BACKGROUND:
Although chronological age is associated with mortality and morbidity after surgery for unruptured cerebral aneurysms (UCAs), there is little evidence regarding an association between the simplified 5-factor modified frailty index (mFI-5) and postoperative outcomes based on age group.
OBJECTIVE:
To investigate the association of the mFI-5 score with worse outcomes, mortality, and complications in patients after surgery for UCA by chronological age groups using a Japanese national database.
METHODS:
This study included 32 902 patients with UCAs enrolled in a Japanese national database between 2011 and 2015. Age group (younger than 65 years, 65-74 years, and 75 years or older), sex, UCA location, treatment, medications, Barthel Index (BI), medical history, mFI-5 score, and in-hospital mortality and complications were evaluated. We identified risk factors for worsening BI score, in-hospital mortality, and overall postoperative complications in each age group.
RESULTS:
In total, 14 465 patients were enrolled in this study. Multivariable analysis showed that elderly groups and patients with an mFI-5 score ≥2 items were associated with worsening BI scores (odds ratio 1.95; 95% CI 1.52-2.51) and in-hospital complications (odds ratio 1.79; CI 1.49-2.15), despite having no association with in-hospital mortality. Multivariable analysis by age groups showed that the mFI-5 score ≥2 items was associated with in-hospital complications in all age groups, unlike chronological age in patients younger than 74 years.
CONCLUSION:
The mFI-5 score was a more useful associated factor of in-hospital complications than chronological age in patients younger than 74 years undergoing surgery for UCA.
KEY WORDS: Elderly patient, Functional outcome, Nationwide database, Unruptured cerebral aneurysm, 5-Factor modified frailty index
ABBREVIATIONS:
- ACA
anterior cerebral artery
- ACoA
anterior communicating artery
- BA
basilar artery
- BI
Barthel index
- MCA
middle cerebral artery
- mFI-5
5-factor modified frailty index
- UCA
unruptured cerebral aneurysm
- VA
vertebral artery.
Japan reportedly has the highest percentage of elderly people and the longest average life expectancy in the world.1 Therefore, research about aging is appropriate in Japan. With increasing life expectancy, asymptomatic unruptured cerebral aneurysms (UCAs) have been detected in elderly persons worldwide through neuroimaging techniques such as brain magnetic resonance imaging. UCAs are 2.8 times more likely to rupture in Japanese individuals than in Europeans and Americans,2 and older age is reportedly a risk factor for UCA rupture.3 Therefore, in Japan, many elderly patients older than 65 years receive surgical treatment for UCAs. Owing to the wide variance between the decline in physical and cognitive functions associated with aging, there is dissociation between the chronological and biological ages. Although both the chronological and biological ages contribute to postoperative mortality and in-hospital complications in patients with UCA, their usefulness when examined by age groups is not known.4-6 The treatment risk in patients with UCAs is related to aging and the presence of comorbidities, indicating that careful follow-up is an alternative in older patients (older than 65 years) (class IIa; level of evidence B); however, the exact indications for surgery in elderly patients with UCAs are hitherto unclear.7
Currently, the 5-factor modified frailty index (mFI-5), which is more convenient than the 11-factor modified frailty index,8-10 has been introduced in various areas of surgery and contains the following 5 factors: hypertension, diabetes mellitus history, functional dependence prevalence, congestive heart failure, and chronic obstructive pulmonary disease history.11-14 Complementary to chronological age, these frailty indexes reflect biological age. Few reports have examined the usefulness of the mFI-5 score for its association with activities of daily living deterioration, in-hospital complications, and mortality in patients after UCA surgery.15 It is unclear whether chronological or biological age could be a more appropriate indicator for UCA surgery outcomes based on age groups. Therefore, the purpose of this study was to investigate and compare the associations of the mFI-5 score and chronological age with worsening outcomes, in-hospital complications, and mortality after UCA surgery by age groups in Japan.
METHODS
Ethical Statements
We obtained approval from the Institutional Review Board of our university before study initiation. Because all data in this study were anonymous, informed consent from each of the eligible patients was substituted by the opt-out method.
Data Source
We conducted our study using the diagnosis procedure combination (DPC) database, a nationwide database in Japan,16 which includes data on patient admission and discharge; other studies have explained this database in detail.5,6,17 All data were coded according to the International Classification of Diseases and Related Health Problems, Tenth Revision. The validity of the DPC in-patient database is well established, and both sensitivity and specificity of the procedure records have been reported to be greater than 90%.18 Details of the DPC database are presented in Text 1 of Supplemental Digital Content 1, http://links.lww.com/NEU/D429.
Selection of Patients
In total, 32 902 patients treated for UCAs between January 2011 and March 2015 were enrolled in this study. The diagnosis of UCAs was identified as International Classification of Diseases and Related Health Problems, Tenth Revision, code I671. In this study, we included initially admitted patients and those 18 years or older who underwent a single treatment with surgical clipping or endovascular coiling (Figure). The treatment selection procedure is presented in Text 2 of Supplemental Digital Content 1, http://links.lww.com/NEU/D429.
FIGURE.
Selection process for patients with unruptured cerebral aneurysms.
The data examined included patient age, sex, treatment method (clipping or coiling), comorbidities, medications at admission, aneurysm location, mFI-5 score, hospital volume, Barthel index (BI) score at admission and discharge, and in-hospital complications. Moreover, aneurysm location was examined and classified as follows: internal carotid artery, anterior communicating artery, middle cerebral artery, anterior cerebral artery, basilar artery (BA), vertebral artery (VA), and others. The specific comorbidities considered were diabetes mellitus, hypertension, dyslipidemia, stroke, angina pectoris, and congestive heart failure. Regarding medications taken at the time of hospitalization, antiplatelet agents, anticoagulants, and statins were recorded. Complications such as intracerebral hemorrhage, cerebral infarction, subarachnoid hemorrhage (SAH), congestive heart failure, and pneumonia were specifically examined.
