More Information

Submitted: 30 March 2020 | Approved: 27 April 2020 | Published: 28 April 2020

How to cite this article: Droth K, El-Armouche A, Andresen D, Bestehorn K. Influence of implantation access on complication rates in pacemaker implantation: An analysis of quality insurance data. J Cardiol Cardiovasc Med. 2020; 5: 095-100.

DOI: 10.29328/journal.jccm.1001093

ORCiD: orcid.org/0000-0002-6791-5366

Copyright: © 2020 Droth K, et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Keywords: Pacemaker; Complication; Implantation access; Subclavian vein; Cephalic vein

 FullText PDF

Influence of implantation access on complication rates in pacemaker implantation: An analysis of quality insurance data

Kristina Droth1, Ali El-Armouche1, Dietrich Andresen2 and Kurt Bestehorn1*

1Institute of Clinical Pharmacology, TU Dresden, Dresden, Germany
2Evangelisches Krankenhaus Hubertus, Berlin, Germany

*Address for Correspondence: Kurt Bestehorn, MD, Ph.D, Institute of Clinical Pharmacology, TU Dresden, Dresden, Germany, Tel: +49 8178 95493; Email: kurt.bestehorn@googlemail.com

Purpose: According to older reports for implantation of pacemakers the access via vena cephalica (VC) is associated with less complications than the access via vena subclavia (VS), but the VS is still preferred. Valid data on head-to head comparisons between access via VC and VS regarding serious complications are missing. The aim of the analysis was to analyze potential associations between type of implantation access and complication rates on a representative dataset.

Methods: The dataset of the German Federal Council containing all procedures in Germany in 2013 was analyzed. The analysis included hospitalized 75,575 patients with primary pacemaker implantation. The statistical evaluation of the data set was based on a multivariate analysis.

Results: Overall, the rate of implantations via VS was two times higher than via VC (64% vs. 36%). There was significantly higher risk of overall complications (3.0% vs. 2.6%), i.e. surgical complications (1.0% vs. 0.5%), asystole (0.2% vs. 0.1%) and pneumothorax (0.7% vs. 0.1%) at the VS access; the VC access had a greater risk of dislocated atrial leads (0.8% vs. 0.6%).

Conclusion: Our analysis of a large-scale database shows that the overall rate of complications is low in primary pacemaker implantation.

Considering the rate of implantations via subclavian vein being twice as high as via cephalic vein, the clinical implications should be considered when choosing the access route and cephalic vein access should be preferred.

The implantation access either via vena cephalica (VC) or vena subclavia (VS) is the usual procedure in pacemaker implantation [1]. Each procedure has advantages and disadvantages such as success rate, procedure time incl. radiation exposure and complication rate.

In Germany, more implantations are performed via VS than via VC [2]. Reasons are the reported higher success rate, the shorter operation and fluoroscopy times, lower blood loss and the simple puncture of the vein [3-7]. Additionally, the access via VS is associated with a higher rate of pneumothorax [6,7].

Although access via VC is associated with less complications and therefore represents an overall more gentle method for implantation of pacemakers [8], the VS is still preferred.

Valid data on head-to-head comparisons between access via VC and VS regarding serious complications is not available. Therefore, we analyzed potential associations between the type of implantation access and complication rates in a real world data set with 75,575 cases.

Dataset

The retrospective analysis is based on the 2013 dataset of the German Federal Council, held by the AQUA Institute (Göttingen, Germany) [9] containing information about all inpatient primary pacemaker implantations performed in Germany during the registration year and consists data of 75,575 patients. The population could be recorded completely and therefore the specification of confidence intervals was not necessary.

Study population

A study population was recruited which included only patients who were implanted either exclusively via VS or exclusively via VC. Patients who used either both or a different implant access were excluded from further analysis. Consequently, the study population contains 69,957 patients (Table 1).

