論文ID: CJ-23-0103
Background: Because the penetration of transvenous lead extraction (TLE) for cardiac implantable electronic device (CIED) infection has not been investigated in Japan, we conducted a population-based, retrospective, descriptive study to evaluate regional disparities in the use of TLE for CIED infection and the potential undertreatment of CIED infection using a nationwide insurance claims database.
Methods and Results: Patients who underwent CIED implantation or generator exchange and TLE between April 2018 and March 2020 were identified. Moreover, the penetration ratio of TLE for CIED infection in each prefecture was estimated. CIED implantation and TLE were most prevalent in the age categories of 80–89 years (40.3%) and 80–89 years (36.9%), respectively. There was no correlation between the number of CIED implantations and that of TLE (rho=−0.087, 95% confidence interval −0.374 to 0.211, P=0.56). The median penetration ratio was 0.00 (interquartile range 0.00–1.29). Of the 47 prefectures, 6, comprising Okinawa, Miyagi, Okayama, Fukuoka, Tokyo, and Osaka, showed a penetration ratio ≥2.00.
Conclusions: Our study data indicated great regional disparities in the penetration of TLE and potential undertreatment of CIED infection in Japan. Additional measures are needed to address these issues.
Transvenous lead extraction (TLE) is performed for various infectious and noninfectious indications. In particular, it is a potentially life-saving procedure for patients with cardiac implantable electronic device (CIED) infection. In fact, the 1-year mortality rate of CIED infection reached 38.2%,1 which decreased to 8.5–17% when TLE was properly performed.2,3 Powered sheaths, such as laser sheaths, were first introduced into the Japanese market in July 2010, and the first national practice guideline in March 20194,5 defined TLE as a Class I recommendation for CIED infection. Recent single-center studies have reported high clinical success rates of TLE in Japan.6–10
However, the penetration of TLE for CIED infection relative to the number of eligible patients in Japan has not been investigated in detail. TLE is not only an invasive procedure that has been linked to rare but catastrophic complications.11 but also requires qualified personnel and specific equipment.12 Therefore, we conducted this study to evaluate if there are regional disparities in the penetration of TLE and to estimate the potential undertreatment of CIED infection by calculating the penetration ratio of TLE in Japan.
We used a nationwide insurance claims database to conduct this population-based, retrospective descriptive study, which was registered in the University Hospital Medical Information Network-Clinical Trial Registry (UMIN-CTR ID: UMIN000049496) and approved by the Institutional Review Board of Sendai Kousei Hospital (approval number, 4-29; approval date, September 16, 2022). The requirement for informed consent was waived because our study used anonymized data that were provided in accordance with legal provisions. We reported according to the Reporting of studies Conducted using Observational Routinely Collected Health Data statement13 (Supplementary Table 1).
Data SourceWe used the National Database of Health Insurance Claims and Specific Health Checkups of Japan (NDB) Open Data, which was created and is maintained by the Ministry of Health, Labor and Welfare based on the Elderly Health Care Security Act.14 The NDB includes >95% of current insurance claims information, because Japan has a public medical and social insurance system15 that covers almost the entire Japanese population. The NDB Open Data of Japan, which is a fundamental table that summarizes the nationwide universal healthcare insurance claims, contains anonymized and publicly disclosed information.16 We linked the NDB Open Data with the population data of each prefecture based on the vital statistics from 2019.17 According to the NDB reporting guidelines,18 if the number of patients who underwent TLE in a prefecture was <10 per year, the number of patients for that prefecture is not disclosed to protect personal information. Therefore, the number 0 was assigned if the number of patients who underwent TLE in a prefecture was <10.
Patient SelectionPatients who underwent CIED implantation or generator exchange for the 2 years between April 2018 and March 2020 in Japan were included in this study. CIED implantation and generator exchange were identified based on the procedure codes (Supplementary Table 2). Two researchers (ST and IT) and a medical affairs division staff member searched for and reviewed the procedure codes. We assumed that <5% of the CIED and TLE procedure data were misclassified as having measurement errors, which were documented in the database but not performed in clinical practice. Currently, there have been no validation studies on the procedure data for CIED and TLE. However, based on previous validation studies on surgery and procedure data,19,20 validity was inferred to be high for invasive procedures.
