Sally C. Howard

Change in stroke incidence, mortality, case-fatality, severity, and risk factors in Oxfordshire, UK from 1981 to 2004 (Oxford Vascular Study)

BACKGROUND The incidence of stroke is predicted to rise because of the rapidly ageing population. However, over the past two decades, findings of randomised trials have identified several interventions that are effective in prevention of stroke. Reliable data on time-trends in stroke incidence, major risk factors, and use of preventive treatments in an ageing population are required to ascertain whether implementation of preventive strategies can offset the predicted rise in stroke incidence. We aimed to obtain these data. METHODS We ascertained changes in incidence of transient ischaemic attack and stroke, risk factors, and premorbid use of preventive treatments from 1981-84 (Oxford Community Stroke Project; OCSP) to 2002-04 (Oxford Vascular Study; OXVASC). FINDINGS Of 476 patients with transient ischaemic attacks or strokes in OXVASC, 262 strokes and 93 transient ischaemic attacks were incident events. Despite more complete case-ascertainment than in OCSP, age-adjusted and sex-adjusted incidence of first-ever stroke fell by 29% (relative incidence 0.71, 95% CI 0.61-0.83, p=0.0002). Incidence declined by more than 50% for primary intracerebral haemorrhage (0.47, 0.27-0.83, p=0.01) but was unchanged for subarachnoid haemorrhage (0.83, 0.44-1.57, p=0.57). Thus, although 28% more incident strokes (366 vs 286) were expected in OXVASC due to demographic change alone (33% increase in those aged 75 or older), the observed number fell (262 vs 286). Major reductions were recorded in mortality rates for incident stroke (0.63, 0.44-0.90, p=0.02) and in incidence of disabling or fatal stroke (0.60, 0.50-0.73, p<0.0001), but no change was seen in case-fatality due to incident stroke (17.2% vs 17.8%; age and sex adjusted relative risk 0.85, 95% CI 0.57-1.28, p=0.45). Comparison of premorbid risk factors revealed substantial reductions in the proportion of smokers, mean total cholesterol, and mean systolic and diastolic blood pressures and major increases in premorbid treatment with antiplatelet, lipid-lowering, and blood pressure lowering drugs (all p<0.0001). INTERPRETATION The age-specific incidence of major stroke in Oxfordshire has fallen by 40% over the past 20 years in association with increased use of preventive treatments and major reductions in premorbid risk factors.

From subgroups to individuals: general principles and the example of carotid endarterectomy

Clinicians often have to make treatment decisions based on the likelihood that an individual patient will benefit. In this article we consider the relevance of relative and absolute risk reductions, and draw attention to the importance of expressing the results of trials and subgroup analyses in terms of absolute risk. We describe the limitations of univariate subgroup analysis in situations in which there are several determinants of treatment effect, and review the potential for targeting treatments with risk models, especially when benefit is probably going to be dependent on the absolute risk of adverse outcomes with or without treatment. The ability to systematically take into account the characteristics of an individual patient and their interactions, to consider the risks and benefits of interventions separately if needed, and to provide patients with personalised estimates of their likelihood of benefit is shown using the example of endarterectomy for symptomatic carotid stenosis.

Reproducibility of Measures of Visit-to-Visit Variability in Blood Pressure after Transient Ischaemic Attack or Minor Stroke