Table 1: Characteristics of total and study population
  total population study population
    Total VC VS
number of patients (n) 75,575 69,957 25,217 (36%) 44,740 (64%)
mean age 76.18 years 76.3 years 76.67 years 76.18 years
women (n) 34,252 (45.3%) 31,695 (45.3%) 11,081 (43.9%) 20,614 (46.1%)
men (n) 41,323 (54.5%) 38,262 (54.7%) 14,136 (56.1%) 24,126 (53.9%)
Statistical analysis

First, both groups of the study population (VC, VS) were analyzed and compared by using descriptive statistical methods. Categorial variables were recorded by absolute and relative data. Ratio-level data was described by using absolute and relative values, averages, standard deviations and range.

Subsequently, a multivariate analysis (FBSA = finest balancing score algorithm [10]) was applied to adjust the risk in order to exclude the influence of possible independent variables on the complication rates as far as possible. In a first step two cohorts of patients treated via VC or VS access were compared for all relevant variables by descriptive statistics. Secondly all patient-clusters with exactly the same risk-/condition-vector were identified based on the following variables: age, sex, ASA classification (ASA 1 – ASA 5), leading symptom, indication for implantation, etiology, ejection fraction (EF), estimated need of stimulation. All possible 1:1 exact matchings were included in these clusters and the complication rates for all possible exact matchings of both cohorts, which converge to the result of the weighting for the identified clusters, were compared.

A χ2-test was applied to compare the complication rates of both groups for statistically significant differences. A p - value < 0.05 was considered as statistically significant.

Complications

The number of patients with at least one complication occurring at the time of discharge from hospital was recorded. The complications that appeared in this patient group were subsequently analyzed and described in more detail.

Table 2 shows a summary of the precisely recorded complications.

Table 2: Complete list of complications within the dataset.
Intraoperative or perioperative complications Arrhythmias Asystole
Ventricular fibrillation
Complications requiring intervention Pneumothorax
Haematothorax
Pericardial effusion
Pocket hematoma
Postoperative complications Lead dysfunction Atrial lead
Ventricular lead
Lead dislocation Atrial lead
ventricular lead
Wound infection CDC 1
CDC 2
CDC 3
Other complications requiring intervention Not specified in dataset

The complication rates before and after risk adjustment were recorded. After risk adjustment, the influence of the independent variables (age, sex, ASA classification, leading symptom, indication for implantation, etiology, ejection fraction (LVEF), estimated need of stimulation) can be largely excluded and the observed differences can be attributed to the implantation approach.

Characteristics of matched patient pairs

Out of the total cohort of 69,957 patients 19,643 pairs with identical profile of all factors could be identified. The profile of all 1:1 matching pairs is summarized in table 3. The characteristic values marked in bold were recorded most frequently.

Table 3: Profile of all 1:1 matching pairs.
Number of patient pairs 19,943
Mean age 77.07 years
ASA classification ASA 1: 6.3% ASA 2: 45.0% ASA 3: 46.9% ASA 4: 1.7%
Leading symptoms Dizziness: 51.9% Singular cardiac syncope: 10.3% Recurring syncope: 28.7%
Pacemaker indication AV-` II: 11.6% AV-Block III: 26.3% Sick sinus syndrome: 39.8% Bradycardia in atrial fibrillation: 20.0%
Etiology of rhythm isturbances unknown: 90.3%
Ejection fraction (LVEF) < 35%: 0.4% > 35 – 50%: 9.1% > 50%: 79.9% unknown: 10.5%
Estimated need of stimulation permanent (> 90%): 20.9% frequent (5% – 90%): 70.2% rare (< 5%): 8.9%
Complications: primary endpoint

The overall complication rates were 3.0% in the total population and 2.9% in the study population (VC 2.6%, VS 3.0%).

The complications most frequently recorded in the total population - listed in descending frequency -were lead dislocation (1.4%), pneumothorax (0.5%), lead dysfunction (0.4%), asystole (0.2%), pericardial effusion (0.2%), pocket hematoma (0.2%), unspecified (0.2%), ventricular fibrillation (0.1%), haematothorax (0.04%), wound infection (0.03%). Rate of in-hospital mortality was 1.4% in both groups.

