full cohort; the number of patients included is detailed in Table 1. One-year survival was similar in the severely obese (86.0%) and control groups (83.2%), and did not differ between groups across two-years of follow up (p=0.45) (Figure 2). There were differences in rates of adverse events between patients with and without severe obesity (Table 2). While rates of bleeding were similar between groups, severely obese patients were less likely to require re-operation for bleeding (0.06 EPPY) than the control group (0.18 EPPY) (p=0.01). Infection was more common in patients with severe obesity: 1.55 compared to 1.08 EPPY (p=0.03). In Figure 3, the one-year freedom from driveline exit site infection was 60.5% (severely obese) versus 81.3% (controls). The rate of sepsis was similar between groups. Acute and chronic renal dysfunction were more commonly reported in patients with severe obesity. BMI increased in both patients with and without severe obesity during the first 24 months of therapy, although this change was only significant for those with baseline BMI ≤ 35 kg/m 2 (Figure 4A). The increase in BMI from baseline to one year was 0.7 ± 5.7 kg/m2 (p=0.56) in the severely obese cohort and 1.8 ± 3.9 kg/m2 in the control group (p<0.0001). Compared to baseline, albumin increased significantly in both the obese (3.5 ± 0.74 g/dL vs. 3.9 ± 0.48 g/dL, p=0.02) and control (3.5 ± 0.70 g/dL vs. 3.9 ± 0.60 g/dL, p<0.0001) groups at 6 months
9 Kiernan HVAD BMI following HVAD implantation. These changes were sustained through 2 years of follow up (Figure 4B). At six months, both severely obese patients and the control group had marked improvements in HRQL as measured by both the EQ-5D and KCCQ scores (Figure 5A and 5B). HRQL improved similarly between groups across follow-up and these improvements were sustained through 2 years of support. Average increase in KCCQ from baseline to six months was 31.6 ± 27.2 (p<0.0001) and 31.0 ± 24.3 (p<0.0001) for severely obese and control groups respectively (p=0.27 for between group comparison at six months). Using mixed-effects ANOVA, there were no statistically significant interaction terms between BMI and KCCQ scores. While change in EQ-5D was significant overall (p=0.02), no single time point was significant. Functional capacity as measured by 6MWD also improved significantly in both groups compared to baseline with sustained improvements through 2 years of follow up. The control group, however, had greater improvements compared to the severely obese patients. At 12 months, the increase in 6MWD was 106.0 ± 191.8m (p=0.009) for severely obese patients and 199.2 ± 216.2m (p<0.0001) for controls (Figure 5C). The actual twelve-month 6MWD was 268.1 ± 189.6m for controls compared to 164.9 ± 176.0m for those with severely obesity (p=0.008). The interaction term between BMI and 6MWD was not significant in mixed-effects ANOVA analyses. Patients with BMI > 35 and ≤ 40 kg/m2 (n=34) were generally similar to those with BMI > 40 kg/m2 (n=14) with the exception that those with BMI > 40 kg/m2 were younger (age 46 ± 4.96 yrs versus 51.3 ± 10.83 yrs; p=0.03) and less commonly diabetic (21.4% versus 58.8%; p=0.03). Other baseline characteristics as well as LVAD parameters were similar between
10 Kiernan HVAD BMI groups (data not shown). Outcomes were also similar between these groups with no differences observed in survival (Supplementary Figure 1), QOL metrics or 6MWD, although analysis is constrained by the limited number of patients in each group. Incidence of driveline exit site infection and right heart failure were also similar (Supplementary figures 2-3). Rate of localized non-device infection was lower in the BMI > 40 kg/m2 group (0.23 vs. 0.75 EPPY: p=0.02) as was bleeding (0.41 vs. 1.1 EPPY; p=0.05). There were no differences in other adverse events.
Discussion While other studies have included either mixed LVAD populations or exclusively axial flow device recipients(9,10), this report specifically examines outcomes of severely obese patients receiving a centrifugal flow LVAD. The proportion of patients with severe obesity undergoing LVAD implantation in the study population (12.6%) was similar to what has been previously reported(8,9). The primary findings of this study can be summarized as following: 1) survival is similar for those with and without severe obesity following CF-LVAD implantation, 2) severely obese patients are at higher risks of certain adverse events, 3) severely obese patients generally do not lose weight with LVAD therapy, 4) severely obese patients experience similar improvements in HRQL outcomes as the non-severely obese, 5) functional capacity improves in the severely obese, although to a lesser extent than those without severe obesity. The proportion of patients with severe obesity undergoing LVAD implantation in the study population (12.6%) was similar to what has been reported in earlier studies. Among 3,856 patients bridged to transplant with a continuous-flow LVAD in the UNOS registry, 10.2% had BMI ≥ 35 kg/m2 (9). Similarly, of 896 undergoing implantation of an axial-flow LVAD, 10% had BMI ≥ 35 kg/m2 (8). BTT patients with severe obesity in the present study were younger than
11 Kiernan HVAD BMI controls, which may speak to the willingness to support younger patients awaiting transplantation despite severe obesity representing a relative contraindication to transplantation. Pump parameters were significantly higher in severely obese patients, which likely points toward the assumption that larger patients require higher LVAD flow. Independent of baseline severe obesity, survival in this cohort of centrifugal flow LVAD recipients exceeded 83% for both groups at one year. The finding that severe obesity does not impact long-term survival is consistent with reports from other present-day cohorts(8-10,12-14). While one year survival was similar between groups, two-year survival was 61% for severely obese patients compared to 74% for patients with BMI ≤ 35 kg/m2. While statistically not significant, decreased long-term survival for the severely obese cannot be excluded. An analysis from the UNOS registry found that after adjusting for other risk factors, there was a trend toward increased risk of death or delisting among CF-LVAD recipients with a BMI ≥ 35 kg/m2 (9). The two-year survival on centrifugal devices was comparable to survival in the HeartMate II clinical trials where normal weight subjects had two-year survival of 60% compared to 68% in those with BMI ≥35 kg/m2 (8). Obese patients had approximately twice the rate of driveline exit site infection compared to the control group (0.40 EPPY compared to 0.22 EPPY). The multicenter HeartMate II experience reported similar increased infections among severely obese patients(8). Similarly in a recent analysis from the ISHLT Mechanically Assisted Circulatory Support (IMACS) registry, patients with severe obesity (defined as BMI ≥ 40 kg/m2) were less likely to be free from a device-related infection at two years compared to patients in lower BMI strata(12). Not all studies, however, have identified obesity to be associated with increased risk of infection(10,11,14,18). One explanation for this difference in some analyses may be attributed
