Browse the corpus

Introduction Prompt treatment with intravenous alteplase improves independent survival after acute ischaemic stroke.1,2 However, concerns about the risk of symptomatic intracranial haemorrhage and whether patients with early signs of ischaemia on CT should receive alteplase could be deterring use of this treatment.3 Early signs of ischaemia on non-enhanced brain CT include tissue hypoattenuation, lesion swelling, and arterial hyperattenuation from occlusive thrombus. Tissue hypoattenuation might represent irreversible tissue damage4 and has been associated with increased risk of symptomatic intracranial haemorrhage in some,5 but not all,6,7 studies. These signs are rarely seen alone, but the association between combinations of signs and prognosis after stroke or their interaction with alteplase is unknown. Moreover, in some alteplase trials, patients with specific findings (eg, extensive tissue hypoattenuation) were excluded.5,8 Other pathological features—such as leukoaraiosis, cerebral atrophy, and old infarcts—are often seen on brain CT in patients with acute stroke. These pre-existing signs might indicate brain frailty9,10 (ie, a vulnerability to ischaemia) or increased risk of symptomatic intracranial haemorrhage.11 The association of pre-existing signs with response to alteplase, with or without early ischaemic signs, is unknown.

n seen on brain CT in patients with acute stroke. These pre-existing signs might indicate brain frailty9,10 (ie, a vulnerability to ischaemia) or increased risk of symptomatic intracranial haemorrhage.11 The association of pre-existing signs with response to alteplase, with or without early ischaemic signs, is unknown. The third International Stroke Trial (IST-3)12 is the largest (n=3035) randomised controlled trial of alteplase in acute stroke to date1,2 and was designed to test whether a wider range of patients could benefit from alteplase up to 6 h after stroke. The primary endpoint of IST-3—an increase in independent survival (defined as an Oxford Handicap Scale [OHS] score of 0–2 at 6 months)—did not differ between alteplase and control groups (odds ratio 1·13, 95% CI 0·95–1·35; p=0·181);12 however, in prespecified secondary analyses (ordinal shift analysis of the OHS and an OHS score of 0–1), alteplase increased functional outcome in all patients randomised up to 6 h and increased the primary outcome in the patients treated less than 3 h after stroke. The effect of alteplase 18 months after stroke has been published elsewhere.13 The findings of IST-3 accord with those of previous alteplase trials, at all time windows.2 A secondary objective of IST-3 was to ascertain whether early ischaemic or pre-existing brain CT signs—individually or in combination—were associated with response to alteplase on several clinically relevant early and late outcomes.14 Here, we present that secondary analysis.

The third International Stroke Trial (IST-3)12 is the largest (n=3035) randomised controlled trial of alteplase in acute stroke to date1,2 and was designed to test whether a wider range of patients could benefit from alteplase up to 6 h after stroke. The primary endpoint of IST-3—an increase in independent survival (defined as an Oxford Handicap Scale [OHS] score of 0–2 at 6 months)—did not differ between alteplase and control groups (odds ratio 1·13, 95% CI 0·95–1·35; p=0·181);12 however, in prespecified secondary analyses (ordinal shift analysis of the OHS and an OHS score of 0–1), alteplase increased functional outcome in all patients randomised up to 6 h and increased the primary outcome in the patients treated less than 3 h after stroke. The effect of alteplase 18 months after stroke has been published elsewhere.13 The findings of IST-3 accord with those of previous alteplase trials, at all time windows.2 A secondary objective of IST-3 was to ascertain whether early ischaemic or pre-existing brain CT signs—individually or in combination—were associated with response to alteplase on several clinically relevant early and late outcomes.14 Here, we present that secondary analysis. Methods Study design and participants IST-3 was an international, multicentre, open-label, prospective blinded endpoint, randomised controlled trial done at 156 centres in 12 countries (appendix pp 2–5). Patients were eligible for the study if the treating clinician felt there was no clear indication for or contraindication to alteplase and judged the treatment promising but unproven for that individual. Eligibility criteria were age 18 years or older (no upper age limit) and symptoms of cortical, lacunar, and posterior circulation stroke,15 of all severities, according to the National Institutes of Health Stroke Scale (NIHSS). All participants had to be able to undergo randomisation and start treatment within 6 h of stroke. We did not exclude patients with early ischaemic or pre-existing imaging signs on brain CT. We excluded individuals younger than 18 years, people with standard contraindications to alteplase, patients with established visible infarction (ie, indicating stroke onset was likely to be more than 6 h previously), and individuals with haemorrhagic stroke or non-stroke lesion as the cause of stroke.

gns on brain CT. We excluded individuals younger than 18 years, people with standard contraindications to alteplase, patients with established visible infarction (ie, indicating stroke onset was likely to be more than 6 h previously), and individuals with haemorrhagic stroke or non-stroke lesion as the cause of stroke. The IST-3 protocol16 and statistical analysis plan14 have been published elsewhere. The study was approved by the Scotland A research ethics committee and by ethics committees and other regulatory bodies of all participating countries, hospitals, and institutions. All patients, or a relative if the patient lacked capacity, provided written informed consent before randomisation into the trial.

sewhere. The study was approved by the Scotland A research ethics committee and by ethics committees and other regulatory bodies of all participating countries, hospitals, and institutions. All patients, or a relative if the patient lacked capacity, provided written informed consent before randomisation into the trial. Randomisation and masking Full details of the IST-3 randomisation procedure, including the minimisation process, are presented elsewhere.12,16 Briefly, the randomising clinician recorded baseline data via a central, secure, telephone-based or web-based system. The system randomly allocated patients to either standard best medical care plus immediate thrombolysis with intravenous alteplase (0·9 mg/kg bodyweight; maximum 90 mg; 10% bolus with the remainder over 1 h) or control (standard best medical care alone) within 6 h of ischaemic stroke. Patients allocated to the control group received the same stroke care as did individuals allocated to the alteplase group. The first 276 patients enrolled to the study were treated in a placebo-controlled, double-blind phase (the placebo was an inert product identical in appearance to alteplase); thereafter, patients were randomised to either immediate alteplase or open control.

