Cardiac rhythm devices in the pediatric population: Utilization and complications

Cardiac rhythm devices in the pediatric population: Utilization and complications Richard J. Czosek, MD,* Karthikeyan Meganathan, MS,† Jeffrey B. Anderson, MD,* Timothy K. Knilans, MD, FHRS,* Bradley S. Marino, MD, MPP, MSCE,* Pamela C. Heaton, PhD‡ From the *The Heart Institute, Cincinnati Children’s Hospital Medical Center, Department of Pediatrics, Cincinnati Children’s Hospital Medical Center, †Department of Public Health Sciences, College of Medicine, University of Cincinnati, ‡Division of Pharmacy Practice and Administrative Sciences, College of Pharmacy, University of Cincinnati, Cincinnati, Ohio. BACKGROUND Cardiac rhythm devices are important in the management of pediatric patients with rhythm abnormalities, although factors driving utilization are poorly understood. OBJECTIVE This study sought to evaluate utilization trends, complication rates, and cost associated with device implantation in the pediatric population. METHODS Device implantation was analyzed using the Kids’ Inpatient Database from 1997 to 2006. The type of device implantation, patient demographics, hospital characteristics, acute inhospital complications, cost, and length of stay (LOS) were analyzed. ␹2 tests were used to test association between categorical variables, and logistic regression analysis was performed to evaluate risk factors associated with complications. RESULTS There were 5788 hospitalizations with device implantations. Although there was a significant increase in defibrillator implantation, there was no significant increase in the number of pacemaker implantations over this time period. Patient- and device-related complications were relatively common in all device cohorts (pacemaker 11.2%, 7.2%; defibrillator 5.9%, 11.5%; and biventricular device 19.4%, 26.7%). Type of complication was

Introduction During the past 3 decades, implanted cardiac pacing and defibrillation devices have become mainstays in the treatment of pediatric patients with acquired and congenital heart disease (CHD). Specifically, patients with congenital complete heart block, those who have undergone Fontan or Senning/Mustard palliations, repair of tetralogy of Fallot, and patients with cardiac ion channel abnormalities and cardiomyopathies that carry an associated risk of sudden death contribute to the current demand for permanently implanted pacemakers and defibrillators.1– 4 In addition to chronic pacing in patients with heart block, continually advancing technologies such as biventricular pacing (BiV) and miniaturization of implantable Address reprint requests and correspondence: Dr. Richard J. Czosek, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue MLC, Cincinnati, OH 45229. E-mail address: [email protected].

dependent on device type. Increased risk of complication was evident in the pacemaker cohort, patients with congenital heart disease, cardiomyopathy, previous cardiac arrest, and other heart operations. Patient-related complications increased cost and LOS regardless of patient or procedural characteristics. Device implantation in patients ⬍5 years old was associated with increased LOS and cost but was not associated with increased risk of complication. CONCLUSION Device utilization in pediatrics is increasing due to escalating defibrillator implantation and biventricular pacing. Cost and LOS are significantly increased by patient complications. Reduction in these complications would improve patient care and lower medical costs. KEYWORDS Pediatric; Electrophysiology; Cardiology; Pacemaker; Defibrillator; Complications; United States ABBREVIATIONS BiV ⫽ biventricular pacing; CHD ⫽ congenital heart disease; CI ⫽ confidence interval; ICD ⫽ implantable cardioverter-defibrillator; LOS ⫽ length of stay; OR ⫽ odds ratio (Heart Rhythm 2012;9:199 –208) © 2012 Heart Rhythm Society. All rights reserved.

cardioverter-defibrillators (ICD) have also impacted the landscape of pediatric device utilization.5 Together with evolving device implant indications, improvements in device technology are likely to increase device demand in the pediatric population. Recently, Burns et al documented an increase in defibrillator utilization in pediatric patients; however, utilization and associated complications involved with pacing and biventricular systems were not evaluated.5 The increasing demand for cardiac pacing devices is associated with complications due to acute surgical implantation and postimplantation complications in younger individuals as well as increased cost and length of stay (LOS).6,7 Although implant strategies continue to improve, pediatric device implantation remains an evolving field and continues to utilize equipment designed for adult populations. Whether these implant strategies impact hospital cost or LOS has not been evaluated.

1547-5271/$ -see front matter © 2012 Heart Rhythm Society. All rights reserved.

doi:10.1016/j.hrthm.2011.09.004

200 The ability to evaluate trends in the use of pediatric pacing devices faces a challenge similar to that limiting study in other fields of pediatric cardiology, namely the lack of patient cohort size. The Kids’ Inpatient Database (KID) is a large national database designed to evaluate hospital outcomes and resource utilization of pediatric patients in the United States. The purpose of this study was to evaluate utilization trends and complications associated with pacemaker, defibrillator, and biventricular device implantation in the pediatric population. Additionally, we expand the recent literature by studying pacemakers and BiVs as well as defibrillators, and importantly, identify risk factors associated with acute complications and associated effects on hospital costs and LOS.

