2021 Volume 85 Issue 8 Pages 1341-1348
Background: Although patients with poor ability to perform activities of daily living, such as those with high Clinical Frailty Score (CFS), will often receive a cardiac implantable electric device (CIED), the indications for implantation in these patients have not been clearly defined. We investigated the association between CFS and prognosis in patients with a CIED.
Methods and Results: We retrospectively enrolled 323 consecutive patients who underwent initial device implantation (age, 77 (70–83) years; male, 181 [56%] patients; high-voltage device, 49 [15%] patients), and the CFS was retrospectively estimated. Primary outcome was all-cause death, and the secondary outcome was hospitalization due to heart failure (HF). Median CFS was 4 (3–5) points. During 2 years’ follow-up, all-cause death occurred in 32 patients (10%). Freedom from all-cause death was significantly lower in patients with a high CFS than in those with a low score (1–2 points: 100%, 3–4 points: 92.9%, 5–9 points: 77.3%, P<0.01). After adjustment for age and sex, the CFS was an independent predictor of the primary outcome (hazard ratio [HR] 2.0, 95% confidence interval [CI] 1.6–2.5, P<0.01), and of the secondary outcome (HR 1.6 [95% CI 1.2–2.0], P<0.01).
Conclusions: The CFS is an independent predictor of both death and hospitalization due to HF in patients with a CIED.
Recently, use of a cardiac implantable electric device (CIED) has become established as an effective strategy for bradycardia, ventricular tachyarrhythmia and heart failure (HF).1,2 However, life expectancy in CIED recipients with comorbidities is poorer than in healthy populations and the incidence of HF is higher.3–5 Indeed, HF is a major cause of death and hospitalization in these patients.5
The aging of populations worldwide has been accompanied by a marked increase in the use of CIEDs.6,7 However, CIED usage in aged patients is generally performed without regard to performance of activities of daily living (ADL), and thus recipients often include patients with poor ability to perform ADLs, such as those with a high Clinical Frailty Score (CFS).3 A comprehensive understanding of prognosis in these patients would facilitate decision making on the indications for a CIED, and improve post-implantation management and outcomes. In the case of patients aged ≥75 years, a previous study showed that daily physical activity time, which was determined by pacemakers after implantation, was associated with mortality.7
To our knowledge, however, no study has investigated both life expectancy and the frequency of hospitalization due to HF after several types of CIED in these patients, and data in support of decision making are unavailable.
Here, as part of efforts to establish indications for CIED treatment in aged patients with low ADL status, we investigated the association between CFS and prognosis in patients with a CIED.
The retrospective study was conducted in 323 consecutive patients who underwent CIED implantation between January 2014 and December 2018. CIEDs were defined as pacemakers and high-voltage devices, such as implantable cardioverter defibrillators, cardiac resynchronization therapy defibrillators, and cardiac resynchronization therapy pacemakers. Exclusion criteria were implantation of a leadless pacemaker and prior pacemaker implantation.
We collected data on comorbidities, laboratory findings, and echocardiography before the implantation procedure. Periprocedural ADLs were assessed using the clinical frailty scale,8 and the CFS was retrospectively estimated from the medical records. The albumin to globulin ratio (A/G ratio) was calculated as a nutritional and inflammatory marker using serum albumin and total protein levels, excluding 3 patients who did not have data for either. The cutoff A/G ratio was set at 1.2, which is the lower limit of normal. On the basis of a Japanese guideline, we set cutoff values for B-type natriuretic peptide (BNP) and N-terminal pro-BNP of 100 pg/mL and 400 pg/mL, respectively.9
We also collected data on intrinsic or paced QRS duration on the 12-lead ECG. Similarly to a previous study, the intrinsic QRS was defined as the width of the escape rhythm before CIED implantation or, if not available, the width of non-paced ventricular rhythm closest to the CIED implantation. Paced QRS was defined as paced ventricular rhythm closest to the CIED implantation.10
Rates of mortality and hospitalization due to HF were followed for 2 years. In accordance with a previous study, the pacing percentage of the right ventricle was obtained at the end of follow-up, censored to an earlier date if events occurred, and then the cutoff value for pacing percentage of the right ventricle was set at 40%.11,12
This study complied with the Declaration of Helsinki. Written informed consent for CIED implantation and participation in the study were given by all patients, and the protocol was approved by the Kansai Rosai Hospital Institutional Review Board.
