Long-term follow-up of unruptured intracranial aneurysms repaired in California

Clinical article

Full access

Object

Using a database that enabled longitudinal follow-up, the authors assessed the long-term outcomes of unruptured cerebral aneurysms repaired by clipping or coiling.

Methods

An observational analysis of the California Office of Statewide Health Planning and Development (OSHPD) database, which follows patients longitudinally in time and through multiple hospitalizations, was performed for all patients initially treated for an unruptured cerebral aneurysm in the period from 1998 to 2005 and with follow-up data through 2009.

Results

Nine hundred forty-four cases (36.5%) were treated with endovascular coiling, 1565 cases (60.5%) were surgically clipped, and 76 cases were treated with both coiling and clipping. There was no significant difference in any demographic variable between the two treatment groups except for age (median: 55 years for the clipped group, 58 years for the coiled group, p < 0.001). Perioperative (30-day) mortality was 1.1% in patients with coiled aneurysms compared with 2.3% in those with clipped aneurysms (p = 0.048). The median follow-up was 7 years (range 4–12 years). At the last follow-up, 153 patients (16.2%) in the coiled group had died compared with 244 (15.6%) in the clipped group (p = 0.693). The adjusted hazard ratio for death at the long-term follow-up was 1.14 (95% CI 0.9–1.4, p = 0.282) for patients with endovascularly treated aneurysms. The incidence of intracranial hemorrhage was similar in the two treatment groups (5.9% clipped vs 4.8% coiled, p = 0.276). One hundred ninety-three patients (20.4%) with coiled aneurysms underwent additional hospitalizations for aneurysm repair procedures compared with only 136 patients (8.7%) with clipped aneurysms (p < 0.001). Cumulative hospital costs per patient for admissions involving aneurysm repair procedures were greater in the clipped group (median cost $98,260 vs $81,620, p < 0.001) through the follow-up.

Conclusions

For unruptured cerebral aneurysms, an observed perioperative survival advantage for endovascular coiling relative to that for surgical clipping was lost on long-term follow-up, according to data from an administrative database of patients who were not randomly allocated to treatment type. A cost advantage of endovascular treatment was maintained even though endovascularly treated patients were more likely to undergo subsequent hospitalizations for additional aneurysm repair procedures. Rates of aneurysm rupture following treatment were similar in the two groups.

Abbreviations used in this paper:ISAT = International Subarachnoid Aneurysm Trial; ISUIA = International Study of Unruptured Intracranial Aneurysms; OSHPD = Office of Statewide Health Planning and Development.

Abstract

Object

Using a database that enabled longitudinal follow-up, the authors assessed the long-term outcomes of unruptured cerebral aneurysms repaired by clipping or coiling.

Methods

An observational analysis of the California Office of Statewide Health Planning and Development (OSHPD) database, which follows patients longitudinally in time and through multiple hospitalizations, was performed for all patients initially treated for an unruptured cerebral aneurysm in the period from 1998 to 2005 and with follow-up data through 2009.

Results

Nine hundred forty-four cases (36.5%) were treated with endovascular coiling, 1565 cases (60.5%) were surgically clipped, and 76 cases were treated with both coiling and clipping. There was no significant difference in any demographic variable between the two treatment groups except for age (median: 55 years for the clipped group, 58 years for the coiled group, p < 0.001). Perioperative (30-day) mortality was 1.1% in patients with coiled aneurysms compared with 2.3% in those with clipped aneurysms (p = 0.048). The median follow-up was 7 years (range 4–12 years). At the last follow-up, 153 patients (16.2%) in the coiled group had died compared with 244 (15.6%) in the clipped group (p = 0.693). The adjusted hazard ratio for death at the long-term follow-up was 1.14 (95% CI 0.9–1.4, p = 0.282) for patients with endovascularly treated aneurysms. The incidence of intracranial hemorrhage was similar in the two treatment groups (5.9% clipped vs 4.8% coiled, p = 0.276). One hundred ninety-three patients (20.4%) with coiled aneurysms underwent additional hospitalizations for aneurysm repair procedures compared with only 136 patients (8.7%) with clipped aneurysms (p < 0.001). Cumulative hospital costs per patient for admissions involving aneurysm repair procedures were greater in the clipped group (median cost $98,260 vs $81,620, p < 0.001) through the follow-up.

Conclusions

For unruptured cerebral aneurysms, an observed perioperative survival advantage for endovascular coiling relative to that for surgical clipping was lost on long-term follow-up, according to data from an administrative database of patients who were not randomly allocated to treatment type. A cost advantage of endovascular treatment was maintained even though endovascularly treated patients were more likely to undergo subsequent hospitalizations for additional aneurysm repair procedures. Rates of aneurysm rupture following treatment were similar in the two groups.

