Numerical results are expressed as mean (standard deviation) or median [interquartile range], depending on the distribution of the variable. Continuous quantitative variables were compared with analysis of variance or the Kruskal-Wallis test. Qualitative variables were compared with the chi-square test. The relationships between the number of implantations and the number of implantation centers per million population and between the total number of implantations and the number of implantations for primary prevention in each center were evaluated using linear regression models.

A total of 153 centers performing ICD implantation in Spain participated in the registry (8 fewer than in 2012) (Table 1). This decrease was due to the grouping of several centers into consortia, which provided pooled data. Of the respondents, 90 were public health care centers (103 in 2012). Figure 1 shows the total number of participating centers, the implantation rate per million population, and the total number of implantations by autonomous community according to data collected by the registry. In 2013, only 14 centers implanted > 100 devices; 76 < 10 and, of these, 26 implanted only 1.

Figure 1.

Distribution implantation activity in 2013 by autonomous community: number of implantation centers/implantation rate per million population/total number of implantations.

(0.35MB).

In total, 4772 first and replacement implantations were performed in 2013, more than in 2012 (4216). According to EUCOMED data,14 this figure represents 85.6% of the total of 5573 ICD implantations carried out in 2013 in Spain. Figure 2 shows the total number of implantations reported to the registry and the number estimated by EUCOMED over the last 11 years.

Figure 2.

Total number of implantations reported to the registry and total number estimated by the European Medical Technology Industry Association (2003-2013).

(0.18MB).

The implantation rate recorded in the registry was 102,1 per million population, while the rate according to the EUCOMED data was 120 per million population. Figure 3 shows the changes occurring in the implantation rate per million population over the last 11 years according to the registry and EUCOMED data. Table 1 shows the number of implantations reported to the registry by each participating center. Table 2 lists the number of implantations performed in each province and the rate per million population for the corresponding autonomous community.

Figure 3.

Total number of implantations reported to the registry per million population and as estimated by the European Medical Technology Industry Association (2003-2013).

(0.16MB).

The name of the participating center was reported in 99% of the cases. Most procedures, 4398 (92%), were performed in public centers.

This information was reported in 4334 data forms sent to the SEC (91%). There were 2981 first implantations, representing 68.8% of the total (69.4% in 2012; 70.2% in 2011 and 73.8% in 2010). The rate of first implantations in 2013 was 63.8 per million population (64 per million population in 2012, equalling that in 2011).

The mean age (standard deviation) [range] of patients receiving an ICD or replacement device was 62.6 (13.4) [7-90] years. In first implantations, the mean age was 61.3 (13.4) years. Most of the patients were men, representing 83.3% of all patients, and 83.7% of patients receiving a first implantation.

The most common underlying cardiac condition in first implantations was ischemic heart disease (53.4%), followed by dilated cardiomyopathy (28.4%), hypertrophic cardiomyopathy (7.1%), primary conduction abnormalities (Brugada syndrome, idiopathic ventricular fibrillation and long QT syndrome) (7.5%) and, in lower percentages, valvular heart disease and arrhythmogenic dysplasia (Figure 4).

Figure 4.

Type of heart disease prompting device implantation (first implantations).

(0.18MB).

Regarding left ventricular function in patients undergoing a first implantation, left ventricular ejection fraction was < 30% in 50.0%; 30%-40% in 25.6%; 41%–50% in 7.8%, and > 50% in 16.2%. The distribution was similar in patients receiving a replacement ICD (Figure 5). This information was reported in 85.5% of the data collection forms.

Figure 5.

Left ventricular ejection fraction in registry patients (all implantations and first implantations). LVEF, left ventricular ejection fraction.

(0.07MB).

Regarding New York Heart Association (NYHA) functional class, most patients were in NYHA II (47.2%), followed by NYHA III (34.3%), NYHA I (16.6%), and NYHA IV (1.9%). In this variable, the distribution between the total number of implantations and first implantations was also very similar (Figure 6). This information was reported in 75.6% of the data collection forms.