Methods Study design and study period A retrospective, observational analysis was performed at 4 annual time points over a 10-year period from 1997 to 2006 to study the use of implanted cardiac pacing devices in pediatric hospitalizations and to analyze acute complications and costs associated with device implantation. This study was performed in accordance with the institutional review board at Cincinnati Children’s Hospital Medical Center. (IRB# 2010-0589).

Data source The 1997, 2000, 2003, and 2006 KID data were used to examine trends in implantation of cardiac pacing device– related pediatric hospitalizations. The KID, developed by the Healthcare Cost and Utilization Project (HCUP), which was sponsored by the Agency for Healthcare Research and Quality (AHRQ), was designed to analyze inpatient visits by children, age 20 years or less, in the United States. The KID includes a sample of discharges of children, extracted from over 2,500 hospitals. The hospitals were communitybased nonrehabilitation hospitals and included stand-alone pediatric centers. Due to its large sample size, the KID is ideal for analyzing rare conditions, procedures, and subpopulations that cannot be studied at a single institution. The KID contains discharge weights or multipliers that, when multiplied for each visit, produce national estimates. The unit of analysis is the hospital visit.

Visit selection Visits during which cardiac pacing devices were implanted were identified in patients 20 years old or younger. All diseases were identified using primary and secondary International Classification of Diseases (ICD-9) procedure codes. This study evaluated new implantation-related hospitalizations in patients with pacemakers, biventricular pacing systems, and defibrillator systems. ICD-9 codes identified for device procedures were as follows: new implantation of a nonbiventricular pacemaker (37.80, 37.81, 37.82, and 37.83), new implantation of an ICD (37.94, 37.95, and 00.51), and new implantation of a biventricular pacemaker/defibrillator (00.50). There were no BiVrelated hospitalizations in 1997 or 2000. Because of the small

Heart Rhythm, Vol 9, No 2, February 2012 number of BiV-related hospitalizations in 2003 and 2006, these frequencies were collapsed and not reported by year.

Covariate description Patient and hospital characteristics Patient-related variables were gender, age, comorbidities, and payer type. Age was categorized into 4 groups: 0 to ⬍5 years; 5 to ⬍12 years; 12 to ⬍18 years, and 18 to 20 years. Because data on race were missing for a significant portion of records in the dataset, race was not used in the analysis. Comorbidities were categorized into 7 categories: CHD, cardiomyopathy, cardiac dysrhythmia, cardiac arrest, channelopathy, heart block, and other heart operations (refer to Appendix for ICD-9 codes). Patients identified as having cardiac arrest experienced a life-threatening event as a reason for hospitalization prior to device implantation. A patient could have more than 1 comorbidity. ICD-9 codes for cardiac ion channel abnormality were new in 2006 and did not exist in the 1997, 2000, and 2003 data years. Payer type was classified as private insurance, government insurance including Medicaid and Medicare, and other, which included self-pay, no charge, and charity. Hospital-related variables included information such as hospital bed size, teaching versus nonteaching status, National Association of Children’s Hospitals and Related Institutions type, location, and region. A small-bed-size hospital is defined as rural with 1 to 49 beds, urban and nonteaching with 1 to 99 beds, or urban and teaching with fewer than 300 beds; a medium-bed-size hospital is defined as rural with 50 to 99 beds, urban and nonteaching with 100 to 199 beds, or urban and teaching with 300 to 499 beds; a large-bed-size hospital is defined as rural with more than 99 beds, urban and nonteaching with more than 199 beds, or urban and teaching with more than 499 beds. National Association of Children’s Hospitals and Related Institutions type identified if a hospital was defined as not a children’s hospital, a children’s general or specialty hospital, or a children’s unit in a general hospital. Hospital region was defined as northeast, midwest, south, or west.

Complications Complications were categorized as device-related complications and patient-related complications. Device-related complications were defined as mechanical device issues for either a pacemaker or an ICD system and included issues with the system generator, lead body, or lead electrode. Patient-related complications included pneumothorax, pericardial effusion/tamponade, hematoma, endocarditis/pericarditis, surgical wound infection, and death (refer to Appendix for ICD-9 codes). Any complication included the number of patients with 1 or more complications during the hospitalization. All complications were acute, occurring during the same hospitalization as the implant procedure.

Cost and LOS analysis The KID database reports charges for each visit. Additionally, AHRQ provides a cost-to-charge multiplier that, when

Czosek et al

Pediatric Cardiac Rhythm Devices

201

Figure 1 Trends in visits involving device implantation by year and comorbidity. A: Trends in pacemaker implantation. B: Trends in ICD implantation. ICD ⫽ implantable cardioverter-defibrillator. *Estimates based on 10 or fewer observations were suppressed according to Healthcare Cost and Utilization Project data use agreement. †ICD-9 coding not available prior to 2006 data point.

applied to the data, estimates the costs associated with each visit. Cost from year 2003 was inflated to year 2006 cost using the consumer price index for inpatient hospital services. LOS is a continuous variable in the data.