oke care as did individuals allocated to the alteplase group. The first 276 patients enrolled to the study were treated in a placebo-controlled, double-blind phase (the placebo was an inert product identical in appearance to alteplase); thereafter, patients were randomised to either immediate alteplase or open control. Procedures To be approved for inclusion in IST-3, all participating centres had to pass minimum image acquisition standards16 to ensure that brain scans were of diagnostic quality for acute stroke (for CT) and included the minimum correct sequences (for MRI). Before randomisation, all patients had either a CT or MRI brain scan, repeated 24–48 h after stroke and again if any neurological deterioration arose in the first 7 days. The present analysis concerns brain images taken before randomisation, up to 6 h after stroke.

and included the minimum correct sequences (for MRI). Before randomisation, all patients had either a CT or MRI brain scan, repeated 24–48 h after stroke and again if any neurological deterioration arose in the first 7 days. The present analysis concerns brain images taken before randomisation, up to 6 h after stroke. Participating centres sent all brain images to the IST-3 coordinating centre for adjudication and quality checking; all scans were anonymised. Adjudicators were masked to all information, except whether the scan was prerandomisation or follow-up. They viewed the images and recorded their ratings on a secure web-based system (systematic image review system [SIRS]). All adjudicators were neuroradiologists (JMW, RvK, AvH, LC, NB, ZM, AF, GP) or stroke neurologists (AP, AA) with experience in stroke imaging; they rated 60 scans from the ACCESS study17,18 and 25 scans from IST-3 that were selected at random, to ensure satisfactory agreement (defined as κ>0·70) for presence of early ischaemic signs and no more than one category difference for pre-existing signs. Adjudicators completed all analyses before the database was locked and the randomisation code was broken. Every scan was read by one adjudicator. Any discrepancies between the scan rating and clinical data (eg, side of brain affected, presence of haemorrhage) identified during data cleaning before data lock were cross-checked by JMW, who assigned a final rating before unmasking took place and the randomisation code was broken.

y scan was read by one adjudicator. Any discrepancies between the scan rating and clinical data (eg, side of brain affected, presence of haemorrhage) identified during data cleaning before data lock were cross-checked by JMW, who assigned a final rating before unmasking took place and the randomisation code was broken. We first checked brain CTs for early ischaemic signs (tissue hypoattenuation, lesion size, swelling, and hyperattenuated artery); we used the term visible infarct to represent any of these signs. We then looked for pre-existing structural signs (old infarcts, leukoaraiosis, and atrophy).19 We classified images using validated scores.17,18 We classed scans as normal only if no early ischaemic changes or pre-existing changes were present.

artery); we used the term visible infarct to represent any of these signs. We then looked for pre-existing structural signs (old infarcts, leukoaraiosis, and atrophy).19 We classified images using validated scores.17,18 We classed scans as normal only if no early ischaemic changes or pre-existing changes were present. We defined the presence and degree of hypoattenuated tissue as either mild (grey matter attenuation equal to normal white matter) or severe (grey and white matter attenuation slightly less than normal white matter but still consistent with onset of stroke within 6 h).20 We classified the extent of acute ischaemic lesions in three ways: with the one-third middle cerebral artery (MCA) method;21,22 with the IST-3 method;17,23 and with the Alberta Stroke Program Early CT Stroke (ASPECTS) score.24 The IST-3 score for infarct extent reflects all arterial territories, whereas the ASPECTS score and one-third MCA method focus only on the MCA territory. Therefore, we used the IST-3 score as the primary measure of infarct size in analyses, condensing the full IST-3 lesion extent score into four groups for analysis: small infarcts, which we defined as lacunar, small cortical, small cerebellar, less than half of brainstem, or less than half of the anterior cerebral artery (ACA) or posterior cerebral artery (PCA) territory; medium infarcts, classed as striatocapsular, the anterior or posterior half of the peripheral MCA territory, or more than half the ACA or PCA territory; large infarcts, defined as the whole of the peripheral MCA territory or all the MCA territory; and very large infarcts, which comprised the whole MCA and PCA territory, all the MCA and ACA territory, or all three territories.24 Small infarcts on the IST-3 score were equivalent to ASPECTS 8–10, medium infarcts corresponded to ASPECTS 5–7, and large and very large infarcts were similar to ASPECTS 0–4. MCA involvement in small and medium infarcts (according to IST-3 score) was equivalent to less than one-third MCA, and in large and very large infarcts it corresponded to more than one-third MCA. We graded ischaemic lesion swelling on a seven-point scale.23 We noted the presence or absence and location of any hyperattenuated artery.17,25

small and medium infarcts (according to IST-3 score) was equivalent to less than one-third MCA, and in large and very large infarcts it corresponded to more than one-third MCA. We graded ischaemic lesion swelling on a seven-point scale.23 We noted the presence or absence and location of any hyperattenuated artery.17,25 We recorded the location of old infarcts (eg, cortical, lacunar, border zone, and brainstem or cerebellar).23 We noted the presence and severity of leukoaraiosis on CT;26 if brain MRI was done, we used the Fazekas scale to grade changes.27 We classified atrophy as none, moderate, or severe when compared against standard examples. We also classed follow-up scans for any haemorrhage (petechial, haematoma in or remote from infarct, intraventricular, or subarachnoid blood), including whether the haemorrhage was likely to worsen neurological status, as part of the outcome assessment for symptomatic intracranial haemorrhage. Outcomes The primary outcome of IST-3 was the proportion of patients alive and independent, defined by an OHS score of 0–2, at 6 months. The OHS28 is similar to the modified Rankin scale29,30 and has values from 0 to 6, with 0 representing no symptoms or completely independent and 6 signifying the patient has died. OHS score 0–2 indicates good functional outcome. We also analysed two prespecified secondary endpoints:14 an ordinal analysis of the OHS; and OHS score of 0–1 (ie, favourable outcome). Other secondary outcomes were symptomatic intracranial haemorrhage within 7 days,14 death within 7 days, and death by 6 months.

