Papers

Prevalence and outcome of symptomatic carotid lesions in young adults

BMJ 1995; 310 doi: https://doi.org/10.1136/bmj.310.6991.1363 (Published 27 May 1995) Cite this as: BMJ 1995;310:1363
  1. Antonio Carolei, professor of neurologya,
  2. Carmine Marini, consultant neurologista,
  3. Patrizia Nencini, consultant neurologistb,
  4. Carlo Gandolfo, professor of neurologyc,
  5. Cristina Motto, research fellowd,
  6. Elietta Zanette, professor of neurologye,
  7. Massimiliano Prencipe, full professor of neurologye,
  8. Cesare Fieschi, full professor of neurology, and the National Research Council Study Groupe
  1. a Clinica Neurologica, Dipartimento di Medicina Interna e Sanita Pubblica, Universita degli Studi di L'Aquila, Ospedale S Maria di Collemaggio, 67100 L'Aquila-Collemaggio, Italy
  2. b Clinica Neurologica, Firenze, Italy
  3. c Clinica Neurologica, Genoa, Italy
  4. d Clinica Neurologica, Ospedale Policlinico, Milan, Italy
  5. e Dipartimento di Scienze Neurologiche, Policlinico Umberto I, Rome, Italy
  1. Members of the National Research Council Study Group who participated in the study are listed at the end of this report. Correspondence to: Professor Carolei.
  • Accepted 24 March 1995

Abstract

Objective: To estimate the prevalence and outcome of symptomatic internal carotid artery lesions in young adults.

Design: Multicentre hospital based observational study with five year follow up.

Setting: Seven neurological departments in northern and central Italy.

Subjects: 240 patients (115 men) aged 15-44 with a recent transient ischaemic attack or stroke in the carotid territory.

Main outcome measures: (a) Prevalence of symptomatic internal carotid artery stenosis or occlusion detected by continuous wave Doppler ultrasonography at entry; (b) incidence rates of cerebral, cardiac, and non-vascular death; non-fatal stroke; and non-fatal myocardial infarction.

Results: Carotid stenoses of 50-99% and occlusions were found in 38 patients (15.8%). Both conditions were significantly more frequent in patients aged over 35 and in those with hypertension, diabetes mellitus, and stroke at entry. The standardised mortality ratio at five years was 10.5 (95% confidence interval 5.0 to 19.3). Survival of patients with stenoses of 0-49% and occlusions was significantly better than that of patients with stenoses of 50-99%. Carotid stenosis of 50-99% was an independent predictor of death (hazard ratio 7.9; 95% confidence interval 2.2 to 29) and non-fatal stroke (hazard ratio 7.4; 1.5 to 37.4).

Conclusions: The prevalence of carotid stenosis or occlusion in young adults after a cerebrovascular event is low. Though patients with high grade symptomatic carotid stenosis are at risk of non-fatal and fatal events, patients with internal artery occlusion apparently have a benign prognosis.

Key messages

  • Key messages

  • Young adults with a history of transient ischaemic attack and stroke have a low prevalence of high grade symptomatic carotid artery stenosis and occlusion

  • The five year prognosis is apparently benign in patients aged 15-44 with internal carotid artery occlusion but poor (in terms of fatal and non-fatal events) in the presence of high grade carotid stenosis

  • Primary and secondary medical prevention of cerebral ischaemia based on routine Doppler examinations should be initiated early in subjects at risk

  • Endarterectomy should be considered even in young adults with ipsilateral carotid artery stenosis of 70-99% to halt the progression of arterial lesions and improve prognosis

Introduction

High grade stenoses and occlusions of the internal carotid artery occur in 30-50% of patients with cerebral ischaemia in the carotid territory.1 Both are commonly associated with age, male sex, family history of stroke, obesity, smoking, hyperlipidaemia, diabetes mellitus, and coronary heart disease.2 3 The prognosis of severe stenoses is worse than that of the less tight lesions, with a 14% annual recurrence rate of stroke and a 9% mortality.4