Case Follow-upLong-term outcomes were followed for 24 months; 5 patients were lost to follow-up, which was performed at routine visits, usually every 6 months. Patients who could not visit the hospital were contacted by a physician and asked about their symptoms. Follow-up of the surgical wound was performed at 1 week after discharge; pacemaker check, ECG and chest X-ray were performed at 1 month after discharge; and pacemaker check and ECG were performed every 6 months thereafter.
OutcomesThe primary outcome measure was all-cause death, and the secondary outcome measure was hospitalization due to HF. Similarly to a previous study, hospitalization due to HF was defined as acute and symptomatic HF that required hospitalization.13 Symptomatic HF was diagnosed based on the Framingham criteria.14
Statistical AnalysisContinuous data are expressed as the mean±standard deviation or the median (interquartile range). Categorical data are presented as absolute values and percentages. In accordance with a previous study, patients were classified into 3 categories using the clinical frailty scale.15 Tests for significance were conducted using the one-way analysis of variance or the Kruskal-Wallis test for continuous variables, and the chi-squared test for categorical variables. Unadjusted and adjusted Cox proportional hazards regression analyses were used to investigate the association between the CFS and the primary or secondary outcome. The CFS and variables with a P value ≤0.05 in the unadjusted analysis were adjusted for age and sex in the adjusted analysis. Kaplan-Meier analysis and the log-rank test were performed to investigate the association between outcomes and risk factors. All analyses were performed using commercial software (SPSSTM, Chicago IL, USA).
Patient characteristics are shown in Table 1. CIED implantation was successfully completed in all patients, consisting of a pacemaker in 274 (84%); implantable cardioverter defibrillator in 34 (11%); cardiac resynchronization therapy defibrillator in 11 (3%); and cardiac resynchronization therapy pacemaker in 4 (1%). Characteristics of the pacemaker patients are shown in Table 2, and those of the high-voltage device patients are shown in Table 3. Regarding implantable cardioverter defibrillator and cardiac resynchronization therapy defibrillator patients, 20 (44%) underwent device implantation for primary prevention and 25 (56%) for secondary prevention.
| Variable | Clinical Frailty Score | P value | ||
|---|---|---|---|---|
| 1–2 points (n=16) |
3–4 points (n=217) |
5–9 points (n=90) |
||
| Age (years) | 51 (38–65) | 76 (71–82) | 81 (76–85) | <0.01 |
| Male, n (%) | 12 (75) | 122 (56) | 47 (52) | 0.24 |
| Body mass index (kg/m2) | 23±4 | 23±4 | 22±4 | <0.01 |
| Dementia, n (%) | 0 (0) | 3 (1) | 16 (18) | <0.01 |
| High-voltage device, n (%) | 9 (56) | 31 (14) | 9 (10) | <0.01 |
| Atrial tachyarrhythmia, n (%) | 3 (19) | 65 (30) | 29 (32) | 0.56 |