12 Kiernan HVAD BMI to a lower BMI cutoff of 30 mg/kg2 with fewer patients with a BMI ≥ 35 mg/kg2, as mild obesity may not be a risk factor for infectious complications11. The increased risk of infection in patients with severe obesity may be attributable to difficulties in maintaining cleanliness of the exit site, increased strain on the percutaneous lead at the exit site associated with positional changes, and differences in immunity associated with obesity(19). We did not find a difference in devicerelated infection between patients with BMI > 35 and ≤ 40 kg/m2 compared to those with BMI > 40 kg/m2, but these comparisons are limited by a small samples size. Renal failure was also higher for severely obese patients, an observation also made in the Novacor clinical trials(18). Other investigators have reported obesity as an independent predictor of renal complications(20) and have hypothesized that obese patients have more renal complications because of the increased presence of type II diabetes and hypertension in those patients(21). A higher proportion of severely obese patients not surprisingly also had diabetes in the present analyses. Another possible mechanistic explanation for the increased renal failure is the link between obesity and right heart failure (RHF). Severely obese VAD candidates have been observed to have elevated right atrial pressure(14), which has been linked to cardiorenal disease(22,23). Likewise, a single-center study of 383 LVAD recipients found that RHF was more common in patients with BMI ≥ 35 kg/m2 (14). Baseline central venous pressure was 14 mmHg in obese subjects compared to 10 mmHg in the non-obese. While severely obese patients in the present study did not have higher rates of RHF based on the INTERMACS definition, this definition captured only those with severe RHF events (severe RHF requiring RVAD, inhaled vasodilators, or prolonged inotropic support). It is conceivable that there is a cohort of patients with less severe, yet clinically significant RHF that may not have been captured as a RHF adverse event. These patients may be at risk for subsequently developing worsening renal
13 Kiernan HVAD BMI function. Paired with the recent temporal analysis of adverse events showing RHF is most common in the first 30 days post implantation(24), this is a potential explanation for the increased acute renal dysfunction observed in the obese group, although these observations require further exploration. In contrast to one prior investigation(18), we found bleeding requiring reoperation was less common in the severely obese cohort. Elsewhere, an IMACS analysis as well as another smaller study reported that early bleeding is more common in underweight individuals rather than the obese(8,12). The present study utilized a binary BMI threshold to examine adverse event rates in the severely obese. It is possible that increased bleeding in those patients at the lower BMI extreme led to the finding of less bleeding requiring reoperation among the severely obese. Novel to the present study is the reported impact of severe obesity on both quality of life and functional capacity outcomes. A prior study also found no impact of BMI on quality of life in HeartMate II recipients(25). These outcomes are particularly relevant given that patients who may be contemplating DT therapy emphasize the importance of improvements in quality of life and functionality even if at the expense of longevity(26,27). Severely obese patients listed for transplantation will likely have increased wait times until transplant likewise making quality of life and functional capacity very relevant to their care. While severely obese patients had improvement in 6MWD, these gains were on average lower than patients without severe obesity (Figure 5C). This suggests that severe obesity may be associated with a lower ceiling for functional capacity recovery following LVAD implantation similar to other comorbidities(25). Severely obese patients should be counseled that while their functional capacity should increase, the magnitude of this improvement will likely be limited by persistent obesity. Furthermore,
14 Kiernan HVAD BMI while exercise tolerance defined 6MWD increased in obese subjects, this did not translate to weight loss and obese subjects had an average increase in BMI over the first year of therapy of 0.7 kg/m2. These data are important as collectively they demonstrate that despite differences in flow dynamics and unloading characteristics(28), centrifugal flow LVADs adequately support large patients, who benefit from similar improvements in HRQOL, functional capacity and survival as those without severe obesity. Heightened understanding of the risk profile associated with severe obesity is important to better inform patient selection. Counseling patients about these outcomes prior to surgery sets appropriate expectations for patients and caregivers regarding the anticipated benefits of LVAD therapy. As has been previously reported, despite the increased functional capacity afforded by LVAD therapy, obese patients rarely lose weight during LVAD support(9,29). The greater room for improvement in non-severely obese LVAD recipients may provide credence for the use of weight loss surgery in these patients(30). Further investigation into the combination of weight loss strategies and LVAD therapies deserves consideration as case studies of patients with bariatric surgery following LVAD implantation have reported positive results(31,32). With careful patient selection, severe obesity should generally not represent a contraindication to LVAD therapy. However, the increased risk of adverse events and smaller anticipated improvements in functional capacity should be discussed during the shared decision-making process.
Limitations There are several inherent limitations to post-hoc analyses of clinical trial data, including that these analyses may be underpowered to detect clinically relevant differences. Until longer
15 Kiernan HVAD BMI term outcomes can be verified from larger multi-center patient registries, these findings should be interpreted thoughtfully and in context of the broader literature. Furthermore, these data were from a BTT population, and thus the most obese HF patients may have been excluded from this study cohort. Few patients had BMI > 40 kg/m2 and the findings of this study cannot be extrapolated to those with more extreme obesity. Furthermore, as would be expected, obese patients were more likely to have diabetes. It is possible that diabetes or other comorbid conditions such as hypertension, that are more common in obese patients, were responsible for the higher rates of renal dysfunction and driveline exit site infection in this cohort. The small sample size of severely obese patients limits statistical modeling to adjust for potential confounders when attempting to clarify the independent association between severe obesity and these adverse events. Also, although cardiac cachexia has been identified as a risk factor for mortality in some studies following LVAD surgery, this was not the focus of the present study and other BMI stratums were not investigated(18). Had underweight patients been excluded from the control group, it is possible that observed differences may have been greater between the severely obese and control patients. While patients were dichotomized based on a BMI threshold of 35 kg/m2, whereas there is more likely a spectrum of risk associated with increasing BMI as it relates to specific adverse events. Nevertheless, utilizing the NIH BMI thresholds for severe obesity (Obesity Class 2 - BMI 35-39 kg/m2 and Class 3 - BMI > 40 kg/m2) helps to contextualize these findings to better inform the patient selection process for LVAD therapy. Finally, obesity-specific HRQL tools were also not available for comparison and findings from HF-specific tools may not be sensitive to evaluate QOL changes relevant to obese subjects. The EQ-5D, however, is a generic HRQL tool that assesses five dimensions relevant across diagnoses: mobility, self-care, usual activities, pain/discomfort and anxiety/ depression. Changes
16 Kiernan HVAD BMI in the HF-specific KCCQ and generic EQ-5D were similar, increasing confidence that the improvements of QOL observed in obese subjects are clinically significant.
Conclusions In this large, multi-center trial of centrifugal flow LVAD recipients, BMI > 35 kg/m2 was not associated with reduced survival although there was a greater risk of certain adverse events. Severely obese patients experienced similar quality of life improvements as the control group. While the severely obese group also experienced increased functional capacity, this improvement was generally less pronounced compared to those patients without severe obesity. Awareness of the increased risk of certain adverse events is vital to guide shared decision making.
Acknowledgements The authors acknowledge Ming-Jay Chow, Medtronic Inc., for his assistance in the preparation of the manuscript.
17 Kiernan HVAD BMI References 1.
2.
3.
4. 5. 6.
7.
8.
9.
10. 11. 12.
13.
14. 15. 16.
17.
