Catheterization and Cardiovascular Interventions 83:E221–E226 (2014)
VALVULAR AND STRUCTURAL HEART DISEASES Original Studies Blood Loss and Transfusion Rates Associated with Transcatheter Aortic Valve Replacement: Recommendations for Patients Who Refuse Blood Transfusion Ronald K. Binder, MD, Marco Barbanti, MD, Jian Ye, MD, Stefan Toggweiler, MD, John Tan, MD, Melanie Freeman, MBBS, Anson Cheung, MD, David A. Wood, MD, and John G. Webb,* MD Background: Transcatheter aortic valve replacement (TAVR) is less invasive than surgical aortic valve replacement and may be preferred for patients who refuse blood transfusions. Objectives: Our study sought to define transfusion rates in TAVR, identify predictors and develop recommendations for patients who refuse transfusions. Methods: A large cohort of consecutive patients undergoing TAVR was prospectively studied. Blood loss and transfusion rates were observed in patients undergoing transfemoral (TF-TAVR) and transapical TAVR (TA-TAVR). Predictors for transfusion were investigated in a multivariate model. Results: Of 373 consecutive patients, 270 underwent TF-TAVR and 103 TA-TAVR. Transfusion rates were significantly lower in TF-TAVR than TA-TAVR (11.1 vs 41.7%; P < 0.001). In patients who did not receive transfusions, blood loss was significantly lower in TF-TAVR than TA-TAVR (23.6 6 12.2 g/l vs 28.9 6 13.7 g/l; P 5 0.004), but did not differ in transfused patients (36.9 6 12.9 g/l vs 33.4 6 21.2 g/l; P 5 0.428). Predictors for transfusions were low baseline hemoglobin, female sex, low body weight and decreased renal function. In seven Jehovah’s Witness patients, who refused transfusions, no vascular complications occurred and clinical outcome was excellent. Conclusion: In patients, who refuse transfusions, TAVR may be performed with good clinical outcomes provided a high baseline hemoglobin level, careful management of the access site and strict measurements to reduce blood loss. TF-TAVR is associated with less blood loss and a lower rate of transfusions than TA-TAVR, and may be the preferred option for patients, who refuse transfusions. VC 2014 Wiley Periodicals, Inc. Key words: bleeding; stenosis; transfemoral; transapical
INTRODUCTION
Patients who refuse transfusion of allogenic blood, for example, Jehovah’s Witnesses (JW), pose a challenge, when in need for interventions associated with blood loss [1–3]. Most JW patients decline blood transfusions even when exsanguination is imminent [1,2]. Symptomatic severe aortic stenosis is the most common indication for heart valve replacement. Surgical aortic valve replacement has been successfully performed in JW [4,5], but mortality has been documented as a consequence of procedural bleeding [6]. Transcatheter aortic valve replacement (TAVR) is a less C 2014 Wiley Periodicals, Inc. V
St. Paul’s Hospital, University of British Columbia, British Columbia, Vancouver, Canada Conflict of interest: Drs. Binder, Cheung, Webb, Wood, and Ye are consultants to Edwards Lifesciences Inc. Drs. Binder and Toggweiler received unrestricted research grants from the Swiss National Foundation. *Correspondence to: Dr. John G. Webb; St. Paul’s Hospital, 1081 Burrard Street, V6Z 1Y6, Vancouver, BC, Canada. E-mail: [email protected] Received 9 September 2013; Revision accepted 3 January 2014 DOI: 10.1002/ccd.25389 Published online 9 January 2014 in Wiley Online Library (wileyonlinelibrary.com)
E222
Binder et al.
invasive treatment for aortic stenosis [7] and increasing numbers of JW are referred for TAVR. However, decision making in patients who require aortic valve replacement and refuse blood transfusions is not backed by guidelines and there is scarce scientific evidence guiding their management. This study sought to define blood loss and transfusion rates in a large, consecutive cohort of patients undergoing TAVR, to elucidate TAVR outcomes in the sub-group of JW, and to develop recommendations for patients who undergo TAVR and refuse blood transfusions. METHODS
Patients who underwent TAVR at St. Paul’s Hospital, Vancouver, Canada, were prospectively entered into a dedicated database. Patients gave written informed consent for prospective data acquisition approved by the local ethics board. An interdisciplinary team including cardiologists and cardiac surgeons guided their management. We analyzed the in-hospital serum hemoglobin course, transfusion rates, bleeding sources and in-hospital and 30-day outcomes of consecutive patients who underwent TAVR between January 2010 and December 2012. Patients were excluded if they underwent TAVR with an access other than transfemoral or transapical, or who underwent procedures for the treatment of aortic regurgitation with investigational devices. Endpoints were reported according to standardized definitions by the Valve Academic Research Consortium (VARC) [8]. The corrected hemoglobin drop after TAVR was calculated according to the modified Landefeld equation [9]. One unit of packed red blood cells (RBC) is considered to represent 10 g/l of hemoglobin in this equation. The net hemoglobin drop corresponds to the addition of [the number of packed RBC times 10] to the baseline hemoglobin minus the hemoglobin nadir post TAVR. Data Processing and Statistical Analysis R softAll analyses were performed using the SPSSV R ware (version 20, SPSSV, Chicago, IL). Continuous variables are expressed as mean values 6 standard deviation. Continuous variables were tested for a normal distribution (QQ plot) and compared by the t test or the Withney-U test as appropriate. Categorical variables are described by frequencies and percentages and were compared by the Fisher Exact test. Potential predictors for BT were analyzed in an univariate model, and the significant ones were entered into a multivariate model by binary logistic regression. All statistical tests were two-tailed and a value of P < 0.05 was considered statistically significant.