The BI score was used to evaluate activities of daily livings (Text 3 of Supplemental Digital Content 1, http://links.lww.com/NEU/D429). The patients were classified into 3 age groups based on the classifications of the World Health Organization and the Japanese Geriatrics Society: younger than 65 years (nonelderly), 65 to 74 years (pre-elderly), and 75 years or older (elderly).19 Details of hospital information are presented in Text 4 of Supplemental Digital Content 1, http://links.lww.com/NEU/D429. Finally, mFI-5 scores were classified into 3 categories: 0, 1 item, and ≥2 items. A worsening BI score was defined as a decrease in the BI score of 5 or more points at discharge compared with that at admission, and in-hospital mortality was defined as death from any cause. In-hospital complications due to surgery included stroke, congestive heart failure, pneumonia, worsening BI scores, and in-hospital mortality.
Statistical Analysis
Categorical variables were compared using the Fisher exact test or chi-square test. Continuous variables were compared using the t test or Mann-Whitney U test. Multivariable logistic regression analysis was used to reduce the potential sources of bias and examine risk factors for worsening BI scores, in-hospital mortality, and in-hospital complications from admission to discharge. The same analysis was performed separately for worsening BI scores from admission to discharge and for in-hospital complications in the 3 age groups. However, in-hospital mortality was excluded from the analysis because of the small number of in-hospital mortality patients. Only 0.26% of study data were missing because we previously deleted the missing data in the patient selection shown in Figure. In the logistic regression analysis, independent variables including sex, treatment, age, aneurysm location, mFI-5 score, internal oral medication on admission, and hospital volume were selected based on variables in the literature, and odds ratios (ORs) and 95% CIs were calculated.11-14,20 Stata version 15 (StataCorp, College Station) was used to perform all the statistical analyses. A P value < .05 was considered to indicate statistical significance.
Data Availability
Only the results of the anonymized data for this study will be shared on request from any qualified investigator to the corresponding author.
RESULTS
Ultimately, 14 465 patients were included in this study (Figure), with a mean age of 64.0 (range: 56.0-71.0) years. Surgical clipping and endovascular coiling procedures were performed in 9082 and 5383 patients, respectively. Table 1 presents the baseline characteristics and outcomes of patients with UCAs treated with surgery by age group. The mFI-5 score increased significantly with age, as did a worsening BI score, and in-hospital mortality and complications. The incidences of postoperative complications such as intracerebral hemorrhage, SAH, congestive heart failure, and pneumonia did not increase significantly with age. Furthermore, 0.26% of all patients had missing values of BI scores at admission and discharge (Table 1); however, there were no missing data other than BI scores.
TABLE 1.
Baseline Characteristics of Patients With Surgically Treated UCAs, Based on Age Groups
| Total | Nonelderly, younger than 65 y | Pre-elderly, 65-74 y | Elderly, 75 y or older | ||
|---|---|---|---|---|---|
| No. | 14 465 | 7380 | 5364 | 1721 | P value |
| Sex (male, %) | 4380/14 465 (30.3%) | 2578/7380 (34.9%) | 1445/5364 (26.9%) | 357/1721 (20.7%) | <.001a |
| Age (y), median (IQR) | 64.0 (56.0, 71.0) | 56.0 (49.0, 61.0) | 69.0 (67.0, 72.0) | 77.0/1721 (76.0, 80.0) | <.001a |
| Treatment | <.001a | ||||
| Surgical clipping | 9082/14 465 (62.8%) | 4550/7380 (61.7%) | 3598/5364 (67.1%) | 934/1721 (54.3%) | |
| Endovascular coiling | 5383/14 465 (37.2%) | 2830/7380 (38.3%) | 1766/5364 (32.9%) | 787/1721 (45.7%) | |
| Location of the aneurysms | <.001a | ||||
| ICA | 5543/14 465 (38.3%) | 3009/7380 (40.8%) | 1776/5364 (33.1%) | 758/1721 (44.0%) | |
| ACoA | 2328/14 465 (16.1%) | 1157/7380 (15.7%) | 921/5364 (17.2%) | 250/1721 (14.5%) | |
| MCA | 4536/14 465 (31.4%) | 2176/7380 (29.5%) | 1885/5364 (35.1%) | 475/1721 (27.6%) | |
| ACA | 713/14 465 (4.9%) | 321/7380 (4.3%) | 308/5364 (5.7%) | 84/1721 (4.9%) | |
| BA | 812/14 465 (5.6%) | 371/7380 (5.0%) | 335/5364 (6.2%) | 106/1721 (6.2%) | |
| VA | 477/14 465 (3.3%) | 310/7380 (4.2%) | 126/5364 (2.3%) | 41/1721 (2.4%) | |
| Other | 56/14 465 (0.4%) | 36/7380 (0.5%) | 13/5364 (0.2%) | 7/1721 (0.4%) | |
| Medical history | |||||