s died. OHS score 0–2 indicates good functional outcome. We also analysed two prespecified secondary endpoints:14 an ordinal analysis of the OHS; and OHS score of 0–1 (ie, favourable outcome). Other secondary outcomes were symptomatic intracranial haemorrhage within 7 days,14 death within 7 days, and death by 6 months. Local staff at every centre followed up all patients for 7 days and recorded early outcomes (symptomatic intracranial haemorrhage, death, and likely cause of death before 7 days), which were sent to the central trials office. At 6-month follow-up, central researchers who were masked to group allocations sent questionnaires by post to patients or their carers, to record independence (using the OHS);16 if we knew the patient had died, we sent the questionnaire to their family doctor, to obtain the date of death and likely cause. Statistical analyses Analyses of early ischaemic and pre-existing structural brain signs, their effect on functional outcome, death, and symptomatic intracranial haemorrhage, and interactions with alteplase, were prespecified in the trial protocol and analysis plan.14 We analysed imaging data for patients whose prerandomisation scans were received at the central trials office. We did statistical analyses with SAS version 9.3.

ctional outcome, death, and symptomatic intracranial haemorrhage, and interactions with alteplase, were prespecified in the trial protocol and analysis plan.14 We analysed imaging data for patients whose prerandomisation scans were received at the central trials office. We did statistical analyses with SAS version 9.3. We used logistic regression to ascertain associations between imaging signs and: (a) age, NIHSS score, and time to randomisation; and (b) outcomes at 7 days (symptomatic intracranial haemorrhage) and 6 months (OHS score of 0–2 and OHS score of 0–1), which were judged representative of the most clinically relevant early hazard and late benefit. We did multivariate logistic regression, adjusting for the linear effects of age, NIHSS score, and time to randomisation, to identify whether any combinations of imaging variables were associated with symptomatic intracranial haemorrhage or OHS score of 0–2. We then tested for interactions of imaging signs (presence, absence, or severity) and response to alteplase (adjusting for age, NIHSS score, and time to randomisation) with symptomatic intracranial haemorrhage, death within 7 days, and OHS score of 0–2 at 6 months; as a secondary analysis, we assessed interactions of imaging signs and response to alteplase with an OHS score of 0–1 and by ordinal OHS at 6 months (a more statistically sensitive ordinal model than analysis of dichotomous OHS categories). We tested whether associations between imaging signs and outcomes and imaging signs and alteplase differed in patients randomised within 3 h, 3–4·5 h, or 4·5–6 h of stroke. We tested whether the response to alteplase differed by the extent of early ischaemic signs (IST-3 score and ASPECTS score 0–7 vs 8–10), and we did a prespecified meta-analysis of data from IST-3 with imaging data from other thrombolysis trials. We judged p values less than 0·05 significant, except for analyses of interactions between imaging signs and response to alteplase, for which we used a significance level less than 0·01, to minimise false-positive results.

a prespecified meta-analysis of data from IST-3 with imaging data from other thrombolysis trials. We judged p values less than 0·05 significant, except for analyses of interactions between imaging signs and response to alteplase, for which we used a significance level less than 0·01, to minimise false-positive results. This trial is registered at ISRCTN.com, number ISRCTN25765518. Role of the funding sources The funding sources had no role in study design, data collection, data analysis, data interpretation, or writing of the report. All authors had full access to all data in the study and the corresponding author had final responsibility for the decision to submit for publication. Results Between May 1, 2000, and July 31, 2011, 3035 patients were recruited to IST-3 and underwent randomisation. Brain scans were available for 1507 patients assigned to alteplase and for 1510 controls; scans for 18 patients were not received at the central trials office and, therefore, were excluded from analyses (figure 1). Table 1 shows baseline clinical and imaging characteristics, which were well balanced between arms and no variables were missing.

le for 1507 patients assigned to alteplase and for 1510 controls; scans for 18 patients were not received at the central trials office and, therefore, were excluded from analyses (figure 1). Table 1 shows baseline clinical and imaging characteristics, which were well balanced between arms and no variables were missing. Of 3017 prerandomisation brain scans, 2962 (98%) were obtained by CT and 55 (2%) by MRI. The expert panel judged 269 (9%) scans normal (ie, no early ischaemic or pre-existing signs); 1224 (41%) patients had early ischaemic signs (ie, visible infarct, whether or not pre-existing signs were also present) and 1524 (51%) had pre-existing signs (but no early ischaemic signs). The commonest early ischaemic sign was tissue hypoattenuation, seen in 1203 (40%) patients, and the least frequent sign was hyperattenuated artery, recorded in 735 (24%) patients (table 1). Pre-existing signs were common: 1336 (44%) patients had signs of an old infarct, 1547 (51%) had leukoaraiosis, and 2327 (77%) had evidence of atrophy.