Figure 6.

New York Heart Association functional class in registry patients (all implantations and first implantations). NYHA, New York Heart Association.

(0.06MB).

The baseline cardiac rhythm, recorded in 80.5% of the patients, was mainly sinus rhythm (81.3%), followed by atrial fibrillation (16.1%) and paced rhythm (2.1%); the remaining patients had other rhythms (atrial flutter and other arrhythmias).

This information was reported in 80.7% of the data collection forms. In first implantations, the largest group consisted of patients with no documented clinical arrhythmia (50.8%), followed by those with sustained monomorphic ventricular tachycardia, nonsustained ventricular tachycardia, and ventricular fibrillation (21.3%, 14.1% and 12.3%, respectively). Among the total number of implantation procedures performed, 50.8% of devices were placed in patients without documented clinical arrhythmia (Figure 7). Differences in the type of arrhythmia between patients receiving a first implantation and the overall group of patients were statistically significant for patients without arrhythmia (P<.001) and those with sustained monomorphic ventricular tachycardia (P<.001) but not for patients with nonsustained ventricular tachycardia or ventricular fibrillation. The most common clinical presentation in the overall group and the first-implantation group (62.2% and 68.9% completed responses) was asymptomatic, followed by syncope, sudden cardiac death, and “other symptoms” (Figure 8).

Figure 7.

Distribution of arrhythmias prompting device implantation (first implantations and all implantations). NSMVT, nonsustained monomorphic ventricular tachycardia; PVT, polymorphic ventricular tachycardia; SMVT, sustained monomorphic ventricular tachycardia; VF, ventricular fibrillation. *P<.001.

(0.06MB).

Figure 8.

Clinical presentation of arrhythmia in registry patients (first implantations and all implantations). SCD, sudden cardiac death.

(0.07MB).

Information on electrophysiological studies was available in 2244 patients receiving a first implantation (86.4%). Electrophysiological studies were conducted in only 305 patients (13.6%).The most frequently induced arrhythmia was sustained monomorphic ventricular tachycardia (47.9%), followed by ventricular fibrillation (11.8%) and, to a lesser extent, nonsustained ventricular tachycardia (10.7%), and other arrhythmias (3.7%). No arrhythmia was induced in 26.1% of electrophysiological studies, which were mainly carried out in patients with ischemic heart disease and dilated cardiomyopathy.

Several new fields related to the patients’ clinical history have been incorporated in the data collection form since 2011.

For patients with first implantations, between 64.5% and 80.4% of these fields were completed, with the following results: hypertension, 58.4%; hypercholesterolemia, 50.2%; smoking, 33.2%; diabetes mellitus, 29.8%; a history of atrial fibrillation, 26.4%; kidney failure, 13.8%; a history of sudden cardiac death, 8.5%, and stroke, 6.8%.

The QRS interval was documented in 47.9% of the records (mean value 124.1 [35ms]). The QRS interval was > 140ms in 31.1% of the patients. Of these, 79.6% in the first implantation group (80.2% of the overall group) had a combined resynchronization/ICD (ICD-CRT [cardiac resynchronization therapy]) device.

Table 3 shows changes in first implantations by type of heart disease and form of presentation from 2009 to 2013. This information was reported in 93% of the data collection forms. In ischemic heart disease, the most frequent indication was primary prevention (48.8%), with no increase compared with the previous year (50.5%). The main indication in dilated cardiomyopathy was also primary prevention (52.1% vs 62.3% in 2012 and 53.4% in 2010). In less common heart diseases, the most frequent indication was primary prevention of hypertrophic cardiomyopathy and Brugada syndrome and other channelopathies. In congenital heart disease and valvular heart disease, the most frequent indication was secondary prevention.