Table 4 shows the complications recorded for which a statistically significant difference was determined between the two implant approaches (VC, VS) before risk adjustment.

Table 4: Statistically significant different complications before risk adjustment.
  Total population Study population (n) = 69,957
  (n) = 75,575 VC (n) = 25,217 VS (n) = 44,740 p - value*
Incidence of at least one complication 3.0% 2.6% 3.0% 0.002
Lead dislocation
- atrial lead
1.4%
0.7%
1.5%
0.8%
1.3%
0.6%
0.017
0.013
Pneumothorax 0.5% 0.1% 0.7% < 0.001
Asystole 0.2% 0.1% 0.2% 0.037
Haematothorax 0.04% 0.02% 0.01% 0.035
Wound infection 0.03% 0.01% 0.04% 0.038
*p - value < 0.05 was considered significant.

Table 5 shows the complications recorded for which a statistically significant difference was determined between the two implant approaches (VC, VS) after risk adjustment in matched cohorts.

Table 5: Statistically significant different complications after risk adjustment in matched cohorts.
  VC VS p - value*
Number of patients (n) 19,643 19,643  
Incidence of at least one complication 2.6% 3.0% 0.018
Surgical complicationsa 0.5% 1.0% 0.005
Asystole 0.1% 0.2% 0.047
Pneumothorax 0.1% 0.7% < 0.001
Dislocated atrial lead 0.8% 0.6% 0.027
Intra-hospital mortality 1.4% 1.4% 0.691
*p - value < 0.05 was considered significant.
acontains complication with need for intervention: pneumothorax, haematothorax, pericardial effusion, pocket hematoma.

Of all patients in the total population who faced at least one complication, 49.1% were male and 50.9% were female. In the study population, 49.2% were male and 50.8% were female. Overall, woman had a higher risk of complications in all subgroups, as shown in table 6.

Table 6: General and gender-specific complication rates.
  [n] Patients with at least one complication Complication risk depending on gender
female male
total population 75,575 3.0% 3.3% 2.7%
study population 69,957 2.9% 3.2% 2.6%
VC 25,217 (36%) 2.6% 2.9% 2.4%
VS 44,740 (64%) 3.0% 3.4% 2.7%
Comparison VC and VS male: p = 0.094*
Comparison VC and VS female: p = 0.011*
*p - value < 0.05 was considered significant
Selection of pacemaker type: secondary endpoint

In total, two-chamber systems were most frequently implanted, the DDD system accounting for the largest proportion. CRT systems accounted for the smallest proportion of all implanted systems. The exact proportions of the different systems were: VVI 21.1%, AAI 0.2%, DDD 76.2%, VDD 0.6%, CRT with atrial lead 1.6%, CRT without atrial lead 0.14%, other 0.17%.

All systems, with the exception of the VDD system, were implanted via the VS in more than 50% of the cases. The exact distribution is shown in table 7.

Table 7: Implantation accesses of the pacemaker systems.
System implanted VC VS
VVI 42% 58%
AAI 39% 61%
DDD 35% 65%
VDD 52% 48%
CRT with atrial lead 11% 89%
CRT without atrial lead 15% 85%
Other 33% 67%
Intervention and fluoroscopy time: secondary endpoint

The mean cutting-suture time was numerically 0.7 minutes longer after implantation via the VS than via the VC (VS 53.46 min – VC 52.76 min).

The mean fluoroscopy time was numerically 0.67 minutes longer after implantation via the VS than via the VC (VS 4.82 min – VC 4.15 min).

Mortality: secondary endpoint

As data collection took place at the time of discharge from the hospital, only the intra-hospital mortality rate could be determined.

During the observation period 1.4% of the patients of the total population died in the hospital. 48% of them were female and 52% were male. Altogether 1.5% of all female and 1.3% of all male patients died. Hence the mortality rate of female patients was higher.

In the study population, the mortality rate in both groups was 1.4%. A statistically significant difference did not exist (p = 0.691). The mortality rates of both sexes also showed no statistically significant differences in regards to the implant access method (p (male) = 0.445, p (female) = 0.153).

