We prospectively assessed the outcomes of 664 patients aged 60 years or older who underwent their first PPI due to ASB during 2 periods of identical length: from September 1, 2018 to February 28, 2020, when the implants were performed during working hours only [PPI-WH, n=341]); and from 31 July 2020 to 31 December 2021, when urgent implants were made available 24hours a day, 7 days a week [PPI-24/7, n=323]). The months from March to June 2020 were excluded from the analysis due to the COVID-19 pandemic, as the normal activity of our health care service was interrupted (table 1).

The inclusion criterion was the indication for PPI (class I or IIa) according to the current clinical practice guidelines of the European Society of Cardiology.3,7 Exclusion criteria were scheduled pacemaker implantation, and a class I or IIa indication for advanced pacing strategies (leadless pacemaker, resynchronization therapy, left bundle branch area pacing) according to current European guidelines. Leadless pacemakers were considered when there was no upper extremity venous access and in patients on hemodialysis.3 Resynchronization therapy was indicated for patients with left ventricular ejection fraction <40% who had indication for pacing and a high degree of AV block.3 Left bundle branch area pacing was used in resynchronization therapy candidates when the coronary sinus lead implantation was unsuccessful.3 Patients with exclusion criteria underwent implantation of the indicated device on a scheduled basis and during normal working hours. Finally, according to our institution's protocol, patients younger than 60 years with AV block undergo an etiological study with MRI and subsequent scheduled implantation with left bundle branch pacing.

All patients were informed about the study and provided their written consent. This study was carried out in accordance with internationally accepted recommendations for clinical investigation (Declaration of Helsinki of the World Medical Association). The protocol for the PPI-24/7 service was approved by the administrative authorities of our institution.

After the diagnosis of ASB, which usually occurred in the hospital emergency room, and according to the established protocol, an urgent transthoracic echocardiogram and laboratory tests were performed to rule out a potentially reversible cause and to define the most appropriate pacing strategy. The care of patients with ASB with pacemaker indication during regular working hours was exactly the same in both study periods.

In the PPI-WH period, all pacemaker implants were performed during normal working hours (Monday to Friday, from 8:00 to 15:00hours). When the indication was established during normal working hours, the implantation was performed as soon as an operating room was made available as the rooms could be busy following the usual scheduling of patients. When the indication was established outside normal working hours, the patients were hospitalized under continuous electrocardiographic monitoring in the cardiovascular intensive care unit (in case they showed clinical instability or the need for transvenous transient pacing) or in the conventional hospitalization area. Implantation of the permanent pacemaker was then performed at the beginning of the next normal working day.

For the PPI-24/7 period, the device implant was performed as soon as an operating room was available (if the indication occurred during normal working hours) or immediately if it took place outside these hours. In the latter case, the 24-hour Implant Team was activated comprising an electrophysiologist and a nurse, supported in cases of unstable patients by a physician from the general cardiology on-call service.

Device implant and postimplant patient management were identical for both approaches. The patient's written consent was obtained prior to the procedure and cefazolin or vancomycin (if the patient was allergic to beta-lactam) was administered.8 The surgery was carried out following the current standards for cardiac device implants, with the atrial electrode positioned in the right atrial appendage and the ventricular electrode in the right ventricular apex. Venous access (cephalic, subclavian, or axillary vein) was determined by the operator.

A clinical evaluation by the general cardiology emergency team (including an electrocardiogram and a chest radiograph) was carried out 4hours after pacemaker implantation. Early patient discharge was then allowed in the absence of complications or other clinical processes that prevented their leaving. Patients were treated with the support of a Gilchrist bandage for 24hours and were instructed not to lift their arm homolateral to the implanted device, above shoulder level. In addition, pressure dressing of the pocket was applied for 4hours postintervention in all participants. Figure 1 shows the flow chart of patients in the procedures performed during the 2 periods.

Figure 1.

Flow chart of patients according to the approach under study. Diagram depicting the flow of patients in the procedures performed during PPI-WH (top) and PPI-24/7 (bottom). PPI-24/7, implant of permanent pacemaker 24hours a day, 7 days a week; PPI-WH, implant of permanent pacemaker during working hours.

(0.21MB).

