Abstract
Object
Data evaluating the long-term outcomes, particularly with regard to treatment modality, of octogenerians with aneurysmal subarachnoid hemorrhage (SAH) are limited. The primary objectives were to evaluate the disposition (living at home versus institutional settings) and analyze the predictors of long-term survival and return to home of octogenerians after SAH.
Methods
Patients age 80 and older who underwent microsurgical clipping or endovascular coiling for SAH were extracted from 100% nationwide Medicare inpatient claims and linked with the Minimum Data Set (2008–2011). Patient disposition was tracked for two-years after index SAH admission. Multivariable logistic regression stratified by aneurysm treatment modality, and adjusted for patient factors including SAH severity, evaluated predictors of return to home at 60, 365 days after SAH. Multivariable cox proportional hazards analyzed survival 365 days after SAH.
Results
Among 1,298 patients 1-year following SAH, 56% had died or was in hospice care, 8% resided in institutional post-acute care, and 36% had returned home. Open microsurgical clipping aHR 0.67 (.54–.81), males .70 (CI .57–.87), tracheostomy 0.63 (.47–.85), gastrostomy .60 (CI .48–.76), and higher NIH SAH severity .94 CI (.92–.97) were less likely to ever return home. Older age (Adjusted Hazard Ratio (aHR) 1.09, 95% Confidence Interval (CI) 1.05–1.13), tracheostomy aHR 2.06 (1.46–2.91), and gastrostomy 1.55 (1.14–2.10) male sex aHR 1.66 (1.20–2.23) and worse SAH severity 1.51 (1.04–2.18) were associated with reduced survival.
Conclusions
In this national analysis, 56% of octogenerians with SAH died, while 36% of patients returned home within one year of SAH. Coil embolization predicted returning to home, which may suggest a benefit to endovascular treatment in this patient population.
Keywords: Age, Cerebral Aneurysm, Clipping, Endovascular, Minimum Data Set, Long-term Outcomes, Medicare, Octogenerians, Subarachnoid Hemorrhage
Introduction
The aging of the population in the US is a well-known demographic trend. By 2050, it is projected that at least 19% of the population of the United States will be at least 65 years old—an increase from 12% in 2005—and octogenerians will represent a greater proportion of the national population.15 The demographic shift is also reflected in the age distribution of patients presenting with aneurysmal subarachnoid hemorrhage (SAH),2 and a meta-analysis showed that the mean age of patients with SAH has increased between 1973 and 2002.13
There remains a paucity of data evaluating octogenerians treated for SAH, particularly analyzing the rates of long-term independence, as well as the impact of treatment modality utilized for aneurysm repair on outcomes. This is partially because octogenerians have largely been excluded from randomized clinical trials in the subarachnoid hemorrhage population. For example, patients aged greater than 80 were excluded from the Barrow Randomized Aneurysm Trial,10 and while the international subarachnoid aneurysm trial did not employ an age cutoff, only four octogenerians were enrolled.11 Therefore, the applicability of these trials to octogenarian patients remains unknown. While additional data on the long-term outcomes of octogenerians with SAH would inform treatment decisions and augment prognostication, it is unlikely that future randomized trials will include very elderly adults. Moreover, few single-institution studies accrue a sufficient population to provide an appropriately powered analysis of outcomes among octogenerians. Therefore, large national datasets are of particular importance and provide a niche in examining the outcomes of this patient population.
The Center for Medicare & Medicaid Services (CMS) Fee-for-Service Part A inpatient claims files include data on the hospitalization of all Americans aged at least 65 years. Additionally, the Minimum Data Set accrues detailed assessments of all nursing home residents in the United States. Linking these two data sets provides a unique perspective on the long-term survival and institutional care use of older adults in the United States. Previous analyses have identified time at home, free of institutional care as important patient-centered outcomes among stroke patients.5,16,24 Therefore, the goal of this analysis was to 1) evaluate the disposition (living at home versus institutional settings) of octogenerians at up to two-years after SAH; and 2) analyze the predictors of long-term survival and return to home up to one-year post-SAH. As prior studies have shown that the severity of the initial hemorrhage is the strongest predictor of a poor short-term outcome in older patients with SAH,12 it was hypothesized that patient comorbidities and severity of SAH—but not treatment modality used for aneurysm repair—would be the primary predictors of long-term outcomes.
Methods
Data Source & Inclusion Criteria
Data were extracted from the 100% Fee-for-Service Medicare claims denominator file and the corresponding Part A Inpatient Standard Analytic Files from January 1, 2008 to December 31, 2011; the last year of inclusion was 2011 to allow for two-year follow-up of the entire patient population. Patients were included if they 1) had an International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) diagnosis code of subarachnoid hemorrhage (430); 2) underwent cerebral aneurysm repair by either microsurgical clipping (39.51) or endovascular embolization (39.72, 39.75, 39.76, 39.79) during the same hospitalization; and 3) were at least 80 years old. Patients were excluded with a diagnosis code indicating a congenital cerebrovascular malformation (747.81, often used to indicate arteriovenous malformations), cerebral arteritis (437.4), and Moyamoya disease (437.5) or a procedure code for treatment of an arteriovenous malformation surgically (39.53) or via stereotactic radiosurgery (923.x). Moreover, those who underwent both clipping and coiling during the same hospitalization (n=14) were excluded, to provide a dichotomous analysis of treatment modality.
Thereafter, to determine long-term survival and institutional care utilization, patients from the Medicare Part A files were linked to the Minimum Data Set and the CMS vital status file using Medicare health insurance claim number, Social Security number, sex, and date of birth. The vital status file provides data on the date of death and the Minimum Data Set on nursing home utilization. Detailed methods for creating this per-person longitudinal record of health care utilization have been described previously.7 Analysis was covered by a data use agreement with CMS and performed after approval by our institutional review board.
