- Email: [email protected]
A Weekly Dressing Protocol Reduces the Incidence of Driveline Infection
S188
The Journal of Heart and Lung Transplantation, Vol 34, No 4S, April 2015
frailty: 32 patients (31%) were classified as frail and 47 (46%) were pre-frail. Frailty was associated with being female, anaemia, hypoalbuminaemia, and increased levels of bilirubin. Frailty was independent of age, renal function, cognitive impairment or being depressed. Kaplan-Meier cumulative survival curves identified frailty as being significantly associated with increased mortality: follow-up survival was ~70%, ~75% and ~47% for non-frail, pre-frail and frail patients respectively (figure). Conclusion: Frailty is highly prevalent among patients listed for heart transplantation. There exists a strong potential for frailty assessment to independently risk-stratify patients with advanced heart failure.
1-year predicted survival was significantly reduced by 7% in frail individuals (p= 0.029). Conclusion: The prevalence of frailty is lower in HF patients referred for advanced therapies compared to community cohorts. Frailty is associated with reduced survival predicted by the SHFM in this population. Longitudinal studies to evaluate the prognostic impact of frailty in candidates for advanced HF therapies are warranted.
4( 99) Characteristics and Predictors of Improvement in Patients Delisted for Recovery While Awaiting Heart Transplantation S. Kumar , S. Al-Kindi, M. Ige, M. Ginwalla, C. ElAmm, S. Deo, S. Park, G.H. Oliveira. University Hospitals Case Medical Center, Cleveland, OH.
4( 98) Frailty Phenotype Is Associated With Survival as Predicted By the Seattle Heart Failure Model in Heart Failure Patients Referred for Advanced Therapies L.A. Goldraich , A.C. Alba, F. Foroutan, J. MacIver, H.J. Ross. Heart Failure and Cardiac Transplant Programs, Peter Munk Cardiac Center, University of Toronto., Toronto, ON, Canada. Purpose: Frailty has been associated with adverse prognosis in chronic diseases such as heart failure (HF). The impact of frailty remains unexplored in patients being considered for advanced HF therapies. Objective: To evaluate the association between frailty and 1-year survival predicted by Seattle Heart Failure Model (SHFM) in patients being assessed for heart transplant or mechanical circulatory support. Methods: Consecutive HF patients referred for advanced therapies underwent cardiopulmonary testing and 5-domain frailty phenotype assessment (unintentional weight loss, weak grip strength, physical exhaustion, slowness and low physical activity). Frailty was defined by presence of ≥ 3 domains. The SHFM was used to predict 1-year survival. The association between frailty and 1-year predicted survival was analyzed by multivariable linear regression adjusting for creatinine, B-type natriuretic peptide (BNP) and body mass index. Results: Seventy patients were included [53±12 years; 75% male; 34% ischemic etiology; 56% New York Heart Association 3-4; left ventricular ejection fraction 28±13%; BNP 626±822pg/mL; pVO2 13.5±4.0mL/Kg/min; 73% implantable cardiac defibrillator and/or resynchronization therapy]. The prevalence of frailty was 15.7%. There was no difference in age, sex and guideline directed medical therapy between frail and non-frail patients. Frail patients had lower estimated 1-year survival (SHFM 81±13% vs. 91±10%; Figure) in comparison to non-frail individuals. In multivariable analysis,
Purpose: A small proportion of patients listed for heart transplantation improve and do not require transplantation. The characteristics of this population, and predictors of delisting are unknown. Methods: The UNOS registry was queried for all adults (≥ 18 years) listed for heart transplantation between 2008 and 2013. Cox proportional hazard model was performed censoring for delisting for other reasons, death, transplantation and loss to follow-up. Results: Of the 14144 listed patients, 565 (3.3%) were delisted for improvement (DFI). DFI decreased from 4.8% (2008) to 1.72% (2013). The DFI cohort was predominantly male (67.6%), Caucasian (69.7%), mean (SD) age of 52.2 (12.7) with a diagnosis of ischemic cardiomyopathy (29.6%), and status at listing 2 (57%). Delisting happened after a mean of 16.3 months. In a multivariate model (n= 14068), DFI was predicted by UNOS 1a (OR 2.2, p< 0.001), Hispanic ethnicity (OR 1.5, p= 0.007), female gender (1.3, p= 0.011), Non-ischemic cardiomyopathy (OR 1.3, p= 0.018), lack of mechanical circulatory support (OR 2.5, p< 0.001), and lack of ICD (OR 2.1, p< 0.001). Conclusion: Only 3.3% of patients on