The clinical management of unruptured intracranial aneurysms is controversial. Studies looking at the natural history of unruptured aneurysms suggest that the rupture risk for most aneurysms is quite low.9,11 Nevertheless, the high morbidity associated with rupture has led to a steady increase in the volume of procedures performed for unruptured aneurysms in the United States.8 A prospective randomized trial comparing different surgical treatments and conservative management is unlikely given concerns over clinical equipoise.11

The perioperative risks of endovascular coiling have been repeatedly demonstrated to be less than those associated with aneurysm clipping.1,3,5–7,10,13 However, the durability of endovascular treatment for long-term protection of aneurysm rupture has been questioned. Endovascular coiling is known to be associated with a higher rate of subsequent retreatment episodes. Whether the short-term advantages of endovascular treatment relative to clipping are maintained in the long-term remains unknown.

Large scale reviews of clinical databases like the Nationwide Inpatient Sample database have confirmed the short-term outcome advantages of endovascular therapy seen in clinical trials but have been limited to the assessment of outcome at discharge.1,3,5–7 The Office of Statewide Health Planning and Development (OSHPD) of California maintains a database of all patient admissions to nonfederal hospitals within the state. Patients are assigned unique identifiers that allow them to be followed longitudinally in time over multiple admissions and across different hospitals until death. In the present study, we used the OSHPD data to evaluate the long-term outcome differences between endovascular coiling and surgical clipping for treating unruptured cerebral aneurysms in California in the period from 1998 to 2005.

Methods

Study Cohort

The OSHPD database was searched for all patients treated for unruptured cerebral aneurysms between January 1998 and December 2005. All other hospital admissions from 1995 through 2009 for these same patients were collected to obtain pre- and postoperative histories. Deaths were ascertained using the Social Security Death Index, which was linked to the OSHPD records. We searched the OSHPD discharge database for patients with a diagnosis of “unruptured cerebral aneurysm” (ICD-9-CM code 437.3) and a procedure code for aneurysm repair during their admission. The ICD-9-CM procedure code of “clipping of aneurysm” (ICD-9-CM code 39.51) was used to identify surgically repaired aneurysms. Endovascular repair of an aneurysm was identified by the procedure codes “other repair of aneurysm” (ICD-9-CM code 39.52), “endovascular repair or occlusion of head and neck vessels” (ICD-9-CM code 39.72), and “other endovascular repair (of aneurysm) of other vessels” (ICD-9-CM code 39.79). Patients were excluded from the study if they had a diagnosis code of “subarachnoid hemorrhage” (ICD-9-CM code 430) or “intracerebral hemorrhage” (ICD-9-CM code 431) on the index admission or any admission previous to the index admission. Patients were also excluded if they had one of the procedure codes for aneurysm repair prior to the 1998 time period to limit outcome analysis to newly treated aneurysms.

Independent variables, including insurance status, race, sex, age, hospital of admission, area of patient residence, admission type, and source of admission, were coded within the OSHPD. Median household incomes by zip codes were collected from the 2011 United States Census Bureau. They were categorized to the nearest $5000 increment that approximated quartile distributions. The volume of unruptured aneurysms treated at each hospital was calculated based on the number of procedures for aneurysm repair with the concurrent ICD-9-CM code for unruptured aneurysm over the time period investigated averaged out per year. Hospitals were divided into three groups: those performing, on average, 1 or fewer unruptured aneurysm repairs per year (97 hospitals); those performing 2–14 per year (49 hospitals), and those performing more than 15 per year (7 hospitals). Hospital volume categories were constructed to balance both the number of aneurysms treated and the number of hospitals. Age was evaluated as a continuous variable. The Charlson Comorbidity Index was used to assess patient medical comorbidities according to the method established by Romano and colleagues.16

Study End Points

The primary end point examined was death from any cause through the follow-up period ending on December 31, 2009. We also examined the 30-day perioperative mortality associated with the different treatments. Secondary end points included the development of intracranial hemorrhage subsequent to aneurysm repair and the number of additional hospitalizations for aneurysm repair procedures following initial treatment.

Statistical Analysis

For statistical comparisons between the treatment groups of coiling and clipping, Pearson chi-square tests were used to compare categorical variables, and Student t-tests were used to compare continuous variables. Patients who underwent both clipping and coiling during their initial hospitalization for aneurysm repair were not included in the demographic statistical analyses. Kaplan-Meier curves were calculated for each treatment group, and these groups were compared using the log-rank test. Hazard ratios were calculated using a univariate and multivariate Cox-proportional hazards model adjusted for treatment type, sex, age, duration of hospital stay, race, insurance status, median income by zip code of residence, admission type, admission source, Charlson Comorbidity Index, and hospital volume of unruptured aneurysm treatment. Patients with missing data (78 patients [3.1%]) were not included in the multivariate analysis. A p value < 0.05 was considered statistically significant. All p values are 2-tailed. Statistical tests were performed using STATA software (STATA/SE 11.2, StataCorp LP) and R programming (version 2.144.1) with the Survival package.