| Diabetes mellitus | 1297/14 465 (9.0%) | 504/7380 (6.8%) | 601/5364 (11.2%) | 192/1721 (11.2%) | <.001a |
| Hypertension | 5987/14 465 (41.4%) | 2680/7380 (36.3%) | 2447/5364 (45.6%) | 860/1721 (50.0%) | <.001a |
| Cerebral infarction | 1500/14 465 (10.4%) | 769/7380 (10.4%) | 539/5364 (10.0%) | 192/1721 (11.2%) | .411 |
| Angina pectoris | 527/14 465 (3.6%) | 179/7380 (2.4%) | 240/5364 (4.5%) | 108/1721 (6.3%) | <.001a |
| Congestive heart failure | 176/14 465 (1.2%) | 57/7380 (0.8%) | 79/5364 (1.5%) | 40/1721 (2.3%) | <.001a |
| Myocardial infarction | 11/14 465 (0.07%) | 5/7380 (0.07%) | 3/5364 (0.06%) | 3/1721 (0.2%) | .282 |
| Dyslipidemia | 2644/14 465 (18.3%) | 1119/7380 (15.2%) | 1153/5364 (21.5%) | 372/1721 (21.6%) | <.001a |
| mFI-5 score | <.001a | ||||
| 0 item | 7457/14 465 (51.6%) | 4292/7380 (58.2%) | 2483/5364 (46.3%) | 682/1721 (39.6%) | |
| 1 item | 5921/14 465 (40.9%) | 2702/7380 (36.6%) | 2408/5364 (44.9%) | 811/1721 (47.1%) | |
| ≥2 items | 1087/14 465 (7.5%) | 386/7380 (5.2%) | 473/5364 (8.8%) | 228/1721 (13.2%) | |
| Internal oral medication on admission | |||||
| Antiplatelet | 5340/14 465 (36.9%) | 2756/7380 (37.3%) | 1847/5364 (34.4%) | 737/1721 (42.8%) | <.001a |
| Anticoagulant | 280/14 465 (1.9%) | 79/7380 (1.1%) | 133/5364 (2.5%) | 68/1721 (4.0%) | <.001a |
| Statin | 1377/14 465 (9.5%) | 570/7380 (7.7%) | 581/5364 (10.8%) | 226/1721 (13.1%) | <.001a |
| Hospital volume | <.001a | ||||
| 1 | 4651/14 465 (32.2%) | 2267/7380 (30.7%) | 1815/5364 (33.8%) | 569/1721 (33.1%) | |
| 2 | 4767/14 465 (33.0%) | 2439/7380 (33.0%) | 1707/5364 (31.8%) | 621/1721 (36.1%) | |
| 3 | 5047/14 465 (34.9%) | 2674/7380 (36.2%) | 1842/5364 (34.3%) | 531/1721 (30.9%) | |
| Barthel index at admission (points) | <.001a | ||||
| 0-95 | 929/14 443 (6.4%) | 273/7367 (3.7%) | 373/5356 (7.0%) | 283/1720 (16.5%) | |
| 100 | 13 514/14 443 (93.6%) | 7094/7367 (96.3%) | 4983/5356 (93.0%) | 1437/1720 (83.5%) | |
| Barthel index at discharge (points) | <.001a | ||||
| 0-95 | 894/14 428 (6.2%) | 263/7368 (3.6%) | 357/5348 (6.7%) | 274/1712 (16.0%) | |
| 100 | 13 534/14 428 (93.8%) | 7105/7368 (96.4%) | 4991/5348 (93.3%) | 1438/1712 (84.0%) | |
| Complications | |||||
| Intracerebral hemorrhage | 36/14 465 (0.2%) | 16/7380 (0.2%) | 15/5364 (0.3%) | 5/1721 (0.3%) | .730 |
| Cerebral infarction | 867/14 465 (6.0%) | 424/7380 (5.7%) | 329/5364 (6.1%) | 114/1721 (6.6%) | .330 |
| Subarachnoid hemorrhage | 0/14 465 (0%) | 0/7380 (0%) | 0/5364 (0%) | 3/1721 (0.2%) | |
| Congestive heart failure | 67/14 465 (0.5%) | 31/7380 (0.4%) | 27/5364 (0.5%) | 9/1721 (0.5%) | .730 |
| Pneumonia | 265/14 465 (1.8%) | 127/7380 (1.7%) | 97/5364 (1.8%) | 41/1721 (2.4%) | .181 |
| Worsening BI scores | 631/14 465 (4.4%) | 184/7380 (2.5%) | 263/5364 (4.9%) | 184/1721 (10.7%) | <.001a |
| In-hospital mortality | 40/14 465 (0.3%) | 13/7380 (0.2%) | 17/5364 (0.3%) | 10/1721 (0.6%) | .012a |
| In-hospital complications | 1648/14 465 (11.4%) | 709/7380 (9.6%) | 638/5364 (11.9%) | 301/1721 (17.5%) | <.001a |
ACA, anterior cerebral artery; ACoA, anterior communicating artery; BA, basilar artery; BI, Barthel index; ICA, internal carotid artery; MCA, middle cerebral artery; mFI-5, 5-factor modified frailty index; UCAs, unruptured cerebral aneurysms; VA, vertebral artery.
P < .05.
Table 2 presents the results of the multivariable logistic regression analysis of risk factors for worsening BI scores, in-hospital mortality, and in-hospital complications. The significant risk factors for worsening BI scores were the pre-elderly group (OR 1.84, 95% CI 1.52-2.25), elderly group (4.33, 3.47-5.39), mFI-5 score ≥2 items (1.95, 1.52-2.51), the use of anticoagulant and antiplatelet medications at admission, and aneurysm location (BA). However, endovascular coiling had a statistically inverse relationship as a risk factor compared with surgical clipping (0.41, 0.31-0.54). The elderly age group (3.26, 1.38-7.71), male sex (1.95, 1.00-3.78), and aneurysm location (BA and VA) were significant risk factors for in-hospital mortality. Concerning in-hospital complications, the significant risk factors were the pre-elderly group (1.21, 1.08-1.36), elderly group (1.76, 1.51-2.06), mFI-5 score of 1 item (1.31, 1.17-1.47), mFI-5 score ≥2 items (1.79, 1.49-2.15), anticoagulant and antiplatelet medication use, statin use, and aneurysm location (middle cerebral artery, BA, and VA). Conversely, endovascular coiling had a statistically inverse relationship as a risk factor compared with surgical clipping (0.52, 0.44-0.62).