TABLE I. Baseline and Procedural Characteristics Patient characteristics Age (years) Gender Height (cm) Weight (kg) BMI (kg/m2) Diabetes Hypertension Dyslipidemia Smoking History COPD NYHA - Class I - Class II - Class III - Class IV Prosthesis model - SAPIEN - SAPIEN XT - SAPIEN 3 - CoreValve - CENTERA - SJM Portico Prosthesis size - 20 mm - 23 mm - 26 mm - 29 mm - 31 mm Jehovah’s Witness Pre-procedural anemia Baseline hemoglobin (g/l) Prior cerebrovascular accident Prior open heart surgery Porcelain aorta Prior permanent Pacemaker Prior myocardial infarction Hemodialysis Pulmonary hypertension Peripheral vascular disease Atrial fibrillation STS PROM (%) GFR (ml/min)
Transfemoral N ¼ 270
Transapical N ¼ 103
82.1 6 8.2 61.9% (167) male 169.2 6 9.6 76.8 6 18.7 26.9 6 6.4 32.6% (88) 81.9% (221) 66.3% (179) 30.7% (83) 25.2% (68)
79.8 6 8.9 50.2% (52) male 165.5 6 11.4 70.1 6 18.9 25.4 6 5.8 31.1% (32) 83.5% (86) 69% (67) 48.5% (50) 33% (34)
1.1% 15.9% 76.7% 6.3%
(3) (43) (207) (17)
0% 6.8% 81.6% 11.7%
(0) (7) (84) (12)
1.9% 81.5% 4.8% 4.8% 3.7% 3.3%
(5) (220) (13) (13) (10) (9)
27.2% 69.9% 1% 0% 0% 1.9
(28) (72) (1) (0) (0) (2)
P-value 0.024 0.06 0.003 0.004 0.057 0.805 0.763 0.899 0.002 0.153 0.033
VALVULAR AND STRUCTURAL HEART DISEASES Original Studies Blood Loss and Transfusion Rates Associated with Transcatheter Aortic Valve Replacement: Recommendations for Patients Who Refuse Blood Transfusion Ronald K. Binder, MD, Marco Barbanti, MD, Jian Ye, MD, Stefan Toggweiler, MD, John Tan, MD, Melanie Freeman, MBBS, Anson Cheung, MD, David A. Wood, MD, and John G. Webb,* MD Background: Transcatheter aortic valve replacement (TAVR) is less invasive than surgical aortic valve replacement and may be preferred for patients who refuse blood transfusions. Objectives: Our study sought to define transfusion rates in TAVR, identify predictors and develop recommendations for patients who refuse transfusions. Methods: A large cohort of consecutive patients undergoing TAVR was prospectively studied. Blood loss and transfusion rates were observed in patients undergoing transfemoral (TF-TAVR) and transapical TAVR (TA-TAVR). Predictors for transfusion were investigated in a multivariate model. Results: Of 373 consecutive patients, 270 underwent TF-TAVR and 103 TA-TAVR. Transfusion rates were significantly lower in TF-TAVR than TA-TAVR (11.1 vs 41.7%; P < 0.001). In patients who did not receive transfusions, blood loss was significantly lower in TF-TAVR than TA-TAVR (23.6 6 12.2 g/l vs 28.9 6 13.7 g/l; P 5 0.004), but did not differ in transfused patients (36.9 6 12.9 g/l vs 33.4 6 21.2 g/l; P 5 0.428). Predictors for transfusions were low baseline hemoglobin, female sex, low body weight and decreased renal function. In seven Jehovah’s Witness patients, who refused transfusions, no vascular complications occurred and clinical outcome was excellent. Conclusion: In patients, who refuse transfusions, TAVR may be performed with good clinical outcomes provided a high baseline hemoglobin level, careful management of the access site and strict measurements to reduce blood loss. TF-TAVR is associated with less blood loss and a lower rate of transfusions than TA-TAVR, and may be the preferred option for patients, who refuse transfusions. VC 2014 Wiley Periodicals, Inc. Key words: bleeding; stenosis; transfemoral; transapical
INTRODUCTION
Patients who refuse transfusion of allogenic blood, for example, Jehovah’s Witnesses (JW), pose a challenge, when in need for interventions associated with blood loss [1–3]. Most JW patients decline blood transfusions even when exsanguination is imminent [1,2]. Symptomatic severe aortic stenosis is the most common indication for heart valve replacement. Surgical aortic valve replacement has been successfully performed in JW [4,5], but mortality has been documented as a consequence of procedural bleeding [6]. Transcatheter aortic valve replacement (TAVR) is a less C 2014 Wiley Periodicals, Inc. V
St. Paul’s Hospital, University of British Columbia, British Columbia, Vancouver, Canada Conflict of interest: Drs. Binder, Cheung, Webb, Wood, and Ye are consultants to Edwards Lifesciences Inc. Drs. Binder and Toggweiler received unrestricted research grants from the Swiss National Foundation. *Correspondence to: Dr. John G. Webb; St. Paul’s Hospital, 1081 Burrard Street, V6Z 1Y6, Vancouver, BC, Canada. E-mail: [email protected] Received 9 September 2013; Revision accepted 3 January 2014 DOI: 10.1002/ccd.25389 Published online 9 January 2014 in Wiley Online Library (wileyonlinelibrary.com)
E222
Binder et al.
