ASAIO Journal 2014
Anticoagulation Therapy Trends in Children Supported by Ventricular Assist Devices: A Multi-Institutional Study Brady S. Moffett,*† Antonio G. Cabrera,† Jun Teruya,‡ and Lisa Bomgaars§
Ventricular assist device (VAD) use in children has increased dramatically. There are currently few data regarding trends in anticoagulation management for pediatric VADs. A retrospective cohort study was conducted for patients with an International Classification of Diseases, Ninth Revision (ICD-9) code for VAD implantation from 2000 to 2011 from the Pediatric Health Information System database. Patient demographics, use of extracorporeal membrane oxygenation, orthotopic heart transplantation (OHT), disease states, and medications pertinent to the management of VAD anticoagulation were queried. Patients were grouped into 3 year time periods to evaluate trends in medication use over time. Four hundred sixty-six patients were identified with a median length of VAD therapy of 21 days (range 1–362 days). In-hospital mortality was 31.9%, and 54.5% underwent OHT. Length of VAD therapy and patients undergoing OHT increased, while mortality decreased. Patients received a median of two anticoagulant medications (range 0–6), one (range 0–4) antiplatelet medications, three (range 0–5) procoagulant medications, and one (range 0–3) antifibrinolytics. Patients received greater mean numbers of anticoagulant, procoagulant, antifibrinolytic, and antiplatelet agents, and the use of oral medications increased more than twofold over time. There is wide variability for in-hospital pediatric VAD anticoagulation management, with a significant increase in the use of oral agents in more recent years. ASAIO Journal 2014; 60:211–215.
One of the main components to successful VAD therapy is the management of anticoagulation, which can be challenging in infants, neonates, and children. Subsequently, bleeding and thrombotic events are common with pediatric VAD use.2–6 Indications for VAD support include myocarditis, inherited cardiac myopathies, dilated cardiomyopathy, acute rejection of heart transplantation, and congenital heart disease.2,4,7,9–13 Anticoagulation in pediatric VAD patients has likely undergone an evolution as best practices, and ideal methods are elucidated based on clinical experience. Due to the relative paucity of data with regard to VAD anticoagulation therapy in children, an evaluation of pharmacotherapy used to manage anticoagulation in these patients is warranted. Identification of trends in pharmacotherapy can be useful to evaluate the current status of anticoagulation management in pediatric VAD patients. In addition, data from a large multicenter review of anticoagulation therapy can be used to identify areas for potential future investigations or quality improvement initiatives. Purpose The purpose of this study was to evaluate trends in medications used for the management of pediatric VAD anticoagulation over a 12-year period. Methods
Key Words: ventricular assist device, anticoagulation, antiplatelet, pediatric
A retrospective, descriptive study was designed, and Institutional Review Board exemption was obtained. Data for this study were obtained from the Pediatric Health Information System (PHIS), an administrative database that contains inpatient, emergency department, ambulatory surgery, and observation data from 43 not-for-profit, tertiary care pediatric hospitals in the United States. These hospitals are affiliated with the Child Health Corporation of America (Shawnee Mission, KS), a business alliance of children’s hospitals. Data quality and reliability are assured through a joint effort between the Child Health Corporation of America and participating hospitals. The data warehouse function for the PHIS database is managed by Thomson Reuters (Ann Arbor, MI). For the purposes of external benchmarking, participating hospitals provide discharge/encounter data including demographics, diagnoses, and procedures. Forty-two of these hospitals also submit resource utilization data (e.g., pharmaceuticals, imaging, and laboratory) into PHIS. Data are de-identified at the time of data submission, and data are subjected to a number of reliability and validity checks before being included in the database. For this study, patient data from 32 hospitals met study criteria and were included. The PHIS database was queried for patients discharged from January 1, 2000, to December 31, 2011. Patients were
R
ecent reports have established that using a ventricular assist device (VAD) in a pediatric patient with severe heart failure is a feasible option for long-term myocardial support or bridge to heart transplantation.1–6 While the use of VADs in adults has been well established and large numbers of adult patients have received VADs, the use of a VAD in a pediatric patient currently occurs less frequently.2,7,8 From the *Department of Pharmacy, Texas Children’s Hospital, Houston, Texas; †Section of Pediatric Cardiology, Department of Pediatrics, Baylor College of Medicine, Houston, Texas; ‡Departments of Pathology and Immunology, Pediatrics, and Medicine, Baylor College of Medicine, Houston, Texas; and §Section of Hematology and Oncology, Department of Pediatrics, Baylor College of Medicine, Houston, Texas. Submitted for consideration July 2013; accepted for publication in revised form November 2013. Disclosure: Dr. Cabrera is the Site PI for the PumpKin trial. The other authors have no conflicts of interest to report. Reprint Requests: Brady S. Moffett, PharmD, MPH, Department of Pharmacy, Texas Children’s Hospital, 6621 Fannin Street, Suite WB 1120, Houston, TX 77030. Email: [email protected]. Copyright © 2014 by the American Society for Artificial Internal Organs DOI: 10.1097/MAT.0000000000000037
211
