Author + information
- Received February 8, 2016
- Revision received May 19, 2016
- Accepted May 26, 2016
- Published online January 16, 2017.
- Kasra Moazzami, MD, MPHb,
- Elena Dolmatova, MDb,
- Neil Kothari, MDb,
- Victor Mazza, MDa,b,
- Marc Klapholz, MDa,b and
- Alfonso H. Waller, MDa,b,∗ ()
- aDivision of Cardiology, Rutgers New Jersey Medical School, Rutgers, State University of New Jersey, Newark, New Jersey
- bDepartment of Medicine, Rutgers New Jersey Medical School, Rutgers, State University of New Jersey, Newark, New Jersey
- ↵∗Reprint requests and correspondence:
Dr. Alfonso H. Waller, Division of Cardiology, Department of Medicine, Rutgers New Jersey Medical School Rutgers, The State University of New Jersey, 185 South Orange Avenue, Newark, New Jersey 07103.
Objectives The aim of this study was to describe the trends and predictors of cardiac tamponade among permanent pacemaker (PPM) recipients in the United States between 2008 and 2012.
Background Limited data exist regarding the burden, trend, and predictors of tamponade in patients following PPM implantation.
Methods The National (Nationwide) Inpatient Sample database was used to identify PPM implantations between 2008 and 2012.
Results Among 922,549 patients who received PPM devices between 2008 and 2012, cardiac tamponade occurred in 2,595 patients (0.28%). Overall, in-hospital cardiac tamponade rates increased by 35% among recipients of PPMs. The incidence rate steadily increased from 0.26% in 2008 to 0.35% in 2012 (p < 0.0001). Although the mean age (p = 0.28) and sex distribution (p = 0.25) did not change over the years, the rate of in-hospital mortality increased among patients who developed tamponade from 2008 to 2012 (p = 0.014). After multivariate adjustment for patient and hospital characteristics, female sex (odds ratio [OR]: 1.23; 95% confidence interval [CI]: 1.04 to 1.54; p = 0.011), dual-chamber pacemakers (OR: 1.68; 95% CI: 1.17 to 2.41; p < 0.004), and chronic liver disease (OR: 3.18; 95% CI: 1.92 to 5.64; p < 0.001) were found to be independently associated with a greater odds of cardiac tamponade. Conversely, hypertension (OR: 0.71; 95% CI: 0.45 to 0.94; p = 0.021) and atrial fibrillation (OR: 0.78; 95% CI; 0.61 to 0.96; p = 0.002) were associated with lower odds of tamponade.
Conclusions The burden of cardiac tamponade associated with PPM implantation has steadily increased in the United States. Specific patient factors were identified that could predict the risk for developing tamponade among PPM recipients.
Cardiac perforation resulting in tamponade is a severe complication of permanent pacemaker (PPM) implantations leading to significant morbidity and mortality. Previous studies have estimated the prevalence of cardiac perforation to be 0.1% to 0.8% for PPM implantations (1,2). However, most studies included limited numbers of patients and applied older lead technologies. Therefore, recent large-scale studies evaluating the prevalence and trend of cardiac tamponade are lacking. Moreover, even fewer studies have investigated predictors of in-hospital cardiac tamponade from PPM implantations and their in-hospital outcomes. Knowledge regarding potential predictors of tamponade in this population helps physicians identify patients at highest risk for development of this potentially catastrophic complication and facilitate their decision making.
Therefore, in the present study, we used data from the National (Nationwide) Inpatient Sample (NIS) to describe the trends of cardiac tamponade among PPM recipients in the United States between 2008 and 2012 and also to determine the predictors of tamponade in this population.
The NIS was queried to identify patient demographics and risk profile for PPM recipients in the United States between 2008 and 2012 using the International Classification of Diseases-Ninth Revision-Clinical Modification (ICD-9-CM). The NIS contains deidentified patient-level data and is a subset of the Agency for Healthcare Research and Quality. The database is robust and is a nationally representative survey of hospitalizations designed to annually compile a representative sample of hospital discharge records in the United States. It is the largest all-payer inpatient database, representing approximately 20% of all hospitals in the United States, regardless of patient characteristics or payment source. National estimates can be produced using sampling weights provided by the sponsor. Previous studies have shown that the results of the database correlate well with other hospitalization discharge databases in the United States, and therefore it has been used to explain trends in other acute medical and surgical conditions (3).
Study design and cohorts
The study population included all patients who underwent primary PPM implantation from 2008 to 2012 who are included in the NIS database. ICD-9-CM codes were used to identify the study cohort. Primary procedural codes 37.80 to 37.83 and 00.50 were used to identify the population of PPM recipients. Patients undergoing replacement of any type of PPM device (ICD-9-CM procedural codes 37.85 to 37.89) or insertion or replacement of PPM leads (ICD-9-CM procedure codes 37.75 to 37.77) were excluded from the analysis. The type of PPM implanted was characterized as single-chamber ventricular, dual chamber (DDD), or biventricular. The diagnostic code 423.3 was used to identify patients who developed cardiac tamponade during hospital admission in the PPM population. These ICD-9-CM codes have been consistent during this time period, thereby allowing the analysis of longitudinal trends in the data for prevalence of PPM implantations. The NIS discharge records were queried to identify demographics, including age, sex, race, insurance status, primary and secondary procedures, hospitalization outcome, and length of stay.