Independent Variables
Patient age (analyzed continuously), sex, race or ethnicity (non-Hispanic Caucasian, African-American, Hispanic and Asian, other, or not recorded) and socioeconomic status (evaluated by quartile of income in the patient’s Zone Improvement Plan (ZIP) code of residence) were extracted. Comorbidities were assessed as the total number of comorbidities documented using the Elixhauser score (range 0 to 28, with higher scores indicating a greater comorbidity).3
The Nationwide Inpatient Sample-SAH severity scale (NIS-SAH)22 was calculated for each patient and utilized to adjust for the severity of presentation with SAH. The validated NIS-SAH severity scale ranges from 0 to 15, with higher scores indicating greater severity for use with administrative claims databases, and prior publications have shown that higher score has a strong concordance with greater Hunt-Hess grade.22 The scale is constructed using the ICD9-CM diagnosis codes for coma, hydrocephalus, hemiparesis, aphasia, and cranial nerve deficits, and procedure codes for cerebrospinal fluid diversion procedures (ventriculostomy and ventricular shunt placement) and mechanical ventilation. Prior studies have shown that the mean NIS-SAH severity scale of patients with Hunt-Hess Grade 3 was 4, Hunt-Hess Grade 4 was 7.5, and Hunt-Hess Grade 5 was 9.22 The treatment modality used for aneurysm obliteration (microsurgical clipping or endovascular embolization), placement of a tracheostomy (311, 312.1, 312.9) and gastrostomy or jejunostomy (431.1, 431.9, 463.2) during the SAH hospitalization, as well as the length of stay of the SAH hospitalization, were analyzed. Finally, any prior hospitalization in the year before the index admission for SAH was extracted from Medicare Part A claims.
Outcomes
The primary outcome for the analysis was the disposition of patients at 30, 60, 365, and 730 days after admission for SAH. Disposition was defined as the proportion of patients in the following categories: 1) death or hospice enrollment 2) inpatient acute care hospital 3) institutional care (inclusive of all rehabilitation and skilled nursing facilities), and 4) returned home. The time points were chosen to provide a description of both short and long-term outcomes. For multivariable regression, short and long-term outcomes were modeled at 60 and 365 days, with the dependent variable dichotomized as either 1) death/hospice/institutionalization, or 2) returned home. Cox modeling of survival was limited to 365-days after index SAH.
Sensitivity Analysis
A sensitivity analyses was performed to further evaluate the severity adjustment estimated by the NIS-SAH severity scale. This was performed using the individual components of the NIS-SAH severity scale (rather than the composite score) as covariates in regression models.
Statistical Analysis
Statistical analyses were conducted in STATA 13 (StataCorp, College Station, TX) and SAS 9.3 (Carey, NC). Descriptive statistics were performed. Thereafter, multivariable models were constructed: due to their clinical relevance, all potential predictors evaluated were included in models as covariates. Furthermore, interaction terms accounting for the treatment modality used for aneurysm repair and NIS-SAH severity scale, dichotomoized as poor grade (NIS-SAH severity scale greater than 7) or good grade. Cox proportional hazards models analyzed survival (after accounting for time to death), while logistic regression assessed the dichotomous outcome of return to home; both models were constructed at the three selected time points (30, 60, 365 days) after the SAH admission. Patients who died during the subarachnoid hemorrhage hospitalization were not included in the risk set for the Cox proportional hazards models, and therefore only post-hospitalization survival was analyzed in such models. Concordance statistics (C-statistics) evaluated model discriminatory capacity. A probability value less than 0.05 was deemed statistically significant.
Results
Demographics of the Study Population
A total of 1,298 patients with SAH who were aged at least 80 years were evaluated, in whom 77.8% (n=1,010) underwent coil embolization. Among the total cohort, 77.1% were good grade (NIS-SAH severity < 7), and 22.9% were poor grade (NIS-SAH ≥7). The demographics of the study population are depicted and stratified by the treatment modality used for aneurysm repair in Table 1. The most common comorbidities from the Elixhauser index were hypertension (47.4%), fluid and electrolyte disorders (45.6%), weight loss (13.6%) congestive heart failure (11.0%), diabetes mellitus (10.4%), deficient anemia (9.4%), hypothyroidism (9.0%), and chronic pulmonary disease (8.6%). The proportion of patients with cerebrospinal fluid diversion procedures (included in the NIS-SAH severity scale) was 38.7% for ventriculostomy, and 16.8% with ventricular shunt placement.
Table 1.
Demographics of the study population stratified by the treatment modality used for aneurysm repair.
| Characteristic % (No.) |
Total Population (n=1,298) | Coil Embolization (n=1,010) | Microsurgical Clipping (n=288) | P-value |
|---|---|---|---|---|
|
| ||||
| Age, Mean(Standard Deviation)* | 84.7 (3.5) | 84.9 (3.5) | 84.2 (3.3) | 0.002† |
|
| ||||
| Age group | 0.01 | |||
| 80–84 | 59 (771) | 57 (578) | 67(193) | |
| 85–89 | 32 (425) | 35 (352) | 26 (73) | |
| 90–94 | 7 (88) | 7 (68) | 7 (20) | |
| 95–100 | 1 (13) | 1 (12) | -- (--) | |
|
| ||||
| Female Sex | 82 (1,068) | 82 (831) | 82 (237) | 0.99 |
|
| ||||
| Race or Ethnicity | 0.08 | |||
| Caucasian | 77 (996) | 78 (785) | 73 (211) | |
| African-American | 9 (115) | 9 (93) | 7 (22) | |
| Hispanic | 7 (93) | 7 (66) | 9 (27) | |
| Other or Unknown | 7 (93) | 7 (66) | 10 (27) | |
|
| ||||
| Socioeconomic Status | 0.11 | |||
| Lower Quartile | 24 (312) | 24 (238) | 26 (74) | |
| Second Quartile | 25 (319) | 24 (242) | 26 (77) | |
| Third Quartile | 24 (319) | 26 (261) | 19 (58) | |
| Upper Quartile | 27 (353) | 27 (269) | 29 (84) | |
|
| ||||
|
Number of Comorbidities Mean(Standard Deviation) |
2.1 (1.2) | 2.1 (1.2) | 2.1 (1.2) | 0.64 |
|
| ||||
|
NIS-SAH Severity Scale Mean(Standard Deviation) |
6.2 (10.6) | 6.1 (10.6) | 6.5 (10.8) | 0.59 |
|
| ||||
| Poor grade (NIS-SAH ≥ 7) | 23 (296) | 23.9 (241) | 19 (55) | 0.16 |
|
| ||||
| Tracheostomy | 12 (153) | 10 (102) | 18 (51) | <0.001 |
|
| ||||
| Gastrostomy | 14 (180) | 13 (127) | 19 (53) | 0.008 |
|
| ||||
| Hospitalization During the Prior Year | 32 (421) | 34 (341) | 27 (80) | 0.04 |
|
| ||||
|
Length of Stay during the SAH Hospitalization Mean (Standard Deviation) |
15 (10.4) | 14.3 | 17.3 | <0.001 |
NIS-SAH indicates National Inpatient Sample Subarachnoid Hemorrhage Severity Scale. SAH, subarachnoid hemorrhage.