Patients under 45 have a low prevalence of carotid lesions.5 6 7 8 9 Early atherosclerotic disease is the cause of most such lesions, though rare non-atherosclerotic causes are seldom represented.1 5 6 7 8 Long term follow up studies of young adults with focal cerebral ischaemia show the good outcome of patients without atherosclerotic lesions of neck arteries.5 7 10 11 In the presence of carotid stenosis the stroke recurrence rate increases to 20% after an average of 5.6 years of follow up.12

This report describes the prevalence and outcome of symptomatic internal carotid artery stenoses and occlusions, assessed by continuous wave Doppler ultrasonography, in a prospective hospital based series of young adults (age range 15-44) with either transient ischaemic attack or stroke at entry.

Patients and methods

Between April 1984 and March 1988 all patients aged 15-44 who had their first transient ischaemic attack or stroke in the ipsilateral carotid artery territory within eight weeks before admission were investigated in seven neurological departments in northern and central Italy.9 The study protocol included a clinical and laboratory evaluation of vascular risk factors, brain computed tomography, cardiac evaluation with chest radiography, electrocardiography, and transthoracic echocardiography. Selective carotid angiography of the symptomatic vascular territory was performed in most cases.

The presence of internal carotid artery stenosis or occlusion was assessed by continuous wave Doppler ultrasonography (DUD 800, Delalande, France). On the basis of the results patients were divided into three categories: 0-49% stenosis (<6 kHz) of lumen diameter, 50-99% stenosis (6-20 kHz) of lumen diameter, and occlusion.13 Carotid occlusion was diagnosed in the presence of low flow velocities in the common carotid artery and no detectable Doppler signals distal to the occlusion. Interrater agreement among the trained examiners (verified before the study) resulted in the substantial range ((kappa) index 0.75; 95% confidence interval 0.48 to 1.00).

All patients were followed up for five years with an annual clinical evaluation or a complete phone interview by means of a semistructured questionnaire. Hypertension was defined as diastolic blood pressure over 90 mm Hg (when measured out of the acute phase); diabetes mellitus was noted from the medical records before or at recruitment; obesity referred to body weight exceeding 20% of ideal (Lorentz formula).9 Primary end points included cerebral, cardiac, and non-vascular death; non-fatal stroke; and non-fatal myocardial infarction.14 Details of all deaths were assessed from hospital records, necropsy reports, or general practitioners' reports.

Cerebral death was defined as that occurring within 30 days after a severe qualifying stroke with clinically proved deterioration. Cardiac death was defined as death occurring unwitnessed or without any preceding symptoms (sudden death) or associated with cardiac symptoms suggestive of myocardial infarction, severe congestive heart failure, and acute pulmonary oedema. Stroke was defined as a focal neurological deficit of sudden onset persisting beyond 24 hours in a surviving patient.15 Myocardial infarction was diagnosed by the presence of at least two of the following: characteristic ischaemic chest pain for more than 30 minutes, typical increase in serum enzyme activities, and new pathological Q waves on electrocardiography. The only secondary end point considered was transient ischaemic attack, defined as a focal neurological deficit resolving completely within 24 hours.15 Its recurrence was evaluated only in patients with transient ischaemic attack on admission.

During follow up no compulsory treatment was adopted. Strict medical control and optimal treatment of the recognised vascular risk factors were encouraged.

Statistical analysis—Univariate statistical analysis was by Pearson's χ2 test. χ2 Test for trend was also performed to emphasise the linear relation between vascular risk factors and severity of carotid disease. Patients' event free survival was assessed by the Kaplan-Meier method according to the degree of carotid disease. Comparisons between the survival distribution for the different subgroups were performed by log rank test. Survival rate was also compared with that of a standard population, matched for age and sex, according to Italian life tables,16 the standardised mortality ratio being computed with exact 95% confidence interval. Univariate hazard ratios were calculated separately for death, non-fatal stroke, and non-fatal myocardial infarction by the Cox proportional hazards regression, stenosis of 0-49% being taken as the reference category. Adjusted hazard ratios were obtained with age, sex, cardiac disease, and transient ischaemic attack or stroke at entry being included as covariates.