ischaemic sign was tissue hypoattenuation, seen in 1203 (40%) patients, and the least frequent sign was hyperattenuated artery, recorded in 735 (24%) patients (table 1). Pre-existing signs were common: 1336 (44%) patients had signs of an old infarct, 1547 (51%) had leukoaraiosis, and 2327 (77%) had evidence of atrophy. Some strong associations were noted between individual imaging variables and age, NIHSS score, and time to randomisation (table 2; appendix pp 20–22). Every point increase in NIHSS score was associated with a roughly 10% increase in the odds of tissue hypoattenuation, swelling, hyperattenuated artery, or large lesion. Every delay of 1 h increased the odds of tissue hypoattenuation, but not other early or pre-existing structural signs. Every increase in age by 1 year decreased the odds of tissue hypoattenuation, large lesion, swelling, or hyperattenuated artery by about 2% (table 2), but the odds of old infarct, atrophy, and leukoaraiosis were increased.

creased the odds of tissue hypoattenuation, but not other early or pre-existing structural signs. Every increase in age by 1 year decreased the odds of tissue hypoattenuation, large lesion, swelling, or hyperattenuated artery by about 2% (table 2), but the odds of old infarct, atrophy, and leukoaraiosis were increased. Individually, all early ischaemic signs predicted worse outcomes (table 3) with the exception of severe hypoattenuation, although relatively few patients had this sign. Individually, of all pre-existing signs, only old infarct predicted symptomatic haemorrhage (table 2); no pre-existing signs predicted death within 7 days. Leukoaraiosis and severe atrophy predicted death by 6 months, all pre-existing structural signs predicted reduced chance of being alive and independent (OHS score of 0–2) at 6 months, and leukoaraiosis and atrophy predicted diminished chance of a favourable outcome (OHS score of 0–1) at 6 months. Similar associations between imaging signs and outcomes were seen for patients randomised within 3 h, 3–4·5 h after stroke, and 4·5–6 h after stroke (appendix pp 6–7).

endent (OHS score of 0–2) at 6 months, and leukoaraiosis and atrophy predicted diminished chance of a favourable outcome (OHS score of 0–1) at 6 months. Similar associations between imaging signs and outcomes were seen for patients randomised within 3 h, 3–4·5 h after stroke, and 4·5–6 h after stroke (appendix pp 6–7). The multivariate logistic regression model for symptomatic intracranial haemorrhage included age, NIHSS score, time to randomisation, individual imaging signs, alteplase, and use of antiplatelet drugs immediately before stroke.12 Increasing NIHSS score, alteplase, antiplatelet treatment, and pre-existing old infarcts predicted a significantly higher risk of symptomatic intracranial haemorrhage (table 4). With standard stepwise model selection methods, retaining age, NIHSS score, time to randomisation, and treatment group, both old infarcts and hyperattenuated arteries were potentially significant predictors of symptomatic intracranial haemorrhage (table 5). The multivariate logistic regression model for good functional outcome (OHS score of 0–2) at 6 months showed that increasing age and NIHSS score were the strongest adverse predictors, but hyperattenuated arteries, large lesion, and leukoaraiosis each individually predicted less chance of a good outcome by about 25–30% (table 4). With standard stepwise regression methods, the selected model retained large lesion, hyperattenuated arteries, and leukoaraiosis, with coefficients very similar to those in the full model (table 5). Similar effects were seen in patients randomised 0–3 h, 3–4·5 h, and 4·5–6 h after stroke (appendix pp 8–9). However, potential differences—eg, that tissue hypoattenuation (rather than old infarcts or hyperattenuated arteries) predicts symptomatic intracranial haemorrhage in patients randomised 4·5–6 h after stroke—should be interpreted with caution because of the weak significance level and smaller sample.

(appendix pp 8–9). However, potential differences—eg, that tissue hypoattenuation (rather than old infarcts or hyperattenuated arteries) predicts symptomatic intracranial haemorrhage in patients randomised 4·5–6 h after stroke—should be interpreted with caution because of the weak significance level and smaller sample. No interaction was recorded between any individual or combined imaging variable and alteplase, for either functional outcome (OHS score of 0–2, figure 2) or symptomatic intracranial haemorrhage (figure 3). Furthermore, no interaction was noted between any individual or combined image variable and alteplase in the ordinal shift analysis or in analyses restricted to patients randomised within 3 h of stroke or 3–4·5 h or 4·5–6 h after stroke (appendix pp 10–17).

figure 2) or symptomatic intracranial haemorrhage (figure 3). Furthermore, no interaction was noted between any individual or combined image variable and alteplase in the ordinal shift analysis or in analyses restricted to patients randomised within 3 h of stroke or 3–4·5 h or 4·5–6 h after stroke (appendix pp 10–17). Despite the absence of definite interactions between imaging signs and alteplase, the absolute increase in symptomatic intracranial haemorrhage after alteplase with combined imaging signs was substantial. The combination of old infarcts and hyperattenuated arteries (adjusting for age, NIHSS score, and time to randomisation) predicted nearly three-fold increased odds of symptomatic intracranial haemorrhage (odds ratio 2·98, 95% CI 1·71–5·16) versus patients with neither sign (both signs present, absolute excess of events with alteplase 13·8%, 95% CI 6·9–20·7; both signs absent, absolute excess with alteplase 3·2%, 1·4–5·1; appendix p 18). Similar absolute effects on symptomatic intracranial haemorrhage were seen for patients randomised within 3 h of stroke (appendix pp 12–14). A difference in absolute effects was not seen for functional outcome at 6 months (OHS score of 0–2) in the whole study group (appendix p 19) and by separate time windows (data not shown).