Results

Patient Population

There were 3198 index admissions for the repair of unruptured intracranial aneurysms in California between January 1998 and December 2005. Six hundred eighty-nine were excluded because of hospitalizations prior to 1998 with diagnostic codes for subarachnoid hemorrhage or procedure codes indicating prior repair of the aneurysm. The median duration of follow-up for the cohort overall was 7.2 ± 2.3 years (range 4–12 years). Seventy-six patients (3%) were treated with both clipping and coiling during their initial hospitalization for aneurysm repair. Of the aneurysm cases included for analysis, 1565 (60.5%) were treated with surgical clipping and 944 (36.5%) were treated with endovascular coiling. Age was the only significant demographic difference between these two treatment groups (Table 1). The average age of patients with coiled and clipped aneurysms was 58 ± 14.1 and 55 ± 12.6 years, respectively (p < 0.001). An actuarial analysis of the probability of death during the follow-up period based on age revealed no statistical difference between the two treatment groups (average probability 0.101 for clipped aneurysms and 0.112 for coiled lesions, p = 0.489, Mann-Whitney test).2 Thus, age differences between the two treatment groups did not statistically bear on the primary outcome measure.

TABLE 1:

Clinical characteristics of patients with unruptured intracranial aneurysms who underwent either surgical or endovascular treatment in California between 1998 and 2005*

VariableNo. (%)p Value
ClippingCoilingBoth
no. of cases156594476
male sex397 (25.4)240 (25.4)20 (26.3)
age<0.001
 median555856
 range0–1030–9224–85
race0.220
 white992 (63.4)641 (67.9)53 (69.7)
 African American116 (7.4)50 (5.3)7 (9.2)
 Hispanic304 (19.4)131 (13.9)9 (11.8)
 Asian/Pacific Islander99 (6.3)75 (7.9)5 (6.6)
 Native American35 (2.2)35 (3.7)1 (1.3)
primary payer0.224
 Medicare358 (22.9)311 (32.9)19 (25.0)
 Medicaid154 (9.8)45 (4.8)7 (9.2)
 private884 (56.5)518 (54.9)46 (60.5)
 self-pay15 (1.0)13 (1.4)2 (2.6)
 other74 (4.7)39 (4.1)2 (2.6)
median household income0.213
 <$45,000274 (17.5)155 (16.4)18 (23.7)
 $45,000–$59,999469 (30.0)224 (23.7)19 (25.0)
 $60,000–$74,999337 (21.5)219 (23.2)14 (18.4)
 >$75,000456 (29.1%)330 (35.0)22 (28.9)
admission type0.096
 scheduled1161 (74.2)723 (76.6)61 (80.3)
 unscheduled402 (25.7)221 (23.4)15 (19.7)
 unknown1 (0.1)0 (0.0)0 (0.0)
admission source0.2133
 home1465 (93.6)845 (89.5)71 (93.4)
 acute care hospital70 (4.5)68 (7.2)4 (5.3)
 long-term care facility10 (0.6)2 (0.2)0 (0.0)
 other20 (1.3)29 (3.1)1 (1.3)
CCI score0.220
 0–11117 (71.4)697 (73.8)53 (69.7)
 2265 (16.9)157 (16.6)10 (13.2)
 3125 (8.0)59 (6.3)6 (7.9)
 442 (2.7)21 (2.2)6 (7.9)
 5+16 (1.0)10 (1.1)1 (1.3)
hospital aneurysm vol per yr0.1991
 1186 (11.9)16 (1.7)1 (1.3)
 2–15543 (34.7)272 (28.8)19 (25.0)
 15+836 (53.4)656 (69.5)56 (73.7)

CCI = Charlson Comorbidity Index.

Comparison between clipping and coiling.

Primary End Point

There were no significant long-term differences in overall mortality between the two treatment groups. Two hundred forty-four patients (15.6%) with clipped aneurysms were dead at the end of the follow-up compared with 153 dead patients (16.2%) whose aneurysms were originally coiled. While the patients with endovascularly treated aneurysms had improved short-term outcomes (30-day survival, 98.9% compared with 97.7% in the clipped cohort, p = 0.048), t he survival curves between the two groups converged, with prolonged follow-up even demonstrating a beneficial trend toward clipping (Fig. 1). The adjusted hazard ratio for death at the long-term follow-up was 1.14 (95% CI 0.9–1.4, p = 0.254) for endovascularly treated aneurysms (Table 2). There was no difference between the groups in rates of mortality due to subarachnoid hemorrhage or cerebrovascular accidents (p = 0.8426; Table 3). Endovascularly treated patients were more likely to die of cancer than patients who had undergone clipping (p = 0.0424).