Rogers JG, Aaronson KD, Boyle AJ et al. Continuous Flow Left Ventricular Assist Device Improves Functional Capacity and Quality of Life of Advanced Heart Failure Patients. J Am Coll Cardiol 2010;55:1826-1834. Slaughter MS, Pagani FD, McGee EC et al. Use of the HeartWare® Ventricular Assist System for Bridge to Transplant: Combined Results of the ADVANCE and CAP Trial. Journal of Heart and Lung Transplant 2013;32:675-683. Mozaffarian D, Benjamin EJ, Go AS et al. Executive Summary: Heart Disease and Stroke Statistics--2016 Update: A Report From the American Heart Association. Circulation 2016;133:447-54. Reddy YN, Melenovsky V, Redfield MM, Nishimura RA, Borlaug BA. High-Output Heart Failure: A 15-Year Experience. J Am Coll Cardiol 2016;68:473-82. Macha M, Molina EJ, Franco M et al. Pre-transplant obesity in heart transplantation: are there predictors of worse outcomes? Scand Cardiovasc J 2009;43:304-10. Russo MJ, Hong KN, Davies RR et al. The effect of body mass index on survival following heart transplantation: do outcomes support consensus guidelines? Ann Surg 2010;251:144-52. Mehra MR, Canter CE, Hannan MM et al. The 2016 International Society for Heart Lung Transplantation listing criteria for heart transplantation: A 10-year update. J Heart Lung Transplant 2016;35:1-23. Brewer RJ, Lanfear DE, Sai-Sudhakar CB et al. Extremes of body mass index do not impact mid-term survival after continuous-flow left ventricular assist device implantation. J Heart Lung Transplant 2012;31:167-72. Clerkin KJ, Naka Y, Mancini DM, Colombo PC, Topkara VK. The Impact of Obesity on Patients Bridged to Transplantation With Continuous-Flow Left Ventricular Assist Devices. JACC Heart Fail 2016;4:761-8. Mohamedali B, Yost G, Bhat G. Obesity as a Risk Factor for Consideration for Left Ventricular Assist Devices. J Card Fail 2015;21:800-5. Musci M, Loforte A, Potapov EV et al. Body Mass Index and Outcome After Ventricular Assist Device Placement. The Annals of Thoracic Surgery 2008;86:1236-1242. Forest SJ, Xie R, Kirklin JK et al. Impact of body mass index on adverse events after implantation of left ventricular assist devices: An IMACS registry analysis. J Heart Lung Transplant 2018;37:1207-1217. Volkovicher N, Kurihara C, Critsinelis A et al. Effect of obesity on outcomes in patients undergoing implantation of continuous-flow left ventricular assist devices. J Artif Organs 2018;21:180-187. Yost G, Coyle L, Gallagher C et al. The impact of extreme obesity on outcomes after left ventricular assist device implantation. J Thorac Dis 2017;9:4441-4446. Kirklin JK, Naftel DC, Pagani FD et al. Seventh INTERMACS annual report: 15,000 patients and counting. J Heart Lung Transplant 2015;34:1495-504. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults--The Evidence Report. National Institutes of Health. Obes Res 1998;6 Suppl 2:51S-209S. Holman WL, Pae WE, Teutenberg JJ et al. INTERMACS: Interval Analysis of Registry Data. Journal of the American College of Surgeons 2009;208:755-761.
18 Kiernan HVAD BMI 18. 19. 20.
21.
22.
23.
24.
25.
26.
27. 28. 29.
30. 31.
32.
Butler J, Howser R, Portner PM, Pierson RN, 3rd. Body mass index and outcomes after left ventricular assist device placement. Ann Thorac Surg 2005;79:66-73. Falagas ME, Kompoti M. Obesity and infection. The Lancet Infectious diseases 2006;6:438-46. Calder CL, Ortega G, Vij A et al. Morbid obesity is an independent risk factor for postoperative renal dysfunction in young adults: a review of the American College of Surgeons National Surgical Quality Improvement Program database. The American Journal of Surgery 2016;211:772-777. Ejerblad E, Fored CM, Lindblad P, Fryzek J, McLaughlin JK, Nyren O. Obesity and risk for chronic renal failure. Journal of the American Society of Nephrology : JASN 2006;17:1695-702. Mullens W, Abrahams Z, Francis GS et al. Importance of venous congestion for worsening of renal function in advanced decompensated heart failure. J Am Coll Cardiol 2009;53:589-96. Testani JM, McCauley BD, Kimmel SE, Shannon RP. Characteristics of patients with improvement or worsening in renal function during treatment of acute decompensated heart failure. Am J Cardiol 2010;106:1763-9. Maltais S, Aaronson KD, Teuteberg JJ et al. Adverse Event Rates Change Favorably Over Time for Patients Bridged With the HeartWare Left Ventricular Assist Device. ASAIO journal (American Society for Artificial Internal Organs : 1992) 2017. Kiernan MS, Sundareswaran KS, Pham DT et al. Preoperative Determinants of Quality of Life and Functional Capacity Response to Left Ventricular Assist Device Therapy. J Card Fail 2016;22:797-805. Lewis EF, Johnson PA, Johnson W, Collins C, Griffin L, Stevenson LW. Preferences for quality of life or survival expressed by patients with heart failure. J Heart Lung Transplant 2001;20:1016-24. Stevenson LW, Hellkamp AS, Leier CV et al. Changing preferences for survival after hospitalization with advanced heart failure. J Am Coll Cardiol 2008;52:1702-8. Lim HS, Howell N, Ranasinghe A. The Physiology of Continuous-Flow Left Ventricular Assist Devices. Journal of cardiac failure 2017;23:169-180. Emani S, Brewer RJ, John R et al. Patients with low compared with high body mass index gain more weight after implantation of a continuous-flow left ventricular assist device. The Journal of Heart and Lung Transplantation 2013;32:31-35. Zarraga IGE, Schwarz ER. Impact of Dietary Patterns and Interventions on Cardiovascular Health. Circulation 2006;114:961-973. Lockard KL, Allen C, Lohmann D et al. Bariatric surgery for a patient with a HeartMate II ventricular assist device for destination therapy. Progress in transplantation (Aliso Viejo, Calif) 2013;23:28-32. Leviner DB, Keidar A, Ben-Gal T, Medalion B. Cardiac function recovery following LVAD implantation and bariatric surgery in a morbidly obese patient. J Card Surg 2014;29:740-2.
19 Kiernan HVAD BMI Figure 1: Distribution of Baseline Body Mass Index - The number of patients in each BMI stratum is provided.
20 Kiernan HVAD BMI Figure 2: Kaplan Meier Survival of Morbidly Obese and Control LVAD Patients. (P-value denotes comparison of two years of support.)
Figure 3: Freedom from Driveline Exit Site Infection for Morbidly Obese and Control LVAD Patients. (P-value denotes comparison of two years of support.)
21 Kiernan HVAD BMI Figure 4: BMI (4a) and albumin (4b) in Morbidly Obese and Control LVAD Patients across Time. (P-values for within group comparisons to the baseline measurements are shown in the columns and with asterisks denoting a significant difference from baseline measurements (p<0.05). Brackets and p-values above pairs of columns denote between group comparisons at that time point.)
22 Kiernan HVAD BMI Figure 5: Quality of Life and Functional Capacity of Morbidly Obese and Control LVAD Patients. (P-values for within group comparisons to the baseline measurements are shown in the columns with asterisks denoting a significant difference from baseline measurements (p<0.05). Brackets and p-values above pairs of columns denote between group comparisons at that time point.)