All data are presented as percentages rounded to the nearest whole number, with the exception of patient age, number of comorbidities, NIS-SAH severity scale, and Length of Stay. Column percentages may not add to 100 due to rounding.
Statistically significant differences in bivariate analyses are bolded.
-- data suppressed in accordance with Centers for Medicare and Medicaid Services cell size suppression policy.
Short-Term and Long-Term Disposition
At 30-days after SAH, 32.7% of the cohort had died or was in hospice care, 28.3% remained hospitalized, 21.1% were in institutional care, and 16.7% had returned to home. The 60-, 365-, and 730-day disposition of patients is presented for the total study population as well as stratified by both the treatment modality used for aneurysm treatment and by patient grade in Table 2. Institutional care of octogenarian patients at one and two-years from SAH was rare. This is illustrated by Figure 1, displaying primary outcomes of patient death/hospice care, return to home, or institutional care (i.e. inpatient acute hospital, rehabilitation, skilled nursing facility) at presentation, 30-, 60-, 365-, and 730-days after admission. At two-years after SAH, 34.1% of patients had returned to home, including 35.0% of patients treated with coil embolization and 30.9% of those who underwent microsurgical clipping. Figure 2 displays Kaplan-Meier time to event curves comparing time to return home between microsurgical clipping versus endovascular coiling. Time to return home was significantly less for the endovascular coiling group compared to the microsurgical clipping group (P <.0001).
Table 2.
Disposition of the study population at 60, 365, and 730 days after subarachnoid hemorrhage admission, stratified by the treatment modality used for aneurysm repair.
| Characteristic % (No.) |
Total Population (n=1,298) | Coil Embolization (n=1,010) | Microsurgical Clipping (n=288) |
|---|---|---|---|
| All Patients | |||
|
| |||
| Disposition at 60 Days | |||
| Died | 42 (549) | 42 (426) | 43 (123) |
| Home | 26 (349) | 29 (292) | 19 (56) |
| Inpatient | 7 (100) | 6 (64) | 13 (36) |
| Institutional Care | 22 (284) | 21 (215) | 24 (69) |
| Hospice | 1 (12) | -- (--) | -- (--) |
| Disposition at 365 Days | |||
| Died | 55 (713) | 54 (543) | 59 (169) |
| Home | 37 (476) | 38 (382) | 32 (94) |
| Inpatient | -- (--) | -- (--) | -- (--) |
| Institutional Care | 7 (87) | 7 (66) | 8 (22) |
| Hospice | 1 (12) | 1 (11) | 0 (0) |
| Disposition at 730 Days | |||
| Died | 60 (774) | 59 (598) | 61 (175) |
| Home | 34 (443) | 35 (354) | 31 (89) |
| Inpatient | -- (--) | -- (--) | -- (--) |
| Institutional Care | 5 (64) | 4 (42) | 7 (21) |
| Hospice | -- (--) | 1 (11) | -- (--) |
|
| |||
| Good Grade Patients* | |||
|
| |||
| Characteristic | Good Grade (n=1,001) | Coil Embolization (n=769) | Microsurgical Clipping (n=232) |
|
| |||
| Disposition at 60 Days | |||
| Died | 37 (368) | 36 ( 274) | 41 (94) |
| Home | 33 (331) | 37 (281) | 23 (53) |
| Inpatient | 6 (63) | 5 (39) | 10 (24) |
| Institutional Care | 23 (234) | 23 (173) | 26 (60) |
| Hospice | -- (--) | -- (--) | 0 (0) |
| Disposition at 365 Days | |||
| Died | 49 (492) | 47 (364) | 55 (128) |
| Home | 43 (426) | 45 (343) | 36 (83) |
| Inpatient | -- (--) | -- (--) | 0 (0) |
| Institutional Care | 7 (68) | 6 (48) | 9 (20) |
| Hospice | -- (--) | -- (--) | 0 (0) |
| Disposition at 730 Days | |||
| Died | 54 (539) | 53 (406) | 57 (132) |
| Home | 39 (394) | 41 (316) | 34 (78) |
| Inpatient | -- (--) | -- (--) | -- (--) |
| Institutional Care | 5 (52) | 4 (33) | 8 (19) |
| Hospice | 1 (11) | -- (--) | -- (--) |
|
| |||
| Poor Grade Patients | |||
|
| |||
| Characteristic | Poor Grade (n=297) | Coil Embolization (n=241) | Microsurgical Clipping (n=56) |
|
| |||
| Disposition at 60 Days | |||
| Died | 61 (181) | 63 (152) | 52 (29) |
| Home | 7 (22) | 8 (18) | -- (--) |
| Inpatient | 6 (19) | 5 (12) | -- (--) |
| Institutional Care | 17 (51) | 17 (42) | -- (--) |
| Hospice | -- (--) | -- (--) | -- (--) |
| Disposition at 365 Days | |||
| Died | 74 (221) | 74 (179) | 75 (42) |
| Home | 18 (53) | 17 (42) | 20 (11) |
| Inpatient | -- (--) | -- (--) | -- (--) |
| Institutional Care | 7 (20) | 6 (15) | -- (--) |
| Hospice | -- (--) | -- (--) | 0 (0) |
| Disposition at 730 Days | |||
| Died | 79 (235) | 80 (192) | 77 (43) |
| Home | 16 (48) | 15 (37) | 20 (11) |
| Inpatient | 0 (0) | -- (--) | 0 (0) |
| Institutional Care | 4 (11) | -- (--) | -- (--) |
| Hospice | -- (--) | -- (--) | 0 (0) |
All data are presented as percentages, rounded to the nearest whole number, of the entire cohort (complete case analysis) without missing data at each follow-up interval. Column percentages may not add to 100 due to rounding.