absolute effects on symptomatic intracranial haemorrhage were seen for patients randomised within 3 h of stroke (appendix pp 12–14). A difference in absolute effects was not seen for functional outcome at 6 months (OHS score of 0–2) in the whole study group (appendix p 19) and by separate time windows (data not shown). Discussion To date, IST-3 is the largest randomised controlled trial of thrombolysis versus control after ischaemic stroke in patients for whom alteplase was judged promising but unproven. Although IST-3 was neutral on the primary endpoint (OHS score of 0–2),12 this finding does not preclude the presence of clinically relevant interactions with treatment. Therefore, we planned a priori a detailed secondary analysis of the association of imaging signs with thrombolysis effects. Imaging signs are powerful prognostic markers; however, we recorded no unequivocal evidence that any individual imaging sign modified response to alteplase in patients presenting up to 6 h after ischaemic stroke. In this population, no interaction was noted between extensive early ischaemia—commonly cited as an exclusion for thrombolysis treatment—and alteplase, which accords with previous findings (panel, figure 4).6,8,12,22,31–38 Some combinations of imaging features (eg, old infarct and hyperattenuated artery) were associated with an increased absolute excess of symptomatic intracranial haemorrhage after alteplase.

d as an exclusion for thrombolysis treatment—and alteplase, which accords with previous findings (panel, figure 4).6,8,12,22,31–38 Some combinations of imaging features (eg, old infarct and hyperattenuated artery) were associated with an increased absolute excess of symptomatic intracranial haemorrhage after alteplase. The plain CT scans obtained in IST-3 reflect imaging done in practice, with speed and accessibility being important factors in view of the steep time dependency of alteplase benefit.1,2,12,39 The scans reflected common problems, such as movement of the patient or oblique positioning. To date, the brain imaging dataset in IST-3 is the largest in an acute stroke trial (more than 7000 scans) to be read centrally. Scan reading was structured and masked, and follow-up was complete. We pretested the imaging rating system thoroughly17,18 and the assessors were skilled in acute stroke imaging. Although the expert panel might have detected subtle changes in tissue attenuation, we showed previously that those who were less experienced could still identify hyperattenuated arteries, established tissue ischaemia, and pre-existing structural changes with good reliability.17 These results should inform clinical practice.

gh the expert panel might have detected subtle changes in tissue attenuation, we showed previously that those who were less experienced could still identify hyperattenuated arteries, established tissue ischaemia, and pre-existing structural changes with good reliability.17 These results should inform clinical practice. Although more patients were randomised within 3 h of stroke in IST-3 (n=846) than in any other previous trial,32 we might have missed an interaction between an imaging sign and alteplase in the subgroup of patients treated within 3 h, because very few symptomatic intracranial haemorrhage events arose in controls, and the estimate of effect on OHS score had wide CIs in each subgroup. Associations between clinical and imaging signs reflect the specific IST-3 population; inference to the wider population with ischaemic stroke—including individuals not judged eligible for the trial—remains speculative. Few patients with obvious tissue hypoattenuation (ie, in whom the acute ischaemic tissue was of lower attenuation than normal grey and white matter) were included in the trial; severe tissue hypoattenuation, including appearances suggesting a lesion older than 6 h, was an exclusion criterion. Additionally, patients with extensive ischaemic lesions (ASPECTS 0–7) comprised only a quarter of the study population, although this proportion is higher than in any previous trial. Hence, although IST-3 was larger than previous trials, it might not have had enough statistical power to ascertain whether alteplase treatment in this subgroup of patients adds risk or benefit. We might have identified a spurious association, but we have been cautious in the interpretation, including 99% CIs for tests of interaction. The use of stepwise methods to identify parsimonious predictive models should not be interpreted as ruling out effects of those imaging signs that were excluded from the models. Different subsets of the full set of predictors could produce almost equally good results in this dataset; therefore, further research to assess the contribution of different signs to the risk of adverse outcomes would be valuable.

rpreted as ruling out effects of those imaging signs that were excluded from the models. Different subsets of the full set of predictors could produce almost equally good results in this dataset; therefore, further research to assess the contribution of different signs to the risk of adverse outcomes would be valuable. Previous analyses of imaging predictors of the response to alteplase focused on individual acute ischaemic signs.20,40 Here, we have shown that pre-existing signs are important. The association between symptomatic intracranial haemorrhage and old infarcts could account for why the risk of symptomatic intracranial haemorrhage does not change with increasing time to alteplase up to 6 h;1,2 another factor present before the stroke, or that is independent of time after stroke, might increase risk. Worsening of functional outcome with old infarcts, atrophy, and leukoaraiosis might indicate increased susceptibility to acute ischaemia or alteplase hazards and could provide useful markers of so-called brain frailty.41 Further research is warranted to ascertain the strength of association between pre-existing signs on brain CT and clinical markers of brain frailty. Our findings also emphasise the value to future trials of minimising easy-to-detect imaging prognostic signs to avoid baseline imbalances.

rs of so-called brain frailty.41 Further research is warranted to ascertain the strength of association between pre-existing signs on brain CT and clinical markers of brain frailty. Our findings also emphasise the value to future trials of minimising easy-to-detect imaging prognostic signs to avoid baseline imbalances. The presence of combinations of imaging signs in patients after stroke might provide additional information to decision making when clinical uncertainty exists about the likely benefit of alteplase—eg, in a patient presenting close to the latest time window or for whom the likelihood of benefit was marginal.2 Additional factors affecting risk of symptomatic intracranial haemorrhage (eg, taking antiplatelet drugs) might contribute to decision making in such patients. A hint, in stepwise modelling, that early ischaemic tissue hypoattenuation might be associated with increased risk of symptomatic intracranial haemorrhage with alteplase 4·5–6 h after stroke (appendix pp 8–9) should be treated with caution, because this finding was not confirmed in the formal test of interaction (appendix pp 12–17). Helpfully for routine practice, the key prognostic CT imaging variables identified here are easy to detect: hyperattenuated artery has the best observer reliability of all early ischaemic signs20 across a wide range of observers,17 and old infarcts, atrophy, and leukoaraiosis26,42 are also easy to detect.17 Perceived difficulties in detecting early tissue hypoattenuation might have reduced confidence in use of CT scanning before alteplase.3 However, our findings confirm that neither early tissue hypoattenuation nor large infarct extent on ASPECTS score should exclude patients from alteplase, hopefully improving confidence in use of CT scanning in acute stroke.