The indication for implantation was reported in 82.1% of the data collection forms. Most first implantations were indicated for primary prevention (53.0%); this was the second year that the number of implantations decreased for this indication. This trend has been increasing and was statistically significant (P<.01) until 2008, between 2009 and 2010, and became significant again in 2013 (Table 4).

Information on these items was reported in 82.5% of the data collection forms. Implantations were mainly (79.8%) carried out in the electrophysiology laboratory (81.4% in 2012, 76.4% in 2011), followed by the operating room (18.4%). The procedure was carried out by electrophysiologists in 80.7% of implantations (81.0% in 2012, 78.4% in 2011), surgeons in 13.8% (14.0% in 2012, 15.5% in 2011), both specialists in 2.0%, and other specialists and intensivists in 2.0% and 1.6%, respectively.

Information on the generator implantation site in first implantations was reported in 2575 (28.8%) data collection forms. The generator was placed subcutaneously in 94.6% of procedures and in a subpectoral position in the remaining 5.4%. Among the total number of devices implanted, these percentages were 93.5% and 6.5%, respectively.

This information was reported in 91.4% of the data collection forms and is summarized in Table 5. Single/chamber ICDs were implanted in 48.2% (49.4% in 2012 and 46.7% in 2011). Dual-chambers were implanted in 18.9% (18.0% in 2012 and 18.4% in 2011) and ICD-CRT were implanted in 32.9% (32.5% in 2012 and 34.9% in 2011).

In patients with ischemic heart disease, 74.8% of the devices were single- or dual-chamber devices (72.3% in 2012) and 25.5% were ICD-CRT (27.7% in 2012). In patients with dilated cardiomyopathy, 51.7% of the devices were ICD-CRTs (56.5% in 2012 and 59.7% in 2011).

Among 1,353 replacement procedures, information was available for 1,035 (86.9%). The most common reason was battery depletion in 86.9%, followed by complications in 7.9% and change of indication in 5.1%. Of the 69 cases of early replacement procedures reported, 11.6% were carried out within 6 months of the implantation procedure.

In 65% of the replacement reports, information was provided on the status of the leads; 9.3% (82 records) were malfunctioning. Among cases in which malfunctioning was reported, lead extraction was performed in 46%.

Information on this ítem was reported in 73.5% of the data collection forms. The most commonly used pacing was VVI mode (56.6%), followed by DDD mode (29.3%), VVIR mode (7.0%), DDDR mode (4.9%), and other pacing modes, mainly preventive algorithms for ventricular pacing (2.2%)

Information on induction of ventricular fibrillation was reported in 3850 data collection forms and 197 patients (5.1%) underwent this procedure (6.7% in 2012). The mean threshold was 20.4 (6.5) (20.5 [7.1] in 2012) and the mean number of shocks delivered was 1.3.

Information on complications was provided in 79.4% of the data collection forms. In total, 26 complications were described: 6 coronary sinus dissections, 3 pneumothorax, 2 deaths, 2 cases of tamponade, and 13 cases of unspecified complications. The mortality rate was 0.05%, similar to that reported for the previous year (0. 08%).

The first Spanish Implantable Cardioverter-defibrillator Registry was published in 2005 and presented the results for the period 2002-2004.4 The number of ICDs implanted increased each year until 2010,5–10 and decreased in 2011 and 2012. This decrease was observed in the total number of devices implanted in Spain, both in the registry data,11,12 and in the results reported by EUCOMED. In 2013, the number of implantations increased again, exceeding the number reported to the registry in 2010. In Europe, a slight increase has been maintained, both in the number of ICD implantations and in ICD-CRTs.14

Despite the increase in the number of devices implanted, for the second consecutive year, the number of implantations indicated for primary prevention decreased compared with 201212 (53.0% vs 58.1%).