23 Kiernan HVAD BMI Table 1: Baseline Characteristics and Pump Parameters for Morbidly Obese Patients and Controls Morbidly Obese BMI > 35 kg/m2 (N= 48) 49.7±9.75 25.0
Control BMI ≤ 35 kg/m2 (N=334) 53.7±11.90 29.3
64.6
68.6
35.4 0.0
25.1 6.3
175.7±9.80 119.9±14.40 38.8±3.16 2.4±0.20 29.2 3.5±0.74 (44)* 47.9 56.5 (23)*
174.0±10.30 81.2±17.92 26.7±4.69 2.0±0.26 39.2 3.5±0.70 (321)* 33.2 51.4 (111)*
1.4±0.48
1.3±0.49
2.1 35.4 41.7 20.8
6.0 34.7 40.4 18.9
Hemodynamics (Pre-Implant) MAP (mmHg) Cardiac Index (L/min/m^2)
76.5±8.54 (27)* 2.3±0.63 (33)*
77.7±11.29 (210)* 2.2±0.61 (246)*
0.62 0.37
PCWP (mmHg)
23.7±10.50 (29)*
22.4±8.98 (204)*
0.47
RAP or CVP (mmHg)
13.4±5.94 (32)*
10.6±6.59 (222)*
0.02
Flow (L/min)
5.7±0.9
5.0±0.9 (333)*
<0.0001
Speed (RPM)
2828±198
2729±172 (333)*
0.0003
Power (Watts)
5.0±1.0
4.2±0.8 (333)*
<0.0001
Baseline Characteristics Age (years) Female sex (%) Race (%) Caucasian Black/African American Other Height (cm) Weight (kg) BMI (kg/m^2) Body Surface Area (m^2) Ischemic Cause of Heart Failure (%) Albumin (g/dL) Diabetes (%) Insulin Dependent Diabetes (%) Creatinine (mg/dL) INTERMACS Patient Profile (%) 1 2 3 4-7
P value 0.03 0.61 0.08
0.28 <0.0001 <0.0001 <0.0001 0.21 0.52 0.05 0.82 0.27 0.82
VAD Parameters
24 Kiernan HVAD BMI BMI = body mass index, MAP = mean arterial blood pressure, PCWP = pulmonary capillary wedge pressure, RAP = right atrial pressure, CVP = central venous pressure. * the number of patients with data for a given parameter is in the parentheses. All other values were calculated from the full (n=48) morbidly obese and (n=334) control groups.
Table 2: INTERMACS Adverse Events for Obese and Not-Obese Patients Morbidly Obese Control INTERMACS Category EPPY no. of EPPY no. of Adverse Events (67.09 PYtotal) events (339.54 PYtotal) events Bleeding 0.88 59 0.92 313 Re-Operation 0.06 4 0.18 60 GI 0.31 21 0.26 87 Cardiac Arrhythmia 0.54 36 0.50 169 Device Malfunction/Failure 0.28 19 0.27 90 Hemolysis 0.04 3 0.06 21 Infection 1.55 104 1.08 368 Sepsis 0.31 21 0.21 71 Driveline Exit Site 0.40 27 0.22 75
P value* 0.83 0.01 0.46 0.75 0.80 0.58 0.03 0.12 0.01 0.32 0.58 0.53
Neurological 0.18 12 0.24 82 Ischemic CVA 0.04 3 0.06 21 Hemorrhagic CVA 0.10 7 0.08 27 Renal Dysfunction 0.25 17 0.09 29 0.0009 Acute 0.24 16 0.09 29 0.002 Chronic 0.01 1 0.00 0 0.058 Right Heart Failure 0.37 25 0.37 124 0.94 EPPY = events per patient year, GI = gastrointestinal, CVA = cerebrovascular accident, *p-values are for comparisons of EPPY
Journal Pre-proof
Outcomes of Severely Obese Patients Supported by a Centrifugal-Flow Left Ventricular Assist Device Michael S. Kiernan MD, MS , Samer S. Najjar MD , Amanda R. Vest MBBS, MPH , Emma J. Birks MBBS, PhD, BSc, FRCP , Nir Uriel MD , Gregory A. Ewald MD , Katrin Leadley MD , Chetan B. Patel MD PII: DOI: Reference:
S1071-9164(19)30502-0 https://doi.org/10.1016/j.cardfail.2019.10.013 YJCAF 4439
To appear in:
Journal of Cardiac Failure
Received date: Revised date: Accepted date:
8 May 2019 16 October 2019 29 October 2019
Please cite this article as: Michael S. Kiernan MD, MS , Samer S. Najjar MD , Amanda R. Vest MBBS, MPH , Emma J. Birks MBBS, PhD, BSc, FRCP , Nir Uriel MD , Gregory A. Ewald MD , Katrin Leadley MD , Chetan B. Patel MD , Outcomes of Severely Obese Patients Supported by a Centrifugal-Flow Left Ventricular Assist Device, Journal of Cardiac Failure (2019), doi: https://doi.org/10.1016/j.cardfail.2019.10.013
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.
HIGHLIGHTS
CF-LVAD recipients, regardless of severe obesity, had improvements in quality of life
Body mass index > 35 kg/m2 was not associated with reduced survival
Severe obesity is associated with increased risk of driveline infection
Functional capacity gains were more limited for the severely obese, while improvements in quality of life were similar to patients without severe obesity
Awareness of increased adverse event risk associated with severe obesity is necessary to better inform shared decision making
2 Kiernan HVAD BMI Title:
Outcomes of Severely Obese Patients Supported by a Centrifugal-Flow Left Ventricular Assist Device
Authors:
Michael S Kiernan1, MD, MS, Samer S Najjar2, MD, Amanda R Vest1, MBBS, MPH, Emma J Birks3, MBBS, PhD, BSc, FRCP, Nir Uriel4, MD, Gregory A Ewald5, MD, Katrin Leadley6, MD, Chetan B Patel7, MD
Author affiliations: 1
Tufts Medical Center, Boston, MA; 2MedStar Washington Hospital Center, Washington,
DC; 3University of Louisville, Louisville, KY; 4University of Chicago, Chicago, IL; 5
Washington University/Barnes-Jewish Hospital, St. Louis, MO; 6Endotronix Inc. Lisle, IL;
7
Duke University, Durham, NC
Short Title:
Outcomes of HVAD Patients with Severe Obesity
Address for correspondence: Michael S. Kiernan, MD, MSc Department of Cardiology, Tufts Medical Center 800 Washington Street, Box 5931 Boston, MA 02111 Telephone:
617-636-8068
Fax:
617-636-6030
E-mail:
[email protected]
3 Kiernan HVAD BMI Relationships with Industry MSK is a consultant for Medtronic and has received travel grant from Abbott. SSN receives research grant support from and is on the advisory board for Medtronic. EJB receives research grant support from Medtronic and Abbott. NU is a consultant for Medtronic, Abbott, and XDx Inc. GAE is a consultant for Medtronic. KL was an employee of Medtronic. CBP receives conference fee support from Medtronic and Abbott. None of the other authors have a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose.
4 Kiernan HVAD BMI Abstract: Background: Ventricular assist devices provide improved outcomes for advanced heart failure patients, but their benefit in the severely obese is not well documented. Methods: Patients enrolled in the HeartWare ADVANCE trial (n=382) were divided into two body mass index (BMI) groups. Patients with severe obesity (> 35 kg/m2) were compared to a control group with BMI ≤ 35 kg/m2. The association of BMI with survival was tested using Kaplan-Meier analysis and major adverse events were compared. Results: At implantation, 48 (13%) of patients were severely obese. There was no difference in survival through two years of support between severely obese patients and the control group. Severely obese patients were at higher risk of driveline infection (p=0.01) and acute renal dysfunction (p=0.002).
Both groups experienced similar improvements in quality of life.
Functional capacity improved in both severely obese and control patients, although severely obese patients had smaller improvements than controls in their six-minute walk scores. Conclusions: Despite an increased risk of adverse events, severe obesity was not associated with reduced survival or quality of life. A better understanding of the risks and benefits of LVAD therapy in obese patients will help in the shared decision-making of the patient selection process.