the transplant list improve to not requiring heart transplantation. This is predicted by status 1a at listing, Hispanic ethnicity, female gender, non-ischemic etiology, lack of MCS and ICD. 5( 00) A Weekly Dressing Protocol Reduces the Incidence of Driveline Infection M. Puhlman , L. Wang, R. Sullivan, K. Evenson, J. Remick, G. Ott, J. Abraham. Center for Advanced Heart Disease, Providence St Vincent Medical Center, Portland, OR. Purpose: Driveline infections (DI) remain an important source of morbidity for patients supported by durable ventricular assist devices (VAD). There is no universally accepted protocol for management of the driveline exit site, and daily or thrice weekly dressing change protocols are used at many programs. Our center has utilized a standardized implant procedure and weekly outpatient driveline management protocol since 2011 with low rates of driveline infection. Methods: Single center, retrospective review of patients receiving the Heart Mate II VAD (Thoratec Corp, Pleasanton, CA, USA). A standard operative approach was taken in all patients (pre-peritoneal implant, driveline exit from
Abstracts S189 left abdomen, surgical drain in pump pocket until < 100 cc output/24 hours). Sterile, daily dressing changes are performed until the exit site has healed. Weekly changes are then performed by the patient or caregiver using a driveline management system (Centurian Medical Products Corp, Williamson, MI, USA), consisting of a Biopatch, Sorbaview Shield dressing and Centurian foley catheter anchor . Using the INTERMACS definition of DI (clinical signs of infection, a positive culture from the surrounding skin and/or tissue, coupled with the need for antimicrobial therapy), the rate of DI was computed and compared to INTERMACs. Results: Between 2011 and 2014 during 701 months of patient support, 2 out of 25 patients (8%) experienced DI: one S. aureus DI requiring multiple debridements and ultimately device explant; a second Candida Parapsilosis DI managed with chronic suppressive antimicrobials. The crude rate of DI is 0.29/100 months of patient support, significantly lower than the INTERMACS rate of about 1.55/100 patient months (p = 0.001). Conclusion: A strategy of initial daily dressing changes followed by a weekly outpatient driveline management protocol substantially lowers the rate of DI. A prospective multi-center trial of this driveline management strategy is warranted. 5( 01) Extremes of Obesity and LVAD Patient Morbidity and Mortality C. Henderson,1 K. Patel,1 G. Sayer,1 S. Fedson,1 G. Kim,1 T. Ota,2 C. Juricek,2 V. Jeevanandam,2 N. Uriel .1 1Medicine, University of Chicago, Chicago, IL; 2Surgery, University of Chicago, Chicago, IL. Purpose: In continuous flow left ventricular assist device (CF-LVAD) approval studies, a Body Mass Index (BMI) > 40 was considered a contraindication to implantation. However, in the post-approval era, device therapy is not limited by BMI. Yet, there is limited data examining the outcomes of LVAD implantation in patients with extreme obesity. This study aims to assess the impact of severe obesity at the time of LVAD implantation on morbidity and mortality. Methods: A retrospective chart review was conducted for all CF-LVAD patients in a large center between 2008-2014. Patients with a BMI> 35 were included in the study. Data was collected from electronic medical record. Primary outcome was 2-year survival, while secondary outcomes were LVAD-related complications, including infection, thromboembolic events and strokes. Data are presented as mean + SD and survival analysis was performed with Kaplan-Meier curves. Results: 56 patients were identified and stratified into three groups based on pre-implant BMI: 35-40 (N= 35), 40.01-45 (N= 12) and > 45 (N= 9)). The highest BMI group was found to be younger in age (54+12.2, 55+11.3, 43+12.1, p< 0.05) with a higher incidence of non-ischemic cardiomyopathy (63%, 83%, 100%, p = 0.05). The combined 2-year survival of all patients was low at 52%. No significant differences in survival were seen between the 3 groups (55.3% v. 53.6% v. 44.4% [p = 0.72]) (Figure 1). There was a high prevalence of device-related infections (43%, 58%, 56%), intracranial hemorrhage (14%, 17%, 11%), and device thrombosis (20%, 8%, 33%). 33% of patients in the BMI > 45 group underwent device exchange. Conclusion: Morbidly obese patients experience a higher rate of morbidity and mortality with destination therapy than the rates reported in the current literature. The success of LVAD therapy in this population may be limited, and warrants further study.