-- Data suppressed in accordance with Centers for Medicare and Medicaid Services cell size suppression policy.
Figure 1.
Disposition of the entire study population over the two-year study period. Black indicates acute hospital care. Grey indicates institutional care (i.e. rehabilitation, nursing home/skilled nursing facility). Dark grey indicates return to home. Lightest grey indicates death/hospice care.
Figure 2.
Kaplan-Meier time to event curves comparing microsurgical clipping vs. endovascular coiling. Entire cohort is hospitalized at time of aneurysm treatment. Mortality and institutional care at any location other than home (acute hospital, rehabilitation, nursing home/skilled nursing facility) were combined for the analysis. Time to return home was significantly less for the endovascular coiling group compared to the microsurgical clipping group (P <.0001).
Among patients institutionalized 60 days after SAH, the long-term disposition was specifically evaluated, to analyze how many of these patients ultimately returned home. At 365-days after SAH 36.6% had died or was in hospice, 43.3% had returned home, 19.3% remained in institutional care, and 1.0% were an inpatient. While at 730-days after hemorrhage, 46.1% had died or was in hospice, 40.0% had returned home, 13.5% were institutionalized, and 0.5% were inpatient.
Survival Analysis
Multivariable Cox proportional hazards models evaluated the predictors of mortality 365-days after admission for SAH (Table 3). Older patient age, male sex, tracheostomy, and gastrostomy placement during the SAH hospitalization were associated with significantly greater hazard of mortality at both one year after SAH. A sensitivity analysis used the components of the NIS-SAH severity scale as individual covariates, which showed similar statistical associations with the primary analysis (data not shown), and none of the individual components of the scale were significantly associated with survival.
Table 3.
Cox proportional hazard model evaluating predictors of mortality at 365 days after admission for subarachnoid hemorrhage.
| Predictors of Mortality | |||
|---|---|---|---|
|
| |||
| Characteristic | Hazard Ratio | 95% Confidence Interval | P-value |
|
| |||
| 365 Days after SAH | |||
|
| |||
| Treatment Modality | 0.11 | ||
| Endovascular Coiling | Ref. | -- | |
| Microsurgical Clipping | 1.27 | (0.95–1.68) | |
|
| |||
| Age | 1.09 | (1.05–1.12) | <0.001 |
|
| |||
| Female Sex | 0.63 | (0.46–0.86) | 0.003 |
|
| |||
| Race or Ethnicity | 0.60 | ||
| Caucasian | Ref. | -- | |
| African-American | 0.77 | (0.49–1.23) | |
| Hispanic | 0.80 | (0.51–1.24) | |
| Other or Unknown | 0.96 | (0.59–1.57) | |
|
| |||
| Socioeconomic Status | 0.35 | ||
| Lower Quartile | Ref. | -- | |
| Second Quartile | 0.75 | (0.51–1.11) | |
| Third Quartile | 1.05 | (0.73–1.51) | |
| Upper Quartile | 0.88 | (0.62–1.29) | |
|
| |||
| Number of Comorbidities | 1.07 | (0.96–1.20) | 0.24 |
|
| |||
| NIS-SAH Severity Scale | 1.01 | (0.99–1.03) | 0.19 |
|
| |||
| Tracheostomy | 2.11 | (1.49–2.98) | <0.001 |
|
| |||
| Gastrostomy | 1.54 | (1.13–2.09) | 0.006 |
|
| |||
| Hospitalization During the Prior Year | 0.95 | (0.72–1.26) | 0.72 |
|
| |||
| Length of Stay during the SAH Hospitalization | 1.02 | (1.00–1.03) | 0.009 |
NIS-SAH indicates National Inpatient Sample Subarachnoid Hemorrhage Severity Scale. SAH, subarachnoid hemorrhage.
Statistically significant differences in multivariable cox proportional hazard model are bolded.
Return to Home
Cox proportional hazard modeling identified microsurgical clipping, males, greater severity, tracheostomy, gastrostomy, and longer length of stay to be associated with reduced likelihood of returning home by 365-days after subarachnoid hemorrhage (Table 4). Multivariable logistic regression models evaluated the predictors of return to home at 60 and 365-days after admission for SAH (Table 5). The final models included interaction terms for treatment modality used for aneurysm repair and NIS-SAH severity scale. The same associations are provided without interaction terms in Table 6. Microsurgical clipping was associated with reduced odd of return to home among good grade patients, but treatment modality used for aneurysm repair was not significantly associated with disposition among poor grade patients. In addition, older patient age, tracheostomy, and shorter SAH hospitalization were associated with reduced odds of return to home at 60 and 365 days after SAH with moderate discriminatory capacity (C-statistic: 0.75 and 0.69 respectively).
Table 4.