Statistical analysis Descriptive analyses involving frequencies and percentages were used to describe the trends in device implantation– related hospitalizations and associated complications. Associations between categorical variables were tested using Rao-Schott ␹2 tests, and risk factors for complications were evaluated by multivariate logistic regression. All analyses were conducted using weighted estimates according to HCUP documentation to obtain national estimates.8 Estimates based on 10 or fewer unweighted observations were not reported. SAS version 9.2 (SAS Institute Inc., Cary, NC) was utilized for all calculations.

Results Overall trends in device implantation There were more than 28.9 million pediatric hospitalizations for the years 1997, 2000, 2003, and 2006 combined. Of these hospitalizations, 5,788 (0.02%) involved a devicerelated implantation procedure. There was an increase in

total number of device implants from 1997 to 2006, with an overall increase of 30% in pacemaker and defibrillator devices from 1997 to 2006 (Figure 1). The total number of pacemaker implants over this time period was flat and decreased as a percentage of total device implants (87.4% to 61.5%). Conversely, the absolute number of defibrillator implantations increased over this time period and increased as a percentage of total device implants. Associated patient comorbidities were analyzed for the 3 device types and trends and are depicted in Figure 1.

Pacemaker implantation There were 4,077 hospitalizations during which pacemakerrelated devices were implanted. Tables 1 and 2 show the trend in pacemaker implants according to patient and hospital characteristics. A significant increase in patients age 18 years or older and slight decreases in other age groups were observed. The magnitude of male predominance decreased during the study period. There was no change in the proportion of patients with particular insurance types or hospital bed size. The majority of procedures were increasingly being performed in teaching institutions similar to the trend in total hospital admission in this dataset. Interestingly, there was a decrease in pacemaker hospitalizations in the

202

Table 1

Patient characteristics of visits involving device implant Pacemaker

Visits with device implant Age 0–⬍5 5–⬍12 12–⬍18 18⫹ Gender Male Female Insurance Type Private insurance Government Self-pay/Other

Defibrillator

1997

2000

2003

2006

Total

1997

2000

2003

2006

Total

BIV Total

1076

951

1062

987

4077

155

350

507

618

1629

82

488 268 278 41

424 182 245 100

420 251 272 119

421 198 211 155

(43.0) (22.1) (24.7) (10.2)

—* —* 111 (71.7) 24 (15.4)

—* 58 (16.5) 164 (46.9) 116 (33.1)

17 86 235 163

(45.4) (24.9) (25.8) (3.8)

(44.6) (19.2) (25.8) (10.5)

(39.5) (23.6) (25.6) (11.2)

(42.7) (20.1) (21.4) (15.7)

1753 900 1006 415

(3.4) (17.1) (47.0) (32.6)

35 88 282 211

(5.7) (14.3) (45.8) (34.3)

69 246 792 514

(4.3) (15.2) (48.9) (31.7)

30 (38.4) 25 (32.2) —* 16 (20.5)

594 (55.2) 482 (44.8)

526 (55.3) 425 (44.7)

577 (54.7) 478 (45.3)

516 (52.8) 461 (47.2)

2214 (54.5) 1846 (45.5)

105 (68.0) 50 (32.0)

215 (61.3) 135 (38.7)

300 (60.3) 198 (39.7)

373 (60.9) 239 (39.1)

992 (61.5) 622 (38.5)

43 (54.4) 36 (45.6)

619 (57.6) 365 (34.0) 90 (8.4)

573 (60.5) 303 (32.0) 71 (7.5)

600 (56.7) 354 (33.4) 105 (9.9)

529 (53.5) 358 (36.3) 101 (10.2)

2321 (57.1) 1381 (33.9) 366 (9.0)

121 (78.4) 26 (16.5) —*

252 (72.0) 66 (18.9) 32 (9.1)

351 (69.7) 115 (22.8) 38 (7.5)

366 (59.3) 190 (30.8) 62 (10.0)

1090 (67.1) 396 (24.4) 139 (8.6)

38 (46.7) 26 (31.9) 18 (21.4)

Values within parentheses are column percentages. *Estimates based on 10 or fewer observations were suppressed according to Healthcare Cost and Utilization Project documentation.

Table 2

Hospital characteristics of visits involving device implant Pacemaker 1997

2000

2003

2006

Total

1997

2000

2003

2006

Total

BIV Total

1076

951

1062

987

4077

155

350

507

618

1629

82

188 (17.4) 328 (30.5) 560 (52.1)

214 (22.8) 299 (31.8) 426 (45.3)

204 (20.5) 332 (33.5) 456 (46.0)

153 (15.8) 296 (30.5) 519 (53.6)

759 (19.1) 1256 (31.6) 1961 (49.3)

—* 42 (26.9) 100 (64.7)

52 (14.9) 115 (32.9) 183 (52.2)

44 (9.1) 142 (29.2) 300 (61.7)

54 (8.9) 152 (24.8) 406 (66.4)

164 (10.2) 450 (28.1) 990 (61.7)

292 (27.1) 784 (72.9)

70 (7.5) 869 (92.5)