The percentage of ICD-CRT implantations reached a plateau (32.9% vs 32.5% in 2012), as did the percentage of single-chamber ICD implantations (48.2% vs 49.4% in 2012). There was practically no change in the use of dual-chamber ICDs (18.9% vs 18.0%). The rate of ICD-CRT implantation seems to have become stabilized within ICD therapy, and no major changes are expected, unless produced by potential new indications. These data may also be modified in future registries by the arrival of subcutaneous ICDs.

The most common indication for ICD implantation continued to be ischemic heart disease (53.4%), followed by dilated cardiomyopathy (28.2%). As in previous years,12 more than half of the implantations in patients with dilated cardiomyopathy were ICD-CRTs (51.6%), whereas the percentage was lower in patients with ischemic heart disease (25.1%).

The progressive increase in the number of ICD implantations came to a halt in 2011 and 2012. The results of 2013 show a certain recovery: the total number of implantations slightly exceeded the rate per million population for 2010 (102 vs 100).15 In the last few years, no new studies have been published that would modify the indications for ICD implantation. The Multicenter Automatic Defibrillator Implantation Trial II16 was published in 2002, and the Comparison of Medical Therapy, Pacing and Defibrillation in Heart Failure trial17 and the Sudden Cardiac Death in Heart Failure Trial18 in 2005 and 2006, respectively, which laid down the current indications in primary prevention and cardiac resynchronization and led to a progressive rise in the number of implantations during that decade. The indications for ICD and CRT implantation are well supported by clinical practice guidelines.19–22 However, the implantation rate per million population does not correspond to the rate that could be expected from the clinical evidence, either in Spain or in neighboring countries,23 and this tendency has become more marked over time, both in Spain and elsewhere in Europe.

As in previous registries, there are differences in the number of implantations recorded in the present registry and the EUCOMED data, although these differences decreased in 2013 (85.0% vs 80.4% in 2012), despite a slight reduction in the number of participating centers. Most hospitals implanting ICDs send data to the registry, but the rate has not yet reached 100% of implanting centers. Moreover, some data are lost while being sent and processed. All these reasons may explain the differences between the present registry and the EUCOMED data.

The number of participating centers decreased slightly compared with 2012, mainly due to the grouping of several hospitals, which submitted pooled data. None of the hospitals reported an implantation rate > 200; 14 hospitals (8 in 2012 and 11 in 2011) reported > 100 implantations, and 72 centers—mostly private centers—reported < 10 implantations. Some studies show a relationship between implantation volume and the number of complications,24 with a higher implantation volume being associated with a lower complications rate.

There were no significant changes in the patients’ epidemiologic characteristics compared with previous registries. Patients with severe ventricular dysfunction and those in NYHA functional class II or III continued to predominate and a plateau was reached in the growing tendency in the number of implantations carried out in the electrophysiology laboratory (79.8% vs 81.4% in 2012 and 76.4% in 2011) and in those performed by electrophysiologists (80.7% vs 81% in 2012 and 78.4% in 2011).

The differences in ICD implantation rates among the various autonomous communities held steady. The implantation rate was 102 per million population according to the registry and was 120 per million population according to EUCOMED data, with both figures showing an increase compared with 2012 (91.2 and 113.0, respectively). Several autonomous communities showed rates above the average: Principality of Asturias (186), Extremadura (156), Chartered Community of Navarre (137), Valencian Community (127), Galicia (122), Castile and León (122) and Cantabria (105). The autonomous communities with activity below the average included the Community of Madrid (99), Aragon (94), Region of Murcia (91), the Basque Country(85), Castile-La-Mancha (84), Catalonia (84), Andalusia (82), The Canary Islands (77) and the Balearic Islands (71). The difference between the communities with the highest and lowest implantation rates is currently more than double and has increased compared with previous registries (186 vs 71). In general, the percentage of implantations per million population increased in practically all the autonomous communities and decreased only in the Canary Islands.

