Perioperative complications after pacemaker implantation: higher complication rates with subclavian vein puncture than with cephalic vein cutdown

The number of cardiac implantable electronic devices (CIEDs) has steadily increased over the past decade. Between 2007 and 2016, an increase in pacemaker implantation of 20% was reported in the ESC member countries. In 2016, a total of 547,586 PMs were implanted in 4022 centres across the ESC area [1].

There are several international guidelines on the indications for pacing therapy. However, recommendations on the different surgical techniques for pacemaker implantation were not available until recently [2], which is generally learnt by a proctoring process during training [3].

Since Fuhrmann reported on the successful temporary transvenous endocardial pacing via the brachial vein in 1959 [4], the technique for the transvenous endocardial insertion of the cardiac implantable electronic devices (CIEDs) has developed over different phases. As a result, cephalic vein cutdown (CVC) and subclavian puncture (SP) are currently widely used techniques for lead insertion [5, 6]. The choice of the venous access technique used depends on operator experience.

However, which of the two techniques is safer and preferred during CIED implantation is still being debated. In order to evaluate which of the two approaches is safer, we performed this retrospective analysis of a large patient database.

The whole data set consisted of 139,176 patients. Of these 123,693 patients (88.8%) were included in the data analysis as they underwent either CVC or SP. The remaining patients were excluded due to the combination technique or the use of a different access.

In 75,251 (60.8%) patients, the CVC was used and in 48,442 (39.2%) patients the SP was used. Baseline characteristics of both study populations are summarized in Table 1. The mean age of the patients in both groups was 77 years. In both groups, the majority of patients had an ASA classification 2–3 and normal LV function or slight LV dysfunction (EF > 50%). Dual and single chamber pacemakers were the most commonly used devices (98%) in both groups (Table 1).

Although the proportion of implanted CRT systems in the SP group was three times more common than in the CVC group, this difference in the two groups does not affect the overall result because all implanted CRT systems are only 0.68% of all the implanted pacemakers included in the study.

The CVC group had a complication rate of 2.49% (n = 1879). The complication rate in the SP group was 3.64% (n = 1765). The difference of 1.15% had statistical significance with a p-value < 0.0001. The Odds Ratio of SP in comparison to CVC was 1.47 (95% CI: 1.38–1.57, p < 0.001).

All peri-/postoperative complications are summarized in Table 2. In both groups, lead dislocation (CVC group: 1.62%; SP group 1.78%) was the most common complication. Pneumothorax with 0.85% in the SP group comes in second place and is 5 times more common than in the CVC group (0.15%). Ventricular fibrillation, hemothorax, and postoperative wound infection (0.009% in the CVC group and 0.004% in the SP group) were among the rare complications (Fig. 2).

The duration of the surgery and the fluoroscopy time are summarized in Table 3. The pacemaker implantation using the SP technique took a median of 5 min longer than the implantation using the CVC technique. Figure 3 shows the duration of the surgery in both groups. The median time of fluoroscopy in the SP group was also 0.6 min longer than in the CVC group. Figure 4 shows the fluoroscopy time in both groups. Both the surgery time and the fluoroscopy time contain isolated values that appear unrealistic and because these are only individual cases, they have no influence on the overall result.

Transvenous pacing is one of the most important innovations in cardiac pacing [7, 8]. It followed the development of an introducer, a peel-away sheath, described by Littleford et al. that encouraged routine pacemaker implantation without direct approach to any vein and contributed to the fact that non-surgeons increasingly learned and performed pacemaker implants [9].

Further improvements and developments in surgical techniques, pacemaker devices and pacemaker leads [7, 10], as well as an ageing population with the associated expansion of indication spectrum, explain the continuous increase in the annual pacemaker implantations and the increase in the number of implanting centres.

Despite the impressive developments in pacemaker therapy, complications associated with pacemaker implantation are still observed [11‐14]. Analysing big data from quality assurance data or pacemaker register should help in finding out the approach with the lowest complication rate.

Our data in a large, unselected population shows an overall complication rate of 2.9% (3644 events in 123,693 procedures). The rate of all complications was very low in our study compared with previous reports [11‐14]. The reason for this is that only those complications that required specific interventions were reported. Complications, such as pneumothorax, pericardial effusion, or haematoma that were managed conservatively were not reported. Furthermore, our study only represents the short-term complications in the first implantation of pacemakers.

When comparing the complication depending on venous access, our study with 1879 events of 75,251 implantations (2.49%) in the CVC group shows a significantly lower complication rate in the CVC group compared with 1765 events of 48,442 implantations (3.64%) in the SP group. The superiority of the CVC with respect to the complication as pneumothorax and lead failure compared to SP has also been demonstrated in two recently published meta-analyses [15, 16]. Our study with the large data differs in that not only individual complications were considered, but all peri-/postoperative complications that require intervention.

We determined statistical significance for the total complication rate of both groups. The events in the individual complications such as postoperative wound infection (only 2 in the CVC group) are so small that the statistical significance cannot be proved absolutely certain. The most common complication in both groups is lead dislocation, followed by pneumothorax. Both complications occurred more in the SP group, which is in agreement with previous studies [15, 16]. This tendency is also found in the complications, asystole, ventricular fibrillation, hemothorax, pericardial effusion, and other complication requiring intervention. In contrast, there was a higher percentage of lead dysfunction, pocket hematoma, and postoperative wound infection in the CVC group compared to the SP group.

The subgroup analysis of the centres according to their venous access expertise shows that over 50% of the centres had predominant experience with CVC, while the centres with predominant SP experience were presented with only 14%. The statistically significant difference in the complication rate between the CVC and the SP group could no longer be demonstrated in the subgroup with predominantly SP experience, so that the dominant experience could possibly have an influence on the overall result.

In a prospective randomised study, Calkins et al. demonstrated that the placement of endocardial pacemaker and defibrillator leads using the extrathoracic subclavian vein guided by contrast venography had a higher initial success rate compared to CVC with no difference in the incidence of complications [17]. The information in our retrospective studies does not contain any information on whether the subclavian puncture was performed extrathoracically or intrathoracically. In addition, the years of the study (2010–2014) encompass a period before the routine use of venographic methods for axillary vein access and ultrasound-guided venous access. As such the differences between the cutdown and Seldinger methods may be less in the current era.

Recently, the puncture of the axillary vein for lead insertion of cardiac implantable electronic devices has recently become more important, after several studies have shown the superiority of the axillary access in terms of complications compared to the SP and comparable to the CVC with a high success rate of the axillary access [15, 18‐20].

The data collection of the quality assurance programme only included inpatient pacemaker implantations; outpatient pacemaker implantations were not considered. Because the data collection ends with the discharge of the patient from the hospital, the documented complications only represent the short-term complications. The already collected data do not provide us with any information on whether the operator’s expertise, the requirements in their own centre or the anatomical and clinical characteristics of the patient played a role in the choice of venous access. A further potential bias is that, our study has no randomization; therefore we have an imbalance in single observational characteristics (e.g. those undergoing CVC were more likely to have single-lead devices and preserved LVEF). Since our data encompass a period before the routine use of venographic methods for axillary vein access and ultrasound-guided venous access, the differences between CVC and improved puncture techniques may be less in the current era.

Our retrospective data analysis demonstrated that the cephalic vein cutdown (CVC) as a venous access for pacemaker implantation has significantly fewer peri-/postoperative complications compared with the subclavian puncture (SP). Therefore, the CVC should be learned by all pacemaker implanters and appears to be an advantageous technique for the transvenous endocardial insertion of the cardiac implantable electronic devices.