TABLE 2.
Results of Multivariate Logistic Regression Analyses of Worsening BI Scores, In-Hospital Mortality, and In-Hospital Complications in All Cases
| Objective variables (no.) | Worsening BI scores (594/144 407a) | In-hospital mortality (40/13 696a) | In-hospital complications (1648/14 465a) | |||
|---|---|---|---|---|---|---|
| OR (95%CI) | P value | OR (95%CI) | P value | OR (95%CI) | P value | |
| Sex (male) | 1.10 (0.92-1.33) | .270 | 1.95 (1.00-3.78) | .049b | 1.06 (0.94-1.19) | .375 |
| Treatment (endovascular coiling) | 0.41 (0.31-0.54) | <.001b | 2.21 (0.85-5.70) | .102 | 0.52 (0.44-0.62) | <.001b |
| Age group (y) | ||||||
| Nonelderly, younger than 65 | Reference | Reference | Reference | |||
| Pre-elderly, 65-74 | 1.84 (1.52-2.25) | <.001b | 1.88 (0.90-3.95) | .094 | 1.21 (1.08-1.36) | .001b |
| Elderly, 75 or older | 4.33 (3.47-5.39) | <.001b | 3.26 (1.38-7.71) | .007b | 1.76 (1.51-2.06) | <.001b |
| Location of the aneurysms | ||||||
| ICA | Reference | Reference | Reference | |||
| ACoA | 1.15 (0.89-1.48) | .263 | 1.80 (0.76-4.28) | .185 | 0.89 (0.75-1.05) | .155 |
| MCA | 0.91 (0.73-1.14) | .403 | 0.41 (0.12-1.36) | .145 | 0.74 (0.64-0.86) | <.001b |
| ACA | 0.78 (0.51-1.19) | .252 | N/A | 0.66 (0.50-0.88) | .004b | |
| BA | 2.11 (1.54-2.90) | <.001b | 3.42 (1.37-8.53) | .008b | 1.43 (1.17-1.75) | <.001b |
| VA | 1.54 (0.98-2.43) | .061 | 3.37 (1.05-10.83) | .041b | 1.50 (1.17-1.75) | .002b |
| Other | 3.76 (1.54-9.18) | .004b | N/A | 1.86 (0.94-3.60) | .064 | |
| mFI-5 score | ||||||
| 0 item | Reference | Reference | Reference | |||
| 1 item | 1.13 (0.95-1.36) | .164 | 1.64 (0.82-3.28) | .161 | 1.31 (1.17-1.47) | <.001b |
| ≥2 items | 1.95 (1.52-2.51) | <.001b | 2.32 (0.86-6.26) | .095 | 1.79 (1.49-2.15) | <.001b |
| Internal oral medication on admission | ||||||
| Antiplatelet | 1.67 (1.31-2.12) | <.001b | 0.55 (0.23-1.31) | .177 | 2.91 (2.47-3.40) | <.001b |
| Anticoagulant | 2.01 (1.37-2.96) | <.001b | 1.55 (0.36-6.72) | .557 | 1.84 (1.37-2.47) | <.001b |
| Statin | 1.24 (0.98-1.58) | .072 | 0.64 (0.20-2.14) | .478 | 1.22 (1.04-1.44) | .013b |
| Hospital volume | ||||||
| 1 | Reference | Reference | Reference | |||
| 2 | 0.89 (0.74-1.07) | .218 | 1.34 (0.67-2.69) | .410 | 0.88 (0.78-1.00) | .043b |
| 3 | 0.59 (0.48-0.73) | <.001b | 0.53 (0.21-1.31) | .170 | 0.71 (0.63-0.81) | <.001b |
ACA, anterior cerebral artery; ACoA, anterior communicating artery; BA, basilar artery; BI, Barthel index; ICA, internal carotid artery; MCA, middle cerebral artery; mFI-5, 5-factor modified frailty index; OR, odds ratio; VA, vertebral artery.
Complete cases without missing data.
P < .05.
Table 3 presents the results of multivariable logistic regression analysis of worsening BI scores in all age groups. Chronological age was a significant risk factor for worsening BI scores in all 3 age groups. An mFI-5 score ≥2 items was a statistically significant risk factor for worsening BI scores in all age groups (nonelderly [1.78, 1.05-3.02], pre-elderly [1.76, 1.20-2.60], and elderly [2.00, 1.28-3.14]). In all age groups, high-volume hospitals were inversely associated with worsening BI scores.
TABLE 3.
