Intra-Arterial Thrombolysis for Acute Stroke in Patients 80 and Older: A Comparison of Results in Patients Younger than 80 Years

Subjects and Methods

Results

A total of 114 patients received IA fibrinolytic therapy for acute cerebral ischemia during the study period. Of the cohort, 33 (29%) were very elderly (aged ≥80 years) and 81 (71%) were nonelderly. The mean age for the very elderly group was 86 years (range, 80–97 years) and that for the nonelderly was 64 years (range, 25–79 years). Baseline patient characteristics for the 2 groups are presented in Table 1. Of the 57 baseline characteristics measured, only history of prior stroke ransient ischemic attack (TIA) and weight differed significantly between the 2 groups. The very elderly were more likely to have a history of prior stroke/TIA (42% versus 22%, P = .03) and to weigh less (67 versus 76 kg, P = .02). Median pretreatment NIHSS score was 17 for the very elderly and 16 for the nonelderly.

Treatment variables did not differ significantly between the 2 groups, including no differences in time from last known well to treatment start and frequency of combined IV-IA versus IA therapy alone (Table 2).

Recanalization rates (TIMI 2 or 3) were at least as good in the very elderly (79% versus 68%, P = .25). Across the entire cohort, hemorrhagic transformation occurred in 43 patients (38%). The incidence of any cerebral hemorrhage, asymptomatic hemorrhage, minor symptomatic hemorrhage, and major symptomatic hemorrhage did not differ between the 2 groups (Table 3). There was also no difference between the elderly and nonelderly in hemorrhage location; equal percentages suffered lobar and deep (basal ganglia halamus). The elderly did not have a significantly higher incidence of hemorrhages in territories outside of the zone of infarction.

At day 7, the median NIHSS score was 19 for the elderly and 12 for the nonelderly (P = .01). Mortality during the initial hospitalization was increased 2-fold among the very elderly compared with younger patients (18% versus 10%, P = .01). The very elderly continued to have an increased risk of dying after hospitalization; only 57% of very elderly patients survived to day 90 compared with 80% of the nonelderly (P = .01).

However, a substantial proportion (26%) of the elderly patients achieved an excellent final day 90 outcome (mRS 0–1), though less than younger patients (40%, P = .02; Fig 1). Thirteen percent of very elderly patients had complete recovery with no residual symptoms (mRS = 0). The proportion of patients with a severely disabled (mRS 4–5) final outcome did not differ between the very elderly and younger patients (17% versus 17%).

• Download figure
• Open in new tab
• Download powerpoint

Fig 1.

Day 90 modified Rankin scores in patients aged ≥80 years versus younger patients.

In multivariate analysis, age considered as a continuous variable was an independent predictor of death at day 90 (odds ratio [OR], 1.069; confidence interval [CI], 1.021–1.118; P = .004), along with failure to recanalize (OR, 0.269; CI, 0.099–0.731; P = .01) and pretreatment warfarin use (OR, 0.218; CI, 0.043–1.101; P = .076). The multiple logistic regression model incorporating these variables had a receiver operating characteristic curve (ROC) area of 0.754 and will accurately predict death approximately 75% of the time. In multivariate analysis, age considered as a continuous variable was also an independent predictor of good functional outcome at day 90, along with absence of prior history of stroke/TIA, pretreatment NIHSS score, warfarin use, and highest systolic blood pressure during treatment (Table 4). The multiple logistic regression model incorporating these variables had an ROC area of 0.848 and will accurately predict good functional outcome approximately 85% of the time.

In multivariate analysis, age considered as a dichotomous variable with the division point at 80 years was not a predictor of death or good functional outcome. Graphical analysis of the relation between age quartiles and outcome (Fig 2) suggested a more marked drop in good outcomes at above age 63 than above age 80, and a steady increase in likelihood of fatal outcome across all quartiles.

• Download figure
• Open in new tab
• Download powerpoint

Fig 2.

Day 90 modified Rankin scores by age quartiles.

Discussion

In this large cohort, IA fibrinolysis in patients aged ≥80 years yielded recanalization rates and HT rates equal to that in younger persons, a lower proportion of good functional outcomes, and a higher rate of fatal outcomes.