Cox proportional hazard model evaluating predictors of return to home at 365 days after admission for subarachnoid hemorrhage.
| Characteristic | Hazard Ratio | 95% Confidence Interval | P-value |
|---|---|---|---|
|
| |||
| Treatment Modality | <0.001 | ||
| Endovascular Coiling | Ref. | -- | |
| Microsurgical Clipping | 0.67 | (0.54–.081) | |
|
| |||
| Age | 0.98 | (0.95–1.00) | 0.07 |
|
| |||
| Male Sex | 0.70 | (0.57–0.87) | 0.003 |
|
| |||
| Race or Ethnicity | 0.51 | ||
| Caucasian | Ref. | -- | |
| African-American | 0.91 | (0.69–1.20) | |
| Hispanic | 1.31 | (0.99–1.73) | |
| Other or Unknown | 1.07 | (0.79–1.47) | |
|
| |||
| Socioeconomic Status | <0.05 | ||
| Lower Quartile | Ref. | -- | |
| Second Quartile | 1.21 | (0.96–1.52) | |
| Third Quartile | 1.18 | (0.93–1.50) | |
| Upper Quartile | 1.40 | (1.11–1.78) | |
|
| |||
| Number of Comorbidities | 0.95 | (0.89–1.02) | 0.13 |
|
| |||
| NIS-SAH Severity Scale | 0.94 | (0.92–0.97) | <0.05 |
|
| |||
| Tracheostomy | 0.63 | (0.47–0.85) | <0.001 |
|
| |||
| Gastrostomy | 0.60 | (0.48–0.76) | <0.001 |
|
| |||
| Hospitalization During the Prior Year | 0.93 | (0.79–1.11) | 0.64 |
|
| |||
| Length of Stay during the SAH Hospitalization | 0.96 | (0.95–0.97) | <0.05 |
NIS-SAH indicates National Inpatient Sample Subarachnoid Hemorrhage Severity Scale. SAH, subarachnoid hemorrhage.
Statistically significant differences in multivariable cox proportional hazard model are bolded.
Table 5.
Multivariable logistic regression models evaluating the predictors of return to home at 60 and 365 days after admission for subarachnoid hemorrhage, accounting for the interaction between severity of subarachnoid hemorrhage and treatment modality used for aneurysm repair.
| Predictors of Return to Home | |||
|---|---|---|---|
|
| |||
| Characteristic | Odds Ratio | 95% Confidence Interval | P-value |
|
| |||
| 60 Days after SAH | |||
|
| |||
| Treatment Modality among Good Grade Patients | 0.003* | ||
| Endovascular Coiling | Ref. | -- | |
| Microsurgical Clipping | 0.41 | (0.25–0.67) | |
|
|
|||
| Treatment Modality among Poor Grade Patients | |||
| Endovascular Coiling | Ref. | -- | |
| Microsurgical Clipping | 3.36 | (0.94–12.08) | |
|
|
|||
| Grade among Endovascular Patients | |||
| Poor Grade | Ref. | -- | |
| Good Grade | 6.59 | (3.15–13.78) | |
|
|
|||
| Grade among Microsurgical Patients | |||
| Poor Grade | Ref. | -- | |
| Good Grade | 0.8 | (0.25–2.55) | |
|
| |||
| Age | 0.92 | (0.87–0.96) | <0.001 |
|
| |||
| Female Sex | 1.04 | (0.67–1.61) | 0.86 |
|
| |||
| Race or Ethnicity | 0.50 | ||
| Caucasian | Ref. | -- | |
| African-American | 0.91 | (0.46–1.78) | |
| Hispanic | 1.46 | (0.78–2.74) | |
| Other or Unknown | 0.75 | (0.39–1.47) | |
|
| |||
| Socioeconomic Status | 0.02 | ||
| Lower Quartile | Ref. | -- | |
| Second Quartile | 1.05 | (0.63–1.76) | |
| Third Quartile | 1.62 | (0.99–2.65) | |
| Upper Quartile | 1.84 | (1.13–2.99) | |
|
| |||
| Number of Comorbidities | 1.05 | (0.91–1.21) | 0.49 |
|
| |||
| Tracheostomy | 0.15 | (0.04–0.65) | 0.01 |
|
| |||
| Gastrostomy | 0.17 | (0.06–0.48) | <0.001 |
|
| |||
| Hospitalization During the Prior Year | 1.03 | (0.73–1.46) | 0.85 |
|
| |||
| Length of Stay during the SAH Hospitalization | 0.99 | (0.97–1.08) | 0.18 |
|
| |||
| C-statistic | 0.75 | ||
|
| |||
| 365 Days after SAH | |||
|
| |||
| Treatment Modality among Good Grade Patients | 0.09 | ||
| Endovascular Coiling | Ref. | -- | |
| Microsurgical Clipping | 0.64 | (0.46–0.89) | |
|
|
|||
| Treatment Modality among Poor Grade Patients | |||
| Endovascular Coiling | Ref. | -- | |
| Microsurgical Clipping | 1.33 | (0.62–2.85) | |
|
|
|||
| Grade among Endovascular Patients | |||
| Poor Grade | Ref. | -- | |
| Good Grade | 3.46 | (2.35–5.09) | |
|
|
|||
| Grade among Microsurgical Patients | |||
| Poor Grade | Ref. | -- | |
| Good Grade | 1.67 | (0.80–3.51) | |
|
| |||
| Age | 0.90 | (0.86–0.93) | <0.001 |
|
| |||
| Female Sex | 1.30 | (0.94–1.80) | 0.12 |
|
| |||
| Race or Ethnicity | 0.79 | ||
| Caucasian | Ref. | -- | |
| African-American | 1.20 | (0.73–1.89) | |
| Hispanic | 1.17 | (0.69–1.88) | |
| Other or Unknown | 1.11 | (0.94–1.80) | |
|
| |||
| Socioeconomic Status | 0.11 | ||
| Lower Quartile | Ref. | -- | |
| Second Quartile | 1.40 | (0.98–2.00) | |
| Third Quartile | 1.08 | (0.75–1.56) | |
| Upper Quartile | 1.43 | (1.00–2.05) | |
|
| |||
| Number of Comorbidities | 1.06 | (0.95–1.17) | 0.33 |
|
| |||
| Tracheostomy | 0.532 | (0.32–0.84) | 0.008 |
|
| |||
| Gastrostomy | 0.64 | (0.43–0.96) | 0.03 |
|
| |||
| Hospitalization During the Prior Year | 0.78 | (0.60–1.01) | 0.06 |
|
| |||
| Length of Stay during the SAH Hospitalization | 1.03 | (1.01–1.04) | <0.001 |
|
| |||
| C-statistic | 0.69 | ||
SAH indicates subarachnoid hemorrhage.