The study is based on a completely recorded population in Germany of the year 2013. The data is characterized by high quality and completeness [11]. Therefore, despite some limitations, valid results can be assumed.

Implantation access

Within the observed patient cohort, a total of 92.6% of all implantations were performed either via VS or via VC. The proportion of implantations via VS was almost twice as high as the proportion of implantations via VC (64% vs. 36%).

The results are consistent with those of several other smaller studies in which VS was preferred [3,4,8,12]. Regarding the generally higher implantation rate via VS access, the annual report of the German pacemaker and defibrillator register even speaks of a “trend towards the wrong right” [13].

It is noteworthy that the number of implantations via VS has decreased slightly since 2015 and that the access via VC is used more frequently [2].

Complication risk

The overall complication risk of 3,0% is lower than expected from previous studies [8,12,14–17]. We found a significantly higher risk of any complication after using the VS approach (3.0% vs. VC 2.6%) which confirms results of previous smaller studies [8,18,19].

Furthermore, we found that after risk adjustment, VS access is associated with a significant higher risk for asystole and pneumothorax. The higher risk for the development of pneumothorax after subclavian puncture is sufficiently known and is described in many studies without further investigation of possible reasons [5-7,20]. Within the study population, 90% of the observed pneumothoraces occurred after subclavian puncture. The difference was statistically highly significant (p < 0.001). Regarding the risk of asystole, there are no studies available which investigated the influence of implantation access. Only one study could be found which indicates the risk of asystole after pacemaker implantation [16]. However, this study does not establish a correlation between risk of asystole and implantation access.

The only complication, which occurred significantly more frequently after puncture of the VC, was the dislocation of atrial leads. Previous publications have indicated that the fixation mechanism has a significant influence on the dislocation rates of leads [20,21]. No publication could be found that investigated the association between implantation access and atrial lead dislocations.

According to Link, et al. different complication rates of both access routes depend on patient characteristics [17]. We were able to exclude this influence as far as possible by using an exact matching of all documented variables.

Complications showing a significant difference between the two approaches before but not after risk adjustment are lead dislocations in general, haematothorax and wound infection. This implies that the independent variables age, sex, ASA classification, leading symptom, indication for implantation, etiology, ejection fraction (LVEF) and estimated need of stimulation have an assumingly significant influence in regards to the occurrence of these complications.

We found a significantly higher complication risk in women, regardless of implant access. The higher gender-specific risk has already been described in previous studies [14,22].

Implanted pacemaker types depending on implantation approach

In previous studies it has been described that the VC is often only suitable for the implantation of single-chamber systems due to its small size [4,5]. In contrast to this in our analyzed population all systems could be successfully implanted via both accesses. All systems were implanted via the VS in over 50% of the cases, with exception of the VDD system, which was successfully implanted via the VC in 52% of the cases. The implantation of more than one lead is therefore technically possible.

Intervention and fluoroscopy time

We found shorter intervention and fluoroscopy times after implantation via VC than via vs. The mean cutting-suture time was 0.7 minutes shorter and the mean fluoroscopy time was 0.67 minutes shorter. We can thus agree with the publication by Eberhard, et al. which describes an extendet intervention- and fluoroscopy time after subclavian puncture [23]. Other studies describe a longer implantation time with access via VC [4,5,24].

However, it should be noted that fluoroscopy and intervention time is influenced by several factors. For example, the choice of the implanted system has an influence on the duration of the intervention (the time required increases with the number of leads to be implanted [12,23,25]). The surgeon`s experience also has a significant influence [12]. However, these factors could not be investigated within the framework of data analysis.

Mortality

The mortality rates reported in previous studies range between 0.1% and 1.3% and the determined intra-hospital mortality (1.4%) rate is comparable to previous studies [6,8,14,17]. We recorded intra-hospital mortality at time of discharge from the hospital. A follow-up was not possible due to the data acquisition. Furthermore, it was not possible to conclude the exact cause of death. The reason is that the discharge diagnosis was only death and not its exact cause. Thus, it can only be reported how many intra-hospital deaths occurred, but not exactly why the patients died.