Fig. 1.
Fig. 1.

Kaplan-Meier graphs of cumulative mortality. Hashes indicate the extent of follow-up for individual patients, which act as right-censored events. A p value was obtained from the log-rank test of overall survival. Inset shows a more detailed cumulative mortality through the perioperative period.

TABLE 2:

Univariate and multivariate Cox proportional-hazards modeling for overall survival after treatment for unruptured aneurysm*

VariableUnivariateMultivariate
HR95% CIp ValueHR95% CIp Value
male sex0.8120.656–1.0050.0550.7650.614–0.9530.017
age1.0591.051–1.068<0.0011.0531.043–1.063<0.001
treatment type
 clipping(1.00)(1.00)
 coiling1.2060.984–1.4780.0711.1370.912–1.4180.254
 both1.1120.636–1.9420.7101.1900.673–2.1040.551
race
 white(1.00)(1.00)
 African American0.7820.516–1.1860.2471.0040.651–1.5480.987
 Hispanic0.6870.511–0.9220.0130.7710.565–1.0530.102
 Asian/Pacific Islander1.0770.752–1.5440.6851.2030.834–1.7340.323
 Native American0.5850.276–1.2380.1610.7120.335–1.5160.379
primary payer
 private/Medicare(1.00)(1.00)
 Medicaid0.8020.534–1.2040.2871.2790.827–1.9770.269
 self-pay0.2000.028–1.4210.1080.3490.048–2.5120.296
 other0.8090.482–1.3580.4231.1760.676–2.0470.566
median household income
 <$45,000(1.00)(1.00)
 $45,000–$59,9990.7660.579–1.0130.0620.8190.614–1.0920.174
 $60,000–$74,9990.7590.564–1.0210.0680.7430.548–1.0090.057
 >$75,0000.6950.526–0.9180.0100.7730.578–1.0340.083
admission type
 scheduled(1.00)(1.00)
 unscheduled1.3501.093–1.6660.0051.1580.902–1.4870.250
 unknownNA0.995NANANA
admission source
 home(1.00)(1.00)
 acute care hospital1.1320.755–1.6970.5490.9250.579–1.4770.744
 long-term care facility4.9582.212–11.113<0.0011.4490.527–3.9860.472
 other0.9170.434–1.9380.8210.9040.418–1.9550.798
CCI scores
 0–1(1.00)(1.00)
 22.4131.904–3.057<0.0011.9291.512–2.462<0.001
 33.4342.585–4.561<0.0012.8022.091–3.755<0.001
 44.8083.239–7.139<0.0013.4372.272–5.2<0.001
 5+4.7402.583–8.697<0.0013.2781.724–6.233<0.001
hospital aneurysm vol per yr
 1(1.00)(1.00)
 2–150.7610.555–1.0460.0920.7770.556–1.0860.140
 15+0.5840.431–0.7920.0010.6740.48–0.9460.023

NA = not applicable.

TABLE 3:

Comparison of secondary outcomes and causes of death*

VariableNo. (%)p Value
ClipCoilBoth
no. treated156594476
median FU (yrs)7.86.47.4<0.001
 died (% of treated)244 (15.6)153 (16.2)13 (17.1)0.6929
 adtl repairs136 (8.7)193 (20.4)14 (18.4)<0.001
  >1 adtl repairs10 (0.6)51 (5.4)3 (3.9)<0.001
 ICH92 (5.9)45 (4.8)7 (9.2)0.276
 CVA376 (24.0)213 (22.6)21 (27.6)0.409
cause of death (% of deaths)
 accidental7 (2.9)0 (0.0)0 (0.0)0.0467
 cancer45 (18.4)42 (27.5)5 (38.5)0.0457
 cardiovascular48 (19.7)33 (21.6)2 (15.4)0.7015
 cerebrovascular72 (29.5)41 (26.8)4 (30.8)0.5703
 infection11 (4.5)3 (2.0)0 (0.0)0.2646
 other31 (12.7)22 (14.4)2 (15.4)0.6513
 respiratory23 (9.4)12 (7.8)0 (0.0)0.7167
 suicide7 (2.9)0 (0.0)0 (0.0)0.0467
at 4-year FU
 died (% of treated)145 (9.3)106 (11.2)6 (7.9)0.115
 additional repairs118 (7.5)177 (18.8)14 (18.4)<0.001
 ICH78 (5.0)35 (3.7)6 (7.9)0.164
 CVA313 (20.0)176 (18.6)18 (23.7)0.435

adtl = additional; CVA = cerebrovascular accident; FU = follow-up; ICH = intracranial hemorrhage.