OutcomesPatients who underwent TLE were identified by the procedure code K599-05 (TLE with/without laser sheath) entered in the database between April 2018 and March 2020 (Supplementary Table 2). The regional distribution of the numbers of CIED implantations and TLE per 100,000 population in the 47 prefectures was investigated. In addition, the relationship between the number of CIED implantations and that of TLE per 100,000 population was evaluated. The penetration ratio of TLE in each prefecture was estimated by the following formula: (expected number of patients who underwent TLE for CIED infection) / (expected number of patients who developed CIED infection) = (number of TLE × expected proportion of infectious indications for TLE) / (number of CIED implantation × expected incidence of CIED infection). Based on previous reports from Japan, the expected proportion of infectious indications for TLE ranged from 62.5% to 87.5%, and the expected annual incidence of CIED infection ranged from 0.6% to 1.2%.6,7,9,10,21,22 To deduce the presence of undertreatment, a best case scenario that was defined as the most active performance of TLE according to the current guideline, was calculated using a lower limit of 0.6% for the annual incidence of CIED infections (i.e., 1.2% every 2 years) and an upper limit of 87.5% for the proportion of TLE indication for CIED infection. The same annual incidence rate was applied during the study period of 2 years, because the annual incidence of CIED infection seemed to be in direct proportion to the time elapsed.23
Statistical AnalysisThe descriptive data of the patients who underwent CIED implantation and TLE during the study period were summarized by age and sex. Numbers of patients who underwent CIED implantation and TLE were expressed as respective counts per 100,000 population by prefecture. In addition, we rearranged the respective numbers of CIED implantations and TLE per 100,000 population of each prefecture in ascending order: that for CIED implantation was plotted on the x-axis, and that for TLE was plotted on the y-axis. We performed Spearman’s correlation coefficient test between the number of CIED implantations and that for TLE per 100,000 population and obtained the confidence interval (CI) of rho by 1,000 bootstrap replicates using RVAideMemoire-R-package.24 Data processing and all statistical analyses were performed using Excel for Mac (version 16.66.1, Microsoft, Redmond, WA, USA) and RStudio for Mac (version 4.0.5, R Foundation for Statistical Computing, Vienna, Austria).
Table 1 shows the baseline characteristics of the patients. CIED implantation was most prevalent in the age category of 80–89 years (40.3%), followed by 70–79 years (31.3 %) and ≥90 years (12.2%). TLE was most prevalent in the age category of 80–89 years (36.9%), followed by 70–79 years (34.0%) and 60–69 years (15.9%). The regional distributions of CIED implantation and TLE per 100,000 population in the 47 prefectures are presented in Figure 1. Regional disparities were greater for the number of TLE than for the number of CIED implantation, which showed a steep slope of regional distribution because the number of patients who underwent TLE was 0 in 29 of the 47 prefectures. As shown in the scatter plot (Figure 2), there was no correlation between the number of CIED implantations and the number of TLE (rho=−0.087, 95% CI −0.374 to 0.211, P=0.56).