Background: In certain patients in routine practice, blood pressure (BP) measurements differ substantially from week to week or month to month. Although often assumed to be random, such variability could provide information on underlying pathology or prognosis. In order to be informative, however, visit-to-visit BP variability would have to be neither random (i.e. it should be reproducible over time within individuals) nor artefactual (i.e. it should not be an artefact of the method/timing of measurement, for example). Methods: We quantified visit-to-visit variability in BP and explored potential confounding factors by analysing repeat measurements obtained every few months during follow-up in two large trials in patients with a transient ischaemic attack (TIA) or minor ischaemic stroke: the UK-TIA Aspirin Trial (effect of aspirin, effect of season and day of the week of measurement) and the European Carotid Surgery Trial (ECST – effect of carotid endarterectomy). By comparing different periods of follow-up, we also determined the reproducibilities of mean and several different measures of variability for both systolic (SBP) and diastolic BP (DBP). Results: The mean absolute difference between adjacent SBP readings was 14.7 mm Hg in the UK-TIA Trial and 16.0 mm Hg in ECST. Visit-to-visit variability in both SBP and DBP were independent of the potentially confounding factors studied, but reproducibility of all the variability measures was statistically significantly greater than zero. Reproducibility (intraclass correlation) of standard deviation of SBP was 0.32 (p < 0.0001) in the UK-TIA Trial and 0.18 (p = 0.0007) in ECST. Consequently, classification of patients with high (top quintile) or low (bottom quintile) variability was consistent over time (observed/expected = 2.21, 95% confidence interval 1.71–2.85, p < 0.0001, and 1.65, 1.23–2.21, p = 0.0007, respectively). Reproducibility increased with the number of measurements used to calculate variability, and was independent of any correlation with mean BP. Conclusions: Visit-to-visit variability in BP in these populations was reproducible, independently of any correlation with mean BP, demonstrating that visit-to-visit intra-individual BP variability is not random.

Medium-Term Variability of Blood Pressure and Potential Underdiagnosis of Hypertension in Patients With Previous Transient Ischemic Attack or Minor Stroke

Background and Purpose— Blood pressure (BP) is a major risk factor for stroke. However, the variability of systolic and diastolic BP (SBP and DBP) means that single measurements do not provide a reliable measure of usual BP. Although 24-hour ambulatory BP monitoring can correct for the effects of short-term variation, there is also important medium-term variability. The extent of medium-term variability in BP is most marked in patients with a previous transient ischemic attack (TIA) or stroke. We studied the potential impact of this variability on the likely recognition of hypertension. Methods— We analyzed multiple repeated measurements of BP in 3 large cohorts with a TIA or minor stroke: the UK-TIA trial (n=2098), the Dutch TIA trial (n=2953), and the European Carotid Surgery Trial (ECST; n=2646). Regression dilution ratios and coefficients of variation were calculated for SBP and DBP from baseline and repeated measurements during the subsequent 12 months. Categorization based on single baseline measurements was also compared with categorization based on the subsequent “usual” BP. Results— The correlation between measurements of BP at baseline and 3 to 5 months later was poor (R2 from 0.17 to 0.31 for SBP and from 0.10 to 0.20 for DBP). Categorization of patients by baseline values resulted in substantial misclassification in relation to usual BP. For example, of patients with an SBP <140 mm Hg at baseline, the percentage with a usual SBP ≥140 mm Hg was 31.6% in the UK-TIA trial, 48.2% in the Dutch TIA trial, and 57.7% in the ECST. At least 3 consecutive measurements of SBP <120 mm Hg were required to be >90% certain that subsequent usual SBP would not be ≥140 mm Hg. Conclusions— Given the greater medium-term variability of BP in patients with a previous TIA or stroke than in the general population, single measurements of “normal” or “low” BP will substantially underestimate the true prevalence of hypertension.

Apolipoproteins as predictors of ischaemic stroke in patients with a previous transient ischaemic attack.

Background: Weak associations between total and LDL cholesterol and ischaemic stroke compared with coronary heart disease (CHD) are at odds with the similar effectiveness of statin drugs in preventing ischaemic stroke and CHD, suggesting that other lipid sub-fractions that are affected by statins might be better predictors of ischaemic stroke. Apolipoprotein B levels are reduced by statins and are a stronger predictor of CHD than total and LDL cholesterol in patients both on and off statins. However, there are very few published data on apolipoproteins and stroke risk and no studies in patients with previous transient ischaemic attack (TIA). Methods: We performed a prospective cohort study of the associations of baseline total cholesterol, LDL, HDL, apolipoproteins A1 and B (apo A1; apo B) and risk of ischaemic stroke in 261 patients with previous TIA. Cox proportional hazards models were used to determine crude and multivariate-adjusted hazard ratios (HR) above versus below median values at 10-years follow-up. Results: The apo B/apo A1 ratio was the strongest independent predictor of ischaemic stroke (HR = 2.94, 95% CI 1.43–5.88, p = 0.003) followed by apo B (HR = 2.26, 95% CI 1.16–4.38, p = 0.02). The associations between total cholesterol, LDL, HDL, LDL/HDL ratio and apo A1 and ischaemic stroke risk did not reach statistical significance. Conclusions: Apo B and the apo B/apo A1 ratio are predictive of ischaemic stroke in patients with previous TIA. Further studies are required to determine whether the prognostic value of apolipoprotein levels is maintained in patients on statins.