75 (7.6) 917 (92.4)

82 (8.4) 887 (91.6)

519 (13.1) 3457 (86.9)

40 (26.1) 115 (73.9)

28 (8.1) 322 (91.9)

29 (5.9) 458 (94.1)

63 (10.4) 549 (89.6)

161 (10.0) 1443 (90.0)

204 (19.0) 429 (39.9) 442 (41.1)

152 (16.2) 468 (49.9) 318 (33.9)

118 (12.0) 470 (47.7) 398 (40.4)

118 (12.4) 428 (45.1) 404 (42.5)

592 (15.0) 1795 (45.5) 1562 (39.6)

34 (22.2) 51 (33.2) 69 (44.6)

101 (29.2) 141 (40.9) 103 (29.9)

135 (28.8) 138 (29.4) 196 (41.8)

143 (24.6) 194 (33.3) 246 (42.2)

413 (26.6) 524 (33.8) 614 (39.6)

—* 41 (52.2) 33 (42.1)

283 164 294 335

245 59 316 332

144 249 338 332

149 256 300 282

821 728 1247 1280

44 (28.7) —* 45 (29.3) 41 (26.3)

91 (26.1) —* 138 (39.6) 77 (22.0)

89 142 156 120

113 145 221 139

338 354 561 377

—* 21 (25.3) 17 (20.7) 37 (45.4)

(26.3) (15.3) (27.3) (31.1)

(25.7) (6.2) (33.2) (34.9)

(13.6) (23.4) (31.8) (31.2)

(15.1) (26.0) (30.4) (28.5)

(20.1) (17.9) (30.6) (31.4)

Values within parentheses are column percentages. *Estimates based on 10 or fewer observations were suppressed according to Healthcare Cost and Utilization Project documentation.

(17.6) (27.9) (30.8) (23.7)

(18.4) (23.4) (35.7) (22.5)

(20.8) (21.7) (34.4) (23.1)

—* 30 (37.3) 41 (51.2) —* —*

Heart Rhythm, Vol 9, No 2, February 2012

Visits with device implant Hospital Bedsize Small Medium Large Hospital Teaching Type Non-Teaching Teaching NACHTYPE Not Children’s Hospital Children’s Hospital Children’s Unit in General Hospital Hospital Region Northeast Midwest South West

Defibrillator

Czosek et al

Pediatric Cardiac Rhythm Devices

203

Figure 2 Complications by device type. Estimates based on combined data from years 1997, 2000, 2003, and 2006.

northeast and west and an increase in midwestern locations, which paralleled trends in overall hospital admissions in this dataset.

ses of changes in patient and hospital characteristics were not performed.

Defibrillator implantation

The overall proportion of total complications for any device-related hospitalization was relatively high (17.6%), and there was no change in total complication rate from 2000 to 2006 for either the pacemaker or ICD cohort (P ⫽ .53; P ⫽.56 respectively). Specific complication rates for all device types using combined data from 2000, 2003, and 2006 were as follows: pneumothorax 2.2%, hematoma 3.3%, endocarditis/pericarditis 1.1%, surgical infection 2.4%, and death 1.7%. Pericardial effusion/tamponade estimates were not calculated due to HCUP cell number requirements. BiV devices had the highest percentage of procedures with total acute complications (42.3%), whereas pacemakers (17.3%) and defibrillators (16.8%) were lower (P ⬍ .01). Also, BiV devices had a significantly higher percentage of procedures with device-related complications (26.7% vs. 7.2% and 11.5%, P ⬍ .01) and patientrelated complications (19.4% vs. 11.2% and 5.9%, P ⬍ .01) compared with pacemakers and defibrillators. Type of complication, device-related vs. patient-related, varied by implantation type between pacemakers and defibrillators. Pacemaker-related implantations were associated with increased patient-related complications (11.2% vs. 5.9%, P ⬍ .01), whereas defibrillator-related implantations were associated with device-related complications (11.5% vs. 7.2%, P ⬍ .01). Figure 2 shows the proportion of patientrelated complications associated with visits involving any of the 3 devices. The most common patient-related complications were hematoma and surgical infection. Several patient and hospital characteristics were associated with increased risk for complication (Table 3). Concomitant CHD (adjusted odds ratio [OR] 1.51, 95% confidence interval [CI] 1.03 to 2.21), cardiomyopathy (adjusted OR 1.72, 95% CI 1.22 to 2.42), cardiac arrest upon hospitalization (adjusted OR 2.56, 95% CI 1.53 to 4.28), and

There were 1,629 hospitalizations during which ICD-related devices were implanted. From 1997 to 2006, there was a 4-fold increase in the absolute number of ICD implant procedures as well as an increase in ICD implants as a percentage of total device implants per year (12.6% to 38.5 %). The majority of ICD implantations were performed in an older population relative to the pacemaker group. As in the pacemaker cohort, the magnitude of male predominance decreased during the study period. In 2006, in the pacemaker cohort, patients 0 to ⬍5 years comprised 42.7% of the total population, whereas in the defibrillator group, patients ⬍5 years comprised only 5.7% of the population. Although the youngest age group comprised a smaller percentage of defibrillator implantations, they demonstrated a significant increase during the study time period, as did the cohort of patients ⱖ18 years of age. The majority of patients in this cohort were male (61.5%). There was a significant increase in visits for patients receiving government-provided insurance, as well as an increase in implantation at teaching facilities as compared with nonteaching institutions similar to national trends for all admissions in this dataset. No significant changes in other hospital-related factors changed over this time period.