The comorbidities associated with tamponade development were identified using Agency for Healthcare Research and Quality comorbidity measures. These comorbidity measures use ICD-9-CM diagnoses to identify different comorbidities on the discharge date. The severity of comorbidities were identified by using the Deyo modification of the Charlson Comorbidity Index (4). This index contains 17 comorbid conditions with differential weights, with a total score ranging from 0 to 33. Higher Charlson Comorbidity Index scores correspond to greater burden of comorbid diseases.
Stata IC 13 (StataCorp LP, College Station, Texas) was used for all analyses. For categorical variables, chi-square tests were used, and for continuous variables such as length of stay, Wilcoxon signed rank tests were used, because these variables were not normally distributed. For analysis of trends, the ptrend (using the Cochrane-Armitage test for categorical variables ) and nptrend (using the nonparametric test for trend by Cuzick for continuous variables ) commands in Stata were used. Hierarchical mixed-effects logistic regression models were generated to determine the independent predictors of tamponade in the PPM population. Two-level hierarchical models (with patient-level factors nested within hospital-level factors) were created using the unique hospital identification number incorporated as random effects within the model. The model included patient-level variables such as age, sex, and comorbidities as well as hospital-level variables such as hospital size (number of beds), hospital region, and teaching status. Choice of covariates for the multivariate analyses was based on the plausibility that they could be associated with cardiac tamponade. A p value <0.05 was considered to indicate statistical significance.
Baseline clinical characteristics
A total of 922,549 PPM implantations were estimated in the United States between the years of 2008 and 2012. Among these patients, single-chamber ventricular PPMs were implanted in 178,589 patients (17.6%), while 733,638 (72.3%) and 102,486 (10.1%) patients received DDD and biventricular devices, respectively. A total of 2,595 patients (0.28%) developed in-hospital cardiac tamponade during admission. Patients’ clinical characteristics and associated comorbidities are shown in Table 1. The mean age of patients developing tamponade was 71.41 ± 14.96 years, and most patients were women (54.6%) (Table 1). Compared with PPM recipients who did not develop tamponade, patients with tamponade were younger, were more often women, and had higher rates of congestive heart failure, chronic liver disease, chronic kidney disease, and death and longer lengths of hospital stay (Table 1). Cardiac tamponade occurred less often among PPM recipients with histories of diabetes mellitus and hypertension.
Trends in patients developing cardiac tamponade
As shown in Figure 1, the rate of in-hospital cardiac tamponade increased by 35% among recipients of PPMs. The incidence rate steadily increased from 0.26% in 2008 to 0.35% in 2012 (p < 0.0001). From 2008 to 2012, the mean age as well as the sex distribution did not change among patients with tamponade (Table 2). However, significant increases in the rates of most associated comorbidities were seen (Table 2).
Overall, 177 patients (6.8%) who developed in-hospital tamponade died during admission, which was more than 6 times higher than the number of PPM recipients who did not develop tamponade (1.1%) (p < 0.001). Also, as shown in Table 2, the overall rate of in-hospital mortality among patients who developed tamponade significantly increased between 2008 and 2012 (p = 0.014). After multivariate adjustment for patient and hospital characteristics, patients experiencing cardiac tamponade had greater odds of post-procedural hospital stays >3 days (odds ratio: 5.64; 95% confidence interval: 2.78 to 8.78; p < 0.001) and in-hospital death (odds ratio: 6.41; 95% confidence interval: 4.28 to 9.73; p < 0.001).
Predictors of cardiac tamponade
Table 3 demonstrates the independent factors associated with cardiac tamponade. After multivariable adjustment, female sex, chronic liver disease, and DDD device implantation were found to be independently associated with greater odds of cardiac tamponade. Conversely, hypertension and atrial fibrillation were associated with lower odds of tamponade.
With the use of the largest national inpatient database in the United States, the present study reports contemporary data on cardiac tamponade among patients receiving PPM and identified factors that predict the risk for developing tamponade in this patient population. The major findings of the present study are the following: first, the rates of in-hospital cardiac tamponade and in-hospital mortality increased from 2008 to 2012; second, female sex and chronic liver disease were identified as independent factors that increased the odds of developing tamponade; and finally, hypertension and atrial fibrillation were shown to be associated with lower odds of cardiac tamponade.