There was no statistically significant correlation between implant access and mortality rate.

The data collection only took place in inpatient area, outpatient implantations were not considered. Furthermore, only a limited number of patient characteristics were recorded. It is not possible to conclude on existing risk factors, lifestyle habits or other diseases. There is also no information available on the surgeon`s level of experience and the exact cause of intra-hospital death. Because of pseudonymization, the data set also does not allow conclusions about the number of procedures within the hospital in which the implantation took place and the volume/outcome relation. A follow-up after hospital discharge was also not possible based on the dataset, therefore no mean survival times or long-term complications could be determined.

A more recent and comparable German dataset was not available, but according to the German Heart Report the absolute implantation rates have not changed substantially since 2013 [26].

Our analysis of a large-scale database shows that in patients with primary pacemaker implantation the overall rate of complications is relatively low. But even after adjustment for available risk factors there are differences between the access routes: whereas VS access route is associated with statistically significant higher risk of overall complications, asystole and pneumothorax, the VC access has a greater risk of dislocated atrial lead.

In view of the two times higher rate of implantations via VS (64% vs. 36%) the clinical implications should be considered when choosing the access route and the VC access should be preferred. For some time, leadless pacing for single chamber stimulation is an alternative to conventional implantation via VC or vs. However, long term data are still lacking [27].

Our analysis of a large-scale database shows that in patients with primary pacemaker implantation the overall rate of complications is relatively low. But even after adjustment for available risk factors there are differences between the access routes: whereas VS access route is associated with statistically significant higher risk of overall complications, asystole and pneumothorax, the VC access has a greater risk of dislocated atrial lead.

In view of the two times higher rate of implantations via VS (64% vs. 36%) the clinical implications should be considered when choosing the access route and the VC access should be preferred. For some time, leadless pacing for single chamber stimulation is an alternative to conventional implantation via VC or VS. However, long term data are still lacking [27].