Serum urate predicts long-term risk of acute coronary events in women after a transient ischaemic attack and stroke.

Background: Several studies have shown serum urate to be an independent risk factor for vascular disease, but others have not, although a stronger association in women than in men has been a consistent finding. Studies of stroke patients have shown possible associations between urate level and stroke severity, but there have been no large cohort studies of the effect of urate on the long-term risk of future vascular events in patients with a transient ischaemic attack (TIA) or stroke. We studied this relationship in 2 independent cohorts. Methods: Individual data on 15,483 patient-years of follow-up from the UK-TIA trial (13,182 patient-years) and the Oxford TIA study (2,301 patient-years) were analyzed. Hazard ratios (per unit increase of baseline urate in mg/dl) for the risks of stroke and acute coronary events (ACE) were obtained from Cox models stratified by study, with and without adjustment for potential confounders. Potential interactions between urate and baseline characteristics were also assessed. Results: Linear associations between urate and risk of ACE were found in both studies: pooled age- and sex-adjusted hazard ratio = 1.17, 95% CI 1.06–1.30, per unit increase of urate (p = 0.003). Sex, body mass index and previous myocardial infarction or angina were effect modifiers, but only the effect of sex remained after adjustment for other risk factors (p = 0.002), with a 5th:1st quintile hazard ratio of 4.23 (1.97–9.07, p < 0.0001) in women and 1.09 (0.70–1.71, p = 0.69) in men. These findings were consistent across the 2 studies. No associations were found between urate level and either risk or severity of stroke. Conclusions: High urate levels were independent predictors of long-term risk of ACE in women who had a TIA or stroke, but not in men, in 2 independent studies. Urate levels could be useful in identifying women at high risk of coronary events in routine practice.

Potential consequences for recruitment, power, and external validity of requirements for additional risk factors for eligibility in randomized controlled trials in secondary prevention of stroke.

Background and Purpose— Eligibility criteria determine the external validity (generalizability) of the results of randomized controlled trials. To increase the number of outcome events, and hence statistical power, some recent stroke prevention trials have required additional vascular risk factors for eligibility. Methods— To assess the merits of additional eligibility criteria in stroke prevention trials, we analyzed data from 3 trials and 1 hospital-referred series of patients with a transient ischemic attack or minor ischemic stroke. Patients were stratified according to 2 sets of additional risk factors similar to those used in recent trials (MATCH, SPORTIF and PRoFESS); risk of stroke, myocardial infarction, or vascular death was calculated in relation to the number of risk factors. Results— Although the observed risk during follow-up did increase with the number of risk factors present (P<0.01 for both sets), the risks in patients with ≥1 risk factors were not substantially greater than those in all patients. Consequently, although the proportions of patients with no risk factors in the 4 cohorts differed substantially between the 2 sets of eligibility criteria (21% to 28% versus 56% to 73%), in neither case could their exclusion be justified on statistical grounds. Conclusions— The degree of patient selection introduced by use of additional vascular risk factors as eligibility criteria for trials can differ substantially between apparently similar sets of risk factors. Given that the potential for additional eligibility criteria to undermine generalizability and prolong recruitment outweighs any benefits in terms of statistical power, the exclusion of patients with no risk factors is difficult to justify.

Effect of Age and Glycaemic Control on the Association between Fibrinogen and Risk of Acute Coronary Events after Transient Ischaemic Attack or Stroke