The rate of recanalization in our series was similar to that achieved in PROACT II and in previous IA cohort series reports.⁷ Although no prior study specifically examined recanalization in the very elderly, our findings accord with that of Arnold et al¹⁶ who did not find a general association between age and recanalization. Our findings suggest that well-described age associated alterations in intrinsic clotting and fibrinolytic systems^17–20 do not alter thrombi responsiveness to locally administered fibrinolytic agents in the very old.

Several factors have been postulated to increase the risk of intracerebral hemorrhage after fibrinolytic therapy in elderly patients, including cerebral amyloid angiopathy,^17,21 fragile vasculature, and impaired rate of tPA clearance.^18,22 Several studies have identified age as an independent risk factor for intracerebral hemorrhage after intravenous fibrinolysis, administered for both myocardial and cerebral ischemia.^23–25 However, the risk of HT after IA fibrinolysis was not higher in the very old in our series. Although it did not specifically examine the elderly, the PROACT II trial²⁶ and the large series of Arnold et al¹⁶ similarly found that increasing age was not an independent predictor of symptomatic hemorrhage. It has been shown previously that independent predictors for HT after IA lysis include presenting NIHSS score, platelet count, time to recanalization, and serum glucose level, but not age.¹² There are several potential explanations for the absence of a relationship between age and HT in ours and prior series, including more careful selection of very elderly patients for intervention and less effect of age-related hemorrhage propensity factors when lytic agent is administered locally rather than systematically. In addition, prior IA studies and our study were adequately powered only to exclude a major, not modest, effect of age on hemorrhage risk.

The lower rate of good functional outcome and higher mortality rate observed in the patients aged ≥80 years after IA lytic therapy accord with many prior studies showing reduced rates of good outcome and higher mortality in elderly than in nonelderly patients treated with conventional, noninterventional therapies and with intravenous tPA.^16,27–29 In our series, these outcomes are clearly not due to a poorer recanalization rate or to a higher risk of intra-cranial hemorrhage. A variety of factors probably contribute to reduced functional outcome, despite successful reperfusion in the elderly, including a higher frequency of prior stroke/TIA, higher frequency of prestroke comorbid conditions and poststroke medical complications, impaired collateral circulation, reduced neuronal reserve, and fewer social supports.

Is IA fibrinolytic therapy worth pursuing in patients aged ≥80 years in clinical trials and compassionate care practice, despite the lower rate of good outcomes and increased rate of mortality? This question cannot be answered definitively by an open case series study. Our own impression is that the outcomes observed compare favorably with how these patients would have fared under noninterventional care, but a randomized trial is required to confirm or disconfirm this impression. One important insight offered by our findings is that the frequency of severely disabled outcomes is not increased by pursuing an interventional course. A common concern of patients, families, and physicians is that aggressive intervention in the very elderly will increase the occurrence of severely disabled outcomes that many consider a worse outcome than death.³⁰ Patients and providers often prefer that treatment options will result either in good functional outcome or in death but not in an intermediate state of severe disability. Our findings suggest that IA fibrinolytic therapy in the very elderly does not result in an increased frequency of severely disabled outcomes compared with younger persons and achieves good functional outcome in a quarter of patients. These findings can be used in informing patients and families during compassionate care decision-making and in the design of clinical trials in patients aged 80 years or older.

Limitations of the present study include the retrospective, single center design. The treatment approaches used were somewhat heterogeneous, including different thrombolytic agents and routes over a 9-year period, but treatment strategies did not differ in patients aged ≥80 years and their younger counterparts.

Conclusion

Our results indicate that recanalization is achieved with IA fibrinolytics as frequently in very elderly patients as in younger patients, without higher risks of hemorrhage. Although mortality is higher, a substantial minority of the patients aged ≥80 years achieved good outcomes with IA thrombolysis. Further investigation of IA fibrinolytic therapy in the very elderly is merited to develop improved methods to identify, before therapy starts, which patients are capable of achieving a good functional outcome if recanalization is achieved. Furthermore, our data suggest that patients aged 80 years or older should not be excluded from acute stroke trials using thrombolytic drugs or other strategies.