Biventricular device implantation There were 82 hospitalizations during which biventricular devices were implanted in 2003 and 2006 combined. No biventricular devices were implanted in 1997 or 2000. Patient and hospital demographic characteristics were similar to those found in the pacemaker implantation, with urban teaching institutions performing the vast majority of device implantation procedures. Due to small sample sizes, analy-

Device and patient-related complications

204 Table 3

Heart Rhythm, Vol 9, No 2, February 2012 Predictors of complications by device type Pacemaker Complications (%)

Visits With Any Complication Congenital Heart Disease No Yes Cardiomyopathy No Yes Cardiac Dysrhythmia No Yes Cardiac Arrest No Yes Heart Block No Yes Other Heart Operations No Yes Age 0–⬍5 5–⬍12 12–⬍18 18⫹ Gender Male Female Patient Insurance Private insurance Government Self-pay/Other Hospital Bedsize Small Medium Large Hospital Teaching Status Non-Teaching Teaching Hospital Type Not Children’s Hospital Children’s Hospital Children’s Unit in General Hospital Hospital Region Northeast Midwest South West

Defibrillator Adjusted odds ratio (95% CI)

707

Complications (%)

Adjusted odds ratio (95% CI)

274

236 (12.0) 471 (22.3)

Reference 1.51 (1.03, 2.21)

204 (17.2) 70 (15.8)

Reference 0.78 (0.47, 1.30)

558 (15.7) 149 (28.2)

Reference 1.72 (1.22, 2.42)

156 (17.6) 117 (15.8)

Reference 0.78 (0.47, 1.27)

451 (18.7) 256 (15.3)

Reference 0.85 (0.62, 1.16)

95 (16.7) 179 (16.9)

Reference 1.02 (0.68, 1.54)

662 (16.8) 45 (33.5)

Reference 2.56 (1.53, 4.28)

259 (17.1) 15 (12.5)

Reference 0.72 (0.34, 1.53)

250 (16.5) 457 (17.8)

Reference 1.07 (0.81, 1.42)

246 (16.4) 28 (21.7)

Reference 1.44 (0.75, 2.78)

353 (12.7) 354 (27.2)

Reference 1.97 (1.37, 2.84)

242 (16.0) 32 (26.2)

Reference 1.49 (0.79, 2.82)

319 175 155 57

(18.2) (19.4) (15.4) (13.8)

0.73 (0.45, 1.19) 1.08 (0.64, 1.82) 0.95 (0.56, 1.62) Reference

15 35 133 90

(21.3) (14.0) (16.8) (17.5)

0.97 (0.30, 3.08) 0.86 (0.43, 1.75) 1.02 (0.57, 1.81) Reference

401 (18.1) 304 (16.5)

Reference 0.90 (0.70, 1.15)

179 (18.1) 93 (15.0)

Reference 0.80 (0.54, 1.19)

370 (16.0) 265 (19.2) 72 (19.5)

Reference 1.06 (0.82, 1.38) 1.17 (0.79, 1.72)

183 (16.8) 73 (18.5) 17 (12.5)

Reference 1.08 (0.65, 1.77) 0.77 (0.33, 1.78)

125 (16.5) 207 (16.5) 358 (18.2)

Reference 0.92 (0.66, 1.29) 1.26 (0.76, 2.08)

25 (15.3) 60 (13.3) 182 (18.3)

Reference 1.00 (0.49, 2.03) 1.75 (0.79, 3.91)

64 (12.4) 626 (18.1)

Reference 1.73 (1.16, 2.60)

32 (20.1) 234 (16.3)

Reference 0.93 (0.41, 2.12)

82 (13.9) 310 (17.3) 295 (18.9)

Reference 1.05 (0.70, 1.58) 1.03 (0.67, 1.59)

83 (20.0) 80 (15.2) 93 (15.1)

Reference 1.07 (0.55, 2.11) 0.66 (0.33, 1.29)

129 111 222 244

Reference 0.93 (0.60, 1.45) 1.34 (0.92, 1.96) 1.35 (0.99, 1.84)

47 78 86 63

Reference 1.67 (0.82, 3.40) 1.14 (0.58, 2.21) 1.27 (0.63, 2.58)

(15.8) (15.2) (17.8) (19.1)

(13.8) (22.0) (15.4) (16.7)

*Estimates based on 10 or fewer observations were suppressed according to Healthcare Cost and Utilization Project documentation.

other heart operations (adjusted OR 1.97, 95% CI 1.37 to 2.84) were all associated with increased risk of complication within the pacemaker cohort. Pacemaker implantations performed in teaching institutions were also associated with an increased risk of complication (adjusted OR 1.73, 95% CI 1.16 to 2.60). This increased risk may in part be affected by differences in patient complexity as teaching institutions had higher percentages of patients with associated CHD, other heart operations, and cardiac arrest upon presentation.