Results

BASELINE CHARACTERISTICS

Two hundred and forty patients were included (125 women, 115 men; mean age 36.0 (SD 7.3) years). Most (n=138) presented with stroke as the first episode. Thirty eight patients (15.8%) had ipsilateral internal carotid artery stenosis of 50-99% (n=30) or occlusion (n=8) detected by Doppler ultrasonography. Seven stenoses (23.3%) were bilateral. Selective carotid angiography was performed on 175 patients (73%). In 144 (82%) cases there was agreement between the angiographic and ultrasound findings. The coefficient of agreement adjusted for chance ((kappa) index) was 0.62 (95% confidence interval 0.48 to 0.76), indicating substantial concordance between the two methods.

Most patients (73%) were taking antiplatelet drugs (n=153) or anticoagulants (n=22) during follow up. Five patients with transient ischaemic attack and four with stroke at entry underwent carotid endarterectomy because of symptomatic preocclusive carotid stenosis (n=2), ipsilateral stenosis of greater (n=3) or less than 70% (n=1), bilateral stenosis of greater than 70% (n=2), and post-stenotic aneurysm (n=1), all confirmed by angiography.

Table I shows the linear association of age over 35 (P=0.05 (χ2 test); P=0.01 (χ2 test for trend)), hypertension (P=0.0002; P=0.00006), stroke at entry (P=0.012; P=0.003), and diabetes mellitus (P=0.004; P=0.06) with the severity of carotid disease.

TABLE I

Risk factor distribution in 240 patients with internal carotid artery stenosis 0-49% and 50-99% and occlusion. Results expressed as numbers (percentages) of patients

View this table:

ANALYSIS OF OUTCOME EVENTS

The five year clinical follow up (table II), based on information reported at the annual clinical evaluation (162 patients; 68%) or collected by phone interview (76; 32%), was available in 238 patients. Eight men and two women died (4.2%). Of the nine patients who underwent endarterectomy, one had a contralateral transient ischaemic attack five years after surgery, one died in a car accident, and one with bilateral stenosis had a stroke eight months after surgery and committed suicide four months later. The observed survival rate (95.8%) was worse than expected for the age and sex matched standard Italian population (99.6%) (standardised mortality ratio 10.5 (95% confidence interval 5.04 to 19.32)) (fig 1).

TABLE II

Summary of end point occurrences in 240 patients with internal carotid artery stenosis of 0-49% and 50-99% and occlusion. Results expressed as numbers (percentages) of patients

View this table:
FIG 1
FIG 1

Observed and expected survival curves

Table III shows the annual recurrence rates for cerebral and cardiac events in the three subgroups of carotid lesions and in the overall series. The cumulative risk of death among patients with carotid stenoses of 50-99% was higher (P=0.0012 (log rank test)) than among patients with stenoses of 0-49% and occlusions (fig 2).

TABLE III

Percentage annual recurrence rates (95% confidence intervals) of new transient ischaemic attack, non-fatal and fatal stroke, and cardiac event in 238 patients with internal carotid artery stenosis of 0-49% and 50-99% and occlusion

View this table:
FIG 2
FIG 2

Kaplan-Meier cumulative risk of death in patients with internal carotid artery stenoses of 0-49% and 50-99% and occlusion

Carotid stenosis of 50-99% was associated with significant hazard ratios—namely, 7.2 (95% confidence interval 2.1 to 25.0) for all deaths and 7.5 (1.5 to 37.1) for non-fatal stroke (table IV). These hazard ratios remained significant after adjusting for other potential outcome predictors.