Statistically significant differences by multivariable logistic regression are bolded.
Table 6.
Logistic regression model evaluating predictors of return to home at 60 and 365 days after admission for subarachnoid hemorrhage.
| Return to Home | |||
|---|---|---|---|
|
| |||
| Characteristic | Odds Ratio | 95% Confidence Interval | P-value |
|
| |||
| 60 Days after SAH | |||
|
| |||
| Treatment Modality | |||
| Endovascular Coiling | Ref. | -- | -- |
| Microsurgical Clipping | 0.50 | (0.31–0.80) | 0.004* |
|
| |||
| Age | 0.92 | (0.88–0.97) | 0.001 |
|
| |||
| Female Sex | 1.01 | (0.65–1.57) | 0.96 |
|
| |||
| Race or Ethnicity | 0.48 | ||
| Caucasian | Ref. | -- | |
| African-American | 0.91 | (0.46–1.81) | |
| Hispanic | 1.51 | (0.80–2.87) | |
| Other or Unknown | 0.76 | (0.38–1.50) | |
|
| |||
| Socioeconomic Status | 0.04 | ||
| Lower Quartile | Ref. | -- | |
| Second Quartile | 1.01 | (0.60–1.70) | |
| Third Quartile | 1.61 | (0.98–2.66) | |
| Upper Quartile | 1.72 | (1.05–2.82) | |
|
| |||
| Number of Comorbidities | 1.01 | (0.87–1.16) | 0.93 |
|
| |||
| NIS-SAH Severity Scale | 0.75 | (0.69–0.81) | <0.001 |
|
| |||
| Tracheostomy | 0.23 | (0.05–0.98) | 0.047 |
|
| |||
| Gastrostomy | 0.18 | (0.06–0.52) | 0.001 |
|
| |||
| Hospitalization During the Prior Year | 0.98 | (0.69–1.40) | 0.92 |
|
| |||
| Length of Stay during the SAH Hospitalization | 0.99 | (0.97–1.01) | 0.39 |
|
| |||
| Concordance Statistic | 0.79 | ||
|
| |||
| 365 Days after SAH | |||
|
| |||
| Treatment Modality | 0.04 | ||
| Endovascular Coiling | Ref. | -- | |
| Microsurgical Clipping | 0.72 | (0.53–0.98) | |
|
| |||
| Age | 0.90 | (0.86–0.93) | <0.001 |
|
| |||
| Female Sex | 1.28 | (0.92–1.79) | 0.14 |
|
| |||
| Race or Ethnicity | 0.83 | ||
| Caucasian | Ref. | -- | |
| African-American | 1.17 | (0.71–1.85) | |
| Hispanic | 1.15 | (0.69–1.84) | |
| Other or Unknown | 1.13 | (0.92–1.79) | |
|
| |||
| Socioeconomic Status | 0.18 | ||
| Lower Quartile | Ref. | -- | |
| Second Quartile | 1.35 | (0.94–1.93) | |
| Third Quartile | 1.05 | (0.73–1.52) | |
| Upper Quartile | 1.37 | (0.95–1.96) | |
|
| |||
| Number of Comorbidities | 1.04 | (0.93–1.16) | 0.50 |
|
| |||
| NIS-SAH Severity Scale | 0.86 | (0.83–0.90) | <0.001 |
|
| |||
| Tracheostomy | 0.61 | (0.37–0.99) | 0.04 |
|
| |||
| Gastrostomy | 0.65 | (0.43–0.97) | 0.04 |
|
| |||
| Hospitalization During the Prior Year | 0.76 | (0.58–0.99) | 0.04 |
|
| |||
| Length of Stay During the SAH Hospitalization | 0.9003 | (0.86–0.93) | <0.001 |
|
| |||
| Concordance Statistic | 0.72 | ||
NIS-SAH indicates National Inpatient Sample Subarachnoid Hemorrhage Severity Scale. SAH, subarachnoid hemorrhage.
Statistically significant differences from multivariable logistic regression model are bolded.
A sensitivity analysis using the individual components of the NIS-SAH severity scale as covariates, rather than the composite score revealed similar associations with for the primary analysis among covariates (data not shown).
Discussion
This nationwide study demonstrates a low rate of chronic institutionalization and high rate of mortality among octogenerians after SAH. Although a substantial proportion of patients were institutionalized or died during the early post-acute care period, long-term trends reveal most patients ultimately return home or die by one and two-years of follow up. Stratified by treatment modality, the independent variables of greater age, male sex, SAH severity, tracheostomy and gastrostomy were all were associated with lower odds of survival and returning home. Coil embolization was a predictor of returning home, suggesting a benefit to endovascular treatment in this non-randomized analysis of this patient population.
This study enhances understanding where few previous studies have specifically examined the outcomes of very elderly patients undergoing treatment of SAH. Some authors have reported comparatively favorable outcomes among this patient population. Horiuchi et al. reported a 54.1% rate of good short-term outcome (as measured by the Glasgow Outcome Scale score) and predictors of a poor outcome were greater SAH grade and anterior cerebral arterial location, but not patient age.6 Likewise, others have observed high rates of good outcomes among octogenerians treated endovascularly .23.17
However, not all authors argue that very elderly patients with SAH should be offered treatment. In 2011, Koffijberg et al. proposed a Markov model of the cost-effectiveness of aneurysm treatment in elderly patients, and argued that beyond 80 years, the balance of risk and benefit did not favor aneurysm occlusion.9, Others have found low rates of survival and favor caution in pursuing surgical treatment.4,14,18 Prior analysis of the Nationwide Inpatient Sample has shown that very elderly adults are increasingly being treated with endovascular embolization in the United States,1 and in this setting additional data evaluating the long-term outcomes of octogenerians are needed.