The primary safety endpoint was established as the cumulative 180-day incidence of complications related to the index arrhythmia and device implant. In addition, complications occurring before implantation and those related to implantation (peri- and postimplantation) were analyzed separately. Major complications following implantation included cardiac perforation, pericardial tamponade, valve damage, hemothorax, pneumothorax, hematoma requiring surgery, myocardial infarction, peripheral embolus, ictus, and death. The primary efficacy endpoint was determined as the average number of hospital stays per patient due to the index arrhythmia. A hospital stay was defined as an overnight stay plus a main meal (lunch or dinner).9,10 Two secondary endpoints were predefined: cumulative survival at 180 days, and cumulative survival free from cardiovascular mortality at 180 days. We performed a cost analysis, defined as the savings from the use of hospital resources (stay in conventional hospitalization, stay in the cardiovascular intensive care unit and hospital emergency) in the 6 months following the index clinical event. The costs generated by the use of hospital resources are regulated by the National Health System:11,12 €642.2 for a stay in conventional hospitalization, €1053 for a stay in an intensive care unit and €119.14 for a stay in an emergency room. Likewise, the remuneration of the extra staff of the 24/7 service (on call physician and nurse) is legally established.13 Finally, an incremental cost-effectiveness analysis (ICE) was performed, following the formula: ICE=C PPI-WH – CB PPI-24/7 / E PPI-WH -E PPI-24/7, where C is cost, and E is efficacy (survival at 6 months)14 (table 1).

Follow-up was performed by face-to-face consultation: 10 days after implantation, 180 days after implantation, and then annually or biannually thereafter, depending on the patients’ clinical characteristics. For the endpoint analysis, follow-up began at the time of diagnosis of the arrhythmia leading to pacemaker implantation.

The analysis was performed using SPSS 25.0 for Windows (SPSS Inc, United States). Normally distributed continuous variables are described by mean and standard deviation, while categorical variables are expressed by absolute number and percentage. Continuous variables without normal distribution (determined by the Kolgomorov-Smirnov test) are described by the median and interquartile range (IQR). A comparison of categorical variables was performed using the chi-square test (or Fisher exact test if n <5). A comparison of 2 continuous variables with normal distribution was performed using the Student t test. The comparison of t2 continuous variables without normal distribution was performed using the Mann-Whitney U test. The log-rank method was used to compare the cumulative incidences at 180 days, expressed graphically with Kaplan-Meier curves. Multivariable analysis of the incidence of complications at 180 days was performed using the Cox regression method, including the following variables: age, sex, hypertension, diabetes, previous ischemic heart disease, atrial fibrillation, chronic oral anticoagulation, cognitive impairment, dual-chamber pacemaker implantation, and study period (PPI-WH vs PPI-24/7). A value of P <.05 was considered statistically significant.

As depicted in table 1, the baseline characteristics of the patients, the prevalence of wide QRS complex and venous access were similar in the 2 periods analyzed. Table 2 shows the distribution of pacing mode according to clinical arrhythmia and age. There were no significant differences in pacing mode in PPI-WH vs PPI-24/7.

Diagnosis of ASB was most frequently established in the emergency department (76%). For PPI-24/7, 178 (55%) device implants were performed out of working hours: 113 in the afternoon (from 15:00 to 21:59h), 50 at night (22:00 to 07:59h), and 15 in the morning on a nonworking day. This strategy was associated with a significant reduction of 13hours in the average delay from diagnosis to implantation (table 1).

The cumulative incidence of patients with complications at 180 days was significantly lower in the PPI-24/7 period: 9% vs 17% (P=.002; log-rank test) (figure 2). In a multivariate analysis (Cox regression), age (adjusted odds ratio, 1.04, per year) and PPI-24/7 (adjusted odds ratio, 0.5) were found to be independent predictors of complications (table 3).

Figure 2.

Cumulative incidence of complications at 180 days. Kaplan-Meier plot showing the lowest incidence of complications for PPI-24/7. All complications from diagnosis of bradycardia are included. PPI-24/7, permanent pacemaker implantation 24hours a day, 7 days a week; PPI-WH, permanent pacemaker implantation during working hours.

(0.2MB).