212 MOFFETT et al. included if they had an International Classification of Diseases, Ninth Revision (ICD-9) procedure code for placement of a VAD (37.52, 37.60, 37.62, 37.65, 37.66, 37.68) during the inpatient admission. Patients were excluded if they were 19 years of age or greater or had missing or incomplete data. Patient demographic information, disease state information, use of extracorporeal membrane oxygenation (ECMO), heart transplantation, and days of VAD therapy were collected. Medications were identified as anticoagulants, procoagulants (including antifibrinolytic agents), and antiplatelet agents, and utilization of each of these was collected for the duration of VAD therapy. Age was segmented into the following groups: neonates (1–30 days), infants (31 days–2 years), children (>2–12 years), and adolescents (13–18 years). Patients were grouped into 3 year periods by date of discharge: 2000–2002, 2003–2005, 2006–2008, and 2009–2011 for analysis of trends. Mean, standard deviation, median, and range were used to describe the population. Chi-squared analysis, Student’s t-test, analysis of variance, and Kruskal-Wallis test were used to determine differences in groups. Stata IC v.12 (StataCorp, College Station, TX) was used for all analyses. Results A total of 466 patients were identified from the database. Median length of VAD therapy was 21 days (range 1–362 days). Patient demographics are listed in Table 1. Mortality was 31.9% for the entire time period, and 54.5% underwent orthotopic heart transplant. No patients had a documented intrinsic bleeding disorder (e.g., hemophilia or von Willebrand’s disease). Mean length of VAD therapy and percentage of patients undergoing orthotopic heart transplantation (OHT) increased over time, while mortality trended toward decreasing over the four time periods (Figure 1). Patients who underwent ECMO before VAD placement had a significantly lower median age (2.5 years [IQR 0.4–10.6 years] vs. 10.0 years [IQR 1.0–14.6 years]) and significantly higher percent mortality (21.8% vs. 43.6%, p < 0.01). Table 1. Patient Demographics Category (n = 466) Male (%) Age at VAD implant (median, range) Neonates (%) Infants (%) Child (%) Adolescents (%) Race/ethnicity Caucasian (%) African American (%) Asian (%) Hispanic (%) American Indian (%) Length of stay (days) (median, range) Underlying condition Cardiomyopathy (%) Congenital heart disease (%) Myocarditis (%) Previous extracorporeal membrane oxygenation (%) VAD, ventricular assist device.
Result 54.1 5.8 years (1 day–18.8 years) 9.4 33.9 29.2 27.5 59.4 21 2.6 13.4 1.3 60 (1–476) 66.5 44.2 14.8 46.8
Overall, patients received a median of two anticoagulant medications (range 0–6), a median of one (range 0–4) antiplatelet medications, a median of three (range 0–5) procoagulant medications, and a median of one (range 0–2) antifibrinolytic medications. Over the four time periods, patients increasingly received a statistically greater mean number of anticoagulant (p < 0.05), procoagulant (p < 0.05), and antiplatelet agents (p < 0.05) (Figure 2). No statistically significant change was noted in antifibrinolytic use (p > 0.05). Percentage of patients receiving oral medications for the prevention of thrombus more than doubled over the study period from 32.1% receiving an oral medication in 2000–2002 to 68.2% receiving an oral medication in 2009–2011 (p < 0.01). The median time to first oral medication after VAD implantation was 4 days (range 1–134 days). Anticoagulant Medications Overall, the use of anticoagulant medications (unfractionated heparin [UFH], enoxaparin, argatroban, bivalirudin, lepirudin, alteplase, warfarin) occurred in 98.3% of patients. Unfractionated heparin was used in 98.3% of patients, and this trend did not differ widely between time periods (p = 0.11). Enoxaparin use increased significantly over time with no patient use in the 2000–2002 period to 38.6% of patients receiving enoxaparin in the 2009–2011 time period (p < 0.01). Warfarin was used in 26.2% of patients, and this did not change significantly over time (p = 0.97). Alteplase was used in 31.8% of patients, with a significantly higher percentage of patients receiving alteplase in the 2009–2011 time period (41.8%) when compared with the 2000–2002 time period (7.1%) (p < 0.01). Antithrombin was also used significantly more in the most recent time period (2009–2011, 44.1%) when compared with 2000–2002 (0%, p < 0.01). A direct thrombin inhibitor was used rarely (argatroban [0.9%], bivalirudin [0.6%], lepirudin [0.2%]). Patients who previously underwent ECMO were more likely to receive antithrombin III (22.9% vs. 33.0%, p = 0.01) and less likely to receive warfarin (33.5% vs. 17.9%, p < 0.01). The use of UFH did not differ between age groups, but infants were more likely to receive enoxaparin, alteplase, and antithrombin. Older patients were more likely to receive warfarin (p < 0.05) (Table 2). Procoagulant Medications The use of procoagulant/antifibrinolytic medications during VAD therapy in patients increased by 44.5% over the study period (p < 0.01). Cryoprecipitate was used frequently in patients (86.3%), and fresh frozen plasma was also used frequently (90.8%). Recombinant factor VIIa was the most frequently used of all the exogenous coagulation factors (23.2%), with factor VIII (0.2%) and factor IX (4.7%) used less often. Protamine was used in 65.7% of patients and vitamin K in 16.1% of patients. Aminocaproic acid was used in 34.1% of patients, tranexamic acid in 13.3%, and aprotinin, which is currently no longer available in the United States, in 25.1% of patients. Patients who previously underwent ECMO were less likely to receive vitamin K (19.4% vs. 12.4%, p = 0.04) and factor IX (7.7% vs. 1.4%, p < 0.01). Vitamin K was used more frequently in the older patient age groups when compared with the younger age groups. All other
ANTICOAGULATION THERAPY IN CHILDREN SUPPORTED BY VAD
213
Figure 1. Patient mortality, heart transplantation, and ventricular assist device (VAD) length of therapy trends. OHT, orthotopic heart transplantation.