Key Words: Left Ventricular Assist Device, Patient Outcomes, Obesity
5 Kiernan HVAD BMI Introduction With advancements in technology leading to improvements in survival as well as quality of life(1,2), left ventricular assist devices (LVADs) are increasingly used to support patients with end-stage heart failure as either a bridge to cardiac transplantation (BTT) or as destination therapy (DT). Concurrent with the epidemic of heart failure(3), is also the problem of obesity which is present in greater than one third of U.S. adults when defined as a body mass index [BMI] > 30 kg/m2. Obesity is a risk factor for the development of heart failure(4) and many obese patients will progress to end-stage disease requiring evaluation for cardiac transplantation and LVAD therapy. Among heart transplant candidates, obesity is associated with increased wait times and may lead to worse outcomes following cardiac transplantation(5,6). Accordingly, society guidelines recommend that patients achieve a BMI ≤ 35 kg/m2 prior to listing for cardiac transplantation (7). In contrast to the heart transplant literature, differences in long-term survival have generally not been seen between obese and non-obese patients undergoing LVAD implantation(8-14). There are mixed findings as to whether obesity affects the rates of certain adverse events such as device infection, but differences in results may be due to varying definitions of obesity utilized as well as the heterogeneity of the populations previously studied. With the expanding use of LVADs(15), appropriate patient selection is important to maintain optimal outcomes. While outcomes of obese patients undergoing implantation of an axial flow LVAD have been reported(8), it remains unclear if these findings are device specific or applicable to all technologies. We performed a retrospective analysis of the HeartWare ADVANCE trial and continued access protocol (CAP) to further explore outcomes of severely obese patients undergoing
6 Kiernan HVAD BMI implantation of a centrifugal flow LVAD as a bridge to cardiac transplantation. The primary aim was to determine the impact of severe obesity on survival, adverse events, health related quality of life (HRQL), and functional capacity in CF-LVAD recipients.
Methods The HeartWare HVAD system was evaluated as a BTT therapy in the United States in a prospective multicenter trial. The study design of the ADVANCE bridge-to-transplant and CAP trials have been previously described(2). A total of 382 patients received an HVAD between August 2008 and November 2012. The National Institute of Health (NIH) classifies obesity as follows: Class I – 30.0 to 34.9 kg/m2; Class II – 35.0 – 39.9 kg/m2; Class III - ≥ 40 kg/m2 (16). Patients from the ADVANCE trial were divided into two groups. The severely obese cohort was comprised of patients with body mass index > 35 kg/m2 at the time of implant and the nonseverely obese (control group) included patients with BMI ≤ 35 kg/m2. Adverse events were defined using Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) definitions(17) and patients were followed to transplant, death, or alive at last follow up. The studies were conducted in compliance with FDA regulations for Good Clinical Practices and approved by the Institutional Review Board designated by each participating study site. Statistical analysis Descriptive statistics were used to evaluate baseline clinical and demographic characteristics as well as LVAD parameters. Results are reported as mean ± standard deviation for continuous variables and as percent (count x 100/sample size) for binary variables. Comparisons between the obese and non-obese patients are made with a two-sample t-test for
7 Kiernan HVAD BMI continuous variables and Fisher’s exact test for categoric variables. Adverse events are reported as events per patient year (number of events) and comparisons between the severely obese and control groups were made with Poisson regression. Kaplan-Meier methodology is used to generate survival estimates and freedom from driveline infection curves for the two groups. Several HRQL and functional capacity measures including Kansas City Cardiomyopathy Questionnaire (KCCQ) Overall Summary Score, EuroQOL-5D (EQ-5D) Overall Health State Score, and six minute walk distance (6MWD) were assessed at baseline, 6, 12, and 24 month visits. A score of zero was imputed if the site indicated the six minute walk test could not be performed. Comparisons of the follow up visits to the baseline measurements were made with a paired t-test while differences between the obese and control groups were assessed with a two sample t-test. Following the same methodology, patient characteristics of BMI and serum albumin levels were also compared. To account for multiple hypothesis testing and thereby reduce the risk of Type I error, mixed effect ANOVA modeling was also used to allow for variability across patients and the fixed effects of BMI and time. A sensitivity analysis was also performed to compare characterisitics as well as outcomes between patients with a BMI > 35 and ≤ 40 kg/m2 versus > 40 kg/m2. All statistical analyses are performed with SAS v. 9.2 software (SAS Institute, Cary, NC).
Results The distribution of baseline body mass index of the 382 patients enrolled in the ADVANCE and CAP trials ranged from 14.89 to 48.09 (Figure 1). A total of 48 patients had BMI > 35 kg/m2 (severely obese) while 334 had BMI ≤ 35 kg/m2 (controls). Baseline characteristics and LVAD parameters of severely obese patients and controls are presented in
8 Kiernan HVAD BMI Table 1. The severely obese group has significantly higher weight and body surface area (BSA) compared to controls. The severely obese group was younger, more commonly diabetic, and had a higher proportion of black patients. Right atrial (RAP) was higher in severely obese patients: 13.4 ± 5.94 mmHg versus 10.6 ± 6.59 mmHg (controls). LVAD parameters were also different between groups. Speed settings were higher in severely obese patients compared to controls, as were measured power and calculated flow. Preoperative hemodynamic data were not available for the full cohort; the number of patients included is detailed in Table 1. One-year survival was similar in the severely obese (86.0%) and control groups (83.2%), and did not differ between groups across two-years of follow up (p=0.45) (Figure 2). There were differences in rates of adverse events between patients with and without severe obesity (Table 2). While rates of bleeding were similar between groups, severely obese patients were less likely to require re-operation for bleeding (0.06 EPPY) than the control group (0.18 EPPY) (p=0.01). Infection was more common in patients with severe obesity: 1.55 compared to 1.08 EPPY (p=0.03). In Figure 3, the one-year freedom from driveline exit site infection was 60.5% (severely obese) versus 81.3% (controls). The rate of sepsis was similar between groups. Acute and chronic renal dysfunction were more commonly reported in patients with severe obesity. BMI increased in both patients with and without severe obesity during the first 24 months of therapy, although this change was only significant for those with baseline BMI ≤ 35 kg/m 2 (Figure 4A). The increase in BMI from baseline to one year was 0.7 ± 5.7 kg/m2 (p=0.56) in the severely obese cohort and 1.8 ± 3.9 kg/m2 in the control group (p<0.0001). Compared to baseline, albumin increased significantly in both the obese (3.5 ± 0.74 g/dL vs. 3.9 ± 0.48 g/dL, p=0.02) and control (3.5 ± 0.70 g/dL vs. 3.9 ± 0.60 g/dL, p<0.0001) groups at 6 months
9 Kiernan HVAD BMI following HVAD implantation. These changes were sustained through 2 years of follow up (Figure 4B). At six months, both severely obese patients and the control group had marked improvements in HRQL as measured by both the EQ-5D and KCCQ scores (Figure 5A and 5B). HRQL improved similarly between groups across follow-up and these improvements were sustained through 2 years of support. Average increase in KCCQ from baseline to six months was 31.6 ± 27.2 (p<0.0001) and 31.0 ± 24.3 (p<0.0001) for severely obese and control groups respectively (p=0.27 for between group comparison at six months). Using mixed-effects ANOVA, there were no statistically significant interaction terms between BMI and KCCQ scores. While change in EQ-5D was significant overall (p=0.02), no single time point was significant. Functional capacity as measured by 6MWD also improved significantly in both groups compared to baseline with sustained improvements through 2 years of follow up. The control group, however, had greater improvements compared to the severely obese patients. At 12 months, the increase in 6MWD was 106.0 ± 191.8m (p=0.009) for severely obese patients and 199.2 ± 216.2m (p<0.0001) for controls (Figure 5C). The actual twelve-month 6MWD was 268.1 ± 189.6m for controls compared to 164.9 ± 176.0m for those with severely obesity (p=0.008). The interaction term between BMI and 6MWD was not significant in mixed-effects ANOVA analyses. Patients with BMI > 35 and ≤ 40 kg/m2 (n=34) were generally similar to those with BMI > 40 kg/m2 (n=14) with the exception that those with BMI > 40 kg/m2 were younger (age 46 ± 4.96 yrs versus 51.3 ± 10.83 yrs; p=0.03) and less commonly diabetic (21.4% versus 58.8%; p=0.03). Other baseline characteristics as well as LVAD parameters were similar between
10 Kiernan HVAD BMI groups (data not shown). Outcomes were also similar between these groups with no differences observed in survival (Supplementary Figure 1), QOL metrics or 6MWD, although analysis is constrained by the limited number of patients in each group. Incidence of driveline exit site infection and right heart failure were also similar (Supplementary figures 2-3). Rate of localized non-device infection was lower in the BMI > 40 kg/m2 group (0.23 vs. 0.75 EPPY: p=0.02) as was bleeding (0.41 vs. 1.1 EPPY; p=0.05). There were no differences in other adverse events.