As compared with the pacemaker cohort, there were no patient or hospital characteristics found to be associated with increased risk of complication in the defibrillator cohort.

Cost and hospital LOS Hospital LOS, from combined years 2003 and 2006, was analyzed for the pacemaker and defibrillator cohorts and further stratified by the presence of CHD or additional heart

Czosek et al

Pediatric Cardiac Rhythm Devices

surgeries at the time of device implantation (Figure 3). Patient-related complications increased hospital LOS by 2.0 and 3.3 times, respectively, in the pacemaker and defibrillator cohorts. This increase occurred in patients with or without additional heart operations during their hospitalization as well as with or without CHD. Device-related complications seemed to have less effect on LOS than patientrelated complications and demonstrated no increase in the

Figure 3 Mean hospital cost and mean length of stay, for combined years 2003 and 2006, by device and complication type. Estimates based on combined data from years 2003 and 2006, with 2003 cost inflated to 2006 cost using the consumer price index for inpatient hospital services. *Estimates based on 10 or fewer observations were suppressed according to Healthcare Cost and Utilization Project data use agreement.

205 defibrillator cohort. The impact of implantation complications demonstrated similar effects on cost. Patient-related complications had a more significant cost increase compared with device-related complications. Patient-related complications significantly increased hospital cost in both the pacemaker and defibrillator cohorts (1.4 and 1.3 times, respectively), a finding that was also seen in patients with or without associated CHD or additional heart operations.

206

Figure 4 Mean hospital cost and mean length of stay, for combined years 2003 and 2006, by patient age. Estimates based on combined data from years 2003 and 2006, with 2003 cost inflated to 2006 cost using the consumer price index for inpatient hospital services.

Last, we analyzed the effect of patient age on hospital LOS and cost (Figure 4). Not surprisingly, both hospital LOS and cost were greatest in the youngest population cohort, patient age 0 to ⬍5 years.

Discussion Due to an increase in the number of clinical indications, technological advances, and innovative implantation techniques, clinical demand for implanted cardiac pacing devices in the pediatric population has risen. This study demonstrated that although defibrillator and biventricular pacing system usage is on the rise, pacemaker utilization has not demonstrated a similar increase. Furthermore, we found that associated acute total complications were common and significantly affected both hospitalization cost and LOS. Although there are several mechanistic underpinnings driving the increase in defibrillator implantation, the idea of primary prevention in pediatrics has widespread appeal.1,2,9,10 A second driver is an improved understanding of the implications of identifying a proband within a larger family unit. Genetic as well as other clinical diagnostic testing in otherwise healthy family members of affected individuals with a malignant channelopathy or cardiomyopathy has widened the scope of individuals who are perceived to be at an increased risk of arrhythmia.11–13 Lastly, continuing technological advances, allowing for miniaturization of devices as well as innovative implanta-

Heart Rhythm, Vol 9, No 2, February 2012 tion techniques, has also widened the pool of potential patients in which ICD therapy is possible.7,14,15 Analysis of the comorbidities in this study provides some justification of this rationale. Although the nature of database analysis does not allow us to distinguish primary vs. secondary implantation indications, cardiomyopathies and channelopathies comprised a significant and increasing proportion of the ICD cohort. Although the increasing proportion of ICD implantations could represent an overuse of device technology, the lack of similar increases in pacemaker implantation suggests otherwise. It is our belief that the increasing utilization of defibrillator systems demonstrated in this study likely stems from a thoughtful change in practice based on recent improvements in diagnostic and implant tools. Without similar driving factors, pacemakers have not increased in utilization, and unlike defibrillators, the number of visits for pacemaker implantation were relatively stable over the 9-year study period. The exception to this observation was in the arena of biventricular pacing systems, which like their adult counterparts, became an increasingly popular mode of therapy over the last decade.16,17 As in many other areas of pediatric cardiology, the utilization of biventricular pacing has attempted to use adult criteria in efforts to define pediatric indications.18 Unlike the fairly uniform adult patient population with structurally normal hearts, the pediatric population consists of a heterogeneous cohort of patients ranging from patients with poor function to those with single ventricles with multisite ventricular pacing. Patients with associated CHD composed a significant portion of all device cohorts. Of the 5,788 device-implantation-related visits, 45% had associated CHD. In the BiV cohort, 61% of patient visits had CHD. Utilization of biventricular devices are likely to increase as indications in pediatric populations improve and expand. In this study, there was a surprisingly high percentage of total complications, though similar to recently published data in ICD implantation in pediatric cohorts.5 Recent adult studies have demonstrated lower rates of device-related complications.19 –21 This difference was most pronounced in the small cohort of patients with biventricular devices; the total percentage of complications was ⬎40%. As with many forms of new technology, there are potential learning curves that may result in higher rates of complications. In pediatrics, there is a heightened sense of concern in regard to the risks of placing pacing and defibrillator systems, although this concern typically relates to long-term complications such as venous occlusion in smaller individuals, lead breakdown secondary to patient growth and activity, or inappropriate defibrillator therapy. The basis for the significant difference in acute complications cannot be fully explained in this retrospective analysis, but likely has multifactorial etiologies stemming from relative patient size, complexity of cardiac anatomy, and lack of pediatric device-related technology. It would be expected that a portion of the total complications seen in this cohort resulted secondary to concomitant heart surgeries performed in a subset of the