Comparison between clipping and coiling (chi-square test).

Multivariate analysis showed risk factors for mortality included female sex, age, treatment at a low-volume hospital, increasing Charlson Comorbidity Index, and having Medicaid insurance (Table 2). Race, median income, and urgency of admission were not significantly associated with mortality.

Secondary Outcomes

Patients whose aneurysms were treated by endovascular methods were much more likely to undergo an additional hospitalization for aneurysm repair after the initial treatment (20.4% vs 8.7% of clipped aneurysms, p < 0.001; Table 3). Many of the patients with coiled aneurysms underwent multiple additional hospitalizations for repair procedures. As a group, the 944 cases initially treated with coiling had 265 (28.1%) additional admissions for aneurysm repairs. This compares to only 149 (9.5%) additional admissions in the group of 1565 initially clipped aneurysms. Among the cases with additional hospitalizations for aneurysm repair, the median time to subsequent hospitalization was 216 days for a coiled aneurysm and 108 days for a clipped lesion. Thirty-day mortality rates following subsequent hospitalizations for repair were higher (3.9% after clipping, 1.9% after coiling) than those after initial treatments.

Crossover between treatment groups at subsequent hospitalizations was rare. Only 47 patients (3%) with initially coiled aneurysms underwent clipping procedures at subsequent hospitalizations, and 51 patients (5.4%) with initially clipped aneurysms later underwent coiling procedures. Exclusion of crossover patients did not affect the results of multivariate analysis.

The incidence of aneurysm rupture as estimated by the incidence of nontraumatic intracranial hemorrhage was not significantly different between clipped and coiled aneurysms during the follow-up period (5.9% to 4.8%, p = 0.276). There was no significant difference in the incidence of cerebrovascular accidents between the treatment groups, including ischemic or hemorrhagic complications, nor was there any correlation between cerebrovascular accidents and additional treatments.

The hospital stay was longer in the clipped population than in the coiled population (median 5 vs 2 days, p < 0.001), as was the median hospital charge ($93,070 vs $68,260, p < 0.001). Cumulative hospital costs for admissions involving aneurysm repair procedures were also greater in the clipped population (median $98,260 vs $81,620, p < 0.001) through the follow-up.

Trends in the Treatment of Unruptured Aneurysms

Between 1998 and 2009 there was a steady increase in the number of unruptured aneurysms being treated (Fig. 2). In 1998, 268 unruptured aneurysms were treated as compared with 596 in 2009. The number of aneurysms treated by clipping remained fairly constant, whereas the number treated by endovascular methods increased dramatically. Therefore, the expansion of endovascular coiling did not come at the expense of clipping volume; rather, the overall number of treated aneurysms increased. The percentage of unruptured aneurysms treated by endovascular coiling was 20.5% in 1998 and increased to a relative proportion of 54% of unruptured aneurysms treated in 2009.

Fig. 2.
Fig. 2.

Trends in the treatment of unruptured aneurysms from 1998 to 2009.

Discussion

Clinical studies and retrospective reviews influencing the management of aneurysms have largely relied on data derived from short-term outcomes. The present report provides long-term follow-up data after discharge from aneurysm repair. Our results showed that the early mortality benefits of endovascular coiling relative to clipping are eroded on long-term follow-up posthospitalization and even trend toward worse survival. This finding agrees with the latest reports from the International Subarachnoid Aneurysm Trial (ISAT), which has continued to follow patients enrolled in the original study.14 Results from the most recent ISAT report show that many of the advantages observed in the early follow-up after endovascular coiling are no longer present in the extended follow-up. At 5 years posttreatment, there was no significant difference in the number of patients living independently between treatment groups despite early differences in outcome.14 Five-year ISAT mortality rates, although still significantly favoring endovascular coiling, are substantially smaller than initially described. When all ISAT patients were included in a survival analysis, with follow-up ranging from 6 to 14 years, there was no longer a significant difference in cumulative mortality between the two treatment groups.

No randomized clinical trial has demonstrated a long-term benefit from coiling over clipping in the treatment of unruptured aneurysms. The ISAT was a prospective randomized trial comparing endovascular coiling with clipping of ruptured aneurysms.12 The discrepancy in perioperative mortality was so disproportionately in favor of coiling that the trial was terminated early. Retrospective studies of administrative databases have shown similar improvements in perioperative outcome favoring endovascular coiling.1,3,5–7 The perioperative mortality rates for endovascular coiling (1.1%) and surgical clipping (2.3%) in our study are consistent with other published mortality rates, which range from 0.5% to 1.7% for endovascular coiling and from 1.2% to 3.5% for aneurysm clipping.1,3,5–7,9,10 The superior perioperative outcomes of aneurysm coiling treatment have been an impetus toward adopting endovascular methods as the first choice therapeutic option for most aneurysms.