The incidence of preimplant complications increased significantly with the delay from ASB diagnosis to pacemaker implant (figure 3). PPI-24/7 was associated with a reduction in the incidence of preimplant complications (2.5% vs 12%; log-rank=20; P <.001), but the benefit was concentrated only in patients who underwent the implant within 12hours (igure 3). As shown in table 4, the recurrence of symptomatic bradyarrhythmia and the occurrence of delirium were significantly less frequent in PPI-24/7. As expected, the time from diagnosis to implant was significantly longer in patients with preimplant complications: 24hours (IQR: [12-35] vs 12 [11-13]; nonparametric P <.001.

Figure 3.

Incidence of preimplant complications according to the time from diagnosis of bradycardia. The incidence of preimplant complications increased linearly with time from the clinical onset of bradycardia among the total study population (n=664 patients). PPI-24/7, permanent pacemaker implantation 24hours a day, 7 days a week.

(0.21MB).

A total of 42 (6.3%) patients had an implant-related complication. The cumulative incidence at 180 days was similar in the 2 periods analyzed: 6.7% (PPI-24/7) vs 5.9% (PPI-WH); P=0.6 (log-rank test) (figure 4). As shown in table 1, the distribution of the types of implant-related complications was also similar in the 2 periods. Out of a total of 178 patients receiving the implant outside working hours, 6 developed complications (cumulative incidence at 180 days: 3.9%): 5 lead dislocations and 1 mild hematoma.

Figure 4.

Cumulative incidence of procedure-related complications at 180 days. Kaplan-Meier plot showing that the safety profile of implantation was similar for the 2 approaches under study. PPI-24/7, permanent pacemaker implantation 24hours a day, 7 days a week; PPI-WH, permanent pacemaker implantation during working hours.

(0.2MB).

As shown in table 1, the average number of hospital stays was significantly lower with the PPI-24/7 approach, with a reduction of 2 hospital stays per patient (nonparametric P <.001). In addition, the percentage of participants requiring admission to the cardiovascular intensive care unit (66%) and the number of hours of hospital stays were reduced in PPI-24/7 (table 1). On the other hand, the average number of hospital stays per patient increased by 2 in those with preimplant complications: 3 (IQR [1-7] vs 1 [0-2]; nonparametric P <.001). This was due not only to a delay in implantation but also because the frequency of same-day discharge was significantly lower for these patients: 25% vs 62% (P <.001).

The cumulative survival at 180 days was similar in PPI-WH (96.8%) and PPI-24/7 (95.4%) (figure 5A). However, approximately 3 out of 4 participants died of noncardiovascular causes. Consequently, cardiovascular mortality at 180 days was as low as 1% (figure 5B). Table 1 summarizes the causes of death.

Figure 5.

Mortality and cardiovascular mortality. Kaplan-Meier plots depicting cumulative survival free from death (left) and cardiovascular death (right). No differences for either variable were found between the approaches under study.

(0.23MB).

The average cost per patient was €3237 (PPI-WH) vs €2597 (PPI-24/7); P=.04. Therefore, the cost per patient in the PPI-24/7 was 20% lower, corresponding to an average saving of €640 (95% confidence interval: 497-739) per patient. PPI-24/7 was cost-effective, with an ICE of−457 euros (table 1).

Although several studies have shown that the strategy of same-day discharge after invasive electrophysiology procedures is feasible, both in the settings of catheter ablation15,16 and device implants,5 many hospitals have not yet implemented this strategy in routine practice. In a recent European Heart Rhythm Association survey, only 50% of hospitals (preferably high-volume tertiary centers) used same-day discharge after pacemaker implantation. Although same-day discharge after device implantation is safe4,5 and cost-effective,5 usually leading to patient satisfaction, it does not translate into a clinical benefit.17 On the other hand, most of the morbidity (often of noncardiac origin) in patients with ASB occurs in the time interval between diagnosis and PPI. In a previous and retrospective study, 32% of patients experienced at least 1 adverse event during this waiting period.6 Generally, patients experiencing preimplant complications are not suitable for early discharge. This may explain why, according to our data, the feasibility of same-day discharge is significantly reduced by almost 50% when comparing PPI-24/7 with PPI-WH. Finally, the implant of transvenous temporary pacemakers is associated with longer hospital stays18 and a higher frequency of complications,18–20 again in a time-dependent manner. Our experience with the systematic PPI-24/7 program has allowed us to avoid the implantation of temporary pacemakers in unstable patients and the morbidity that could be associated with their use.