| Temporary cardiac pacing, n (%) | 2 (13) | 52 (24) | 38 (42) | <0.01 |
| Dual-chamber pacemaker, n (%) | 13 (81) | 200 (92) | 67 (74) | <0.01 |
| Diabetes mellitus, n (%) | 1 (6) | 45 (21) | 28 (31) | 0.04 |
| Hemodialysis, n (%) | 0 (0) | 9 (4) | 14 (16) | <0.01 |
| Hypertension, n (%) | 2 (13) | 133 (61) | 46 (51) | <0.01 |
| Stroke, n (%) | 0 (0) | 10 (5) | 15 (17) | <0.01 |
| Coronary artery disease, n (%) | 1 (6) | 63 (29) | 39 (43) | <0.01 |
| Prior hospitalization due to HF, n (%) | 1 (6) | 31 (14) | 16 (18) | 0.45 |
| Antiplatelet agent, n (%) | 2 (13) | 80 (37) | 38 (42) | 0.08 |
| Oral anticoagulant, n (%) | 1 (6) | 57 (26) | 26 (29) | 0.16 |
| Steroids, n (%) | 1 (6) | 8 (4) | 5 (6) | 0.71 |
| Immunosuppressant, n (%) | 0 (0) | 3 (1) | 2 (2) | 0.76 |
| Hemoglobin (g/dL) | 14.4±1.2 | 12.4±1.7 | 11.2±1.9 | <0.01 |
| eGFR (mL/min/1.73 m2) | 76 (60–86) | 53 (40–65) | 41 (19–58) | <0.01 |
| A/G ratio | 1.7±0.2 | 1.3±0.3 | 1.1±0.3 | <0.01 |
| BNP (pg/mL)† | 119 (16–164) | 128 (58–346) | 243 (115–646) | <0.01 |
| NT-proBNP (pg/mL)† | 236 (39–236) | 884 (222–1,781) | 1,438 (592–8,272) | 0.04 |
| C-reactive protein (mg/dL) | 0.1 (0.1–0.3) | 0.1 (0.1–0.4) | 0.5 (0.1–1.4) | <0.01 |
| LVEF (%) | 63±10 | 62±15 | 59±14 | 0.47 |
| Left ventricular mass index (g/m2) | 99 (84–111) | 98 (84–115) | 106 (88–132) | 0.11 |
| Left atrial diameter (mm) | 37±8 | 40±8 | 40±8 | 0.34 |
Values indicate mean±standard deviation or median (1st–3rd quartile). †Patients on hemodialysis were excluded. A/G ratio, albumin to globulin ratio; BNP, B-type natriuretic peptide; eGFR, estimated glomerular filtration rate; HF, heart failure; LVEF, left ventricular ejection fraction; NT-proBNP, N-terminal pro-BNP.
| Variable | Clinical Frailty Score | P value | ||
|---|---|---|---|---|
| 1–2 points (n=7) |
3–4 points (n=186) |
5–9 points (n=81) |
||
| Age (years) | 54 (34–83) | 77 (72–82) | 81 (77–86) | <0.01 |
| Male, n (%) | 4 (57) | 96 (52) | 41 (51) | 0.94 |
| Body mass index (kg/m2) | 21±2 | 24±4 | 22±4 | <0.01 |
| Dementia, n (%) | 0 (0) | 3 (2) | 16 (20) | <0.01 |
| Baseline heart disease | ||||
| Sick sinus syndrome, n (%) | 2 (29) | 69 (37) | 35 (43) | 0.55 |
| Atrioventricular block, n (%) | 5 (71) | 107 (58) | 41 (51) | 0.41 |
| Atrial fibrillation with bradycardia, n (%) | 0 (0) | 10 (5) | 5 (6) | 0.79 |
| Atrial tachyarrhythmia, n (%) | 1 (14) | 53 (29) | 26 (32) | 0.57 |
| Right ventricular pacing ≥40%, n (%) | 5 (83) | 101 (56) | 47 (64) | 0.20 |
| Temporary cardiac pacing, n (%) | 2 (29) | 51 (27) | 37 (46) | 0.01 |
| Dual-chamber pacemaker, n (%) | 6 (86) | 170 (91) | 58 (72) | <0.01 |
| Diabetes mellitus, n (%) | 0 (0) | 37 (20) | 23 (28) | 0.11 |
| Hemodialysis, n (%) | 0 (0) | 8 (4) | 14 (17) | <0.01 |
| Hypertension, n (%) | 0 (0) | 123 (66) | 42 (52) | <0.01 |
| Stroke, n (%) | 0 (0) | 8 (4) | 14 (17) | <0.01 |
| Coronary artery disease, n (%) | 0 (0) | 47 (25) | 33 (41) | <0.01 |