Results of Multivariate Logistic Regression Analyses of Worsening BI Scores Across All Age Groups
| Worsening BI scores | ||||||
|---|---|---|---|---|---|---|
| Age group (no.) | Nonelderly, younger than 65 y (172/7356a) | Pre-elderly, 65-74 y (247/5340a) | Elderly, 75 y or older (175/1711a) | |||
| OR (95%CI) | P value | OR (95%CI) | P value | OR (95%CI) | P value | |
| Sex (male) | 1.24 (0.90-1.69) | .187 | 1.26 (0.95-1.66) | .102 | 0.77 (0.51-1.17) | .222 |
| Treatment (endovascular coiling) | 0.53 (0.30-0.94) | .029b | 0.25 (0.16-0.38) | <.001b | 0.53 (0.33-0.85) | .008b |
| Age | 1.02 (1.00-1.04) | .046b | 1.40 (1.00-1.09) | .047b | 1.11 (1.06-1.17) | <.001b |
| Location of the aneurysms | ||||||
| ICA | Reference | Reference | Reference | |||
| ACoA | 1.35 (0.86-2.13) | .195 | 0.83 (0.56-1.23) | .349 | 1.69 (1.05-2.72) | .031b |
| MCA | 1.01 (0.67-1.54) | .950 | 0.71 (0.51-0.99) | .044b | 1.20 (0.78-1.85) | .395 |
| ACA | 0.91 (0.41-2.06) | .829 | 0.59 (0.31-1.13) | .111 | 0.93 (0.42-3.81) | .862 |
| BA | 2.36 (1.32-4.24) | .004b | 2.22 (1.37-3.58) | .001b | 1.91 (1.01-3.62) | .048b |
| VA | 1.90 (0.96-3.76) | .065 | 1.52 (0.71-3.25) | .285 | 1.28 (0.43-3.81) | .660 |
| Other | 7.64 (2.55-22.87) | <.001b | 1.85 (0.23-15.00) | .563 | 2.05 (0.23-18.63) | .524 |
| mFI-5 score | ||||||
| 0 item | Reference | Reference | Reference | |||
| 1 item | 1.15 (0.84-1.58) | .387 | 1.05 (0.79-1.38) | .747 | 1.13 (0.79-1.62) | .501 |
| ≥2 items | 1.78 (1.05-3.02) | .032b | 1.76 (1.20-2.60) | .004b | 2.00 (1.28-3.14) | .002b |
| Internal oral medication on admission | ||||||
| Antiplatelet | 1.32 (0.78-2.23) | .302 | 2.14 (1.51-3.04) | <.001b | 1.48 (0.98-2.25) | .064 |
| Anticoagulant | 1.94 (0.76-4.94) | .167 | 2.48 (1.44-4.27) | .001b | 1.63 (0.84-3.16) | .151 |
| Statin | 1.42 (0.90-2.23) | .135 | 0.90 (0.60-1.33) | .594 | 1.73 (1.15-2.60) | .008b |
| Hospital volume | ||||||
| 1 | Reference | Reference | Reference | |||
| 2 | 1.01 (0.72-1.41) | .961 | 0.88 (0.65-1.18) | .395 | 0.77 (0.54-1.11) | .162 |
| 3 | 0.54 (0.36-0.80) | .002b | 0.62 (0.46-0.85) | .003b | 0.62 (0.41-0.92) | .018b |
ACA, anterior cerebral artery; ACoA, anterior communicating artery; BA, basilar artery; BI, Barthel index; ICA, internal carotid artery; MCA, middle cerebral artery; mFI-5, 5-factor modified frailty index; OR, odds ratio; VA, vertebral artery.
Complete cases without missing data.
P < .05.
Table 4 presents the results of multivariable logistic regression analysis for in-hospital complications in the 3 age groups. Chronological age was a significant risk factor only in the elderly group, and there was no significant difference in the age group younger than 74 years. On the contrary, an mFI-5 score ≥2 items was a statistically significant risk factor of in-hospital complications in all 3 age groups (nonelderly [1.66, 1.20-2.28], pre-elderly [1.72, 1.30-2.29], and elderly [1.99, 1.36-2.93]). In addition, in the nonelderly and pre-elderly groups, an mFI-5 score of 1 item was associated with in-hospital complications. The use of oral anticoagulants, antiplatelet agents, or statins at the time of admission was a significant risk factor for in-hospital complications in all age groups. By contrast, treatment with endovascular coiling was inversely associated with the occurrence of in-hospital complications in all age groups. Furthermore, the high hospital volume was inversely associated with in-hospital complications. Although chronological age was not associated with the occurrence of in-hospital complications for patients younger than 74 years, an mFI-5 score ≥2 items was a statistically significant risk factor in all age groups.
TABLE 4.
Results of Multivariate Logistic Regression Analyses of In-Hospital Complications Across All Age Groups
| In-hospital complications | ||||||
|---|---|---|---|---|---|---|
| Age group (no.) | Nonelderly, younger than 65 y (709/7380a) | Pre-elderly, 65-74 y (638/5364a) | Elderly, 75 y or older (301/1721a) | |||
| OR (95%CI) | P value | OR (95%CI) | P value | OR (95%CI) | P value | |
| Sex (male) | 1.12 (0.94-1.33) | .190 | 1.06 (0.87-1.29) | .545 | 0.85 (0.61-1.18) | .331 |
| Treatment (endovascular coiling) | 0.66 (0.49-0.88) | .005b | 0.47 (0.36-0.62) | <.001b | 0.37 (0.26-0.56) | <.001b |
| Age | 1.00 (0.99-1.01) | .910 | 1.00 (0.97-1.03) | .924 | 1.08 (1.04-1.13) | <.001b |
| Location of the aneurysms | ||||||
| ICA | Reference | Reference | Reference | |||
| ACoA | 0.92 (0.71-1.18) | .492 | 0.75 (0.58-0.98) | .033b | 1.30 (0.87-1.93) | .196 |
| MCA | 0.77 (0.61-0.97) | .025b | 0.66 (0.52-0.83) | <.001b | 0.93 (0.65-1.33) | .701 |
| ACA | 0.78 (0.50-1.21) | .264 | 0.56 (0.36-1.93) | .011b | 0.64 (0.32-1.28) | .205 |
| BA | 1.33 (0.97-1.83) | .075 | 1.42 (1.04-1.93) | .028b | 1.88 (1.13-3.12) | .015b |
| VA | 1.45 (1.04-2.02) | .028b | 1.60 (1.01-2.52) | .045b | 1.39 (0.61-3.18) | .432 |
| Other | 2.64 (1.21-5.74) | .014b | 0.99 (0.21-4.59) | .986 | 0.96 (0.11-8.64) | .968 |
| mFI-5 score | ||||||
| 0 item | Reference | Reference | Reference | |||
| 1 item | 1.35 (1.14-1.60) | .001b | 1.28 (1.07-1.54) | .008b | 1.32 (0.98-1,77) | .068 |
| ≥2 items | 1.66 (1.20-2.28) | .002b | 1.72 (1.30-2.29) | <.001b | 1.99 (1.36-2.93) | <.001b |
| Internal oral medication on admission | ||||||
| Antiplatelet | 2.52 (1.91-3.32) | <.001b | 3.28 (2.57-4.18) | <.001b | 2.81 (1.97-4.00) | <.001b |
| Anticoagulant | 1.95 (1.10-3.45) | .022b | 1.69 (1.08-2.63) | .021b | 1.96 (1.12-3.41) | .018b |
| Statin | 1.35 (1.04-1.74) | .024b | 1.02 (0.79-1.32) | .852b | 1.57 (1.12-2.22) | .010b |
| Hospital volume | ||||||
| 1 | Reference | Reference | Reference | |||
| 2 | 0.87 (0.72-1.06) | .164 | 0.87 (0.71-1.07) | .190 | 0.89 (0.65-1.22) | .430 |
| 3 | 0.75 (0.62-0.91) | .004b | 0.74 (0.60-0.91) | .004b | 0.58 (0.42-0.82) | .002b |
ACA, anterior cerebral artery; ACoA, anterior communicating artery; BA, basilar artery; ICA, internal carotid artery; MCA, middle cerebral artery; mFI-5, 5-factor modified frailty index; OR, odds ratio; VA, vertebral artery.