Several authors have reported that in participants with acute bradyarrhythmia, there is a positive correlation between waiting time and PPI.6,20,21 Risgaard et al.6 propose that delaying implantation for more than 24hours is associated with a higher incidence of complications. Our data support this proposal, even with waiting times that are less than 24hours, such as in the PPI-WH approach (average of 16). Moreover, an additional 13-hour reduction is associated with a significant decrease (by 80%) in the incidence of preimplantation complications. These preimplant complications, beyond their direct clinical consequences, have an additional impact on hospital management because they prolong hospitalization by 2 hospital stays per patient. Applying these data to all patients and given that PPI-24/7 reduces preimplant complications, this strategy is associated with a significant reduction of 2 hospital stays per patient, which translates into a saving of 20% in the average cost per patient.

The incidence of complications related to pacemaker implantation at 6 months ranges from 1% to 12%.3–5,22–24 The safety of this procedure is highly dependent on the operator's experience,25 the size of the center,25 and the complexity of the implant device.22 In our service, at a university hospital with trainee electrophysiologists, the cumulative 6-month incidence of implant-related complications was 6%, of which about 4% were major. These data are similar to those from international registries of routine clinical practice (ie, in the historical cohort of over 2600 patients from 6 large multicenter studies, the incidence of major complications at 6 months was 7.4%).23 Although the same-day discharge strategy is not associated with an increased incidence of complications,4 there is little information on the safety of implants performed urgently outside normal working hours. Our data indicate that urgent implantation (with a mean delay of 3hours from indication) is not associated with a higher frequency of complications21 and, when performed outside normal working hours, has an excellent safety profile with the lowest incidence of major complications (2.8%). A possible explanation is that the implants placed outside working hours were performed by experienced (> 5 years) specialists working as the first operator in the implantation of the device.

The evidence provided by our study is based on an observational analysis. Because of the lack of randomization, it is impossible to completely rule out bias in any result. In addition, the single-center design limits the generalizability of our findings. However, this is a prospective study with systematic assessment of all consecutive patients (avoiding selection bias), representing a typical population in routine clinical practice with advanced age and a high prevalence of comorbidities. In addition, our hospital is a relatively high-volume center, and the after-hours implants were performed by highly experienced implant physicians. Therefore, our results may not be generalizable and extrapolated to low-volume settings.

In addition, slightly more than half of the implanted devices were VVI. This frequency is higher than in other contemporary series of unselected patients.2 However, in our service, according to the established protocol, we use this pacing mode in patients who, even in sinus rhythm, have sporadic paroxysmal AV block, as well as in the very old (> 80 years), and/or those with advanced comorbidities and advanced AV block. In these patients, this pacing mode shows excellent clinical performance.26,27

The venous access used in most patients was the subclavian vein. Both axillary and cephalic vein access have a lower risk of pneumothorax. The risk of pneumothorax in subclavian access depends on the experience of the institution28 and operators.25 In our series the incidence was only 0.3% (2 patients), probably due to the relatively high volume of cases per operator in our center. On the other hand, retrospective studies have suggested that subclavian vein access may be associated with an increased risk of fracture during follow-up.29

In our institution, we do not use septal pacing/conduction system pacing in patients with normal left ventricular ejection fraction (LVEF). Although this is a promising strategy, which could prevent LVEF deterioration in participants subjected to pacing rate> 20%, there is still insufficient data to implement it systematically in patients with normal LVEF according to current European guidelines.3 If this approach to pacing were to generate enough scientific evidence to be considered by clinical practice guidelines as a priority, it should be included as such in pacemaker implantation programs, whether urgent or not.

Finally, the implementation of this service in health systems may present logistical issues, as it requires specialized personnel working outside of normal hours, as well as financial difficulties due to the extra remuneration of these staff. However, according to the prices of our public system, the PPI-24/7 service is associated with a significant 20% reduction in the average cost per patient.