CIED | TLE | |||||
---|---|---|---|---|---|---|
Overall (n=125,644), n (%) |
Male (n=64,990), n (%) |
Female (n=60,654), n (%) |
Overall (n=1,126), n (%) |
Male (n=885), n (%) |
Female (n=241), n (%) |
|
Age category (years) | ||||||
0–9 | 88 (0.1) | 46 (0.1) | 42 (0.1) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
10–19 | 141 (0.1) | 66 (0.1) | 75 (0.1) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
20–29 | 262 (0.2) | 174 (0.3) | 88 (0.1) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
30–39 | 737 (0.6) | 492 (0.8) | 245 (0.4) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
40–49 | 1,909 (1.5) | 1,333 (2.1) | 576 (0.9) | 25 (2.2) | 25 (2.8) | 0 (0.0) |
50–59 | 4,260 (3.4) | 2,919 (4.5) | 1,341 (2.2) | 47 (4.2) | 47 (5.3) | 0 (0.0) |
60–69 | 12,887 (10.3) | 8,029 (12.4) | 4,858 (8.0) | 179 (15.9) | 168 (19.0) | 11 (4.6) |
70–79 | 39,385 (31.3) | 22,270 (34.3) | 17,115 (28.2) | 383 (34.0) | 305 (34.5) | 78 (32.4) |
80–89 | 50,616 (40.3) | 24,432 (37.6) | 26,184 (43.2) | 416 (36.9) | 304 (34.4) | 112 (46.5) |
≥90 | 15,359 (12.2) | 5,229 (8.0) | 10,130 (16.7) | 76 (6.7) | 36 (4.1) | 40 (16.6) |
Note: numbers that were <10 in the original raw data were handled as 0. CIED, cardiac implantable electronic device; TLE, transvenous lead extraction.
Regional distribution per 100,000 population in the 47 prefectures of Japan from April 2018 to March 2020. (A) CIED implantation and (B) TLE. The error bar indicates 95% confidence interval. CIED, cardiac implantable electronic device; TLE, transvenous lead extraction.
Scatter plot of the relationship between the number of CIED implantations and that of TLE per 100,000 population from April 2018 to March 2020. The x-axis on the top shows a histogram of the number of CIED implanted per 100,000 population, and the y-axis on the right shows a histogram of the number of TLE per 100,000 population. CI, confidence interval; CIED, cardiac implantable electronic device; TLE, transvenous lead extraction.
Table 2 and Supplementary Table 3 show the penetration ratio of TLE in each prefecture. The median penetration ratio of TLE was 0.00 (interquartile range, 0.00–1.29). Therefore, <10 cases per year were performed in many prefectures. Notably, 6 prefectures (Okinawa, Miyagi, Okayama, Fukuoka, Tokyo, and Osaka) showed ≥2.00 penetration ratio of TLE for CIED infection. As shown in a choropleth map, there were regional disparities in the penetration ratios for the best case scenario among the 47 prefectures in Japan (Figure 3).
Prefecture | Patients who underwent CIED implantation for 2 years, n |
Patients who underwent TLE for 2 years, n |
Expected number of patients who developed CIED infection for 2 years |
Expected number of patients who underwent TLE for CIED infection for 2 years |
Penetration ratio in “best case” scenario† |
||
---|---|---|---|---|---|---|---|
Incidence: 1.2% every 2 years, n |
Incidence: 2.4% every 2 years, n |
Infectious indication: 62.8%, n |
Infectious indication: 87.5%, n |
Incidence: 1.2% every 2 years, Infectious indication: 87.5% |
|||
Okinawa | 1,229 | 34 | 7.4 | 14.7 | 21.4 | 29.8 | 4.04 |
Miyagi | 2,618 | 54 | 15.7 | 31.4 | 33.9 | 47.3 | 3.00 |
Okayama | 2,227 | 44 | 13.4 | 26.7 | 27.6 | 38.5 | 2.88 |
Fukuoka | 5,748 | 112 | 34.5 | 69.0 | 70.3 | 98.0 | 2.84 |
Tokyo | 12,407 | 215 | 74.4 | 148.9 | 135.0 | 188.1 | 2.53 |
Osaka | 8,432 | 133 | 50.6 | 101.2 | 83.5 | 116.4 | 2.30 |
Kumamoto | 2,071 | 28 | 12.4 | 24.9 | 17.6 | 24.5 | 1.97 |
Ibaraki | 2,446 | 33 | 14.7 | 29.4 | 20.7 | 28.9 | 1.97 |
Hokkaido | 8,294 | 97 | 49.8 | 99.5 | 60.9 | 84.9 | 1.71 |
Kanagawa | 8,093 | 79 | 48.6 | 97.1 | 49.6 | 69.1 | 1.42 |
Hyogo | 5,194 | 50 | 31.2 | 62.3 | 31.4 | 43.8 | 1.41 |
Aichi | 5,822 | 56 | 34.9 | 69.9 | 35.2 | 49.0 | 1.40 |
Kyoto | 2,715 | 22 | 16.3 | 32.6 | 13.8 | 19.3 | 1.18 |