Discussions

Doppler ultrasonography has so far been formally assessed as a diagnostic tool and is likely to remain the first line screening technique for carotid lesions, while the most recently developed ultrasound methods and angiography may clarify the diagnosis in equivocal cases. The conclusions we have drawn apply to patients with continuous wave Doppler evaluation performed according to standardised criteria.13 The substantial agreement ((kappa) =0.62) between selective carotid angiography and ultrasound in 73% of patients (175/240) allowed an unbiased estimate of both prevalence and outcome of symptomatic internal carotid artery lesions in our hospital based series.

CAROTID LESIONS AND ASSOCIATED RISK FACTORS

The 15.8% prevalence (38 cases) of ipsilateral carotid stenoses of 50-99% and occlusions agreed with other reports referring to younger case series5 6 7 8 but was lower than that reported in elderly patients with cerebral ischaemia.1 Bilateral severe carotid stenoses (seven of 30 cases; 23.3%) were more frequent than reported (8.5%).17 The association between the degree of carotid stenosis and the severity of cerebral ischaemia at entry was confirmed in our series.4 In addition, severe carotid stenoses and occlusions were associated with age over 35, hypertension, and diabetes mellitus.

The lack of an association with cigarette smoking, hypercholesterolaemia, and hypertriglyceridaemia, found in elderly patients,2 might be due to lack of sensitivity of continuous wave Doppler ultrasonography, which did not identify non-stenosing carotid plaques and could not rule out carotid dissection or fibromuscular dysplasia. Alternatively, some vascular risk factors might require longer exposure to act as true atherogenic determinants of carotid disease progression. Their association with the extent and severity of carotid disease was stronger in elderly than in young patients.3 Different criteria to define and assess vascular risk factors or to categorise carotid abnormalities might also explain discrepancies with other studies.18

LONG TERM PROGNOSIS

The standardised mortality ratio (10.5) was higher than reported in other studies (ratios between 1.5 and 2.0) with a similar follow up period and in comparisons with age and sex matched population controls, reflecting the high risk of death among young adults with cerebral ischaemia in the carotid territory.14 19

Annual incidence rates of cerebral (0.6%) and cardiac events (0.5%) were lower than reported in elderly people (4.2% and 2.9% respectively)14 but close to those found in patients under 30 (0.7% and nil).2 The five year mortality was 16.7% in patients with severe carotid stenosis. This figure was roughly comparable to the 15% reported in older patients4 and 20% in patients younger than 50 followed up for an average of 5.6 years.12 The small subgroup of young adults with carotid occlusion showed the apparently favourable prognosis reported in elderly subjects with either symptomatic or asymptomatic carotid occlusion.5 20 21 22

In our series severe carotid stenosis of 50-99% contributed independently to increased mortality and the recurrence of non-fatal stroke in young adults (table IV).

TABLE IV

Univariate and adjusted hazard ratios for death, non-fatal stroke, and non-fatal myocardial infarction by Cox regression analysis

View this table:

CONCLUSIONS

This study shows that young adults with high grade symptomatic carotid stenosis are at risk of non-fatal and fatal events, whereas patients with carotid occlusion apparently have a benign prognosis. Therefore, in subjects at risk primary and secondary medical prevention should be initiated early.9 Furthermore, other workers suggest that endarterectomy should be considered, after thorough evaluation, even in young adults with ipsilateral high grade symptomatic carotid stenosis of 70-99% to halt the progression of arterial lesions and improve prognosis.17 23

This study was funded by the National Research Council (grant 94.004671.PF40-95.2.572). We thank Dr N Bornstein, Professor J D Easton, and Professor M Motolese for comments and suggestions during the preparation of the manuscript.

Members of the National Research Council Study Group who participated were: M Frontoni (University of Rome); D Inzitari (University of Florence); C Finocchi (University of Genoa); R Totaro (University of L'Aquila); G Landi (Ospedale Maggiore, Milan); L De Zanche (Padua Hospital); U Scoditti (Parma Hospital).

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
View Abstract