Czosek et al

Pediatric Cardiac Rhythm Devices

patients, and concomitant heart operation was found to be a risk factor in the pacemaker cohort but was not in the ICD cohort. In this pediatric cohort, patient-related complications were associated with significant cost and LOS increases. This occurred regardless of device type or associated patient characteristics, such as associated CHD. Additionally, similar effects occurred in patients admitted for isolated device implant or whether the device was placed during a hospitalization with additional heart surgery. Although cost and length of hospitalization were increased in the younger-age cohorts, we were surprised that younger age was not associated with increased risk of complication in either the pacemaker or the defibrillator groups. Although previous studies have described the frequency of device-related complications, this is the first study to evaluate type of device complication (patient-related vs. device-related) and the relationship of complication type to device type and effects on hospital cost and LOS. Although the overall numbers of defibrillator and BiV implantations were relatively small compared with those of pacemaker implantations, these cohorts comprise an increasing segment of the device population. With continued improvement in genetic and clinical recognition of disease processes that carry an increased association of sudden death, and an aging CHD population with associated heart failure, a continued increase in utilization of these devices would be anticipated. Because of the associated risks of implantation in the pediatric population, especially in patients with BiV devices, this increasing trend in utilization should be tempered with ongoing analysis of the effectiveness and necessity of devices in these diverse populations.18 Additionally, pediatric-specific studies evaluating pediatric-specific risk and implant indications are required to better define factors that may often differ from adult cohorts. Finally, the high percentage of device-related complications underscores the continued need for pediatric-specific devices and device-related technology.

Study limitations This study has several limitations. First, as with all secondary database studies, the potential for misclassification bias existed. Second, the database is cross-sectional, and therefore, patients could not be followed up longitudinally, which limited our evaluation of complications to those detected during the implantation hospitalization. Third, clinical details such as the type of channelopathy or specific device indications were not available. Also, changes to coding schema, such as the introduction of codes for channelopathy in 2006, limited the trend analysis. However, despite these limitations, this database allowed for a population-based approach to examine trends in the utilization of pacing devices.

Conclusions In conclusion, this study demonstrated that although the use of defibrillator and biventricular pacing systems was rising, nonbiventricular pacing system utilization was flat from

207 1997 to 2006. Overall total complications were relatively frequent, and complication type is influenced by device type. Patient-related complications were associated with significant increases in hospital cost and LOS. A reduction in patient-related complications would result in improvement in patient quality of life as well as a reduction the economic burden associated with pediatric device management. Whether the increased utilization of implanted defibrillators and biventricular pacing systems is warranted by an associated increase in survival needs further study.