5( 02) Handgrip Strength Is a Predictor for Length of Stay in Patients Implanted With Left Ventricular Assist Devices G. Yost, M. Gregory, G. Bhat . Center for Heart Transplant and Assist Devices, Advocate Christ Medical Center, Oak Lawn, IL. Purpose: Frailty is a known predictor of survival and clinical outcomes in numerous patient populations, including those with advanced heart failure. Handgrip strength (HS) is an easily obtained measure of frailty which predicts adverse outcomes in elderly and heart failure populations. In this study we investigated the predictive capability of HS in patients undergoing evaluation for left ventricular assist device (LVAD) implantation. Methods: HS was measured preoperatively using a Jamar hand dynamometer in 28 patients between 2013-14. Demographic, nutrition, and length of stay (LOS) data were collected for all patients. Patients were stratified into 3 groups based on average HS. Results: Group A, with mean age 62.3 ± 10.4 years, included patients with HS less than 45 kg. Group B, with mean age 53.9 ± 12 years, included patients with HS between 45 and 69 kg. Group C, with mean age 61.7 ± 11.3 years, included patients with HS 69 kg or greater. Height, weight, and gender were significantly different between groups and exhibited significant correlations against HS (Table 1). Additionally, though differences between groups were non-significant, LOS was found to correlate significantly with HS and decreased as HS improved (Table 1). When the cohort was divided into two groups of HS greater than or less than 55 kg, unique LOS timelines were noted in Kaplan-Meier analysis (Figure 1). Conclusion: HS is a known predictor of adverse outcomes in the heart failure population. Our data suggest, further, that HS can be used to reliably predict LOS following workup for LVAD implantation.
Table 1 Correlation to HS
Height (cm) Weight (kg) Gender LOS
Intergroup Comparison
r-value
Group A p-value (n= 9)
Group B (n= 10)
Group C (n= 9)
0.697 0.437 0.485 -0.352
< 0.001 0.022 0.009 0.072
177.3 ± 6.1 93.1 ± 9.7 90% Male 34.1 ± 24.6 59.75 ± 7.4
182.3 ± 7.2 97.0 ± 7.5 88% Male 24.8 ± 21.6 80.85 ± 6.2
166.3 ± 8.6 78.4 ± 21.9 56% Male 40.1 ± 20.7 37.2 ± 11.6
p-value 0.002 0.009 0.002 0.26 < 0.001
The Journal of Heart and Lung Transplantation, Vol 34, No 4S, April 2015
frailty: 32 patients (31%) were classified as frail and 47 (46%) were pre-frail. Frailty was associated with being female, anaemia, hypoalbuminaemia, and increased levels of bilirubin. Frailty was independent of age, renal function, cognitive impairment or being depressed. Kaplan-Meier cumulative survival curves identified frailty as being significantly associated with increased mortality: follow-up survival was ~70%, ~75% and ~47% for non-frail, pre-frail and frail patients respectively (figure). Conclusion: Frailty is highly prevalent among patients listed for heart transplantation. There exists a strong potential for frailty assessment to independently risk-stratify patients with advanced heart failure.
1-year predicted survival was significantly reduced by 7% in frail individuals (p= 0.029). Conclusion: The prevalence of frailty is lower in HF patients referred for advanced therapies compared to community cohorts. Frailty is associated with reduced survival predicted by the SHFM in this population. Longitudinal studies to evaluate the prognostic impact of frailty in candidates for advanced HF therapies are warranted.
4( 99) Characteristics and Predictors of Improvement in Patients Delisted for Recovery While Awaiting Heart Transplantation S. Kumar , S. Al-Kindi, M. Ige, M. Ginwalla, C. ElAmm, S. Deo, S. Park, G.H. Oliveira. University Hospitals Case Medical Center, Cleveland, OH.