  1. Butter C, Seifert M, Israel CW. Sachkunde. Kardiale Resynchronisationstherapie (CRT)“. Der Kardiol. 2015; 9: 244-252.
  2. Markewitz A. Jahresbericht 2016 des Deutschen Herzschrittmacher- und Defibrillator-Registers – Teil 1: Herzschrittmacher. Herzschrittmachertherapie + Elektrophysiologie. 2019; 30: 102-116.
  3. Calkins H, Ramza BM, Brinker J, Atiga W, Donahue K, et al. Prospective Randomized Comparison of the Safety and Effectiveness of Placement of Endocardial Pacemaker and Defibrillator Leads Using the Extrathoracic Subclavian Vein Guided by Contrast Venography Versus the Cephalic Approach. Pacing Clin Electrophysiol. 2001; 24: 456-464. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11341082  
  4. Kronski D, Haas H. Punktion der Vena subclavia als primärer Zugang bei der Schrittmacher-Sondenimplantation. Herzschrittmachertherapie + Elektrophysiologie. 2001; 12: 204-207.
  5. Chan NY, Kwong NP, Cheong AP. Venous access and long-term pacemaker lead failure: comparing contrast-guided axillary vein puncture with subclavian puncture and cephalic cutdown. Europace. 2016: 19: 1193-1197. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27733455  
  6. Israel CW, Ekosso-Ejangue L. Probleme, Komplikationen und Notfälle bei der Schrittmacherimplantation. Herzschrittmachertherapie + Elektrophysiologie. 2015; 26: 309-319.
  7. Hansky B, Stellbrink C. Kardiale implantierbare elektronische Systeme. Der Kardiol. 2016; 10: 379-394.
  8. van Eck JW, van Hemel NM, Zuithof P, van Asseldonk JP, Voskuil TL, et al. Incidence and predictors of in-hospital events after first implantation of pacemakers. Europace. 2007; 9: 884-889. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/17556376
  9. AQUA - Institut für angewandte Qualitätsförderung und Forschung im Gesundheitswesen GmbH. 09/1 – Herzschrittmacher- Implantation Bundesauswertung 2013. Göttingen; 2014.
  10. Bestehorn F, Bestehorn M, Bestehorn M, Kirches C. A deterministic balancing score algorithm to avoid common pitfalls of propensity score matching. March 2018. http://arxiv.org/abs/1803.02704.
  11. AQUA - Institut für angewandte Qualitätsförderung und Forschung im Gesundheitswesen GmbH. Qualitätsreport 2013. Göttingen; 2014.
  12. Leong KM, Pollard C, Cooke CJ. Cardiology registrars and permanent pacemaker complication rates in a district general hospital--safety and service implications. Clin Med (Northfield Il). 2014;14: 34-37. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24532741  
  13. Markewitz A. Jahresbericht 2015 Des Deutschen Herzschrittmacher- Und Defibrillatorregisters.; 2015.
  14. Kirkfeldt RE, Johansen JB, Nohr EA, Jorgensen OD, Nielsen JC. Complications after cardiac implantable electronic device implantations: an analysis of a complete, nationwide cohort in Denmark. Eur Heart J. 2014; 35: 1186-1194. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24347317  
  15. Pakarinen S, Oikarinen L, Toivonen L. Short-term implantation-related complications of cardiac rhythm management device therapy: a retrospective single-centre 1-year survey. Europace. 2010;12:103-108. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19914920
  16. Nowak B, Misselwitz B. Effects of increasing age onto procedural parameters in pacemaker implantation: results of an obligatory external quality control program. Europace. 2008;11: 75-79. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/18996982
  17. Link MS, Estes NA , Griffin JJ, Wang PJ, Maloney JD, et al. Complications of dual chamber pacemaker implantation in the elderly. Pacemaker Selection in the Elderly (PASE) Investigators. J Interv Card Electrophysiol. 1998;2: 175-179. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/9870010 .
  18. Furman S. Subclavian Puncture for Pacemaker Lead Placement. Furman S, ed. Pacing Clin Electrophysiol. 1986; 9: 467-467. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/2426662
  19. PARSONNET V, ROELKE M. The Cephalic Vein Cutdown Versus Subclavian Puncture for Pacemaker/ICD Lead Implantation. Pacing Clin Electrophysiol. 1999; 22: 695-697. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/10353126
  20. Kolb C. Strategien zur Vermeidung und Behandlung von Komplikationen der Herzschrittmacherimplantation. Herzschrittmacherther Elektrophysiol. 2007;18: 234-242.
  21. Eber B, Lehner S, Lassnig E, et al. Komplikationen bei Schrittmacherimplantation - eine Analyse anlässlich eines Fallberichtes. J für Kardiol - Austrian J Cardiol. 2009;16: 108-111.
  22. Nowak B, Misselwitz B, Erdogan A, Funck R, Irnich W, et al. Do gender differences exist in pacemaker implantation?--results of an obligatory external quality control program. Europace. 2010; 12: 210-215. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19864309
  23. Eberhardt F, Bode F, Bonnemeier H, Boguschewski F, Schlei M, et al. Long term complications in single and dual chamber pacing are influenced by surgical experience and patient morbidity. Heart. 2005; 91: 500-506. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/15772212
  24. Aggarwal RK, Connelly DT, Ray SG, Ball J, Charles RG. Early complications of permanent pacemaker implantation: no difference between dual and single chamber systems. Br Heart J. 1995; 73: 571-575. http://www.ncbi.nlm.nih.gov/pubmed/7626359
  25. Luria DM, Feinberg MS, Gurevitz OT, Bar-Lev DS, Granit C, et al. Randomized Comparison of J-Shaped Atrial Leads with and without Active Fixation Mechanism. Pacing Clin Electrophysiol. 2007; 30: 412-417. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/17367362
  26. Deutscher Herzbericht 2018 Frankfurt am Main December 2018.
  27. Merkel M, Grotherr P, Radzewitz A, Schmitt C. Leadless Pacing: Current State and Future Direction. Leadless Pacing: Current State and Future Direction. Cardiol Ther. 2017; 6: 175-181. PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28707043