procoagulant/antifibrinolytic medications did not have significant differences between age groups (Table 2). Antiplatelet Medications The use of antiplatelet agents (aspirin, dipyridamole, clopidogrel, pentoxifylline, omega-3 fatty acids, abciximab) more than doubled over the study period from 21.4% receiving an antiplatelet medication in 2000–2002 to 67.2% receiving an antiplatelet medication in 2009–2011 (p < 0.01). The most commonly used antiplatelet medication was aspirin in 51.9% of patients, followed by dipyridamole (31.6%), clopidogrel (4.9%), pentoxifylline (2.2%), omega-3 fatty acids (1.3%), and abciximab (0.2%). Aspirin and dipyridamole were used most frequently in the infant age group (Table 2). Patients who
previously underwent ECMO were less likely to receive pentoxifylline (3.6% vs. 0.46%, p = 0.01). Discussion This is the first evaluation of national prescribing trends for hemostatic and antithrombotic agents for pediatric VAD patients, and these data represent the largest description of this patient population to date in the literature. Overall, VAD therapy in pediatric hospitals appears to be improving as noted by the increased length of VAD therapy, the decreased mortality, and the increased number of patients undergoing OHT while on a VAD. As experienced is gained and different strategies for VAD support are evaluated, the outcomes with VAD use in pediatric patients have improved. Over the years of the study
Figure 2. Trend of antiplatelet, procoagulant, antifibrionlytic, and anticoagulant medications in pediatric ventricular assist device patients.
214 MOFFETT et al. Table 2. Ventricular Assist Device Therapy per Age Group Medications (n = 466) Oral medications* Anticoagulant Unfractionated heparin (%) Antithrombin (%)* Alteplase (%)* Enoxaparin (%)* Warfarin (%)* Argatroban (%) Lepirudin (%) Bivalirudin (%) Antiplatelet Abciximab (%) Aspirin (%)* Dipyridamole (%)* Clopidogrel (%) Pentoxifylline (%) Omega-3 fatty acids (%) Procoagulant Factor VIII (%) Factor IX (%) Factor VIIa (%) Vitamin K (%)* Protamine (%) Cryoprecipitate (%) Fresh frozen plasma (%) Antifibrinolytic Aminocaproic acid (%) Aprotinin (%) Tranexamic acid (%)
Neonates (n = 44)
Infants (n = 158)
Children (n = 136)
Adolescents (n = 128)
27.8
68.8
65.4
56.3
100 20.5 13.6 11.4 0 0 0 0
98.7 44.3 36.1 43.7 12 0.63 0.63 0.63
97.8 25.7 25 27.9 36.8 1.5 0 0
97.7 11.7 24.2 16.4 41.4 0.78 0 1.6
0 27.3 11.4 0 0 2.3
0.63 62 48.7 7.6 2.5 1.3
0 58.1 37.5 5.9 0.74 3.7
0 45.3 10.9 2.3 3.9 0
0 6.8 15.9 11.4 56.8 90.9 93.2
0 5.7 18.9 8.9 68.4 88.6 93.7
0 4.4 30.1 18.4 59.6 85.3 88.9
0.78 3.1 23.4 24.2 71.9 82.8 88.3
31.8 20.5 6.8
34.8 24.1 15.8
32.4 22.1 16.2
35.9 31.3 9.4
*One or more age groups significantly different, p < 0.05.
period in this report, trends toward using different antiplatelet and varied medications are readily evident. We believe that this reflects on ongoing learning curve as more and more pediatric VADs are used.2,4,11,12,14,15 It appears that patient age is a significant driver for choice of pharmacotherapy. For example, enoxaparin was more frequently used in younger patients, likely because of the difficulty of warfarin management in this patient group.16 This age group would be more likely to use vitamin K containing formulas and have significant changes in diet during their growth. There is also a lack of robust data for the management of warfarin in infants and no age appropriate formulations (i.e., a suspension). Infants tend to have lower antithrombin levels and can develop acquired antithrombin deficiency rapidly when placed on mechanical circulatory support, which is reflected by increased supplementation of antithrombin in this age group.17,18 Conversely, warfarin use was increased in the older age groups, where warfarin would be less complicated to manage and administer. The increased use of oral medications, including the increased use of antiplatelet medications, reflects a significant change in VAD management strategy in pediatric patients over time. The current protocol for the Berlin Heart recommends the use of oral antiplatelet medications (aspirin, dipyridamole), and the increased use of the Berlin Heart VAD is reflected through the increased use of oral antiplatelet medications.2,4,6 It appears that the management of VADs in the pediatric population moving toward the use of antiplatelet medications for long-term therapy and the role of anticoagulant medications in long-term pediatric VAD patients may be diminished. As previously stated, aspirin and dipyridamole have been frequently
used, and in the future, antiplatelet agents such as clopidogrel or prasugrel may have a role in pediatric VAD management. Unexpectedly, we discovered that many patients received a wide variety of procoagulant and antifibrinolytic medications, such as aminocaproic acid, coagulation factor products, cryoprecipitate, and fresh frozen plasma. The exact indications for each medication used cannot be determined from this database; however, based on clinical experience, some insights can be made. Many of these medications may have been used when implanting or explanting devices, to limit intraoperative or postsurgical bleeding. In addition, the reported incidence of bleeding with pediatric VAD implantation is high, especially during the first 1–2 weeks after implantation, and use of these agents may reflect this morbidity.2,19 These agents represent some of the most costly and resource-intensive medications used in pediatric VAD management, particularly in the case of blood products. Extensive evaluation of the use of these agents in pediatric VAD patients is warranted.20 Limitations to this report are those common to retrospective reviews. The PHIS database is an administrative and billing database that limits the granularity of the data. Clinical end-points, such as bleeding or thrombus, were unable to be evaluated due to the nature of the database and represent important end-points that have been investigated in other publications.1–6,21 We also cannot conclusively state that changes in VAD anticoagulation therapy have resulted in improved outcomes. Improvements in surgical technique, VAD type and construction, and experience gained in the management of pediatric patients on VADs undoubtedly contribute to the improvements seen in this report. Changes in therapy have likely occurred concurrently with other changes to pediatric