invasive treatment for aortic stenosis [7] and increasing numbers of JW are referred for TAVR. However, decision making in patients who require aortic valve replacement and refuse blood transfusions is not backed by guidelines and there is scarce scientific evidence guiding their management. This study sought to define blood loss and transfusion rates in a large, consecutive cohort of patients undergoing TAVR, to elucidate TAVR outcomes in the sub-group of JW, and to develop recommendations for patients who undergo TAVR and refuse blood transfusions. METHODS
Patients who underwent TAVR at St. Paul’s Hospital, Vancouver, Canada, were prospectively entered into a dedicated database. Patients gave written informed consent for prospective data acquisition approved by the local ethics board. An interdisciplinary team including cardiologists and cardiac surgeons guided their management. We analyzed the in-hospital serum hemoglobin course, transfusion rates, bleeding sources and in-hospital and 30-day outcomes of consecutive patients who underwent TAVR between January 2010 and December 2012. Patients were excluded if they underwent TAVR with an access other than transfemoral or transapical, or who underwent procedures for the treatment of aortic regurgitation with investigational devices. Endpoints were reported according to standardized definitions by the Valve Academic Research Consortium (VARC) [8]. The corrected hemoglobin drop after TAVR was calculated according to the modified Landefeld equation [9]. One unit of packed red blood cells (RBC) is considered to represent 10 g/l of hemoglobin in this equation. The net hemoglobin drop corresponds to the addition of [the number of packed RBC times 10] to the baseline hemoglobin minus the hemoglobin nadir post TAVR. Data Processing and Statistical Analysis R softAll analyses were performed using the SPSSV R ware (version 20, SPSSV, Chicago, IL). Continuous variables are expressed as mean values 6 standard deviation. Continuous variables were tested for a normal distribution (QQ plot) and compared by the t test or the Withney-U test as appropriate. Categorical variables are described by frequencies and percentages and were compared by the Fisher Exact test. Potential predictors for BT were analyzed in an univariate model, and the significant ones were entered into a multivariate model by binary logistic regression. All statistical tests were two-tailed and a value of P < 0.05 was considered statistically significant.
TABLE I. Baseline and Procedural Characteristics Patient characteristics Age (years) Gender Height (cm) Weight (kg) BMI (kg/m2) Diabetes Hypertension Dyslipidemia Smoking History COPD NYHA - Class I - Class II - Class III - Class IV Prosthesis model - SAPIEN - SAPIEN XT - SAPIEN 3 - CoreValve - CENTERA - SJM Portico Prosthesis size - 20 mm - 23 mm - 26 mm - 29 mm - 31 mm Jehovah’s Witness Pre-procedural anemia Baseline hemoglobin (g/l) Prior cerebrovascular accident Prior open heart surgery Porcelain aorta Prior permanent Pacemaker Prior myocardial infarction Hemodialysis Pulmonary hypertension Peripheral vascular disease Atrial fibrillation STS PROM (%) GFR (ml/min)
Transfemoral N ¼ 270
Transapical N ¼ 103
82.1 6 8.2 61.9% (167) male 169.2 6 9.6 76.8 6 18.7 26.9 6 6.4 32.6% (88) 81.9% (221) 66.3% (179) 30.7% (83) 25.2% (68)
79.8 6 8.9 50.2% (52) male 165.5 6 11.4 70.1 6 18.9 25.4 6 5.8 31.1% (32) 83.5% (86) 69% (67) 48.5% (50) 33% (34)
1.1% 15.9% 76.7% 6.3%
(3) (43) (207) (17)
0% 6.8% 81.6% 11.7%
(0) (7) (84) (12)
1.9% 81.5% 4.8% 4.8% 3.7% 3.3%
(5) (220) (13) (13) (10) (9)
27.2% 69.9% 1% 0% 0% 1.9
(28) (72) (1) (0) (0) (2)
P-value 0.024 0.06 0.003 0.004 0.057 0.805 0.763 0.899 0.002 0.153 0.033