Discussion While other studies have included either mixed LVAD populations or exclusively axial flow device recipients(9,10), this report specifically examines outcomes of severely obese patients receiving a centrifugal flow LVAD. The proportion of patients with severe obesity undergoing LVAD implantation in the study population (12.6%) was similar to what has been previously reported(8,9). The primary findings of this study can be summarized as following: 1) survival is similar for those with and without severe obesity following CF-LVAD implantation, 2) severely obese patients are at higher risks of certain adverse events, 3) severely obese patients generally do not lose weight with LVAD therapy, 4) severely obese patients experience similar improvements in HRQL outcomes as the non-severely obese, 5) functional capacity improves in the severely obese, although to a lesser extent than those without severe obesity. The proportion of patients with severe obesity undergoing LVAD implantation in the study population (12.6%) was similar to what has been reported in earlier studies. Among 3,856 patients bridged to transplant with a continuous-flow LVAD in the UNOS registry, 10.2% had BMI ≥ 35 kg/m2 (9). Similarly, of 896 undergoing implantation of an axial-flow LVAD, 10% had BMI ≥ 35 kg/m2 (8). BTT patients with severe obesity in the present study were younger than
11 Kiernan HVAD BMI controls, which may speak to the willingness to support younger patients awaiting transplantation despite severe obesity representing a relative contraindication to transplantation. Pump parameters were significantly higher in severely obese patients, which likely points toward the assumption that larger patients require higher LVAD flow. Independent of baseline severe obesity, survival in this cohort of centrifugal flow LVAD recipients exceeded 83% for both groups at one year. The finding that severe obesity does not impact long-term survival is consistent with reports from other present-day cohorts(8-10,12-14). While one year survival was similar between groups, two-year survival was 61% for severely obese patients compared to 74% for patients with BMI ≤ 35 kg/m2. While statistically not significant, decreased long-term survival for the severely obese cannot be excluded. An analysis from the UNOS registry found that after adjusting for other risk factors, there was a trend toward increased risk of death or delisting among CF-LVAD recipients with a BMI ≥ 35 kg/m2 (9). The two-year survival on centrifugal devices was comparable to survival in the HeartMate II clinical trials where normal weight subjects had two-year survival of 60% compared to 68% in those with BMI ≥35 kg/m2 (8). Obese patients had approximately twice the rate of driveline exit site infection compared to the control group (0.40 EPPY compared to 0.22 EPPY). The multicenter HeartMate II experience reported similar increased infections among severely obese patients(8). Similarly in a recent analysis from the ISHLT Mechanically Assisted Circulatory Support (IMACS) registry, patients with severe obesity (defined as BMI ≥ 40 kg/m2) were less likely to be free from a device-related infection at two years compared to patients in lower BMI strata(12). Not all studies, however, have identified obesity to be associated with increased risk of infection(10,11,14,18). One explanation for this difference in some analyses may be attributed
12 Kiernan HVAD BMI to a lower BMI cutoff of 30 mg/kg2 with fewer patients with a BMI ≥ 35 mg/kg2, as mild obesity may not be a risk factor for infectious complications11. The increased risk of infection in patients with severe obesity may be attributable to difficulties in maintaining cleanliness of the exit site, increased strain on the percutaneous lead at the exit site associated with positional changes, and differences in immunity associated with obesity(19). We did not find a difference in devicerelated infection between patients with BMI > 35 and ≤ 40 kg/m2 compared to those with BMI > 40 kg/m2, but these comparisons are limited by a small samples size. Renal failure was also higher for severely obese patients, an observation also made in the Novacor clinical trials(18). Other investigators have reported obesity as an independent predictor of renal complications(20) and have hypothesized that obese patients have more renal complications because of the increased presence of type II diabetes and hypertension in those patients(21). A higher proportion of severely obese patients not surprisingly also had diabetes in the present analyses. Another possible mechanistic explanation for the increased renal failure is the link between obesity and right heart failure (RHF). Severely obese VAD candidates have been observed to have elevated right atrial pressure(14), which has been linked to cardiorenal disease(22,23). Likewise, a single-center study of 383 LVAD recipients found that RHF was more common in patients with BMI ≥ 35 kg/m2 (14). Baseline central venous pressure was 14 mmHg in obese subjects compared to 10 mmHg in the non-obese. While severely obese patients in the present study did not have higher rates of RHF based on the INTERMACS definition, this definition captured only those with severe RHF events (severe RHF requiring RVAD, inhaled vasodilators, or prolonged inotropic support). It is conceivable that there is a cohort of patients with less severe, yet clinically significant RHF that may not have been captured as a RHF adverse event. These patients may be at risk for subsequently developing worsening renal
13 Kiernan HVAD BMI function. Paired with the recent temporal analysis of adverse events showing RHF is most common in the first 30 days post implantation(24), this is a potential explanation for the increased acute renal dysfunction observed in the obese group, although these observations require further exploration. In contrast to one prior investigation(18), we found bleeding requiring reoperation was less common in the severely obese cohort. Elsewhere, an IMACS analysis as well as another smaller study reported that early bleeding is more common in underweight individuals rather than the obese(8,12). The present study utilized a binary BMI threshold to examine adverse event rates in the severely obese. It is possible that increased bleeding in those patients at the lower BMI extreme led to the finding of less bleeding requiring reoperation among the severely obese. Novel to the present study is the reported impact of severe obesity on both quality of life and functional capacity outcomes. A prior study also found no impact of BMI on quality of life in HeartMate II recipients(25). These outcomes are particularly relevant given that patients who may be contemplating DT therapy emphasize the importance of improvements in quality of life and functionality even if at the expense of longevity(26,27). Severely obese patients listed for transplantation will likely have increased wait times until transplant likewise making quality of life and functional capacity very relevant to their care. While severely obese patients had improvement in 6MWD, these gains were on average lower than patients without severe obesity (Figure 5C). This suggests that severe obesity may be associated with a lower ceiling for functional capacity recovery following LVAD implantation similar to other comorbidities(25). Severely obese patients should be counseled that while their functional capacity should increase, the magnitude of this improvement will likely be limited by persistent obesity. Furthermore,
14 Kiernan HVAD BMI while exercise tolerance defined 6MWD increased in obese subjects, this did not translate to weight loss and obese subjects had an average increase in BMI over the first year of therapy of 0.7 kg/m2. These data are important as collectively they demonstrate that despite differences in flow dynamics and unloading characteristics(28), centrifugal flow LVADs adequately support large patients, who benefit from similar improvements in HRQOL, functional capacity and survival as those without severe obesity. Heightened understanding of the risk profile associated with severe obesity is important to better inform patient selection. Counseling patients about these outcomes prior to surgery sets appropriate expectations for patients and caregivers regarding the anticipated benefits of LVAD therapy. As has been previously reported, despite the increased functional capacity afforded by LVAD therapy, obese patients rarely lose weight during LVAD support(9,29). The greater room for improvement in non-severely obese LVAD recipients may provide credence for the use of weight loss surgery in these patients(30). Further investigation into the combination of weight loss strategies and LVAD therapies deserves consideration as case studies of patients with bariatric surgery following LVAD implantation have reported positive results(31,32). With careful patient selection, severe obesity should generally not represent a contraindication to LVAD therapy. However, the increased risk of adverse events and smaller anticipated improvements in functional capacity should be discussed during the shared decision-making process.