ANTICOAGULATION THERAPY IN CHILDREN SUPPORTED BY VAD
VAD management. In addition, the database only codes for VAD implantation/explantation but does not specify type or size of VAD. Inferences to the type of VAD can be made on the age of the patient (i.e., Berlin Heart for neonates and infants), but direct associations of pharmacotherapy and a particular type of VAD cannot be made. Doses of medications and laboratory results are also not included in the database, which limits the use of the database to a broad evaluation of prescribing trends. Despite these limitations, the data evaluated are useful for determining future areas of investigation and quality improvement in pediatric VAD therapy. Conclusions The management of pediatric VAD anticoagulation has changed significantly over the past 12 years, including the use of more and varied agents, and the greater use of oral medications. Pediatric VAD outcomes have improved with changes in anticoagulation management. References 1. Fan Y, Weng YG, Xiao YB, et al: Outcomes of ventricular assist device support in young patients with small body surface area. Eur J Cardiothorac Surg 39: 699–704, 2011. 2. Fraser CD Jr, Jaquiss RD, Rosenthal DN, et al; Berlin Heart Study Investigators: Prospective trial of a pediatric ventricular assist device. N Engl J Med 367: 532–541, 2012. 3. Hill JD, Reinhartz O: Clinical outcomes in pediatric patients implanted with Thoratec ventricular assist device. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 115–122, 2006. 4. Humpl T, Furness S, Gruenwald C, Hyslop C, Van AG: The Berlin Heart EXCOR Pediatrics—The SickKids experience 2004–2008. Artif Organs 34: 1082–1086, 2010. 5. Kaczmarek I, Sachweh J, Groetzner J, et al: Mechanical circulatory support in pediatric patients with the MEDOS assist device. ASAIO J 51: 498–500, 2005. 6. Morales DL, Almond CS, Jaquiss RD, et al: Bridging children of all sizes to cardiac transplantation: The initial multicenter North American experience with the Berlin Heart EXCOR ventricular assist device. J Heart Lung Transplant 30: 1–8, 2011. 7. Hetzer R, Alexi-Meskishvili V, Weng Y, et al: Mechanical cardiac support in the young with the Berlin Heart EXCOR pulsatile
215
ventricular assist device: 15 years’ experience. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 99–108, 2006. 8. Morales DL, Gunter KS, Fraser CD: Pediatric mechanical circulatory support. Int J Artif Organs 29: 920–937, 2006. 9. Hetzer R, Potapov EV, Alexi-Meskishvili V, et al: Single-center experience with treatment of cardiogenic shock in children by pediatric ventricular assist devices. J Thorac Cardiovasc Surg 141: 616–623, 2011. 10. Morales DL, Braud BE, Price JF, et al: Use of mechanical circulatory support in pediatric patients with acute cardiac graft rejection. ASAIO J 53: 701–705, 2007. 11. Sharma MS, Forbess JM, Guleserian KJ: Ventricular assist device support in children and adolescents with heart failure: The Children’s Medical Center of Dallas experience. Artif Organs 36: 635–639, 2012. 12. VanderPluym CJ, Rebeyka IM, Ross DB, Buchholz H: The use of ventricular assist devices in pediatric patients with univentricular hearts. J Thorac Cardiovasc Surg 141: 588–590, 2011. 13. Wilmot I, Morales DL, Price JF, et al: Effectiveness of mechanical circulatory support in children with acute fulminant and persistent myocarditis. J Card Fail 17: 487–494, 2011. 14. Blume ED, Naftel DC, Bastardi HJ, Duncan BW, Kirklin JK, Webber SA; Pediatric Heart Transplant Study Investigators: Outcomes of children bridged to heart transplantation with ventricular assist devices: A multi-institutional study. Circulation 113: 2313– 2319, 2006. 15. Chen JM, Richmond ME, Charette K, et al: A decade of pediatric mechanical circulatory support before and after cardiac transplantation. J Thorac Cardiovasc Surg 143: 344–351, 2012. 16. Streif W, Andrew M, Marzinotto V, et al: Analysis of warfarin therapy in pediatric patients: A prospective cohort study of 319 patients. Blood 94: 3007–3014, 1999. 17. Bassler D, Schmidt B: Antithrombin replacement in neonates: Is there any indication? Thromb Res 118: 107–111, 2006. 18. Nankervis CA, Preston TJ, Dysart KC, et al: Assessing heparin dosing in neonates on venoarterial extracorporeal membrane oxygenation. ASAIO J 53: 111–114, 2007. 19. Arabía FA, Tsau PH, Smith RG, et al: Pediatric bridge to heart transplantation: Application of the Berlin Heart, Medos and Thoratec ventricular assist devices. J Heart Lung Transplant 25: 16–21, 2006. 20. Stiller B, Lemmer J, Merkle F, et al: Consumption of blood products during mechanical circulatory support in children: Comparison between ECMO and a pulsatile ventricular assist device. Intensive Care Med 30: 1814–1820, 2004. 21. Miera O, Potapov EV, Redlin M, et al: First experiences with the HeartWare ventricular assist system in children. Ann Thorac Surg 91: 1256–1260, 2011.