ly tissue hypoattenuation might have reduced confidence in use of CT scanning before alteplase.3 However, our findings confirm that neither early tissue hypoattenuation nor large infarct extent on ASPECTS score should exclude patients from alteplase, hopefully improving confidence in use of CT scanning in acute stroke. Correspondence to: Prof J M Wardlaw, Brain Research Imaging Centre, Centre for Clinical Brain Sciences, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, UK joanna.wardlaw@ed.ac.uk Supplementary Material Supplementary appendix

ly tissue hypoattenuation might have reduced confidence in use of CT scanning before alteplase.3 However, our findings confirm that neither early tissue hypoattenuation nor large infarct extent on ASPECTS score should exclude patients from alteplase, hopefully improving confidence in use of CT scanning in acute stroke. Correspondence to: Prof J M Wardlaw, Brain Research Imaging Centre, Centre for Clinical Brain Sciences, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, UK joanna.wardlaw@ed.ac.uk Supplementary Material Supplementary appendix Acknowledgments We dedicate this report to the memory of Veronica Murray, IST-3 national coordinator for Sweden, stroke expert, close friend, and colleague, whose energy and enthusiasm ensured the success of IST-3 and who died suddenly, shortly after the submission of this paper. IST-3 is an investigator-led trial. We thank all IST-3 collaborators, including the national coordinators, participating centres, steering committee, data monitoring committee, and event adjudication committee (appendix pp 2–5). The IST-3 collaborative group wishes chiefly to acknowledge the support of all patients who participated in the study and the many individuals not mentioned specifically in the report who have supported the study. We thank Lisa Blackwell (Clinical Trial Service Unit, Oxford, UK) for preparing figures 2 and 3. We thank the funding organisations for supporting the trial. The University of Edinburgh and the Lothian Health Board are co-sponsors. The start-up phase was supported by a grant from Stroke Association UK. The expansion phase was funded by Health Foundation UK. The main phase of the trial is funded by the UK MRC and managed by the National Institute for Health Research (NIHR) on behalf of the MRC-NIHR partnership. Further funding by: Research Council of Norway; AFA Insurances (Sweden); Swedish Heart Lung Fund; Foundation of Marianne and Marcus Wallenberg; Stockholm County Council and Karolinska Institute Joint ALF-project grants (Sweden); Government of Poland; Australian Heart Foundation; Australian NHMRC; Swiss National Research Foundation; Swiss Heart Foundation; Foundation for health and cardio-/neurovascular research (Basel, Switzerland); Assessorato alla Sanita (Regione dell'Umbria); and Danube University (Krems, Austria). Alteplase and placebo for 300 patients in the double-blind component of the start-up phase were supplied by Boehringer Ingelheim. IST-3 acknowledges the extensive support of the NIHR Stroke Research Network, National Health Service (NHS) Research Scotland, through the Scottish Stroke Research Network), and the National Institute for Social Care and Health Research Clinical Research Centre.

the start-up phase were supplied by Boehringer Ingelheim. IST-3 acknowledges the extensive support of the NIHR Stroke Research Network, National Health Service (NHS) Research Scotland, through the Scottish Stroke Research Network), and the National Institute for Social Care and Health Research Clinical Research Centre. Imaging work was undertaken at the Brain Imaging Research Centre, a member of the SINAPSE collaboration (Division of Clinical Neurosciences, University of Edinburgh, Edinburgh, UK). SINAPSE is funded by the Scottish Funding Council and the Chief Scientist Office of the Scottish Executive. Additional support was received from Chest Heart and Stroke Scotland, DesAcc, University of Edinburgh, Danderyd Hospital R&D Department, Karolinska Institutet, Oslo University Hospital, and the Dalhousie University Internal Medicine Research Fund. This report presents independent research supported by the NIHR through the UK Stroke Research Network. The views expressed in this publication are those of the authors and not those of the NHS, the NIHR, or the Department of Health.

slo University Hospital, and the Dalhousie University Internal Medicine Research Fund. This report presents independent research supported by the NIHR through the UK Stroke Research Network. The views expressed in this publication are those of the authors and not those of the NHS, the NIHR, or the Department of Health. Contributors PS, RIL, and JMW (co-chief investigators) had the idea for and managed the IST-3 trial. JMW designed the imaging contribution to IST-3, with input from RvK, AF, and AvH. DP programmed the scan reading mechanism, SIRS, and the image data administration. ES managed the image administration and curation. Image assessments were done by JMW, ZM, RvK, AvH, NB, LC, AP, AF, GP, and AA. GC is the study statistician who prepared the analyses for this report. GM advised on statistical aspects. WW helped with recruitment and contributed to the statistical interpretation. JMW drafted the report; all authors commented on drafts and approved the final version. JMW, PS, RIL, WW, and GC had full access to all the data in the study. JMW takes responsibility for the integrity of the data and the accuracy of the data analysis.