References 1. Begley DA, Mohiddin SA, Tripodi D, Winkler JB, Fananapazir L. Efficacy of implantable cardioverter defibrillator therapy for primary and secondary prevention of sudden cardiac death in hypertrophic cardiomyopathy. Pacing Clin Electrophysiol 2003;26:1887–1896. 2. Berul CI, Van Hare GF, Friedman RA, et al. Results of a multicenter retrospective implantable cardioverter-defibrillator registry of pediatric and congenital heart disease patients. J Am Coll Cardiol 2008;51:1685–1691. 3. Maron BJ, Spirito P, Shen WK, et al. Implantable cardioverter-defibrillators and prevention of sudden cardiac death in hypertrophic cardiomyopathy. JAMA 2007;298:405– 412. 4. Silka MJ, Kron J, Dunnigan A, Dick M 2nd. Sudden cardiac death and the use of implantable cardioverter-defibrillators in pediatric patients. The Pediatric Electrophysiology Society. Circulation 1993;87:800 – 807. 5. Burns KM, Evans F, Kaltman JR. Pediatric ICD utilization in the United States from 1997 to 2006. Heart Rhythm 8:23–28. Epub 2010 Sep 29. 6. Alexander ME, Cecchin F, Walsh EP, Triedman JK, Bevilacqua LM, Berul CI. Implications of implantable cardioverter defibrillator therapy in congenital heart disease and pediatrics. J Cardiovasc Electrophysiol 2004;15:72–76. 7. Stephenson EA, Batra AS, Berul CI, et al. A multicenter experience with novel implantable cardioverter defibrillator configurations in the pediatric and congenital heart disease population. J Cardiovasc Electrophysiol 2006;17:41– 46. 8. Agency for Healthcare Research and Quality. HCUP kid database documentation. Healthcare Cost and Utilization Project (HCUP). 2009;2011. 9. Chatrath R, Porter CB, Ackerman MJ. Role of transvenous implantable cardioverter-defibrillators in preventing sudden cardiac death in children, adolescents, and young adults. Mayo Clin Proc 2002;77:226 –231. 10. Khairy P, Harris L, Cecchin F, et al. Implantable cardioverter-defibrillators in tetralogy of Fallot. Circulation 2008;117:363–370. 11. Svendsen JH, Geelen P. Screening for, and management of, possible arrhythmogenic syndromes (channelopathies/ion channel diseases). Europace 2010;12:741– 742. 12. Skinner JR, Crawford J, Love DR, et al. Prospective, population-based long QT molecular autopsy study of post-mortem negative sudden death in 1– 40 year olds. Heart Rhythm 2011; 8: 412-419. 13. Tester DJ, Ackerman MJ. Postmortem long QT syndrome genetic testing for sudden unexplained death in the young. J Am Coll Cardiol 2007;49:240 –246. 14. Berul CI. Defibrillator indications and implantation in young children. Heart Rhythm 2008;5:1755–1757. 15. Silka MJ, Bar-Cohen Y. Pacemakers and implantable cardioverter-defibrillators in pediatric patients. Heart Rhythm 2006;3:1360 –1366. 16. Greene EA, Berul CI. Pacing treatment for dilated cardiomyopathy: optimization of resynchronization pacing in pediatrics. Curr Opin Cardiol 2010;25:95–101. 17. Cecchin F, Frangini PA, Berul CI, et al. Cardiac resynchronization therapy (and multisite pacing) in pediatrics and congenital heart disease: five years experience in a single institution. J Cardiovasc Electrophysiol 2009;20:58 – 65. 18. Misra N, Webber SA, Degroff CG. Adult definitions for dyssynchrony are inappropriate for pediatric patients. Echocardiography 2011;28:468 – 474. 19. Poole JE, Gleva MJ, Holcomb R, et al. Complication rates associated with pacemaker or implantable cardioverter-defibrillator generator replacements and upgrade procedures: results from the replace registry. Circulation 2010; 122:1553–1561. 20. Klug D, Balde M, Kacet S, et al. Risk factors related to infections of implanted pacemakers and cardioverter-defibrillators: results of a large prospective study. Circulation 2007;116:1349 –1355. 21. Moller M, Arnsbo P, Wiggers R, et al. Quality assessment of pacemaker implantations in Denmark. Europace 2002;4:107–112.

208

Heart Rhythm, Vol 9, No 2, February 2012

Appendix

ICD-9 Codes description

Complications Device Complications 996.01 996.72 996.04 E878.1 Pneumothorax 512.1 Pericardial Effusion/Tamponade 423.0 423.3 Hematoma 998.12 998.11 Endocarditis/Pericarditis 421.xx 420.xx Surgical Infection 998.59 996.61 Accidental Puncture/Laceration E870.0 E870.6 998.2 Cardiac Co-morbidities Congenital Heart Disease 745.xx-747.49 except 746.84 Cardiomyopathy 425.1 425.4 746.84 334.0, 452.8 429.3 271.0, 425.7 359.1 277.9 674.5 428.xx 425.xx Cardiac Dysrhythmia 427.xx, except 427.5 Cardiac Arrest 427.5 Channelopathy 426.81, 426.82 Heart Block 426.0, 426.1x, 426.6, 426.9, 746.86 Other Heart Operations 35 – 35.04, 35.1 – 35.14, 35.2 – 35.28, 35.3 – 35.35, 35.39 – 35.42, 35.5 – 35.56, 35.6 – 35.63, 35.7 – 35.73, 35.8 – 35.84, 35.9 – 35.94, 35.98, 35.99, 37.5 – 37.54, 37.6 – 37.66, 37.68, 37.91, 38.3 – 38.35, 38.44, 38.45, 39.00 – 39.02, 39.2 – 39.29, 39.51 – 39.54, 39.6 – 39.63

Pacemaker complication, mechanical Pacemaker complication, other ICD complication, mechanical ICD complication, mechanical Pneumothorax Pericardial effusion Pericardial Tamponade Hematoma Hemorrhage complicating a procedure Endocarditis Pericarditis Surgical Infection Pacemaker complication, inflammatory Accidental cut puncture perforation or hemorrhage during surgical operation Accidental cut puncture perforation or hemorrhage during aspiration of fluid or tissue puncture and catheterization Accidental puncture or laceration during a procedure not elsewhere classified Congenital Heart Disorders Hypertrophic Cardiomyopathy, obstructive Hypertrophic Cardiomyopathy, Non-obstructive Hypertrophic Cardiomyopathy,Congenital Freidrichs Ataxia Dilated Glycogen Storage Disease Muscular Dystrophy Metabolic Peri/Post Partum Heart Failure Cardiomyopathy Cardiac Dysrhythmias Cardiac Arrest Long QT Syndome (LQTS) Heart Block Other Heart Operations