4( 98) Frailty Phenotype Is Associated With Survival as Predicted By the Seattle Heart Failure Model in Heart Failure Patients Referred for Advanced Therapies L.A. Goldraich , A.C. Alba, F. Foroutan, J. MacIver, H.J. Ross. Heart Failure and Cardiac Transplant Programs, Peter Munk Cardiac Center, University of Toronto., Toronto, ON, Canada. Purpose: Frailty has been associated with adverse prognosis in chronic diseases such as heart failure (HF). The impact of frailty remains unexplored in patients being considered for advanced HF therapies. Objective: To evaluate the association between frailty and 1-year survival predicted by Seattle Heart Failure Model (SHFM) in patients being assessed for heart transplant or mechanical circulatory support. Methods: Consecutive HF patients referred for advanced therapies underwent cardiopulmonary testing and 5-domain frailty phenotype assessment (unintentional weight loss, weak grip strength, physical exhaustion, slowness and low physical activity). Frailty was defined by presence of ≥ 3 domains. The SHFM was used to predict 1-year survival. The association between frailty and 1-year predicted survival was analyzed by multivariable linear regression adjusting for creatinine, B-type natriuretic peptide (BNP) and body mass index. Results: Seventy patients were included [53±12 years; 75% male; 34% ischemic etiology; 56% New York Heart Association 3-4; left ventricular ejection fraction 28±13%; BNP 626±822pg/mL; pVO2 13.5±4.0mL/Kg/min; 73% implantable cardiac defibrillator and/or resynchronization therapy]. The prevalence of frailty was 15.7%. There was no difference in age, sex and guideline directed medical therapy between frail and non-frail patients. Frail patients had lower estimated 1-year survival (SHFM 81±13% vs. 91±10%; Figure) in comparison to non-frail individuals. In multivariable analysis,
Purpose: A small proportion of patients listed for heart transplantation improve and do not require transplantation. The characteristics of this population, and predictors of delisting are unknown. Methods: The UNOS registry was queried for all adults (≥ 18 years) listed for heart transplantation between 2008 and 2013. Cox proportional hazard model was performed censoring for delisting for other reasons, death, transplantation and loss to follow-up. Results: Of the 14144 listed patients, 565 (3.3%) were delisted for improvement (DFI). DFI decreased from 4.8% (2008) to 1.72% (2013). The DFI cohort was predominantly male (67.6%), Caucasian (69.7%), mean (SD) age of 52.2 (12.7) with a diagnosis of ischemic cardiomyopathy (29.6%), and status at listing 2 (57%). Delisting happened after a mean of 16.3 months. In a multivariate model (n= 14068), DFI was predicted by UNOS 1a (OR 2.2, p< 0.001), Hispanic ethnicity (OR 1.5, p= 0.007), female gender (1.3, p= 0.011), Non-ischemic cardiomyopathy (OR 1.3, p= 0.018), lack of mechanical circulatory support (OR 2.5, p< 0.001), and lack of ICD (OR 2.1, p< 0.001). Conclusion: Only 3.3% of patients on the transplant list improve to not requiring heart transplantation. This is predicted by status 1a at listing, Hispanic ethnicity, female gender, non-ischemic etiology, lack of MCS and ICD. 5( 00) A Weekly Dressing Protocol Reduces the Incidence of Driveline Infection M. Puhlman , L. Wang, R. Sullivan, K. Evenson, J. Remick, G. Ott, J. Abraham. Center for Advanced Heart Disease, Providence St Vincent Medical Center, Portland, OR. Purpose: Driveline infections (DI) remain an important source of morbidity for patients supported by durable ventricular assist devices (VAD). There is no universally accepted protocol for management of the driveline exit site, and daily or thrice weekly dressing change protocols are used at many programs. Our center has utilized a standardized implant procedure and weekly outpatient driveline management protocol since 2011 with low rates of driveline infection. Methods: Single center, retrospective review of patients receiving the Heart Mate II VAD (Thoratec Corp, Pleasanton, CA, USA). A standard operative approach was taken in all patients (pre-peritoneal implant, driveline exit from
Abstracts S189 left abdomen, surgical drain in pump pocket until < 100 cc output/24 hours). Sterile, daily dressing changes are performed until the exit site has healed. Weekly changes are then performed by the patient or caregiver using a driveline management system (Centurian Medical Products Corp, Williamson, MI, USA), consisting of a Biopatch, Sorbaview Shield dressing and Centurian foley catheter anchor . Using the INTERMACS definition of DI (clinical signs of infection, a positive culture from the surrounding skin and/or tissue, coupled with the need for antimicrobial therapy), the rate of DI was computed and compared to INTERMACs. Results: Between 2011 and 2014 during 701 months of patient support, 2 out of 25 patients (8%) experienced DI: one S. aureus DI requiring multiple debridements and ultimately device explant; a second Candida Parapsilosis DI managed with chronic suppressive antimicrobials. The crude rate of DI is 0.29/100 months of patient support, significantly lower than the INTERMACS rate of about 1.55/100 patient months (p = 0.001). Conclusion: A strategy of initial daily dressing changes followed by a weekly outpatient driveline management protocol substantially lowers the rate of DI. A prospective multi-center trial of this driveline management strategy is warranted. 