with recruitment and contributed to the statistical interpretation. JMW drafted the report; all authors commented on drafts and approved the final version. JMW, PS, RIL, WW, and GC had full access to all the data in the study. JMW takes responsibility for the integrity of the data and the accuracy of the data analysis. Writing committee On behalf of the IST-3 Collaborative Group (appendix pp 2–5): Joanna M Wardlaw, Peter Sandercock, Geoff Cohen, and Andrew Farrall (University of Edinburgh, Edinburgh, UK); Richard I Lindley (Sydney Medical School, Westmead Hospital and The George Institute for Global Health, University of Sydney, Sydney, NSW, Australia); Rudiger von Kummer (Department of Neuroradiology, University Hospital, Technische Universität, Dresden, Germany); Anders von Heijne (Danderyd Hospital, Stockholm, Sweden); Nick Bradey (Department of Neuroradiology, James Cook University Hospital, Middlesbrough, UK); Andre Peeters (Cliniques Universitaires Saint-Luc, Brussels, Belgium); Lesley Cala (School of Pathology and Laboratory Medicine, The University of Western Australia, Crawley, WA, Australia); Alessandro Adami (Stroke Center, Department of Neurology, Ospedale Sacro Cuore-Don Calabria, Verona, Italy); Zoe Morris (NHS Lothian, Edinburgh, UK); Gillian Potter (Salford Royal NHS Foundation Trust, Salford, UK); Gordon Murray (Centre for Population Health Science, University of Edinburgh, Edinburgh, UK); and Will Whiteley, David Perry, and Eleni Sakka (Centre for Clinical Brain Science, University of Edinburgh, Edinburgh, UK).

orris (NHS Lothian, Edinburgh, UK); Gillian Potter (Salford Royal NHS Foundation Trust, Salford, UK); Gordon Murray (Centre for Population Health Science, University of Edinburgh, Edinburgh, UK); and Will Whiteley, David Perry, and Eleni Sakka (Centre for Clinical Brain Science, University of Edinburgh, Edinburgh, UK). Declaration of interests JMW reports grants from the UK Medical Research Council (MRC), Stroke Association UK, Health Foundation UK, Chest Heart Stroke Scotland, and Scottish Funding Council during the study; and non-financial support from Boehringer Ingelheim during the study. PS reports grants from the UK MRC, Stroke Association UK, and Health Foundation UK during the study; non-financial support from Boehringer Ingelheim during the study; and other non-financial support from Boehringer Ingelheim outside the submitted work. RIL reports grants from the Australian National Health and Medical Research Council (NHMRC) and multiple other grant agencies during the conduct of the study; and personal fees and non-financial support from Boehringer Ingelheim and personal fees from Covidien outside the submitted work. GM reports grants from UK MRC during the study. WW reports grants from UK MRC outside the submitted work. GC, RvK, AvH, NB, AP, LC, AA, ZM, AF, GP, DP, and ES declare no competing interests. Figure 1 Trial profile Figure 2 Forest plot showing the adjusted effect of treatment and baseline imaging signs on Oxford Handicap Scale score 0–2 at 6 months Data are adjusted for age, National Institutes of Health Stroke Scale score, and time to randomisation.

Declaration of interests JMW reports grants from the UK Medical Research Council (MRC), Stroke Association UK, Health Foundation UK, Chest Heart Stroke Scotland, and Scottish Funding Council during the study; and non-financial support from Boehringer Ingelheim during the study. PS reports grants from the UK MRC, Stroke Association UK, and Health Foundation UK during the study; non-financial support from Boehringer Ingelheim during the study; and other non-financial support from Boehringer Ingelheim outside the submitted work. RIL reports grants from the Australian National Health and Medical Research Council (NHMRC) and multiple other grant agencies during the conduct of the study; and personal fees and non-financial support from Boehringer Ingelheim and personal fees from Covidien outside the submitted work. GM reports grants from UK MRC during the study. WW reports grants from UK MRC outside the submitted work. GC, RvK, AvH, NB, AP, LC, AA, ZM, AF, GP, DP, and ES declare no competing interests. Figure 1 Trial profile Figure 2 Forest plot showing the adjusted effect of treatment and baseline imaging signs on Oxford Handicap Scale score 0–2 at 6 months Data are adjusted for age, National Institutes of Health Stroke Scale score, and time to randomisation. Figure 3 Forest plot showing the adjusted effect of treatment and baseline imaging signs on symptomatic intracranial haemorrhage within 7 days Data are adjusted for age, National Institutes of Health Stroke Scale score, and time to randomisation.

Data are adjusted for age, National Institutes of Health Stroke Scale score, and time to randomisation. Figure 3 Forest plot showing the adjusted effect of treatment and baseline imaging signs on symptomatic intracranial haemorrhage within 7 days Data are adjusted for age, National Institutes of Health Stroke Scale score, and time to randomisation. Figure 4 Forest plot showing the interaction between response to alteplase and early infarct size in previous trials ASPECTS=Alberta Stroke Programme Early CT Signs. ECASS II=European Co-operative Acute Stroke Study-II. IST-3=third International Stroke Trial. NINDS=National Institute of Neurological Disorders and Stroke. PROACT 2=Prolyse in Acute Cerebral Thromboembolism 2. Table 1 Baseline clinical and imaging variables