Anticoagulation Therapy Trends in Children Supported by Ventricular Assist Devices: A Multi-Institutional Study Brady S. Moffett,*† Antonio G. Cabrera,† Jun Teruya,‡ and Lisa Bomgaars§
Ventricular assist device (VAD) use in children has increased dramatically. There are currently few data regarding trends in anticoagulation management for pediatric VADs. A retrospective cohort study was conducted for patients with an International Classification of Diseases, Ninth Revision (ICD-9) code for VAD implantation from 2000 to 2011 from the Pediatric Health Information System database. Patient demographics, use of extracorporeal membrane oxygenation, orthotopic heart transplantation (OHT), disease states, and medications pertinent to the management of VAD anticoagulation were queried. Patients were grouped into 3 year time periods to evaluate trends in medication use over time. Four hundred sixty-six patients were identified with a median length of VAD therapy of 21 days (range 1–362 days). In-hospital mortality was 31.9%, and 54.5% underwent OHT. Length of VAD therapy and patients undergoing OHT increased, while mortality decreased. Patients received a median of two anticoagulant medications (range 0–6), one (range 0–4) antiplatelet medications, three (range 0–5) procoagulant medications, and one (range 0–3) antifibrinolytics. Patients received greater mean numbers of anticoagulant, procoagulant, antifibrinolytic, and antiplatelet agents, and the use of oral medications increased more than twofold over time. There is wide variability for in-hospital pediatric VAD anticoagulation management, with a significant increase in the use of oral agents in more recent years. ASAIO Journal 2014; 60:211–215.
One of the main components to successful VAD therapy is the management of anticoagulation, which can be challenging in infants, neonates, and children. Subsequently, bleeding and thrombotic events are common with pediatric VAD use.2–6 Indications for VAD support include myocarditis, inherited cardiac myopathies, dilated cardiomyopathy, acute rejection of heart transplantation, and congenital heart disease.2,4,7,9–13 Anticoagulation in pediatric VAD patients has likely undergone an evolution as best practices, and ideal methods are elucidated based on clinical experience. Due to the relative paucity of data with regard to VAD anticoagulation therapy in children, an evaluation of pharmacotherapy used to manage anticoagulation in these patients is warranted. Identification of trends in pharmacotherapy can be useful to evaluate the current status of anticoagulation management in pediatric VAD patients. In addition, data from a large multicenter review of anticoagulation therapy can be used to identify areas for potential future investigations or quality improvement initiatives. Purpose The purpose of this study was to evaluate trends in medications used for the management of pediatric VAD anticoagulation over a 12-year period. Methods
Key Words: ventricular assist device, anticoagulation, antiplatelet, pediatric
A retrospective, descriptive study was designed, and Institutional Review Board exemption was obtained. Data for this study were obtained from the Pediatric Health Information System (PHIS), an administrative database that contains inpatient, emergency department, ambulatory surgery, and observation data from 43 not-for-profit, tertiary care pediatric hospitals in the United States. These hospitals are affiliated with the Child Health Corporation of America (Shawnee Mission, KS), a business alliance of children’s hospitals. Data quality and reliability are assured through a joint effort between the Child Health Corporation of America and participating hospitals. The data warehouse function for the PHIS database is managed by Thomson Reuters (Ann Arbor, MI). For the purposes of external benchmarking, participating hospitals provide discharge/encounter data including demographics, diagnoses, and procedures. Forty-two of these hospitals also submit resource utilization data (e.g., pharmaceuticals, imaging, and laboratory) into PHIS. Data are de-identified at the time of data submission, and data are subjected to a number of reliability and validity checks before being included in the database. For this study, patient data from 32 hospitals met study criteria and were included. The PHIS database was queried for patients discharged from January 1, 2000, to December 31, 2011. Patients were
R
ecent reports have established that using a ventricular assist device (VAD) in a pediatric patient with severe heart failure is a feasible option for long-term myocardial support or bridge to heart transplantation.1–6 While the use of VADs in adults has been well established and large numbers of adult patients have received VADs, the use of a VAD in a pediatric patient currently occurs less frequently.2,7,8 From the *Department of Pharmacy, Texas Children’s Hospital, Houston, Texas; †Section of Pediatric Cardiology, Department of Pediatrics, Baylor College of Medicine, Houston, Texas; ‡Departments of Pathology and Immunology, Pediatrics, and Medicine, Baylor College of Medicine, Houston, Texas; and §Section of Hematology and Oncology, Department of Pediatrics, Baylor College of Medicine, Houston, Texas. Submitted for consideration July 2013; accepted for publication in revised form November 2013. Disclosure: Dr. Cabrera is the Site PI for the PumpKin trial. The other authors have no conflicts of interest to report. Reprint Requests: Brady S. Moffett, PharmD, MPH, Department of Pharmacy, Texas Children’s Hospital, 6621 Fannin Street, Suite WB 1120, Houston, TX 77030. Email: [email protected]. Copyright © 2014 by the American Society for Artificial Internal Organs DOI: 10.1097/MAT.0000000000000037
211