Complete cases without missing data.
P < .05.
Focusing on mFI-5 and chronological age, we found that an mFI-5 score ≥2 items was a significant risk factor for worsening BI scores and in-hospital complications in all age groups. However, chronological age was not associated with in-hospital complications in both the nonelderly and pre-elderly groups. Focusing on treatment, endovascular coiling was a significant inverse risk factor for worsening BI score and in-hospital complications compared with surgical clipping, regardless of age. Supplemental Digital Content 2, Table, http://links.lww.com/NEU/D430, presents the results of multivariable logistic regression analysis for in-hospital mortality in the nonelderly, pre-elderly, and elderly groups.
DISCUSSION
We demonstrated that chronological age was not associated with in-hospital complications in patients younger than 74 years undergoing treatment for UCA, whereas an mFI-5 score ≥2 items was associated with in-hospital complications in all age groups. This means that the mFI-5 score (biological age) may be an independent associated factor of post-UCA treatment outcomes, unlike chronological age. Our finding that frailty significantly affects the occurrence of complications after UCA treatment suggests that routine preoperative and prophylactic assessments of frailty should be considered for patients younger than 74 years before UCA treatment. Systematic screening for frailty would help identify patients at risk for complications after UCA treatment. Preoperative information on patient frailty status may help clinicians provide early treatment and rehabilitation interventions to mitigate the effects of postoperative complications.
With the increasing physical, mental, and social diversity of the elderly population, determining the optimal decision-making process for the treatment of elderly patients based on chronological age alone has become difficult.21-23 The phenomenon of aging is not consistent for all individuals, and thus, it may be inappropriate to associate the physical and cognitive functions of individuals in a particular age group. Because of the great variability in the aging process at the individual level, some younger patients may have higher mFI-5 scores than elderly patients. In fact, in our study, frailty was more prevalent in the elderly, although it was also observed in some younger patients. Frailty has been defined as a multidimensional condition that independently affects a patient's functional prognosis. Therefore, when considering prognostic models, it is suggested that biological age (mFI-5 score), unlike chronological age, should be considered during the initial evaluation of patients undergoing UCA surgery.
In the field of neurosurgery, the mFI-5 score is reportedly associated with postoperative outcomes in patients with pituitary adenoma, meningioma, glioma, and UCAs.13-15,24,25 Hitherto, no large-scale studies have been conducted on patients in separate age groups using nationwide UCA databases; only small-scale studies have been conducted.15,26 Based on separate analyses for each age group (including elderly patients) using nationwide data, the mFI-5 score was found to be a potential associated factor of postoperative outcomes in patients with UCAs, regardless of age group. In this study, patients who underwent surgical clipping were more predisposed to developing worsening BI scores and in-hospital complications in all age groups than those who underwent coil embolization. In actual clinical practice, the mFI-5 score and treatment method should be considered in the decision making for patients undergoing surgery for UCA, irrespective of age.
Patient age is a poor prognostic factor in the treatment of UCA.27,28 Owing to the global differences in population aging, the specific criteria for age vary by era and region. In the previous era, it was reported that age 50 years was the borderline for poor postoperative prognosis.29,30 Japan has one of the fastest aging populations in the world, and because aging in the rest of the world will follow a similar pattern,31 the results of this study may become a global standard.
Although worsening BI scores and in-hospital complications were associated with an mFI-5 score ≥2 items, in-hospital mortality was not associated with an mFI-5 score ≥2 items, probably because only 40 cases of in-hospital mortality were recorded, which were too small for statistical analysis. A hospital that handled a large number of surgical cases had significantly lower numbers of both in-hospital complications and worsening BI scores, probably because large hospitals with a high number of surgical cases are better equipped to prevent perioperative complications.32
Limitations
This study has several limitations. First, this study had a registry-based retrospective design, whereas a planned prospective cohort study or a randomized controlled trial would have offered a higher level of evidence. Furthermore, bias could not be eliminated completely. Second, the DPC database used in this study did not contain information pertaining to certain items. For example, we obtained data concerning aneurysm location but not aneurysm size, aneurysm morphology, SAH history, or family history, which, as shown in other national databases, may be unmeasurable confounders. Moreover, the clinicians who administered treatment to patients registered in this DPC database were working in different hospitals throughout Japan. Third, our data initially contain over half of the missing data; nevertheless, the final missing value is 0.26% for accurate analysis, which is small, and thus is not expected to significantly affect the overall results. Finally, our results should be interpreted with caution because they may not be generalizable to other countries with different health care resources and systems, given that the data in this study were obtained solely from the Japanese population, with the highest proportion of elderly people globally. However, our findings may present a standard for future research in other populations. In this study, we found a higher overall proportion of patients undergoing surgical clipping in Japan, with its unique characteristics.33
CONCLUSION
The mFI-5 score was a more useful associated factor for in-hospital complications than chronological age after UCA treatment in patients younger than 74 years. Considerations of both chronological age and the mFI-5 score may help better optimize therapeutic interventions for patients with UCA.