Our study is unique in its use of 100% Medicare Part A inpatient files linked to the Minimum Data Set, which provides a nationally representative examination of long-term disposition and survival of octogenerians treated for aneurysmal subarachnoid hemorrhage. There is a substantial increase (17%) in mortality between 30 days and one year in this patient population, highlighting that many deaths occur post-hospitalization. However, among survivors of SAH, only a small minority remained institutionalized in the long-term—6.7% at one-year and 4.1% at two-years after SAH. Moreover, 36.7% of patients had returned to home at one-year and 34.1% at two-years, emphasizing that good outcomes in this patient population are possible, albeit in appropriately selected patients. While return to home is not as granular an assessment of independence compared with ordinal scales of neurologic outcome (such as the Glasgow Outcome Scale), it indicates a level of functional independence which does not require the patient to depend on institutional care. Additionally, return to home is an important clinical endpoint of interest to patients and their families when determining the aggressiveness of care to pursue.5,16,24
To identify predictors of long-term survival and return to home, Cox models and multivariable regression models were constructed. Older patient age, male sex, and the placement of tracheostomy or gastrostomy were independently associated with reduced survival at one- and two-years after SAH. Tracheostomy and gastrostomy may be proxies for neurologic function towards the end of the subarachnoid hemorrhage hospitalization, and the NIS-SAH outcome measure, based on mortality, discharge to institutional care, and tracheostomy or gastrostomy placement, has been shown to have excellent concordance with modified Rankin Scale in the subarachnoid hemorrhage population.22 Additionally, older patient age, higher NIS-SAH severity scale (worse SAH grade), the use of microsurgical clipping, and tracheostomy were independently associated with reduced odds of return to home one year after SAH.
Clinical decision making for treating octogenerians with SAH presents several unique considerations. While some studies have suggested that older patients with SAH have a lower incidence of symptomatic delayed cerebral ischemia,21 several authors have emphasized that older patients sustain inferior long-term outcomes compared with younger patients.12.8,19.20 Our findings of SAH clinical grade as an important independent predictor of octogenerians’ ability to return home add an important perspective in interpreting previous work.
Limitations to this analysis include: 1) clinical data are encoded in the Minimum Data Set using ICD9-CM identifiers. While severity adjustment was estimated using the NIS-SAH severity score –which has been externally validated against Hunt-Hess grade—several other noteworthy clinical features in patients with aneurysmal subarachnoid hemorrhage including World Federation of Neurological Surgeons, Fisher grade, aneurysm location or size, and delayed cerebral ischemia were not available due to the lack of corresponding ICD9-CM identifier. 2) Aneurysm size, location and other characteristics were not available in the dataset. If less complicated aneurysms are more frequently selected for endovascular approach, accounting for these factors may reduce the strength of the reported associations between aneurysm treatment and mortality or return to home. 3) Although return to home was analyzed as the primary proxy for a good outcome, the neurologic status of such patients, including the modified Rankin Scale or the Glasgow Outcome Scale score could not be analyzed with the study design., 4) The ICD-CM code for subarachnoid hemorrhage does not specify the etiology as aneurysmal or non-aneurysmal, patients with aneurysmal subarachnoid hemorrhage treated conservatively could not be evaluated.
Conclusions
This nationally-representative analysis of long-term survival and outcomes of very elderly adults with SAH can inform prognostication and may inform treatment decisions. Although rates of long-term institutional care dependence among survivors are low, the high mortality rates emphasize the importance of appropriate patient selection for aneurysm repair. While the treatment modality used for aneurysm occlusion was not associated with survival, endovascular coil embolization was associated with greater odds of return to home among good grade patients at one- and two-years after SAH. Nevertheless, treatment modality did not significantly impact the outcomes of poor grade patients. Although, this analysis did not control for aneurysm architecture and location, our results favor endovascular treatment when otherwise clinically appropriate to good grade octogenarian patients. The decisions for managing poor grade SAH patients should be based upon a comprehensive discussion between the patient’s family and clinicians, considering low probability of long-term survival and return to home.
Footnotes
Disclosures:
Sources of Funding: Dr. Mor is supported by National Institutes of Health (NIH) R01HL111032 and K24AG049057. Dr. Mitchell is supported by NIH–National Institute on Aging K24AG033640. Dr. Fehnel is supported by the American Academy of Neurology Clinical Research Training Fellowship.
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.