Background: Fibrinogen is an independent risk factor for acute vascular events, but there is uncertainty as to whether it is causal. One potential causal mechanism is the formation of low permeability fibrin clot in association with raised fibrinogen. We hypothesised that if this effect of fibrinogen were causally related to risk of vascular events, the risk relationship would be affected by the two other factors that affect fibrin clot permeability – age and glycaemic control. Methods: We studied the relationship between fibrinogen and risk of incident coronary events by age and baseline glucose levels in pooled data from three cohorts of patients with known cerebrovascular disease (UK-TIA Aspirin trial; Dutch TIA trial; Oxford TIA Study) during 23,272 patient-years of follow-up. Results: Risk of coronary events increased linearly with fibrinogen, but there was a significant interaction with age (p = 0.01 across tertiles of age), with the association being strongest for individuals aged <60 years (upper fibrinogen quintile hazard ratio = 3.95, 95% CI = 2.67–5.85, p < 0.0001). The risk relation was diminished in individuals with impaired glucose tolerance or diabetes. The effects of age and glycaemic control were independent, such that there was an almost fivefold increase in risk across quintiles of fibrinogen in patients aged <60 years with below median normal glucose levels (upper quintile hazard ratio = 4.90, 95% CI = 2.79–8.58, p < 0.0001). Conclusions: The effect of age and glycaemic control on the relationship between fibrinogen and risk of acute coronary events supports the hypothesis of a causal effect of fibrinogen mediated via the permeability of fibrin clot.

Leukoaraiosis on MRI in patients with minimally symptomatic obstructive sleep apnoea.

Background: Obstructive sleep apnoea (OSA) is associated with hypertension, nocturnal blood pressure (BP) surges, and increased risk of stroke. It may therefore also be associated with a higher risk of developing leukoaraiosis. Only few data about the prevalence of leukoaraiosis in patients with OSA, and any association between degrees of severity of either condition, exist. Methods: We studied patients who were part of a clinical trial (MOSAIC) in minimally symptomatic OSA. All patients had brain MRI (T2, FLAIR) at baseline. A single observer assessed the images for the presence and severity of leukoaraiosis (ARWMC-score). We related the extent of leukoaraiosis to the severity of OSA (measured by oxygen desaturation index [ODI]) and the presence of other vascular risk factors. Results: 183 patients (156 men, 85.2%; mean age ± SD = 57.7 ± 7.4 years; median oxygen desaturation index = 9.6, interquartile range = 4.6-16.0) took part in the study. Although 135 (74%) patients had some leukoaraiosis, this was generally mild. We confirmed the well-known risk factor associations between leukoaraiosis, increasing age (p < 0.0001) and hypertension (p = 0.003), but we did not find any association between OSA and leukoaraiosis (p = 0.33), despite both conditions being associated with increasing current BP and a history of hypertension. Conclusion: Our data confirm the well-known association between leukoaraiosis, age and increasing BP. However, we found no association between OSA and leukoaraiosis despite some shared risk factor associations. Our findings suggest that OSA is not a strong independent risk factor for leukoaraiosis. Confounding by hypertension may explain any apparent association in previously reported studies of patients with severer OSA.

Reporting of consistency of blood pressure control in randomized controlled trials of antihypertensive drugs: a systematic review of 1372 trial reports.

Objective: Hypertension is a powerful treatable risk factor for stroke. Reports of randomized controlled trials (RCTs) of antihypertensive drugs rightly concentrate on clinical outcomes, but control of blood pressure (BP) during follow-up is also important, particularly given that inconsistent control is associated with a high risk of stroke and that antihypertensive drug classes differ in this regard. Methods: We performed a systematic review of reporting of BP control in RCTs of antihypertensive drugs. We searched bibliographic databases (1950–2009) for systematic reviews of RCTs of BP-lowering and identified the main report of all trials. Results: We identified 94 larger trials (>100 participants/arm, >1-year follow-up) and 1278 smaller/shorter trials. Ninety-one (96.8%) larger trials reported some data on mean BP during follow-up, but none reported effects on the consistency of control of BP over time. Although 81 (86.2%) larger trials reported group distribution of BP at baseline (usually SD), only 22 (23.4%) reported such data at any follow-up visit. Eleven (11.7%) larger trials reported group distribution of the change in BP from baseline to follow-up, but 61 (64.9%) reported no data at all on group distribution of BP at follow-up. Thirty-nine (41.5%) trials reported the proportion of patients reaching some BP target during follow-up, but no trial reported data on the consistency of control to target within individuals over time. Similar proportions were observed in the 1278 smaller/short trials. Conclusion: Reporting of BP control is limited in RCTs of BP-lowering drugs. We suggest reporting guidelines.