Limitations There are several inherent limitations to post-hoc analyses of clinical trial data, including that these analyses may be underpowered to detect clinically relevant differences. Until longer
15 Kiernan HVAD BMI term outcomes can be verified from larger multi-center patient registries, these findings should be interpreted thoughtfully and in context of the broader literature. Furthermore, these data were from a BTT population, and thus the most obese HF patients may have been excluded from this study cohort. Few patients had BMI > 40 kg/m2 and the findings of this study cannot be extrapolated to those with more extreme obesity. Furthermore, as would be expected, obese patients were more likely to have diabetes. It is possible that diabetes or other comorbid conditions such as hypertension, that are more common in obese patients, were responsible for the higher rates of renal dysfunction and driveline exit site infection in this cohort. The small sample size of severely obese patients limits statistical modeling to adjust for potential confounders when attempting to clarify the independent association between severe obesity and these adverse events. Also, although cardiac cachexia has been identified as a risk factor for mortality in some studies following LVAD surgery, this was not the focus of the present study and other BMI stratums were not investigated(18). Had underweight patients been excluded from the control group, it is possible that observed differences may have been greater between the severely obese and control patients. While patients were dichotomized based on a BMI threshold of 35 kg/m2, whereas there is more likely a spectrum of risk associated with increasing BMI as it relates to specific adverse events. Nevertheless, utilizing the NIH BMI thresholds for severe obesity (Obesity Class 2 - BMI 35-39 kg/m2 and Class 3 - BMI > 40 kg/m2) helps to contextualize these findings to better inform the patient selection process for LVAD therapy. Finally, obesity-specific HRQL tools were also not available for comparison and findings from HF-specific tools may not be sensitive to evaluate QOL changes relevant to obese subjects. The EQ-5D, however, is a generic HRQL tool that assesses five dimensions relevant across diagnoses: mobility, self-care, usual activities, pain/discomfort and anxiety/ depression. Changes
16 Kiernan HVAD BMI in the HF-specific KCCQ and generic EQ-5D were similar, increasing confidence that the improvements of QOL observed in obese subjects are clinically significant.
Conclusions In this large, multi-center trial of centrifugal flow LVAD recipients, BMI > 35 kg/m2 was not associated with reduced survival although there was a greater risk of certain adverse events. Severely obese patients experienced similar quality of life improvements as the control group. While the severely obese group also experienced increased functional capacity, this improvement was generally less pronounced compared to those patients without severe obesity. Awareness of the increased risk of certain adverse events is vital to guide shared decision making.
Acknowledgements The authors acknowledge Ming-Jay Chow, Medtronic Inc., for his assistance in the preparation of the manuscript.
17 Kiernan HVAD BMI References 1.
2.
3.
4. 5. 6.
7.
8.
9.
10. 11. 12.
13.
14. 15. 16.
17.
Rogers JG, Aaronson KD, Boyle AJ et al. Continuous Flow Left Ventricular Assist Device Improves Functional Capacity and Quality of Life of Advanced Heart Failure Patients. J Am Coll Cardiol 2010;55:1826-1834. Slaughter MS, Pagani FD, McGee EC et al. Use of the HeartWare® Ventricular Assist System for Bridge to Transplant: Combined Results of the ADVANCE and CAP Trial. Journal of Heart and Lung Transplant 2013;32:675-683. Mozaffarian D, Benjamin EJ, Go AS et al. Executive Summary: Heart Disease and Stroke Statistics--2016 Update: A Report From the American Heart Association. Circulation 2016;133:447-54. Reddy YN, Melenovsky V, Redfield MM, Nishimura RA, Borlaug BA. High-Output Heart Failure: A 15-Year Experience. J Am Coll Cardiol 2016;68:473-82. Macha M, Molina EJ, Franco M et al. Pre-transplant obesity in heart transplantation: are there predictors of worse outcomes? Scand Cardiovasc J 2009;43:304-10. Russo MJ, Hong KN, Davies RR et al. The effect of body mass index on survival following heart transplantation: do outcomes support consensus guidelines? Ann Surg 2010;251:144-52. Mehra MR, Canter CE, Hannan MM et al. The 2016 International Society for Heart Lung Transplantation listing criteria for heart transplantation: A 10-year update. J Heart Lung Transplant 2016;35:1-23. Brewer RJ, Lanfear DE, Sai-Sudhakar CB et al. Extremes of body mass index do not impact mid-term survival after continuous-flow left ventricular assist device implantation. J Heart Lung Transplant 2012;31:167-72. Clerkin KJ, Naka Y, Mancini DM, Colombo PC, Topkara VK. The Impact of Obesity on Patients Bridged to Transplantation With Continuous-Flow Left Ventricular Assist Devices. JACC Heart Fail 2016;4:761-8. Mohamedali B, Yost G, Bhat G. Obesity as a Risk Factor for Consideration for Left Ventricular Assist Devices. J Card Fail 2015;21:800-5. Musci M, Loforte A, Potapov EV et al. Body Mass Index and Outcome After Ventricular Assist Device Placement. The Annals of Thoracic Surgery 2008;86:1236-1242. Forest SJ, Xie R, Kirklin JK et al. Impact of body mass index on adverse events after implantation of left ventricular assist devices: An IMACS registry analysis. J Heart Lung Transplant 2018;37:1207-1217. Volkovicher N, Kurihara C, Critsinelis A et al. Effect of obesity on outcomes in patients undergoing implantation of continuous-flow left ventricular assist devices. J Artif Organs 2018;21:180-187. Yost G, Coyle L, Gallagher C et al. The impact of extreme obesity on outcomes after left ventricular assist device implantation. J Thorac Dis 2017;9:4441-4446. Kirklin JK, Naftel DC, Pagani FD et al. Seventh INTERMACS annual report: 15,000 patients and counting. J Heart Lung Transplant 2015;34:1495-504. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults--The Evidence Report. National Institutes of Health. Obes Res 1998;6 Suppl 2:51S-209S. Holman WL, Pae WE, Teutenberg JJ et al. INTERMACS: Interval Analysis of Registry Data. Journal of the American College of Surgeons 2009;208:755-761.
18 Kiernan HVAD BMI 18. 19. 20.
21.
22.
23.
24.
25.
26.
27. 28. 29.
30. 31.
32.
