Current issues of ACP Journal Club are published in Annals of Internal Medicine


Prognosis

5 factors predicted all-cause mortality and 4 factors predicted major stroke up to 10 years after minor stroke

ACP J Club. 1998 Jul-Aug;129:16. doi:10.7326/ACPJC-1998-129-1-016


Source Citation

Prencipe M, Culasso F, Rasura M, et al. Long-term prognosis after a minor stroke. 10-year mortality and major stroke recurrence rates in a hospital-based cohort. Stroke. 1998 Jan;29:126-32.


Abstract

Question

What are the long-term risk factors for all-cause mortality and major stroke in adults who have had a minor stroke?

Design

Inception cohort.

Setting

A university hospital in Rome, Italy.

Patients

322 consecutive patients (mean age 55 y, 76% men) who were hospitalized within 48 hours after onset of a minor, first-ever stroke with signs and symptoms that lasted > 24 hours but caused little or no lifestyle change (< 3 on the modified Rankin scale) in the first 30 days. Patients with deficits caused by new stroke, cerebral angiography, or carotid endarterectomy within 30 days of stroke were excluded. 94% of patients were followed for 10 years or until death or a major stroke.

Assessment of prognostic factors

Age, sex, type of stroke at baseline, disability at discharge, hypertension, diabetes mellitus, hypercholesterolemia, history of migraine headache or myocardial infarction, use of oral contraceptives, recurrent minor strokes, and nonvalvular atrial fibrillation.

Main outcome measures

All-cause mortality and major stroke.

Main results

96 deaths occurred: 39 from cardiovascular causes (19 from sudden death), 24 from recurrent stroke, 31 from other causes, and 2 from unknown causes. 69 patients had new cerebrovascular events (37 major strokes). Multivariate analysis identified 5 factors that predicted all-cause mortality and 4 that predicted major stroke (Table).

Conclusions

All-cause mortality was increased in adults with minor stroke who were ≥ 65 years of age and had hypercholesterolemia, previous myocardial infarction, nonvalvular atrial fibrillation, and minor disability at discharge. Subsequent major stroke was predicted by recurrent minor strokes, previous myocardial infarction, hypertension, and nonlacunar stroke at baseline.

Source of funding: National Research Council.

For correspondence: Dr. M. Prencipe, Dipartimento di Scienze Neurologiche, Universita degli Studi "La Sapienza," Viale dell'Università 30, 00185 Roma, Italy. FAX 39-6-445-4275.


Table. Hazards ratios (HRs) for long-term (up to 10 y) risk factors for all-cause mortality and major stroke after minor stroke*

All-cause mortality risk factor HR (95% CI)
Age ≥ 65 y 1.07 (1.05 to 1.09)
Hypercholesterolemia 1.8 (1.2 to 2.7)
Previous MI 1.8 (1.1 to 3.1)
Nonvalvular AF 2.0 (1.1 to 3.7)
Minor disability 3.4 (2.2 to 5.2)

Major stroke risk factor HR (CI)
Recurrent minor strokes 2.8 (1.3 to 6.2)
Previous MI 2.9 (1.3 to 6.8)
Hypertension 3.0 (1.4 to 6.4)
Nonlacunar stroke 3.1 (1.9 to 4.6)

*AF = atrial fibrillation; MI = myocardial infarction.


Commentary

The study by Prencipe and colleagues provides useful long-term follow-up data on a well-characterized group of patients with minor stroke. However, as with other hospital-based studies, the external validity of the findings is limited by selection effects. Neither the mean age of 55 years nor the preponderance of men is typical of patients with stroke, even those without substantial disability. Thus, the absolute risks for both death (3.2%/y) and recurrent stroke (1.4%/y) are considerably lower than in other community-based studies (about 8% and 7%/y) (1) and clinic-based studies (about 5% and 4%/y) (2). Such risks are also likely to be higher in patients with residual disability.

During the 10-year follow-up period, the risk for death was nearly 3 times that for disabling stroke, and deaths caused by nonvascular events were more common than fatal strokes. Thus, in this relatively young population, the potential benefits of specific, secondary preventive measures for stroke (especially expensive interventions, such as carotid surgery) are limited. Such interventions should certainly target groups at highest risk (e.g., patients with hypertension, nonlacunar stroke, and recurrent minor stroke), although simple measures, such as aspirin and smoking cessation, should be generally recommended.

If the priority is to prevent premature death rather than stroke, the best strategy seems to be to control general cardiovascular risk factors, such as hypercholesterolemia. Although patients with major disabilities were excluded, those with even minimal lifestyle restriction from their initial stroke were at substantially increased risk for death.Therefore, such patients may derive greater absolute benefits from preventive measures.

Many of the risk factors studied were highly interrelated. Therefore, physicians should not rely too heavily on the specific variables identified as significant predictors by the multivariate (Cox) model, especially when these variables differ from those identified in univariate (log-rank) analysis.

David Barer, MSc, DM
Queen Elizabeth HospitalSheriff Hill, Gateshead, England, UK


References

1. Dennis MS, Burn JP, Sandercock PA, et al. Long-term survival after first-ever stroke: the Oxfordshire Community Stroke Project. Stroke. 1993;24:796-800.

2. Hankey GJ, Slattery JM, Wardlow CP. Long-term survival after first-ever stroke: the Oxfordshire Community Stroke Project. J Neurol Neurosurg Psychiatry. 1991;54:793-802.

3. Burn J, Dennis M, Bamford J, et al. Long-term risk of recurrent stroke after a first-ever stroke. The Oxfordshire Community Stroke Project. Stroke. 1994;25:333-7.