Gunma | 2,059 | 14 | 12.4 | 24.7 | 8.8 | 12.3 | 0.99 |
Nagano | 2,457 | 11 | 14.7 | 29.5 | 6.9 | 9.6 | 0.65 |
Hiroshima | 3,185 | 10 | 19.1 | 38.2 | 6.3 | 8.8 | 0.46 |
Chiba | 5,380 | 16 | 32.3 | 64.6 | 10.0 | 14.0 | 0.43 |
Saitama | 5,234 | 10 | 31.4 | 62.8 | 6.3 | 8.8 | 0.28 |
All the other prefectures (29 prefectures), median (min.−max.) |
1,332 [530–3,521] |
0 | 8.0 [3.3–31.4] |
16.0 [6.7–62.8] |
N/A | N/A | N/A |
Summary in Japan, median (IQR) |
1,735 [1,225–2,666.5] |
0 [0–20] |
10.5 [7.4–16.1] |
21.0 [14.8–32.3] |
0.0 [0.0–16.7] |
0.0 [0.0–23.2] |
0.00 [0.00–1.29] |
†The penetration ratio of TLE in each prefecture was estimated by the following formula: (expected number of patients who underwent TLE for CIED infection) / (expected number of patients who developed CIED infection) = (number of TLE × expected proportion of infectious indications for TLE) / (number of CIED implantation × expected incidence of CIED infection). The best case scenario was that with the highest penetration ratio of TLE, wherein the expected proportion of infectious indications was 87.5% and the expected incidence of CIED infection was 1.2% every 2 years. Note: 0 indicates that the number of patients who underwent TLE in a prefecture was <10.
Geographic distribution of the penetration ratio of transvenous lead extraction in the 47 prefectures in Japan in the “best case” scenario, which is that with the highest penetration ratio of TLE. The expected proportion of infectious indications is 87.5% and the expected incidence of CIED infection is 1.2% every 2 years. Note that 0 indicates that the number of patients who underwent TLE in a prefecture was <10. CIED, cardiac implantable electronic device; TLE, transvenous lead extraction.
To our knowledge, this was one of the first studies to clarify the regional disparities in the penetration of TLE and to evaluate the potential undertreatment of CIED infection in Japan. The main findings of this study were that the penetration of TLE for CIED infection differed markedly among the prefectures and that there was potential undertreatment of CIED infection in Japan.
There are a few reasons for the regional disparities in the penetration of TLE for CIED infection across prefectures. TLE requires qualified personnel and special equipment, such as powered sheaths and its accessories,12 and maintaining a hybrid operating room is costly. Another reason could be that it is a lengthy process to become a certified lead extractor.12,25 Moreover, the low awareness of the practice guidelines among noncardiologist physicians in general hospitals may explain the insufficient referrals to lead extraction centers. Notably, of the 6 prefectures with a relatively high penetration ratio, Okinawa and Miyagi have regional hub centers for lead extraction. The high penetration ratio in all 6 prefectures suggested that patients with CIED infection in prefectures without a hub centers were referred to the hub centers in neighboring prefectures. When performing TLE, the balance between the case volume of the lead extraction center and the individual experience of the lead extraction operators is important, because the incidences of major complications and death are related to these factors. This volume-outcome relationship is supported by data from the European Lead Extraction ConTRolled Registry, which used 30 procedures per year as the cutoff to define low- and high-volume centers.26 We believe the same importance should be given to the operator’s individual experience. In Japan, with the exception of densely populated urban areas, 1 or 2 lead extraction centers in each prefecture would be appropriate to maintain enough cases to overcome the learning curve for TLE. Therefore, each prefecture may need to establish a system to centralize patients for TLE from regional hospitals to hub lead extraction centers.