Butler J, Howser R, Portner PM, Pierson RN, 3rd. Body mass index and outcomes after left ventricular assist device placement. Ann Thorac Surg 2005;79:66-73. Falagas ME, Kompoti M. Obesity and infection. The Lancet Infectious diseases 2006;6:438-46. Calder CL, Ortega G, Vij A et al. Morbid obesity is an independent risk factor for postoperative renal dysfunction in young adults: a review of the American College of Surgeons National Surgical Quality Improvement Program database. The American Journal of Surgery 2016;211:772-777. Ejerblad E, Fored CM, Lindblad P, Fryzek J, McLaughlin JK, Nyren O. Obesity and risk for chronic renal failure. Journal of the American Society of Nephrology : JASN 2006;17:1695-702. Mullens W, Abrahams Z, Francis GS et al. Importance of venous congestion for worsening of renal function in advanced decompensated heart failure. J Am Coll Cardiol 2009;53:589-96. Testani JM, McCauley BD, Kimmel SE, Shannon RP. Characteristics of patients with improvement or worsening in renal function during treatment of acute decompensated heart failure. Am J Cardiol 2010;106:1763-9. Maltais S, Aaronson KD, Teuteberg JJ et al. Adverse Event Rates Change Favorably Over Time for Patients Bridged With the HeartWare Left Ventricular Assist Device. ASAIO journal (American Society for Artificial Internal Organs : 1992) 2017. Kiernan MS, Sundareswaran KS, Pham DT et al. Preoperative Determinants of Quality of Life and Functional Capacity Response to Left Ventricular Assist Device Therapy. J Card Fail 2016;22:797-805. Lewis EF, Johnson PA, Johnson W, Collins C, Griffin L, Stevenson LW. Preferences for quality of life or survival expressed by patients with heart failure. J Heart Lung Transplant 2001;20:1016-24. Stevenson LW, Hellkamp AS, Leier CV et al. Changing preferences for survival after hospitalization with advanced heart failure. J Am Coll Cardiol 2008;52:1702-8. Lim HS, Howell N, Ranasinghe A. The Physiology of Continuous-Flow Left Ventricular Assist Devices. Journal of cardiac failure 2017;23:169-180. Emani S, Brewer RJ, John R et al. Patients with low compared with high body mass index gain more weight after implantation of a continuous-flow left ventricular assist device. The Journal of Heart and Lung Transplantation 2013;32:31-35. Zarraga IGE, Schwarz ER. Impact of Dietary Patterns and Interventions on Cardiovascular Health. Circulation 2006;114:961-973. Lockard KL, Allen C, Lohmann D et al. Bariatric surgery for a patient with a HeartMate II ventricular assist device for destination therapy. Progress in transplantation (Aliso Viejo, Calif) 2013;23:28-32. Leviner DB, Keidar A, Ben-Gal T, Medalion B. Cardiac function recovery following LVAD implantation and bariatric surgery in a morbidly obese patient. J Card Surg 2014;29:740-2.
19 Kiernan HVAD BMI Figure 1: Distribution of Baseline Body Mass Index - The number of patients in each BMI stratum is provided.
20 Kiernan HVAD BMI Figure 2: Kaplan Meier Survival of Morbidly Obese and Control LVAD Patients. (P-value denotes comparison of two years of support.)
Figure 3: Freedom from Driveline Exit Site Infection for Morbidly Obese and Control LVAD Patients. (P-value denotes comparison of two years of support.)
21 Kiernan HVAD BMI Figure 4: BMI (4a) and albumin (4b) in Morbidly Obese and Control LVAD Patients across Time. (P-values for within group comparisons to the baseline measurements are shown in the columns and with asterisks denoting a significant difference from baseline measurements (p<0.05). Brackets and p-values above pairs of columns denote between group comparisons at that time point.)
22 Kiernan HVAD BMI Figure 5: Quality of Life and Functional Capacity of Morbidly Obese and Control LVAD Patients. (P-values for within group comparisons to the baseline measurements are shown in the columns with asterisks denoting a significant difference from baseline measurements (p<0.05). Brackets and p-values above pairs of columns denote between group comparisons at that time point.)
23 Kiernan HVAD BMI Table 1: Baseline Characteristics and Pump Parameters for Morbidly Obese Patients and Controls Morbidly Obese BMI > 35 kg/m2 (N= 48) 49.7±9.75 25.0
Control BMI ≤ 35 kg/m2 (N=334) 53.7±11.90 29.3
64.6
68.6
35.4 0.0
25.1 6.3
175.7±9.80 119.9±14.40 38.8±3.16 2.4±0.20 29.2 3.5±0.74 (44)* 47.9 56.5 (23)*
174.0±10.30 81.2±17.92 26.7±4.69 2.0±0.26 39.2 3.5±0.70 (321)* 33.2 51.4 (111)*
1.4±0.48
1.3±0.49
2.1 35.4 41.7 20.8
6.0 34.7 40.4 18.9
Hemodynamics (Pre-Implant) MAP (mmHg) Cardiac Index (L/min/m^2)
76.5±8.54 (27)* 2.3±0.63 (33)*
77.7±11.29 (210)* 2.2±0.61 (246)*
0.62 0.37
PCWP (mmHg)
23.7±10.50 (29)*
22.4±8.98 (204)*
0.47
RAP or CVP (mmHg)
13.4±5.94 (32)*
10.6±6.59 (222)*
0.02
Flow (L/min)
5.7±0.9
5.0±0.9 (333)*
<0.0001
Speed (RPM)
2828±198
2729±172 (333)*
0.0003
Power (Watts)
5.0±1.0
4.2±0.8 (333)*
<0.0001
Baseline Characteristics Age (years) Female sex (%) Race (%) Caucasian Black/African American Other Height (cm) Weight (kg) BMI (kg/m^2) Body Surface Area (m^2) Ischemic Cause of Heart Failure (%) Albumin (g/dL) Diabetes (%) Insulin Dependent Diabetes (%) Creatinine (mg/dL) INTERMACS Patient Profile (%) 1 2 3 4-7
P value 0.03 0.61 0.08
0.28 <0.0001 <0.0001 <0.0001 0.21 0.52 0.05 0.82 0.27 0.82
VAD Parameters
24 Kiernan HVAD BMI BMI = body mass index, MAP = mean arterial blood pressure, PCWP = pulmonary capillary wedge pressure, RAP = right atrial pressure, CVP = central venous pressure. * the number of patients with data for a given parameter is in the parentheses. All other values were calculated from the full (n=48) morbidly obese and (n=334) control groups.
Table 2: INTERMACS Adverse Events for Obese and Not-Obese Patients Morbidly Obese Control INTERMACS Category EPPY no. of EPPY no. of Adverse Events (67.09 PYtotal) events (339.54 PYtotal) events Bleeding 0.88 59 0.92 313 Re-Operation 0.06 4 0.18 60 GI 0.31 21 0.26 87 Cardiac Arrhythmia 0.54 36 0.50 169 Device Malfunction/Failure 0.28 19 0.27 90 Hemolysis 0.04 3 0.06 21 Infection 1.55 104 1.08 368 Sepsis 0.31 21 0.21 71 Driveline Exit Site 0.40 27 0.22 75
P value* 0.83 0.01 0.46 0.75 0.80 0.58 0.03 0.12 0.01 0.32 0.58 0.53
Neurological 0.18 12 0.24 82 Ischemic CVA 0.04 3 0.06 21 Hemorrhagic CVA 0.10 7 0.08 27 Renal Dysfunction 0.25 17 0.09 29 0.0009 Acute 0.24 16 0.09 29 0.002 Chronic 0.01 1 0.00 0 0.058 Right Heart Failure 0.37 25 0.37 124 0.94 EPPY = events per patient year, GI = gastrointestinal, CVA = cerebrovascular accident, *p-values are for comparisons of EPPY