212 MOFFETT et al. included if they had an International Classification of Diseases, Ninth Revision (ICD-9) procedure code for placement of a VAD (37.52, 37.60, 37.62, 37.65, 37.66, 37.68) during the inpatient admission. Patients were excluded if they were 19 years of age or greater or had missing or incomplete data. Patient demographic information, disease state information, use of extracorporeal membrane oxygenation (ECMO), heart transplantation, and days of VAD therapy were collected. Medications were identified as anticoagulants, procoagulants (including antifibrinolytic agents), and antiplatelet agents, and utilization of each of these was collected for the duration of VAD therapy. Age was segmented into the following groups: neonates (1–30 days), infants (31 days–2 years), children (>2–12 years), and adolescents (13–18 years). Patients were grouped into 3 year periods by date of discharge: 2000–2002, 2003–2005, 2006–2008, and 2009–2011 for analysis of trends. Mean, standard deviation, median, and range were used to describe the population. Chi-squared analysis, Student’s t-test, analysis of variance, and Kruskal-Wallis test were used to determine differences in groups. Stata IC v.12 (StataCorp, College Station, TX) was used for all analyses. Results A total of 466 patients were identified from the database. Median length of VAD therapy was 21 days (range 1–362 days). Patient demographics are listed in Table 1. Mortality was 31.9% for the entire time period, and 54.5% underwent orthotopic heart transplant. No patients had a documented intrinsic bleeding disorder (e.g., hemophilia or von Willebrand’s disease). Mean length of VAD therapy and percentage of patients undergoing orthotopic heart transplantation (OHT) increased over time, while mortality trended toward decreasing over the four time periods (Figure 1). Patients who underwent ECMO before VAD placement had a significantly lower median age (2.5 years [IQR 0.4–10.6 years] vs. 10.0 years [IQR 1.0–14.6 years]) and significantly higher percent mortality (21.8% vs. 43.6%, p < 0.01). Table 1. Patient Demographics Category (n = 466) Male (%) Age at VAD implant (median, range) Neonates (%) Infants (%) Child (%) Adolescents (%) Race/ethnicity Caucasian (%) African American (%) Asian (%) Hispanic (%) American Indian (%) Length of stay (days) (median, range) Underlying condition Cardiomyopathy (%) Congenital heart disease (%) Myocarditis (%) Previous extracorporeal membrane oxygenation (%) VAD, ventricular assist device.
Result 54.1 5.8 years (1 day–18.8 years) 9.4 33.9 29.2 27.5 59.4 21 2.6 13.4 1.3 60 (1–476) 66.5 44.2 14.8 46.8
Overall, patients received a median of two anticoagulant medications (range 0–6), a median of one (range 0–4) antiplatelet medications, a median of three (range 0–5) procoagulant medications, and a median of one (range 0–2) antifibrinolytic medications. Over the four time periods, patients increasingly received a statistically greater mean number of anticoagulant (p < 0.05), procoagulant (p < 0.05), and antiplatelet agents (p < 0.05) (Figure 2). No statistically significant change was noted in antifibrinolytic use (p > 0.05). Percentage of patients receiving oral medications for the prevention of thrombus more than doubled over the study period from 32.1% receiving an oral medication in 2000–2002 to 68.2% receiving an oral medication in 2009–2011 (p < 0.01). The median time to first oral medication after VAD implantation was 4 days (range 1–134 days). Anticoagulant Medications Overall, the use of anticoagulant medications (unfractionated heparin [UFH], enoxaparin, argatroban, bivalirudin, lepirudin, alteplase, warfarin) occurred in 98.3% of patients. Unfractionated heparin was used in 98.3% of patients, and this trend did not differ widely between time periods (p = 0.11). Enoxaparin use increased significantly over time with no patient use in the 2000–2002 period to 38.6% of patients receiving enoxaparin in the 2009–2011 time period (p < 0.01). Warfarin was used in 26.2% of patients, and this did not change significantly over time (p = 0.97). Alteplase was used in 31.8% of patients, with a significantly higher percentage of patients receiving alteplase in the 2009–2011 time period (41.8%) when compared with the 2000–2002 time period (7.1%) (p < 0.01). Antithrombin was also used significantly more in the most recent time period (2009–2011, 44.1%) when compared with 2000–2002 (0%, p < 0.01). A direct thrombin inhibitor was used rarely (argatroban [0.9%], bivalirudin [0.6%], lepirudin [0.2%]). Patients who previously underwent ECMO were more likely to receive antithrombin III (22.9% vs. 33.0%, p = 0.01) and less likely to receive warfarin (33.5% vs. 17.9%, p < 0.01). The use of UFH did not differ between age groups, but infants were more likely to receive enoxaparin, alteplase, and antithrombin. Older patients were more likely to receive warfarin (p < 0.05) (Table 2). Procoagulant Medications The use of procoagulant/antifibrinolytic medications during VAD therapy in patients increased by 44.5% over the study period (p < 0.01). Cryoprecipitate was used frequently in patients (86.3%), and fresh frozen plasma was also used frequently (90.8%). Recombinant factor VIIa was the most frequently used of all the exogenous coagulation factors (23.2%), with factor VIII (0.2%) and factor IX (4.7%) used less often. Protamine was used in 65.7% of patients and vitamin K in 16.1% of patients. Aminocaproic acid was used in 34.1% of patients, tranexamic acid in 13.3%, and aprotinin, which is currently no longer available in the United States, in 25.1% of patients. Patients who previously underwent ECMO were less likely to receive vitamin K (19.4% vs. 12.4%, p = 0.04) and factor IX (7.7% vs. 1.4%, p < 0.01). Vitamin K was used more frequently in the older patient age groups when compared with the younger age groups. All other
ANTICOAGULATION THERAPY IN CHILDREN SUPPORTED BY VAD
213
Figure 1. Patient mortality, heart transplantation, and ventricular assist device (VAD) length of therapy trends. OHT, orthotopic heart transplantation.