5( 01) Extremes of Obesity and LVAD Patient Morbidity and Mortality C. Henderson,1 K. Patel,1 G. Sayer,1 S. Fedson,1 G. Kim,1 T. Ota,2 C. Juricek,2 V. Jeevanandam,2 N. Uriel .1 1Medicine, University of Chicago, Chicago, IL; 2Surgery, University of Chicago, Chicago, IL. Purpose: In continuous flow left ventricular assist device (CF-LVAD) approval studies, a Body Mass Index (BMI) > 40 was considered a contraindication to implantation. However, in the post-approval era, device therapy is not limited by BMI. Yet, there is limited data examining the outcomes of LVAD implantation in patients with extreme obesity. This study aims to assess the impact of severe obesity at the time of LVAD implantation on morbidity and mortality. Methods: A retrospective chart review was conducted for all CF-LVAD patients in a large center between 2008-2014. Patients with a BMI> 35 were included in the study. Data was collected from electronic medical record. Primary outcome was 2-year survival, while secondary outcomes were LVAD-related complications, including infection, thromboembolic events and strokes. Data are presented as mean + SD and survival analysis was performed with Kaplan-Meier curves. Results: 56 patients were identified and stratified into three groups based on pre-implant BMI: 35-40 (N= 35), 40.01-45 (N= 12) and > 45 (N= 9)). The highest BMI group was found to be younger in age (54+12.2, 55+11.3, 43+12.1, p< 0.05) with a higher incidence of non-ischemic cardiomyopathy (63%, 83%, 100%, p = 0.05). The combined 2-year survival of all patients was low at 52%. No significant differences in survival were seen between the 3 groups (55.3% v. 53.6% v. 44.4% [p = 0.72]) (Figure 1). There was a high prevalence of device-related infections (43%, 58%, 56%), intracranial hemorrhage (14%, 17%, 11%), and device thrombosis (20%, 8%, 33%). 33% of patients in the BMI > 45 group underwent device exchange. Conclusion: Morbidly obese patients experience a higher rate of morbidity and mortality with destination therapy than the rates reported in the current literature. The success of LVAD therapy in this population may be limited, and warrants further study.
5( 02) Handgrip Strength Is a Predictor for Length of Stay in Patients Implanted With Left Ventricular Assist Devices G. Yost, M. Gregory, G. Bhat . Center for Heart Transplant and Assist Devices, Advocate Christ Medical Center, Oak Lawn, IL. Purpose: Frailty is a known predictor of survival and clinical outcomes in numerous patient populations, including those with advanced heart failure. Handgrip strength (HS) is an easily obtained measure of frailty which predicts adverse outcomes in elderly and heart failure populations. In this study we investigated the predictive capability of HS in patients undergoing evaluation for left ventricular assist device (LVAD) implantation. Methods: HS was measured preoperatively using a Jamar hand dynamometer in 28 patients between 2013-14. Demographic, nutrition, and length of stay (LOS) data were collected for all patients. Patients were stratified into 3 groups based on average HS. Results: Group A, with mean age 62.3 ± 10.4 years, included patients with HS less than 45 kg. Group B, with mean age 53.9 ± 12 years, included patients with HS between 45 and 69 kg. Group C, with mean age 61.7 ± 11.3 years, included patients with HS 69 kg or greater. Height, weight, and gender were significantly different between groups and exhibited significant correlations against HS (Table 1). Additionally, though differences between groups were non-significant, LOS was found to correlate significantly with HS and decreased as HS improved (Table 1). When the cohort was divided into two groups of HS greater than or less than 55 kg, unique LOS timelines were noted in Kaplan-Meier analysis (Figure 1). Conclusion: HS is a known predictor of adverse outcomes in the heart failure population. Our data suggest, further, that HS can be used to reliably predict LOS following workup for LVAD implantation.
Table 1 Correlation to HS
Height (cm) Weight (kg) Gender LOS
Intergroup Comparison
r-value
Group A p-value (n= 9)
Group B (n= 10)
Group C (n= 9)
0.697 0.437 0.485 -0.352
< 0.001 0.022 0.009 0.072
177.3 ± 6.1 93.1 ± 9.7 90% Male 34.1 ± 24.6 59.75 ± 7.4
182.3 ± 7.2 97.0 ± 7.5 88% Male 24.8 ± 21.6 80.85 ± 6.2
166.3 ± 8.6 78.4 ± 21.9 56% Male 40.1 ± 20.7 37.2 ± 11.6
p-value 0.002 0.009 0.002 0.26 < 0.001