The International Study of Unruptured Intracranial Aneurysms (ISUIA) described the natural history of unruptured aneurysms along with the risks associated with aneurysm repair.9 The conclusion of their landmark paper, based on the outcomes of 2621 patients at 53 different centers in North America and Europe, was that the risks related to aneurysm repair greatly exceed the risks of rupture for the majority of unruptured aneurysms. The actual risk of rupture for incidentally discovered aneurysms is the subject of much debate since both clinical practice and studies of patients with subarachnoid hemorrhage suggest that the rupture rates of small aneurysms reported by the ISUIA are artificially low.11,15,18

Despite uncertainty regarding a clinical benefit, the rate of unruptured aneurysm treatment in the United States has dramatically increased over the decade following the ISUIA report,5,8 a trend we also observed in California. Much of the increased volume has come in the form of endovascular coiling, which can be attributed to a combination factors, including an increased rate of aneurysm discovery from advances in imaging technology and the perceived low risks associated with endovascular therapy. Incidentally discovered aneurysms that previously would have been left untreated are now more likely to be considered for treatment through endovascular methods. If the latter is indeed contributing to the increase in unruptured aneurysm treatments, it becomes much more important for the clinician to have an accurate understanding of therapy risks based on long-term results.

One possible reason for the discrepancy between perioperative and long-term mortality rates for clipping and coiling is the higher number of additional repair procedures performed after endovascular coiling. Each subsequent hospitalization for aneurysm repair procedures entails additional perioperative risks in the patient's cumulative treatment course. Another reason relates to differences in underlying comorbidities. The patients treated by endovascular coiling were significantly older than those treated with clipping, although the median age difference of 3 years is not large enough in magnitude to affect observed differences in clinical outcomes. We attempted to limit the potential confounding of comorbidities by calculating each patient's Charlson Comorbidity Index at the time of treatment. There was no significant difference in Charlson scores between the two treatment groups, but it is possible that certain risk factors not captured by the Charlson Morbidity Index, such as aneurysm size, location, and morphology, influenced treatment bias.

We also note that the present work reveals a potential policy implication for the treatment of unruptured aneurysms. We observed that the hospitals that performed higher volumes of unruptured aneurysm treatments had lower mortality for both clipping and coiling, a finding previously noted in the literature.4,17 In the situation of an unruptured aneurysm, more time can be devoted to thinking about evaluation and management in a high-volume center, whereas treatment of a ruptured aneurysm is a more time-sensitive matter.

An advantage of the OSHPD database relative to many other administrative databases, including the Nationwide Inpatient Sample, is the ability to follow patients longitudinally in time through multiple hospitalizations in the state of California. A limitation, however, is that admissions to hospitals outside of California would go unrecognized, as would any death that occurred outside the United States. These occurrences would probably be few and would be randomly and evenly distributed throughout our comparison groups, and thus unlikely to impact our conclusions. Another strength of our study is its access to 100% of the California hospitals and their discharge data. As California is the most populous state in the United States, with a population (37 million) larger than that for the entire country of Canada (34 million), we believe that the findings in our study can be applicable to other populations.

The main limitations of our study are that it is an observational analysis and the patients were not assigned treatments in a randomized manner. Details of aneurysm size, location, morphology, and multiplicity are known risk factors not discernible in the OSHPD database, although they probably significantly influence a selective treatment bias on an individual basis. We also recognize the limitations of administrative databases, which include imperfect coding, reporting inconsistencies, and lack of clinical details.19 However, the agreement in demographic properties, treatment trends, and perioperative outcomes with those in other clinical studies and administrative database reports argues for a reasonable degree of reliability in our data.

Conclusions

This is the first study to evaluate long-term survival outcomes after treatment of unruptured aneurysms on a statewide, multiinstitutional scale, based on data from an administrative database in which patients were not randomly allocated to treatment type. Californians undergoing treatment for unruptured aneurysms between 1998 and 2005 demonstrated that the perioperative survival advantages of endovascular coiling relative to surgical clipping were lost on long-term follow-up (4–12 years). Patients treated by endovascular methods were more likely to undergo subsequent hospitalizations for aneurysm repair procedures, but average cumulative hospital costs per patient for all admissions involving aneurysm repair procedures were less for patients who underwent coiling. Randomized controlled trials that incorporate long-term follow-up are needed to determine the optimal management for unruptured aneurysms.

Disclosure

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Dr. Khalessi is a consultant for Stryker, Penumbra, and Covidien.