| Prior hospitalization due to HF, n (%) | 0 (0) | 18 (10) | 8 (10) | 0.69 |
| Antiplatelet agent, n (%) | 0 (0) | 65 (35) | 32 (40) | 0.11 |
| Oral anticoagulant, n (%) | 0 (0) | 44 (24) | 23 (28) | 0.22 |
| Steroids, n (%) | 0 (0) | 6 (3) | 5 (6) | 0.46 |
| Immunosuppressant, n (%) | 0 (0) | 3 (2) | 2 (3) | 0.83 |
| Hemoglobin (g/dL) | 14.3±1.4 | 12.4±1.8 | 11.0±1.8 | <0.01 |
| eGFR (mL/min/1.73 m2) | 81 (70–93) | 53 (40–65) | 41 (16–59) | <0.01 |
| A/G ratio | 1.7±0.2 | 1.3±0.3 | 1.1±0.3 | <0.01 |
| BNP (pg/mL)† | 92 (15–163) | 94 (81–109) | 222 (110–629) | <0.01 |
| NT-proBNP (pg/mL)† | 432 (432–432) | 872 (199–1,600) | 1,438 (592–8,272) | 0.09 |
| C-reactive protein (mg/dL) | 0.1 (0.1–0.1) | 0.1 (0.1–0.3) | 0.4 (0.1–1.5) | <0.01 |
| LVEF (%) | 67±8 | 66±10 | 62±12 | 0.02 |
| Left ventricular mass index (g/m2) | 92 (71–105) | 94 (81–110) | 102 (84–131) | 0.06 |
| Left atrial diameter (mm) | 35±11 | 40±8 | 40±7 | 0.40 |
Values indicate mean±standard deviation or median (1st–3rd quartile). †Patients on hemodialysis were excluded. Abbreviations as in Table 1.
| Variable | Clinical Frailty Score | P value | ||
|---|---|---|---|---|
| 1–2 points (n=9) |
3–4 points (n=31) |
5–9 points (n=9) |
||
| Age (years) | 44 (38–60) | 68 (63–76) | 69 (67–74) | <0.01 |
| Male, n (%) | 8 (89) | 26 (84) | 6 (67) | 0.41 |
| Body mass index (kg/m2) | 25±3 | 23±4 | 22±3 | 0.11 |
| Dementia, n (%) | 0 (0) | 0 (0) | 0 (0) | – |
| Atrial tachyarrhythmia, n (%) | 2 (22) | 12 (39) | 3 (33) | 0.66 |
| Temporary cardiac pacing, n (%) | 0 (0) | 1 (3) | 1 (11) | 0.46 |
| Dual-chamber pacemaker, n (%) | 7 (78) | 30 (97) | 9 (100) | 0.08 |
| Diabetes mellitus, n (%) | 1 (11) | 9 (26) | 5 (56) | 0.10 |
| Hemodialysis, n (%) | 0 (0) | 1 (3) | 0 (0) | 0.74 |
| Hypertension, n (%) | 2 (22) | 10 (32) | 4 (44) | 0.60 |
| Stroke, n (%) | 0 (0) | 2 (7) | 1 (11) | 0.61 |
| Coronary artery disease, n (%) | 1 (11) | 16 (52) | 6 (67) | 0.04 |
| Prior hospitalization due to HF, n (%) | 1 (11) | 13 (42) | 9 (89) | <0.01 |
| Antiplatelet agent, n (%) | 2 (22) | 15 (48) | 6 (67) | 0.16 |
| Oral anticoagulant, n (%) | 1 (11) | 13 (42) | 3 (33) | 0.23 |
| Steroids, n (%) | 1 (11) | 2 (7) | 0 (0) | 0.61 |
| Immunosuppressant, n (%) | 0 (0) | 0 (0) | 0 (0) | – |
| Hemoglobin (g/dL) | 14.5±1.1 | 12.5±1.7 | 12.0±2.2 | <0.01 |
| eGFR (mL/min/1.73 m2) | 70±17 | 52±25 | 44±20 | 0.054 |
| A/G ratio | 1.7±0.2 | 1.2±0.3 | 1.1±0.3 | <0.01 |
| BNP (pg/mL)† | 130 (19–177) | 345 (111–699) | 352 (185–1,132) | 0.49 |
| NT-proBNP (pg/mL)† | 39 (39–39) | 1,786 (871–5,533) | – | 0.16 |
| C-reactive protein (mg/dL) | 0.3±0.3 | 0.3±0.3 | 0.8±0.7 | 0.02 |
| LVEF (%) | 60±11 | 35±16 | 40±18 | <0.01 |
| Left ventricular mass index (g/m2) | 99 (89–112) | 136 (104–181) | 124 (114–192) | 0.02 |
| Left atrial diameter (mm) | 38±6 | 43±8 | 46±8 | 0.12 |
Values indicate mean±standard deviation or median (1st–3rd quartile). †Patients on hemodialysis were excluded. Abbreviations as in Table 1.