procoagulant/antifibrinolytic medications did not have significant differences between age groups (Table 2). Antiplatelet Medications The use of antiplatelet agents (aspirin, dipyridamole, clopidogrel, pentoxifylline, omega-3 fatty acids, abciximab) more than doubled over the study period from 21.4% receiving an antiplatelet medication in 2000–2002 to 67.2% receiving an antiplatelet medication in 2009–2011 (p < 0.01). The most commonly used antiplatelet medication was aspirin in 51.9% of patients, followed by dipyridamole (31.6%), clopidogrel (4.9%), pentoxifylline (2.2%), omega-3 fatty acids (1.3%), and abciximab (0.2%). Aspirin and dipyridamole were used most frequently in the infant age group (Table 2). Patients who
previously underwent ECMO were less likely to receive pentoxifylline (3.6% vs. 0.46%, p = 0.01). Discussion This is the first evaluation of national prescribing trends for hemostatic and antithrombotic agents for pediatric VAD patients, and these data represent the largest description of this patient population to date in the literature. Overall, VAD therapy in pediatric hospitals appears to be improving as noted by the increased length of VAD therapy, the decreased mortality, and the increased number of patients undergoing OHT while on a VAD. As experienced is gained and different strategies for VAD support are evaluated, the outcomes with VAD use in pediatric patients have improved. Over the years of the study
Figure 2. Trend of antiplatelet, procoagulant, antifibrionlytic, and anticoagulant medications in pediatric ventricular assist device patients.
214 MOFFETT et al. Table 2. Ventricular Assist Device Therapy per Age Group Medications (n = 466) Oral medications* Anticoagulant Unfractionated heparin (%) Antithrombin (%)* Alteplase (%)* Enoxaparin (%)* Warfarin (%)* Argatroban (%) Lepirudin (%) Bivalirudin (%) Antiplatelet Abciximab (%) Aspirin (%)* Dipyridamole (%)* Clopidogrel (%) Pentoxifylline (%) Omega-3 fatty acids (%) Procoagulant Factor VIII (%) Factor IX (%) Factor VIIa (%) Vitamin K (%)* Protamine (%) Cryoprecipitate (%) Fresh frozen plasma (%) Antifibrinolytic Aminocaproic acid (%) Aprotinin (%) Tranexamic acid (%)
Neonates (n = 44)
Infants (n = 158)
Children (n = 136)
Adolescents (n = 128)
27.8
68.8
65.4
56.3
100 20.5 13.6 11.4 0 0 0 0
98.7 44.3 36.1 43.7 12 0.63 0.63 0.63
97.8 25.7 25 27.9 36.8 1.5 0 0
97.7 11.7 24.2 16.4 41.4 0.78 0 1.6
0 27.3 11.4 0 0 2.3
0.63 62 48.7 7.6 2.5 1.3
0 58.1 37.5 5.9 0.74 3.7
0 45.3 10.9 2.3 3.9 0
0 6.8 15.9 11.4 56.8 90.9 93.2
0 5.7 18.9 8.9 68.4 88.6 93.7
0 4.4 30.1 18.4 59.6 85.3 88.9
0.78 3.1 23.4 24.2 71.9 82.8 88.3
31.8 20.5 6.8
34.8 24.1 15.8
32.4 22.1 16.2
35.9 31.3 9.4
*One or more age groups significantly different, p < 0.05.
period in this report, trends toward using different antiplatelet and varied medications are readily evident. We believe that this reflects on ongoing learning curve as more and more pediatric VADs are used.2,4,11,12,14,15 It appears that patient age is a significant driver for choice of pharmacotherapy. For example, enoxaparin was more frequently used in younger patients, likely because of the difficulty of warfarin management in this patient group.16 This age group would be more likely to use vitamin K containing formulas and have significant changes in diet during their growth. There is also a lack of robust data for the management of warfarin in infants and no age appropriate formulations (i.e., a suspension). Infants tend to have lower antithrombin levels and can develop acquired antithrombin deficiency rapidly when placed on mechanical circulatory support, which is reflected by increased supplementation of antithrombin in this age group.17,18 Conversely, warfarin use was increased in the older age groups, where warfarin would be less complicated to manage and administer. The increased use of oral medications, including the increased use of antiplatelet medications, reflects a significant change in VAD management strategy in pediatric patients over time. The current protocol for the Berlin Heart recommends the use of oral antiplatelet medications (aspirin, dipyridamole), and the increased use of the Berlin Heart VAD is reflected through the increased use of oral antiplatelet medications.2,4,6 It appears that the management of VADs in the pediatric population moving toward the use of antiplatelet medications for long-term therapy and the role of anticoagulant medications in long-term pediatric VAD patients may be diminished. As previously stated, aspirin and dipyridamole have been frequently
used, and in the future, antiplatelet agents such as clopidogrel or prasugrel may have a role in pediatric VAD management. Unexpectedly, we discovered that many patients received a wide variety of procoagulant and antifibrinolytic medications, such as aminocaproic acid, coagulation factor products, cryoprecipitate, and fresh frozen plasma. The exact indications for each medication used cannot be determined from this database; however, based on clinical experience, some insights can be made. Many of these medications may have been used when implanting or explanting devices, to limit intraoperative or postsurgical bleeding. In addition, the reported incidence of bleeding with pediatric VAD implantation is high, especially during the first 1–2 weeks after implantation, and use of these agents may reflect this morbidity.2,19 These agents represent some of the most costly and resource-intensive medications used in pediatric VAD management, particularly in the case of blood products. Extensive evaluation of the use of these agents in pediatric VAD patients is warranted.20 Limitations to this report are those common to retrospective reviews. The PHIS database is an administrative and billing database that limits the granularity of the data. Clinical end-points, such as bleeding or thrombus, were unable to be evaluated due to the nature of the database and represent important end-points that have been investigated in other publications.1–6,21 We also cannot conclusively state that changes in VAD anticoagulation therapy have resulted in improved outcomes. Improvements in surgical technique, VAD type and construction, and experience gained in the management of pediatric patients on VADs undoubtedly contribute to the improvements seen in this report. Changes in therapy have likely occurred concurrently with other changes to pediatric