Author contributions to the study and manuscript preparation include the following. Conception and design: Carter. Acquisition of data: Marcus, Chang. Analysis and interpretation of data: Gonda, McCutcheon, Marcus, Noorbakhsh, Chen, Chang, Carter. Drafting the article: Gonda. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Gonda. Statistical analysis: Gonda, Chang. Study supervision: Carter.

Portions of this work were presented in abstract form at the 2013 Congress of Neurological Surgeons Annual Meeting held in San Francisco, California, on October 19–23, 2013.

References

  • 1

    Alshekhlee AMehta SEdgell RCVora NFeen EMohammadi A: Hospital mortality and complications of electively clipped or coiled unruptured intracranial aneurysm. Stroke 41:147114762010

  • 2

    Arias E: United States life tables, 2008. Natl Vital Stat Rep 61:1632012

  • 3

    Barker FG IIAmin-Hanjani SButler WEHoh BLRabinov JDPryor JC: Age-dependent differences in short-term outcome after surgical or endovascular treatment of unruptured intracranial aneurysms in the United States, 1996–2000. Neurosurgery 54:18302004

  • 4

    Barnholtz-Sloan JSYu CSloan AEVengoechea JWang MDignam JJ: A nomogram for individualized estimation of survival among patients with brain metastasis. Neuro Oncol 14:9109182012

  • 5

    Brinjikji WRabinstein AANasr DMLanzino GKallmes DFCloft HJ: Better outcomes with treatment by coiling relative to clipping of unruptured intracranial aneurysms in the United States, 2001–2008. AJNR Am J Neuroradiol 32:107110752011

  • 6

    Higashida RTLahue BJTorbey MTHopkins LNLeip EHanley DF: Treatment of unruptured intracranial aneurysms: a nationwide assessment of effectiveness. AJNR Am J Neuroradiol 28:1461512007

  • 7

    Hoh BLRabinov JDPryor JCCarter BSBarker FG II: In-hospital morbidity and mortality after endovascular treatment of unruptured intracranial aneurysms in the United States, 1996–2000: effect of hospital and physician volume. AJNR Am J Neuroradiol 24:140914202003

  • 8

    Huang MCBaaj AADownes KYoussef ASSauvageau Evan Loveren HR: Paradoxical trends in the management of unruptured cerebral aneurysms in the United States: analysis of nationwide database over a 10-year period. Stroke 42:173017352011

  • 9

    International Study of Unruptured Intracranial Aneurysms Investigators: Unruptured intracranial aneurysms—risk of rupture and risks of surgical intervention. N Engl J Med 339:172517331998. (Erratum in N Engl J Med 340:

  • 10

    Johnston SCZhao SDudley RABerman MFGress DR: Treatment of unruptured cerebral aneurysms in California. Stroke 32:5976052001

  • 11

    Komotar RJStarke RMConnolly ES: The natural course of unruptured cerebral aneurysms. Neurosurgery 71:N7N92012

  • 12

    Molyneux AKerr RStratton ISandercock PClarke MShrimpton J: International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 360:126712742002

  • 13

    Molyneux AKerr RStratton ISandercock PClarke MShrimpton J: International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomized trial. J Stroke Cerebrovasc Dis 11:3043142002

  • 14

    Molyneux AJKerr RSBirks JRamzi NYarnold JSneade M: Risk of recurrent subarachnoid haemorrhage, death, or dependence and standardised mortality ratios after clipping or coiling of an intracranial aneurysm in the International Subarachnoid Aneurysm Trial (ISAT): long-term follow-up. Lancet Neurol 8:4274332009

  • 15

    Morita AKirino THashi KAoki NFukuhara SHashimoto N: The natural course of unruptured cerebral aneurysms in a Japanese cohort. N Engl J Med 366:247424822012

  • 16

    Romano PSRoos LLJollis JG: Adapting a clinical comorbidity index for use with ICD-9-CM administrative data: differing perspectives. J Clin Epidemiol 46:107510901993

  • 17

    Sperduto PWKased NRoberge DXu ZShanley RLuo X: Summary report on the graded prognostic assessment: an accurate and facile diagnosis-specific tool to estimate survival for patients with brain metastases. J Clin Oncol 30:4194252012

  • 18

    Vlak MHAlgra ABrandenburg RRinkel GJ: Prevalence of unruptured intracranial aneurysms, with emphasis on sex, age, comorbidity, country, and time period: a systematic review and meta-analysis. Lancet Neurol 10:6266362011

  • 19

    Woodworth GFBaird CJGarces-Ambrossi GTonascia JTamargo RJ: Inaccuracy of the administrative database: comparative analysis of two databases for the diagnosis and treatment of intracranial aneurysms. Neurosurgery 65:2512572009

Article Information

Address correspondence to: David D. Gonda, M.D., Division of Neurosurgery, University of California, San Diego, 200 West Arbor Dr., #8893, San Diego, CA 92103. email: dgonda@ucsd.edu.