Our study suggested that CIED infections in Japan may have been undertreated during the study period, considering that TLE is the only curative treatment.4 One of the possible reasons for the potential undertreatment of CIED infections was the low penetration of TLE for patients with persistent or recurrent bacteremia or fungemia and no other identifiable source for relapse or continued infection, which are Class I indications.4 In clinical practice, these patients are often examined by noncardiologist physicians, such as general practitioners, intensivists, and oncologists, and not by cardiologists. Therefore, cardiologists and lead extractors may need to intensify activities to raise awareness of the clinical indications among noncardiologist physicians. Another reason may be the inappropriate treatment of cutting the lead short and burying it under the pectoralis major, which is still being performed. Among patients with CIED infection, many of those who have retained materials develop recurrence, whereas the success rate of eradicating infection is reported to be high (>90%) when all the system components are removed.27 Therefore, complete removal should be the standard treatment for any CIED infection, and eradicating inappropriate techniques by raising the awareness of TLE becomes increasingly necessary.
Despite our assumption that the proportions of TLE for infectious and noninfectious indications remained unchanged for the 2-year study period, the actual number of eligible patients with noninfectious indications may have changed, depending on the patient’s preferences and the operator’s technical proficiency in performing TLE. Although the annual number of TLE for both noninfectious and infectious indications has increased,21 the increase in the number of TLE was more significant for noninfectious indications than for infectious indications. Accordingly, the proportion of TLEs for noninfectious indications has increased, whereas that for infectious indications has decreased.6,7,9,10,21 This trend of increasing TLE for noninfectious indications may be attributed to the fact that in patients with noninfectious indications, particularly those with a Class IIb indication, the decision to undergo TLE is determined by the patient’s preference and the operator’s technical proficiency in TLE.
Future DirectionIn Japan, the actual incidence of CIED infections is difficult to determine. Although previous studies have reported it for Japan,22,28 those were conducted by administering questionnaires to physicians in selected hospitals. To accurately evaluate the actual incidence and potential undertreatment of CIED infections, a national database that records the clinical and technical information of all CIED implantations and extractions in all centers may be required.
Study Strengths and LimitationsA major strength of our study was the inclusion of data from essentially all Japanese patients, suggesting little bias in the information, because the database includes almost all the administrative claims and health checkup data from all the insurers across Japan. Therefore, most of the patients who underwent CIED implantation and TLE in Japan during the study period were analyzed; this allowed us to identify the potential undertreatment of CIED infection. Another strength was the minimal misclassification of CIED implantation and TLE, because the validity of operative information in the Japanese administrative data is reported to be high.19
However, there are several potential limitations. First, the national database had no information on the actual indication for TLE. Therefore, using the NDB database, we cannot know the actual number of patients who underwent TLE for CIED infection during the study period. Second, the data on CIED and TLE were not obtained for prefectures registering <10 patients who underwent TLE per year. Third, the results of this study cannot be extrapolated to other countries. Fourth, the incidence of CIED infection was only roughly estimated, because the data on TLE could not differentiate between a high-power device, such as an implantable cardioverter defibrillator or a cardiac resynchronization therapy with defibrillator (CRT-D), and a low-power device, such as a pacemaker. Moreover, we were unable to use previous cohort data, which showed that the incidence of CIED infection was lowest in pacemakers and highest in CRT-Ds.23 Finally, in the absence of unique patient identifiers in the NDB database, some patients may have been duplicated. However, such patients were considered to represent only a minority, because performing ≥2 CIED implantation or TLE during the study period of 2 years would have been rare.
Despite the introduction of powered sheaths and the announcement of the first national practice guideline on lead extraction, the penetration of TLE for CIED infection varied widely across prefectures and thus CIED infection may be undertreated in Japan.
We thank Manami Suzuki of Sendai Kousei Hospital for helping with the research and collection of data.
None.
The authors have no conflicts of interest to declare.
The study protocol was approved by the Institutional Review Board of Sendai Kousei Hospital (approval number, 4-29; approval date, September 16, 2022).
Please find supplementary file(s);
https://doi.org/10.1253/circj.CJ-23-0103