References
- 1.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;34:1022–1027. doi: 10.3174/ajnr.A3321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.de Rooij NK, Linn FHH, van der Plas JA, Algra A, Rinkel GJE. Incidence of subarachnoid haemorrhage: a systematic review with emphasis on region, age, gender and time trends. J Neurol Neurosurg Psychiatry. 2007;78:1365–1372. doi: 10.1136/jnnp.2007.117655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Elixhauser A, Steiner C, Harris RD, Coffey RM. Comorbidity measures for use with administrative data. Medical Care. 1998;36:8–27. doi: 10.1097/00005650-199801000-00004. [DOI] [PubMed] [Google Scholar]
- 4.Fridriksson SM, Hillman J, Saveland H, Brandt L. Intracranial aneurysm surgery in the 8th and 9th decades of life: impact on population-based management outcome. Neurosurgery. 1995;37:627–631. doi: 10.1227/00006123-199510000-00004. discussion 631–622. [DOI] [PubMed] [Google Scholar]
- 5.Hannah D, Lindholm B, Maisch L. Certain uncertainty: life after stroke from the patient’s perspective. Circ Cardiovasc Qual Outcomes. 2014;7:968–969. doi: 10.1161/CIRCOUTCOMES.114.001315. [DOI] [PubMed] [Google Scholar]
- 6.Horiuchi T, Tsutsumi K, Ito K, Hongo K. Results of clipping surgery for aneurysmal subarachnoid hemorrhage in the ninth and tenth decades of life. J Clin Neurosci. 2014;21:1567–1569. doi: 10.1016/j.jocn.2013.11.047. [DOI] [PubMed] [Google Scholar]
- 7.Intrator O, Hiris J, Berg K, Miller SC, Mor V. The residential history file: studying nursing home residents’ long-term care histories(*) Health Serv Res. 2011;46:120–137. doi: 10.1111/j.1475-6773.2010.01194.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Khosla A, Brinjikji W, Cloft H, Lanzino G, Kallmes DF. Age-related complications following endovascular treatment of unruptured intracranial aneurysms. Am J Neuroradiol. 2012;33:953–957. doi: 10.3174/ajnr.A2881. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Koffijberg H, Buskens E, Rinkel GJ. Aneurysm occlusion in elderly patients with aneurysmal subarachnoid haemorrhage: a cost-utility analysis. J Neurol Neurosurg Psychiatry. 2011;82:718–727. doi: 10.1136/jnnp.2009.185660. [DOI] [PubMed] [Google Scholar]
- 10.McDougall CG, Spetzler RF, Zabramski JM, Partovi S, Hills NK, Nakaji P, et al. The Barrow Ruptured Aneurysm Trial. J Neurosurg. 2012;116:135–144. doi: 10.3171/2011.8.JNS101767. [DOI] [PubMed] [Google Scholar]
- 11.Molyneux A, Kerr R, Stratton I, Sandercock P, Clarke M, Shrimpton J, et al. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet. 2002;360:1267–1274. doi: 10.1016/s0140-6736(02)11314-6. [DOI] [PubMed] [Google Scholar]
- 12.Nieuwkamp DJ, Rinkel GJ, Silva R, Greebe P, Schokking DA, Ferro JM. Subarachnoid haemorrhage in patients > or = 75 years: clinical course, treatment and outcome. J Neurol Neurosurg Psychiatry. 2006;77:933–937. doi: 10.1136/jnnp.2005.084350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Nieuwkamp DJ, Setz LE, Algra A, Linn FHH, de Rooij NK, Rinkel GJE. Changes in case fatality of aneurysmal subarachnoid haemorrhage over time, according to age, sex, and region: a meta-analysis. Lancet Neurology. 2009;8:635–642. doi: 10.1016/S1474-4422(09)70126-7. [DOI] [PubMed] [Google Scholar]
- 14.Park J, Woo H, Kang DH, Kim Y. Critical age affecting 1-year functional outcome in elderly patients aged >/= 70 years with aneurysmal subarachnoid hemorrhage. Acta Neurochir (Wien) 2014;156:1655–1661. doi: 10.1007/s00701-014-2133-6. [DOI] [PubMed] [Google Scholar]
- 15.Passel JS, D’Vera C. US Population Projections: 2005–2050. Washington, DC: Pew Research Center; 2008. [Google Scholar]
- 16.Quinn TJ, Dawson J, Lees JS, Chang TP, Walters MR, Lees KR, et al. Time spent at home poststroke: “home-time” a meaningful and robust outcome measure for stroke trials. Stroke. 2008;39:231–233. doi: 10.1161/STROKEAHA.107.493320. [DOI] [PubMed] [Google Scholar]
- 17.Sadamasa N, Koyanagi M, Fukuda H, Chin M, Handa A, Yamagata S. Is aneurysm repair justified for the patients aged 80 or older after aneurysmal subarachnoid hemorrhage? J Neurointerv Surg. 2014;6:664–666. doi: 10.1136/neurintsurg-2013-010951. [DOI] [PubMed] [Google Scholar]
- 18.Scholler K, Massmann M, Markl G, Kunz M, Fesl G, Bruckmann H, et al. Aneurysmal subarachnoid hemorrhage in elderly patients: long-term outcome and prognostic factors in an interdisciplinary treatment approach. J Neurol. 2013;260:1052–1060. doi: 10.1007/s00415-012-6758-1. [DOI] [PubMed] [Google Scholar]
- 19.Sturiale CL, Brinjikji W, Murad MH, Lanzino G. Endovascular treatment of intracranial aneurysms in elderly patients: a systematic review and meta-analysis. Stroke. 2013;44:1897–1902. doi: 10.1161/STROKEAHA.113.001524. [DOI] [PubMed] [Google Scholar]
- 20.Voeks JH, Howard G, Roubin GS, Malas MB, Cohen DJ, Sternbergh WC, 3rd, et al. Age and outcomes after carotid stenting and endarterectomy: the carotid revascularization endarterectomy versus stenting trial. Stroke. 2011;42:3484–3490. doi: 10.1161/STROKEAHA.111.624155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Wachter D, Hans F, Kreitschmann-Andermahr I, Rohde V. Lower incidence of transcranial Doppler and symptomatic vasospasm after aneurysmal subarachnoid hemorrhage and aneurysm clipping in the elderly patient? Neurosurgery. 2011;69:261–266. doi: 10.1227/NEU.0b013e31821d2b49. discussion 266–267. [DOI] [PubMed] [Google Scholar]
- 22.Washington CW, Derdeyn CP, Dacey RG, Jr, Dhar R, Zipfel GJ. Analysis of subarachnoid hemorrhage using the Nationwide Inpatient Sample: the NIS-SAH Severity Score and Outcome Measure. J Neurosurg. 2014;121:482–489. doi: 10.3171/2014.4.JNS131100. [DOI] [PubMed] [Google Scholar]
- 23.Wilson TJ, Davis MC, Stetler WR, Giles DA, Chaudhary N, Gemmete JJ, et al. Endovascular treatment for aneurysmal subarachnoid hemorrhage in the ninth decade of life and beyond. J Neurointerv Surg. 2014;6:175–177. doi: 10.1136/neurintsurg-2013-010714. [DOI] [PubMed] [Google Scholar]
- 24.Xian Y, O’Brien EC, Fonarow GC, Olson DM, Schwamm LH, Hannah D, et al. Patient-Centered Research into Outcomes Stroke Patients Prefer and Effectiveness Research: Implementing the patient-driven research paradigm to aid decision making in stroke care. Am Heart J. 2015;170:36–45. 45 e31–11. doi: 10.1016/j.ahj.2015.04.008. [DOI] [PubMed] [Google Scholar]