The data for 2013 show a widespread increase in Spain, which was more marked in the autonomous communities with ICD implantation rates above the average. No association was found between the gross domestic product of each region and the number of implantations. Curiously, most of the regions with higher per capita incomes had below-average implantation rates. The regions with higher-than-average rates were the least densely populated, except for the Valencian Community. Equally, there was no association between implantation rates and the incidence of ischemic heart disease and heart failure in the distinct regions. These differences could be explained by other reasons, such as the organization of the health system in each region, the number of arrhythmia units, and the distribution of referrals.

The implantation rate in countries participating in the EUCOMED was 289 per million population (273 in 2012), including ICD and ICD-CRT. Germany, with 541 devices, continued to be the country with the highest number of implantations. Spain (120 implantations per million) was the country with the lowest number of implantations. The countries with above-average implantation rates were the Netherlands (388), Italy (381), Denmark (322), and the Czech Republic (351). Countries with below-average implantation rates were Poland (276), Austria (253), Ireland (250), Belgium (237), Sweden (223), Norway (219), France (202), Switzerland (196), Finland (182), the United Kingdom (164), Portugal (148), Greece (141), and Spain (120). The difference in the implantation rate in Spain compared with the mean continued to increase (120 vs 289 in 2013 and 113 vs 269 in 2012). Likewise, the difference between Spain and the next-ranking country also increased (120 vs 141).

The ICD implantation rate was 176 per million population in 2013 (167 in 2012). The country with the highest number of implantations was Germany (331) and that with the lowest was Spain (80).

The ICD-CRT implantation rate was 113 per million population (106 in 2012). The rate continued to be highest in Germany (210 implantations) and lowest in Spain (40).

The proportion of ICD-CRT with respect to the total number of implantations ranged from 26% in Poland and 45% in Italy. The percentages were above 40% in the Czech Republic, France, Switzerland, Ireland, and Italy and were <30% in Norway and Poland. The proportion was in Spain was 33%.

The 2013 data show that implantation rates increased in Europe. The differences in Spain compared with the European mean continued to increase.

Other European countries showed the same regional differences25,26 found in our registry and there is no clear explanation for this finding. A possible explanation involves the number of available arrhythmia units but the relationship is far from clear, at least in Spain, where the regions with the highest number of available units had the lowest implantation rates. Other explanations, such as per capita income, also fail to elucidate these differences; the ICD implantation rate was much higher in countries such as Ireland, the Czech Republic and Poland than in Spain. Other factors that could be related to the implantation rate and its variability in Spain are the prevalence of cardiovascular diseases, access to the health system and its organization, and the degree of acceptance and adherence to clinical practice guidelines.

According to the EUCOMED data, the registry included 85.6% of the implantations performed in Spain. This percentage is higher than those for 2012 (80.8%) and 2011 (83.6%). The number of implantations reported to the registry decreased after 2007, when 90% of implantations were notified, but has risen again and the number reported is still representative of the situation in Spain. The number of participating centers has slightly declined in the last 2 years, mainly because amalgamation among hospitals led to data pooling.

In some hospitals, the true number of implantations differed from the number notified. Some data collection forms were not received or properly registered, since data can be sent and received in several ways. We continue to hope that next year's data will be gathered through the online data collection form, which could improve the results and narrow the gap between the data received and those provided by EUCOMED.

There was excessive variability in the response rate to distinct items on the ICD data collection form, oscillating between 99.0% for the item on the implanting hospital to 47.9% for that on QRS width. Lastly, the percentage of complications reported to the registry does not reflect the true situation, given that the data were completed during or immediately after implantation. Consequently, there was no record of most subacute complications.

The current registry is the X official report of this activity in Spain. All members of the Electrophysiology and Arrhythmias Section of the SEC who participate in this registry can feel proud of its continued existence. Its progressive modernization will allow more and better information to be collected with less effort on the part of all the staff involved in its maintenance. Its future computerization may allow the inclusion of more ambitious clinical objectives and parameters such as mortality, shocks delivered, complications, etc., which will provide important clinical information and represent a substantial improvement.