Supplementary Material
Acknowledgments
We thank all hospitals that participated in the DPC data registry in this study.
Footnotes
Supplemental digital content is available for this article at neurosurgery-online.com.
Contributor Information
Masashi Kuwabara, Email: m214028@gmail.com.
Nobuaki Michihata, Email: michihata-tky@umin.org.
Shigeyuki Sakamoto, Email: shigeyukisakamoto@gmail.com.
Takahito Okazaki, Email: okazaki@hiroshima-u.ac.jp.
Masahiro Hosogai, Email: m11050066@yahoo.co.jp.
Yuyo Maeda, Email: maedayuyo4.18@gmail.com.
Nobutaka Horie, Email: nobstanford@gmail.com.
Kaoru Kurisu, Email: kuka422@gmail.com.
Kiyohide Fushimi, Email: kfushimi.hci@tmd.ac.jp.
Hideo Yasunaga, Email: yasunagah-tky@umin.ac.jp.
Funding
This study was supported by the Pfizer Health Research Foundation Japan 2019 (19-E-01) and the Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research (C) 17K10829.
Disclosures
The authors have no personal, financial, or institutional interests in any of the drugs, materials, or devices described in this article.
Supplemental Digital Content
Supplemental Digital Content 1. Text. Details of diagnosis procedure combination data, patient selection, Barthel index, and hospital information.
Supplemental Digital Content 2. Table. Results of multivariate logistic regression analyses for in-hospital mortality across all age groups.
COMMENTS
This paper demonstrates an association between the modified frailty index (mFI-5 score) and outcome and complications following treatment for un-ruptured intracranial aneurysms in the elderly. The results support the notion that chronological age alone is an inadequate predictor. The results are derived from a nationwide Japanese database and therefore represent a range of institutions and clinicians. Because the Japanese population is renowned for not only lifespan but also health span, it is unclear whether these results are generalizable to other populations. Unfortunately, the database lacks detailed information regarding aneurysm characteristics such as location and morphology, which precludes analysis of treatment specific variables. Also, the authors were forced to exclude 12 905 patients (nearly 25%) due to data integrity issues. Further prospective research may clarify the utility of this tool.
Joel D. MacDonald
Salt Lake City, Utah, USA
REFERENCES
- 1.Morita A, Kirino T, Hashi K, et al. The natural course of unruptured cerebral aneurysms in a Japanese cohort. N Engl J Med. 2012;366(26):2474-2482. [DOI] [PubMed] [Google Scholar]
- 2.Greving JP, Wermer MJ, Brown RD, Jr, et al. Development of the PHASES score for prediction of risk of rupture of intracranial aneurysms: a pooled analysis of six prospective cohort studies. Lancet Neurol. 2014;13(1):59-66. [DOI] [PubMed] [Google Scholar]
- 3.Hishikawa T, Date I, Tokunaga K, et al. Risk of rupture of unruptured cerebral aneurysms in elderly patients. Neurology. 2015;85(21):1879-1885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yamashiro S, Nishi T, Koga K, et al. Postoperative quality of life of patients treated for asymptomatic unruptured intracranial aneurysms. J Neurosurg. 2007;107(6):1086-1091. [DOI] [PubMed] [Google Scholar]
- 5.Ikawa F, Michihata N, Matsushige T, et al. In-hospital mortality and poor outcome after surgical clipping and endovascular coiling for aneurysmal subarachnoid hemorrhage using nationwide databases: a systematic review and meta-analysis. Neurosurg Rev. 2020;43(2):655-667. [DOI] [PubMed] [Google Scholar]
- 6.Ikawa F, Michihata N, Akiyama Y, et al. Treatment risk for elderly patients with unruptured cerebral aneurysm from a nationwide database in Japan. World Neurosurg. 2019;132:e89-e98. [DOI] [PubMed] [Google Scholar]
- 7.Thompson BG, Brown RD, Jr, Amin-Hanjani S, et al. Guidelines for the management of patients with unruptured intracranial aneurysms: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2015;46(8):2368-2400. [DOI] [PubMed] [Google Scholar]
- 8.Rubinfeld I, Farooq M, Velanovich V, Syed Z. Predicting surgical risk: how much data is enough? AMIA Annu Symp Proc. 2010;2010:777-781. [PMC free article] [PubMed] [Google Scholar]
- 9.Wahl TS, Graham LA, Hawn MT, et al. Association of the modified frailty index with 30-day surgical readmission. JAMA Surg. 2017;152(8):749-757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Youngerman BE, Neugut AI, Yang J, Hershman DL, Wright JD, Bruce JN. The modified frailty index and 30-day adverse events in oncologic neurosurgery. J Neurooncol. 2018;136(1):197-206. [DOI] [PubMed] [Google Scholar]
- 11.Weaver DJ, Malik AT, Jain N, Yu E, Kim J, Khan SN. The modified 5-item frailty index: a concise and useful tool for assessing the impact of frailty on postoperative morbidity following elective posterior lumbar fusions. World Neurosurg. 2019;124:e626-e632. [DOI] [PubMed] [Google Scholar]
- 12.Subramaniam S, Aalberg JJ, Soriano RP, Divino CM. New 5-factor modified frailty index using American College of Surgeons NSQIP data. J Am Coll Surg. 2018;226(2):173-182. [DOI] [PubMed] [Google Scholar]