A total of 48 patients (15%) had been hospitalized previously for HF, 19 (40%) with a preserved ejection fraction (LVEF ≥50%) and 28 (58%) with a reduced EF (LVEF <50%); 1 patient with prior hospitalization due to HF did not undergo echocardiography before the procedure. The presumed cause of HF was ischemic cardiomyopathy (14 patients), valvular heart disease (13 patients), dilated cardiomyopathy (8 patients), pulmonary artery hypertension (2 patients), bradycardia (2 patients), ventricular arrhythmia (2 patients), and others (7 patients).
Of the 323 patients in this study, the median CFS was 4 (3–5) points. Duration of post-procedure hospitalization was longer in patients with a high CFS than in those with a low score (1–2 points: 7 [7–8] days, 3-4 points: 7 [6–8] days, 5–9 points: 8 [7–17] days, P<0.01).
In contrast, intrinsic QRS duration (1–2 points: 119±26 ms, 3–4 points: 121±28 ms, 5–9 points: 126±29 ms, P=0.42) and paced QRS duration (1–2 points: 159±13 ms, 3–4 points: 157±18 ms, 5-9 points: 162±22 ms, P=0.16) were similar among the 3 groups. In addition, mean total procedure time, mean wave amplitude and mean pacing threshold were also similar in the 3 groups.
Predictors of All-Cause Death and Hospitalization Due to HFWe assessed clinical risk factors for all-cause death by unadjusted and adjusted analysis (Table 4). In the unadjusted analysis, patients with the primary outcome had a lower body mass index, lower hemoglobin, lower estimated glomerular filtration rate, lower A/G ratio, higher CFS, higher BNP, higher C-reactive protein, higher left ventricular mass index, higher left atrial diameter, and lower LVEF than those without. In addition, patients meeting the primary outcome had a higher prevalence of diabetes mellitus, temporary cardiac pacing and coronary artery disease. After adjustment for age and sex, CFS was an independent predictor of all-cause death (Table 4).
| Variable | Non-survivor (n=32) |
Survivor (n=291) |
Unadjusted | Adjusted‡ | ||
|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | |||
| Age (years) | 77 (73–82) | 77 (70–83) | 1.02 (0.98–1.1) | 0.28 | – | – |
| Male, n (%) | 21 (66) | 160 (55) | 1.6 (0.8–3.3) | 0.23 | – | – |
| Body mass index (kg/m2) | 20 (19–24) | 23 (20–25) | 0.9 (0.8–0.95) | <0.01 | 0.85 (0.8–0.9) | <0.01 |