ANTICOAGULATION THERAPY IN CHILDREN SUPPORTED BY VAD
VAD management. In addition, the database only codes for VAD implantation/explantation but does not specify type or size of VAD. Inferences to the type of VAD can be made on the age of the patient (i.e., Berlin Heart for neonates and infants), but direct associations of pharmacotherapy and a particular type of VAD cannot be made. Doses of medications and laboratory results are also not included in the database, which limits the use of the database to a broad evaluation of prescribing trends. Despite these limitations, the data evaluated are useful for determining future areas of investigation and quality improvement in pediatric VAD therapy. Conclusions The management of pediatric VAD anticoagulation has changed significantly over the past 12 years, including the use of more and varied agents, and the greater use of oral medications. Pediatric VAD outcomes have improved with changes in anticoagulation management. References 1. Fan Y, Weng YG, Xiao YB, et al: Outcomes of ventricular assist device support in young patients with small body surface area. Eur J Cardiothorac Surg 39: 699–704, 2011. 2. Fraser CD Jr, Jaquiss RD, Rosenthal DN, et al; Berlin Heart Study Investigators: Prospective trial of a pediatric ventricular assist device. N Engl J Med 367: 532–541, 2012. 3. Hill JD, Reinhartz O: Clinical outcomes in pediatric patients implanted with Thoratec ventricular assist device. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 115–122, 2006. 4. Humpl T, Furness S, Gruenwald C, Hyslop C, Van AG: The Berlin Heart EXCOR Pediatrics—The SickKids experience 2004–2008. Artif Organs 34: 1082–1086, 2010. 5. Kaczmarek I, Sachweh J, Groetzner J, et al: Mechanical circulatory support in pediatric patients with the MEDOS assist device. ASAIO J 51: 498–500, 2005. 6. Morales DL, Almond CS, Jaquiss RD, et al: Bridging children of all sizes to cardiac transplantation: The initial multicenter North American experience with the Berlin Heart EXCOR ventricular assist device. J Heart Lung Transplant 30: 1–8, 2011. 7. Hetzer R, Alexi-Meskishvili V, Weng Y, et al: Mechanical cardiac support in the young with the Berlin Heart EXCOR pulsatile
215
ventricular assist device: 15 years’ experience. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 99–108, 2006. 8. Morales DL, Gunter KS, Fraser CD: Pediatric mechanical circulatory support. Int J Artif Organs 29: 920–937, 2006. 9. Hetzer R, Potapov EV, Alexi-Meskishvili V, et al: Single-center experience with treatment of cardiogenic shock in children by pediatric ventricular assist devices. J Thorac Cardiovasc Surg 141: 616–623, 2011. 10. Morales DL, Braud BE, Price JF, et al: Use of mechanical circulatory support in pediatric patients with acute cardiac graft rejection. ASAIO J 53: 701–705, 2007. 11. Sharma MS, Forbess JM, Guleserian KJ: Ventricular assist device support in children and adolescents with heart failure: The Children’s Medical Center of Dallas experience. Artif Organs 36: 635–639, 2012. 12. VanderPluym CJ, Rebeyka IM, Ross DB, Buchholz H: The use of ventricular assist devices in pediatric patients with univentricular hearts. J Thorac Cardiovasc Surg 141: 588–590, 2011. 13. Wilmot I, Morales DL, Price JF, et al: Effectiveness of mechanical circulatory support in children with acute fulminant and persistent myocarditis. J Card Fail 17: 487–494, 2011. 14. Blume ED, Naftel DC, Bastardi HJ, Duncan BW, Kirklin JK, Webber SA; Pediatric Heart Transplant Study Investigators: Outcomes of children bridged to heart transplantation with ventricular assist devices: A multi-institutional study. Circulation 113: 2313– 2319, 2006. 15. Chen JM, Richmond ME, Charette K, et al: A decade of pediatric mechanical circulatory support before and after cardiac transplantation. J Thorac Cardiovasc Surg 143: 344–351, 2012. 16. Streif W, Andrew M, Marzinotto V, et al: Analysis of warfarin therapy in pediatric patients: A prospective cohort study of 319 patients. Blood 94: 3007–3014, 1999. 17. Bassler D, Schmidt B: Antithrombin replacement in neonates: Is there any indication? Thromb Res 118: 107–111, 2006. 18. Nankervis CA, Preston TJ, Dysart KC, et al: Assessing heparin dosing in neonates on venoarterial extracorporeal membrane oxygenation. ASAIO J 53: 111–114, 2007. 19. Arabía FA, Tsau PH, Smith RG, et al: Pediatric bridge to heart transplantation: Application of the Berlin Heart, Medos and Thoratec ventricular assist devices. J Heart Lung Transplant 25: 16–21, 2006. 20. Stiller B, Lemmer J, Merkle F, et al: Consumption of blood products during mechanical circulatory support in children: Comparison between ECMO and a pulsatile ventricular assist device. Intensive Care Med 30: 1814–1820, 2004. 21. Miera O, Potapov EV, Redlin M, et al: First experiences with the HeartWare ventricular assist system in children. Ann Thorac Surg 91: 1256–1260, 2011.