- 13.Khalafallah AM, Huq S, Jimenez AE, Brem H, Mukherjee D. The 5-factor modified frailty index: an effective predictor of mortality in brain tumor patients. J Neurosurg. 2020;135(1):78-86. [DOI] [PubMed] [Google Scholar]
- 14.Traven SA, Reeves RA, Sekar MG, Slone HS, Walton ZJ. New 5-factor modified frailty index predicts morbidity and mortality in primary hip and knee arthroplasty. J Arthroplasty. 2019;34(1):140-144. [DOI] [PubMed] [Google Scholar]
- 15.Feghali J, Gami A, Rapaport S, et al. Adapting the 5-factor modified frailty index for prediction of postprocedural outcome in patients with unruptured aneurysms. J Neurosurg. 2021;136(2):456-463. [DOI] [PubMed] [Google Scholar]
- 16.Yasunaga HMH, Horiguchi H, Fushimi K, Matsuda S. Clinical epidemiology and health services research using the Diagnosis Procedure Combination Database in Japan. Asian Pac J Dis Manag. 2013;7:19-24. [Google Scholar]
- 17.Morita K, Matsui H, Michihata N, Fushimi K, Yasunaga H. Association of early systemic corticosteroid therapy with mortality in patients with Stevens-Johnson syndrome or toxic epidermal necrolysis: a retrospective cohort study using a nationwide claims database. Am J Clin Dermatol. 2019;20(4):579-592. [DOI] [PubMed] [Google Scholar]
- 18.Ikawa F, Isobe N, Michihata N, et al. In-hospital complications after surgery in elderly patients with asymptomatic or minor symptom meningioma: a nationwide registry study. World Neurosurg. 2021;148:e459-e470. [DOI] [PubMed] [Google Scholar]
- 19.Orimo H, Ito H, Suzuki T, Araki A, Hosoi T, Sawabe M. Reviewing the definition of “elderly”. Geriatr Gerontol Int. 2006;6(3):149-158. [Google Scholar]
- 20.Cohen-Inbar O, Soustiel JF, Zaaroor M. Meningiomas in the elderly, the surgical benefit and a new scoring system. Acta Neurochir (Wien). 2010;152(1):87-97. [DOI] [PubMed] [Google Scholar]
- 21.Shimamura N, Naraoka M, Katagai T, et al. Analysis of factors that influence long-term independent living for elderly subarachnoid hemorrhage patients. World Neurosurg. 2016;90:504-510. [DOI] [PubMed] [Google Scholar]
- 22.Nieuwkamp DJ, Algra A, Blomqvist P, et al. Excess mortality and cardiovascular events in patients surviving subarachnoid hemorrhage: a nationwide study in Sweden. Stroke. 2011;42(4):902-907. [DOI] [PubMed] [Google Scholar]
- 23.Brinjikji W, Lanzino G, Rabinstein AA, Kallmes DF, Cloft HJ. Age-related trends in the treatment and outcomes of ruptured cerebral aneurysms: a study of the nationwide inpatient sample 2001-2009. Am J Neuroradiol. 2013;34(5):1022-1027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Khalafallah A, Shah P, Huq S. The 5-factor modified frailty index predicts health burden following surgery for pituitary adenomas. Pituitary. 2020;23(6):630-640. [DOI] [PubMed] [Google Scholar]
- 25.Huq S, Khalafallah A, Jimenez A. Predicting postoperative outcomes in brain tumor patients with a 5-factor modified frailty index. Neurosurgery. 2020;88(1):147-154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mori K, Wada K, Otani N, et al. Validation of effectiveness of keyhole clipping in nonfrail elderly patients with unruptured intracranial aneurysms. J Neurosurg. 2017;127(6):1307-1314. [DOI] [PubMed] [Google Scholar]
- 27.Bekelis K, Missios S, MacKenzie TA, et al. Predicting inpatient complications from cerebral aneurysm clipping: the Nationwide Inpatient Sample 2005-2009. J Neurosurg. 2014;120(3):591-598. [DOI] [PubMed] [Google Scholar]
- 28.Brinjikji W, Rabinstein AA, Lanzino G, Kallmes DF, Cloft HJ. Effect of age on outcomes of treatment of unruptured cerebral aneurysms: a study of the National Inpatient Sample 2001-2008. Stroke. 2011;42(5):1320-1324. [DOI] [PubMed] [Google Scholar]
- 29.Wiebers DO. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet. 2003;362(9378):103-110. [DOI] [PubMed] [Google Scholar]
- 30.Brinjikji W, Rabinstein AA, Nasr DM, Lanzino G, Kallmes DF, Cloft HJ. Better outcomes with treatment by coiling relative to clipping of unruptured intracranial aneurysms in the United States, 2001-2008. AJNR Am J Neuroradiol. 2011;32(6):1071-1075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.World Population Prospects 2019 Highlights. Accessed January 11, 2022. https://population.un.org/wpp/Publications/Files/WPP2019_Highlights.pdf. [Google Scholar]
- 32.Malone H, Cloney M, Yang J, et al. Failure to rescue and mortality following resection of intracranial neoplasms. Neurosurgery. 2018;83(2):263-269. [DOI] [PubMed] [Google Scholar]
- 33.Ikawa F, Hidaka T, Yoshiyama M, et al. Characteristics of cerebral aneurysms in Japan. Neurol Med Chir (Tokyo). 2019;59(11):399-406. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Digital Content 1. Text. Details of diagnosis procedure combination data, patient selection, Barthel index, and hospital information.
Supplemental Digital Content 2. Table. Results of multivariate logistic regression analyses for in-hospital mortality across all age groups.
Data Availability Statement
Only the results of the anonymized data for this study will be shared on request from any qualified investigator to the corresponding author.