| Clinical Frailty Score, points | 6 (4–7) | 4 (3–4) | 1.9 (1.5–2.4) | <0.01 | 2.0 (1.6–2.5) | <0.01 |
| Temporary cardiac pacing, n (%) | 13 (41) | 79 (27) | 2.1 (1.01–4.2) | 0.046 | 2.1 (1.03–4.3) | 0.04 |
| Diabetes mellitus, n (%) | 12 (38) | 62 (21) | 2.1 (1.03–4.3) | 0.04 | 2.0 (0.95–4.0) | 0.07 |
| Coronary artery disease, n (%) | 17 (53) | 86 (30) | 2.6 (1.3–5.3) | <0.01 | 1.7 (0.8–3.4) | 0.16 |
| Hemoglobin (g/dL) | 10.8±1.9 | 12.3±1.8 | 0.7 (0.5–0.8) | <0.01 | 0.7 (0.5–0.8) | <0.01 |
| eGFR (mL/min/1.73 m2) | 2.9 (1.3–5.0) | 5.2 (3.9–6.6) | 0.7 (0.6–0.8) | <0.01 | 0.97 (0.95–0.99) | <0.01 |
| A/G ratio | 9.9±2.6 | 12.8±3.0 | 0.7 (0.6–0.8) | <0.01 | – | – |
| A/G ratio <1.2 | 28 (88) | 121 (42) | 9.7 (3.4–28) | <0.01 | 9.0 (3.3–27) | <0.01 |
| BNP (pg/mL)† | 6.8 (1.9–11.0) | 1.5 (0.6–3.6) | 1.2 (1.03–1.22) | <0.01 | 1.1 (1.03–1.2) | <0.01 |
| C-reactive protein (mg/dL) | 0.7 (0.2–1.5) | 0.1 (0.1–0.4) | 1.6 (1.4–1.8) | <0.01 | 1.5 (1.3–1.8) | <0.01 |
| LVEF (%) | 6.1 (3.9–6.8) | 6.5 (5.6–7.2) | 0.8 (0.6–0.9) | 0.01 | 0.7 (0.6–0.9) | <0.01 |
| LVMI (g/m2) | 1.1 (1.0–1.3) | 1.0 (0.8–1.2) | 2.9 (1.2–6.6) | 0.01 | 2.9 (1.2–7.1) | 0.02 |
| LAd (mm) | 4.3 (3.8–4.9) | 3.9 (3.5–4.3) | 1.5 (1.1–2.2) | 0.02 | 1.5 (1.05–2.1) | 0.03 |
Values indicate mean±standard deviation or median (1st–3rd quartile). †Patients on hemodialysis were excluded. ‡Variables adjusted for age and sex. A/G ratio, albumin to globulin ratio in 0.1 increments; BNP, B-type natriuretic peptide in 100-pg/mL increments; CI, confidence interval; eGFR, estimated glomerular filtration rate in 10-mL/min/1.73 m2 increments; HR, hazard ratio; LAd, left atrial diameter in 10-mm increments; LVEF, left ventricular ejection fraction in 10% increments; LVMI, left ventricular mass index in 100-g/m2 increments.
In the unadjusted analysis, patients with hospitalization due to HF had a lower A/G ratio, higher CFS, higher BNP, higher N-terminal pro-BNP, higher left ventricular mass index, higher left atrial diameter, and lower LVEF than those without. In addition, patients requiring hospitalization due to HF had a higher prevalence of high-voltage device implantation, atrial tachyarrhythmia, prior hospitalization due to HF, coronary artery disease and antiplatelet agent usage. After adjustment for age and sex, CFS was an independent predictor of hospitalization due to HF (Table 5).