Please include this information when citing this paper: published online April 11, 2014; DOI: 10.3171/2014.3.JNS131159.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Kaplan-Meier graphs of cumulative mortality. Hashes indicate the extent of follow-up for individual patients, which act as right-censored events. A p value was obtained from the log-rank test of overall survival. Inset shows a more detailed cumulative mortality through the perioperative period.

  • View in gallery

    Trends in the treatment of unruptured aneurysms from 1998 to 2009.

References

1

Alshekhlee AMehta SEdgell RCVora NFeen EMohammadi A: Hospital mortality and complications of electively clipped or coiled unruptured intracranial aneurysm. Stroke 41:147114762010

2

Arias E: United States life tables, 2008. Natl Vital Stat Rep 61:1632012

3

Barker FG IIAmin-Hanjani SButler WEHoh BLRabinov JDPryor JC: Age-dependent differences in short-term outcome after surgical or endovascular treatment of unruptured intracranial aneurysms in the United States, 1996–2000. Neurosurgery 54:18302004

4

Barnholtz-Sloan JSYu CSloan AEVengoechea JWang MDignam JJ: A nomogram for individualized estimation of survival among patients with brain metastasis. Neuro Oncol 14:9109182012

5

Brinjikji WRabinstein AANasr DMLanzino GKallmes DFCloft HJ: Better outcomes with treatment by coiling relative to clipping of unruptured intracranial aneurysms in the United States, 2001–2008. AJNR Am J Neuroradiol 32:107110752011

6

Higashida RTLahue BJTorbey MTHopkins LNLeip EHanley DF: Treatment of unruptured intracranial aneurysms: a nationwide assessment of effectiveness. AJNR Am J Neuroradiol 28:1461512007

7

Hoh BLRabinov JDPryor JCCarter BSBarker FG II: In-hospital morbidity and mortality after endovascular treatment of unruptured intracranial aneurysms in the United States, 1996–2000: effect of hospital and physician volume. AJNR Am J Neuroradiol 24:140914202003

8

Huang MCBaaj AADownes KYoussef ASSauvageau Evan Loveren HR: Paradoxical trends in the management of unruptured cerebral aneurysms in the United States: analysis of nationwide database over a 10-year period. Stroke 42:173017352011

9

International Study of Unruptured Intracranial Aneurysms Investigators: Unruptured intracranial aneurysms—risk of rupture and risks of surgical intervention. N Engl J Med 339:172517331998. (Erratum in N Engl J Med 340:

10

Johnston SCZhao SDudley RABerman MFGress DR: Treatment of unruptured cerebral aneurysms in California. Stroke 32:5976052001

11

Komotar RJStarke RMConnolly ES: The natural course of unruptured cerebral aneurysms. Neurosurgery 71:N7N92012

12

Molyneux AKerr RStratton ISandercock PClarke MShrimpton J: International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 360:126712742002

13

Molyneux AKerr RStratton ISandercock PClarke MShrimpton J: International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomized trial. J Stroke Cerebrovasc Dis 11:3043142002

14

Molyneux AJKerr RSBirks JRamzi NYarnold JSneade M: Risk of recurrent subarachnoid haemorrhage, death, or dependence and standardised mortality ratios after clipping or coiling of an intracranial aneurysm in the International Subarachnoid Aneurysm Trial (ISAT): long-term follow-up. Lancet Neurol 8:4274332009

15

Morita AKirino THashi KAoki NFukuhara SHashimoto N: The natural course of unruptured cerebral aneurysms in a Japanese cohort. N Engl J Med 366:247424822012

16

Romano PSRoos LLJollis JG: Adapting a clinical comorbidity index for use with ICD-9-CM administrative data: differing perspectives. J Clin Epidemiol 46:107510901993

17

Sperduto PWKased NRoberge DXu ZShanley RLuo X: Summary report on the graded prognostic assessment: an accurate and facile diagnosis-specific tool to estimate survival for patients with brain metastases. J Clin Oncol 30:4194252012

18

Vlak MHAlgra ABrandenburg RRinkel GJ: Prevalence of unruptured intracranial aneurysms, with emphasis on sex, age, comorbidity, country, and time period: a systematic review and meta-analysis. Lancet Neurol 10:6266362011

19

Woodworth GFBaird CJGarces-Ambrossi GTonascia JTamargo RJ: Inaccuracy of the administrative database: comparative analysis of two databases for the diagnosis and treatment of intracranial aneurysms. Neurosurgery 65:2512572009

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 96 96 62
PDF Downloads 68 68 47
EPUB Downloads 0 0 0

PubMed

Google Scholar