Multivariate-adjusted analysis revealed female sex, history of chronic liver disease, and DDD device implantation to be independently associated with the development of tamponade. Although older studies did not reveal sex as a potential risk factor for developing tamponade following PPM implantation (7–9), recent evidence suggests female sex to be independently associated with cardiac tamponade among PPM recipients (10,11). Studies investigating the in-hospital adverse events related to implantable cardioverter-defibrillator implantations have also shown that women are more likely than men to develop cardiac tamponade following device implantation (12,13). Although the exact mechanism for a higher adverse event rate, including tamponade, is not fully explained, smaller body size and a thinner myocardial wall have been proposed as possible explanations (14). Because cardiac tamponade and other adverse events that are more common in women may be of a mechanical nature, potential technical challenges associated with smaller body size may potentially predispose women to higher adverse events. In addition to female sex, liver disease was also identified as a risk factor for the development of tamponade in the present study. Increased mortality following elective surgery, colorectal resection, emergency laparotomy, and abdominal aortic aneurysm repair in patients with liver disease has been previously reported (14–16). Also, there is evidence of increased mortality and perioperative adverse events in patients with chronic liver disease who undergo cardiac surgery (17). Impaired coagulation as well as intrinsic cirrhosis-related cardiomyopathy are potential mechanisms that could be partly responsible for the increased rate of tamponade in patients with PPM implants (18). Finally, implantation of DDD PPMs was associated with higher odds of developing tamponade, which is consistent with previous studies suggesting a higher risk for cardiac perforation with the implantation of devices with more leads (13,19).
Histories of atrial fibrillation and hypertension were shown to be associated with lower odds of cardiac tamponade in the present study. In a previous study investigating the predictors of cardiac perforation in implantable cardioverter-defibrillator recipients, atrial fibrillation was also associated with a decreased risk (13). Atrial enlargement, increased degree of fibrosis, decreased contractile ability of the atrium against the tip of the leads, and fewer atrial leads often used in patients with atrial fibrillation have been proposed as possible explanations for the protective effects of atrial fibrillation. Conditions associated with myocardial or pericardial fibrosis, such as diabetes mellitus and coronary artery bypass grafting, have been shown to provide protection against cardiac perforation in patients with device implantations (13). Similarly, hypertension-induced myocardial fibrosis could have a protective role against tamponade development, as shown in our cohort of patients (20). It should be noted, however, that the present study was not equipped to identify the underlying mechanisms of these associations, and therefore future studies are required to elucidate the mechanisms of predictors found to be associated with cardiac tamponade in our study.
First, the NIS is an administrative database, which, like similar databases, is predisposed to errors arising from coding inaccuracies. However, previous studies have indicated that the inaccuracies observed in such databases would likely be random and result in undercoding rather than overcoding (21). Therefore, the true incidence of cardiac tamponade could have been underestimated in the present study as a result of undercoding. Second, the present study reports incidences of in-hospital tamponades among PPM recipients and was unable to capture complications that occurred after hospital discharge. However, previous studies have shown that the majority of complications, including tamponade development, occur before discharge, therefore increasing the applicability of the current findings (22,23). Third, the NIS database does not contain any information regarding the type of leads (active vs. passive), size, or device manufacturer of the PPM devices, and therefore, the associations between these factors and the occurrence of cardiac tamponade could not be investigated. Additionally, detailed information about anticoagulation is not available from the database. These limitations include the type of anticoagulation agent used, duration of the therapy, whether the therapy was uninterrupted, and whether bridging with other agents was used. Finally, the possibility of residual confounding cannot be excluded in the present study. Therefore, the results of the present study should be interpreted in the context of its limitations.
In a large-scale, real-world experience of PPM implantations in the United States, the results indicate an increase in the incidence of in-hospital tamponade development from 2008 to 2012. This increase coincided with an increase in the rate of mortality. Specific patient factors were identified that could predict the risk for developing tamponade among PPM recipients. These findings provide insight into the increasing burden of tamponade among patients receiving PPM devices and helps physicians identify patients at highest risk for development of in-hospital tamponade following the procedure.
COMPETENCY IN MEDICAL KNOWLEDGE: Among PPM recipients, more attention should be paid to women and those with histories of chronic liver disease because these patients are more likely to develop in-hospital tamponade following device implantation.
TRANSLATIONAL OUTLOOK: Further research is required to elucidate the mechanisms of novel predictors found to be associated with cardiac tamponade in our study.
This study was supported by the New Jersey Medical School Department of Medicine and the New Jersey Medical School Hispanic Center of Excellence, Health Resources and Services Administration through grant D34HP26020. The authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Moazzami and Dolmatova contributed equally to this work.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors' institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- International Classification of Diseases-Ninth Revision-Clinical Modification
- National (Nationwide) Inpatient Sample
- permanent pacemaker
- Received February 8, 2016.
- Revision received May 19, 2016.
- Accepted May 26, 2016.
- 2017 American College of Cardiology Foundation
- Cano O.,
- Andres A.,
- Alonso P.,
- et al.
- Schwerg M.,
- Stockburger M.,
- Schulze C.,
- et al.
- Peterson P.N.,
- Daugherty S.L.,
- Wang Y.,
- et al.
- Hsu J.C.,
- Varosy P.D.,
- Bao H.,
- Dewland T.A.,
- Curtis J.P.,
- Marcus G.M.
- Dimarakis I.,
- Grant S.,
- Corless R.,
- et al.
- Dewland T.A.,
- Pellegrini C.N.,
- Wang Y.,
- Marcus G.M.,
- Keung E.,
- Varosy P.D.
- Al-Khatib S.M.,
- Lucas F.L.,
- Jollis J.G.,
- Malenka D.J.,
- Wennberg D.E.