| Variable | With hospitalization due to HF (n=35) |
Without hospitalization due to HF (n=288) |
Unadjusted | Adjusted‡ | ||
|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | |||
| Age (years) | 77 (67–82) | 77 (71–83) | 1.01 (0.98–1.04) | 0.62 | – | – |
| Male, n (%) | 19 (54) | 162 (56) | 0.97 (0.5–1.9) | 0.93 | – | – |
| Clinical Frailty Score, points | 4 (4–7) | 4 (3–5) | 1.5 (1.2–1.9) | <0.01 | 1.6 (1.2–2.0) | <0.01 |
| High-voltage device, n (%) | 12 (34) | 37 (13) | 3.0 (1.5–6.1) | <0.01 | 5.3 (2.3–12) | <0.01 |
| Atrial tachyarrhythmia, n (%) | 16 (46) | 81 (28) | 2.1 (1.1–4.1) | 0.03 | 2.1 (1.1–4.1) | 0.03 |
| Coronary artery disease, n (%) | 18 (51) | 85 (30) | 2.6 (1.3–5.0) | <0.01 | 2.8 (1.4–5.5) | <0.01 |
| Prior hospitalization due to HF, n (%) | 16 (46) | 32 (11) | 5.5 (2.9–11) | <0.01 | 6.7 (3.3–13) | <0.01 |
| Antiplatelet agent, n (%) | 19 (54) | 101 (35) | 2.1 (1.1–4.1) | 0.03 | 2.2 (1.1–4.5) | 0.02 |
| A/G ratio | 11±3 | 13±3 | 0.8 (0.7–0.9) | <0.01 | – | – |
| A/G ratio <1.2, n (%) | 23 (66) | 126 (44) | 2.7 (1.3–5.4) | <0.01 | 2.7 (1.3–5.4) | <0.01 |
| BNP (pg/mL)† | 3.1 (1.3–9.0) | 1.4 (0.6–3.6) | 1.1 (1.04–1.2) | 0.01 | 1.1 (1.04–1.2) | <0.01 |
| NT-proBNP (pg/mL)† | 2.0 (1.2–8.6) | 0.9 (0.2–1.5) | 1.1 (1.001–1.3) | 0.04 | 1.1 (1.004–1.3) | 0.04 |
| Elevated BNP or NT-proBNP, n (%)† | 30 (86) | 172 (63) | 3.3 (1.3–8.4) | 0.02 | 3.2 (1.2–8.3) | 0.02 |
| LVEF (%) | 5.4 (3.7–6.3) | 6.6 (5.7–7.2) | 0.7 (0.6–0.8) | <0.01 | 0.6 (0.5–0.7) | <0.01 |
| LVMI (g/m2) | 1.1 (0.9–1.4) | 1.0 (0.8–1.2) | 1.1 (1.05–1.2) | <0.01 | 3.8 (1.7–8.4) | <0.01 |
| LAd (mm) | 4.3 (3.6–4.9) | 3.9 (3.5–4.3) | 1.6 (1.1–2.1) | <0.01 | 1.6 (1.1–2.2) | <0.01 |
Values indicate mean±standard deviation or median (1st–3rd quartile). †Patients on hemodialysis were excluded. ‡Variables adjusted for age and sex. BNP, B-type natriuretic peptide in 100-pg/mL increments (elevated, ≥100 pg/mL); NT-proBNP, N-terminal pro-BNP in 1,000-pg/mL increments (elevated, ≥400 pg/mL). Other abbreviations as in Table 5.
Freedom from all-cause death during the 24-month follow-up period was significantly lower in patients with a high CFS than in those with a low score (Figure). In contrast, there was no significant difference in freedom from all-cause death during the 24-month follow-up period between patients with a high-voltage device and those with a pacemaker (88.6% vs. 89.1%, P=0.98).
Freedom from all-cause death and hospitalization due to heart failure (HFa) in patients classified by clinical frailty scale. (A) Freedom from all-cause death during the 24-month follow-up period was significantly lower in patients with a high Clinical Frailty Score (CFSd) than in those with a low score. (B) Freedom from hospitalization due to HF during the 24-month follow-up period was significantly lower in patients with a high CFS than in those with a low score. aHeart failure, bhazard ratio, cconfidence interval, dClinical Frailty Score.
Outcomes are detailed in Table 6. Causes of death were HF (8 patients), cancer (5 patients), pneumonia (4 patients), acute myocardial infarction (3 patients), sepsis (2 patients), intestinal necrosis (2 patients), ventricular tachycardia (1 patient), infective endocarditis (1 patient), renal failure (1 patient), critical limb ischemia (1 patient), gastrointestinal bleeding (1 patient), unknown cause of cardiac death (1 patient), and other causes of non-cardiac death (2 patients). Fatal cases of HF (1–2 points: 0 [0%], 3–4 points: 3 [1%], 5–9 points: 5 [6%], P=0.052) and sepsis (1–2 points: 0 [0%], 3–4 points: 0 [0%], 5–9 points: 2 [2%], P=0.06) tended to be more common in patients with a high CFS than in those with a low score.
| Variable | Clinical Frailty Score | P value | ||
|---|---|---|---|---|
| 1–2 points (n=16) |
3–4 points (n=217) |
5–9 points (n=90) |
||
| All-cause death, n (%) | 0 (0) | 13 (6) | 19 (21) | <0.01 |
| Cardiac death, n (%) | 0 (0) | 5 (2) | 9 (10) | <0.01 |
| Non-cardiac death, n (%) | 0 (0) | 8 (4) | 10 (11) | <0.01 |
| Hospitalization due to HF, n (%) | 0 (0) | 19 (9) | 16 (18) | <0.01 |
HF, heart failure.
Freedom from hospitalization due to HF during the 24-month follow-up period was significantly lower in patients with a high CFS than in those with a low score (Figure). Freedom from hospitalization due to HF during the 24-month follow-up was also significantly lower in patients with a high-voltage device than in those with a pacemaker (74.0% vs. 90.2%, P<0.01). In patients with hospitalization due to HF, LVEF was decreased at hospitalization (baseline, 51±18%; at hospitalization, 45±19%, P<0.01).
In this retrospective study, we investigated differences in all-cause death and hospitalization due to HF rate among 323 patients undergoing CIED implantation.
Our main findings were as follows: (1) freedom from all-cause death during the 24-month follow-up period was significantly lower in patients with a high CFS than in those with a low score; (2) freedom from hospitalization due to HF a during the 24-month follow-up period was significantly lower in patients with a high CFS than in those with a low score; and (3) after adjustment for age and sex, the CFS was an independent predictor of all-cause death.
CFS and All-Cause DeathAfter adjustment for age and sex, the CFS was an independent predictor of the primary outcome.
In general, performance of ADLs predicts life prognosis in the elderly.16–18 Previous studies have shown that low ADL status is a predictor of poor life prognosis and impaired wound healing in patients with peripheral artery disease.19,20 Possible explanations for the high mortality rates of patients with a high CFS might be: (1) restricted ability to visit hospital, and so do not receive comprehensive treatment; (2) some conditions with high mortality, such as cerebrovascular disease and HF, also cause muscle weakness, which leads to low ADL status.21,22 Furthermore, the incidence and mortality rates of some fatal illnesses, such as aspiration pneumonia, sepsis and HF, are increased in patients with low ADL status.23–25 Our study also showed more fatal cases of HF and sepsis in patients with a high CFS than in those with a low score.
CFS and Hospitalization Due to HFWith regard to hospitalization due to HF, CFS was an independent predictor after adjustment for age and sex.
Possible explanations for the many hospitalizations due to HF in patients with a high CFS might be: (1) patients with low ADL status tend to have poor medication adherence; and (2) patients with HF and low ADL status are unable to benefit from the antiremodeling effect on the left ventricle and the reduction of plasma levels of proinflammatory cytokines of exercise.26–28
Clinical ImplicationsThere are several potential clinical implications. Life expectancy of CIED recipients with comorbidities is poorer than that of healthy populations, and HF frequently occurs in patients with CIED implantation.3–5 Assessment of the CFS is simple and easy, and is useful for follow-up after CIED implantation.
Study LimitationsSeveral limitations warrant mention. First, we were unable to assess cardiovascular events after CIED implantation, such as de novo coronary artery disease and atrial tachyarrhythmia. Second, we were unable to assess mild HF that did not require hospitalization. Third, the pacemaker implantation procedure might have varied at the discretion of the individual operators, even though we attempted to standardize procedures as far as possible. Fourth, the retrospective nature of the study meant there were some differences in patient characteristics among the 3 groups. Fifth, the CFS was retrospectively estimated from the medical records and thus was not necessarily precise. Finally, the size of the study population, high-voltage device implantations and number of events were relatively small, weakening the statistical analysis.
In patients with CIED implantation, a high CFS was an independent predictor of death and hospitalization due to HF.
We thank Guy Harris DO from DMC Corp. (http://www.dmed.co.jp/) for editing drafts of this manuscript.
Participant data are not available for sharing.
None.
Kansai Rosai Hospital Institutional Review Board. Reference number: 2001010.