Figure 4 Forest plot showing the interaction between response to alteplase and early infarct size in previous trials ASPECTS=Alberta Stroke Programme Early CT Signs. ECASS II=European Co-operative Acute Stroke Study-II. IST-3=third International Stroke Trial. NINDS=National Institute of Neurological Disorders and Stroke. PROACT 2=Prolyse in Acute Cerebral Thromboembolism 2. Table 1 Baseline clinical and imaging variables Alteplase (n=1507) Control (n=1510) Age (years) ≤80 693 (46%) 715 (47%) >80 814 (54%) 795 (53%) 18–50 58 (4%) 68 (5%) 51–60 98 (7%) 102 (7%) 61–70 187 (12%) 175 (12%) 71–80 350 (23%) 370 (25%) 81–90 703 (47%) 697 (46%) >90 111 (7%) 98 (6%) NIHSS score* 0–5 303 (20%) 304 (20%) 6–10 419 (28%) 428 (28%) 11–15 304 (20%) 295 (20%) 16–20 268 (18%) 271 (18%) >20 213 (14%) 212 (14%) Time to randomisation (h) 0 to ≤3 431 (29%) 415 (27%) >3 to ≤4·5 575 (38%) 596 (39%) >4·5 to ≤6 501 (33%) 497 (33%) >6 0 2 (<1%) Randomising clinician's assessment of acute ischaemic change on prerandomisation imaging No change 890 (59%) 892 (59%) Possibly change 359 (24%) 339 (22%) Definitely change 258 (17%) 279 (18%) Expert reader's assessment of acute ischaemic change on prerandomisation imaging Scan completely normal 140 (9%) 129 (9%) Scan not normal but no sign of any early ischaemic change 743 (49%) 781 (52%) Signs of any early ischaemic change 624 (41%) 600 (40%) Early ischaemic lesion territory Indeterminate 885 (59%) 914 (61%) MCA or ACA or border zone 589 (39%) 555 (37%) Posterior 22 (1%) 36 (2%) Lacunar 11 (1%) 5 (<1%) Early ischaemic lesion changes in MCA territory None 925 (61%) 960 (64%) One-third or less 357 (24%) 354 (23%) More than one-third 225 (15%) 196 (13%) Early ischaemic lesion size† None visible 885 (59%) 914 (61%) Small 110 (7%) 97 (6%) Medium 250 (17%) 250 (17%) Large 124 (8%) 137 (9%) Very large 138 (9%) 112 (7%) ASPECTS score‡ 0–4 162 (11%) 138 (9%) 5–7 201 (13%) 228 (15%) 8–10 1144 (76%) 1144 (76%) Early ischaemic lesion depth of tissue hypoattenuation§ None 892 (59%) 922 (61%) Mild 503 (33%) 492 (33%) Severe 112 (7%) 96 (6%) Early ischaemic lesion degree of swelling None 1152 (76%) 1171 (78%) Mild sulcal effacement 283 (19%) 265 (18%) Mild ventricular effacement 71 (5%) 73 (5%) Moderate effacement 1 (<1%) 0 Severe effacement 0 1 (<1%) Location of hyperattenuated arteries None 1131 (75%) 1151 (76%) Anterior circulation 360 (24%) 342 (23%) Posterior circulation 16 (1%) 17 (1%) ICA, or BA, or MCA and ACA 42 (3%) 33 (2%) MCA, or ACA, or PCA main 334 (22%) 326 (22%) Pre-existing brain changes Evidence of atrophy 1161 (77%) 1166 (77%) Evidence of leukoaraiosis 765 (5

hyperattenuated arteries None 1131 (75%) 1151 (76%) Anterior circulation 360 (24%) 342 (23%) Posterior circulation 16 (1%) 17 (1%) ICA, or BA, or MCA and ACA 42 (3%) 33 (2%) MCA, or ACA, or PCA main 334 (22%) 326 (22%) Pre-existing brain changes Evidence of atrophy 1161 (77%) 1166 (77%) Evidence of leukoaraiosis 765 (5 1%) 782 (52%) Evidence of old infarcts 685 (45%) 651 (43%) Evidence of non-stroke lesions 73 (5%) 77 (5%) Data are number of patients (%). ACA=anterior cerebral artery. ASPECTS=Alberta Stroke Program Early CT Stroke. BA=basilar artery. ICA=internal carotid artery. IST-3=third International Stroke Trial. MCA=middle cerebral artery. NIHSS=National Institutes of Health Stroke Scale. PCA=posterior cerebral artery. * Assesses neurological deficit in stroke. † Refers to IST-3 image reading categorisation of lesion extent (all vascular territories). ‡ Assesses the extent of infarct affecting the MCA territory by subtracting a point for each of ten regions that are involved in the acute ischaemic lesion. § Classed as mild (ie, grey matter attenuation had become the same as normal white matter) or severe (ie, grey and white matter attenuation less than normal white matter). Table 2 Logistic linear regression analysis of associations between imaging signs and age, NIHSS score, and time to randomisation

‡ Assesses the extent of infarct affecting the MCA territory by subtracting a point for each of ten regions that are involved in the acute ischaemic lesion. § Classed as mild (ie, grey matter attenuation had become the same as normal white matter) or severe (ie, grey and white matter attenuation less than normal white matter). Table 2 Logistic linear regression analysis of associations between imaging signs and age, NIHSS score, and time to randomisation Age, adjusted for NIHSS score NIHSS score, adjusted for age Delay, adjusted for age and NIHSS score Odds ratio (95% CI) p Odds ratio (95% CI) p Odds ratio (95% CI) p Early ischaemic signs Visible infarct 0·98 (0·97–0·98) <0·0001 1·11 (1·09–1·12) <0·0001 1·08 (1·01–1·16) 0·019 Hypoattenuation 0·98 (0·97–0·98) <0·0001 1·10 (1·09–1·12) <0·0001 1·10 (1·03–1·18) 0·006 Large lesion* 0·98 (0·97–0·99) <0·0001 1·11 (1·09–1·13) <0·0001 1·03 (0·94–1·12) 0·561 Swelling 0·98 (0·97–0·99) <0·0001 1·09 (1·08–1·11) <0·0001 1·04 (0·96–1·13) 0·302 Hyperattenuated artery 0·98 (0·97–0·98) <0·0001 1·10 (1·09–1·12) <0·0001 1·02 (0·94–1·10) 0·664 Pre-existing signs Atrophy 1·11 (1·10–1·12) <0·0001 0·99 (0·98–1·00) 0·179 0·98 (0·89–1·07) 0·621 Leukoaraiosis 1·09 (1·08–1·09) <0·0001 0·99 (0·98–1·00) 0·221 0·99 (0·93–1·06) 0·843 Old infarct 1·03 (1·03–1·04) <0·0001 0·99 (0·98–1·00) 0·017 0·98 (0·92–1·05) 0·566 Associations for visible infarct (the summary variable) are provided for completeness. Odds ratios and 95% CIs indicate the increased or decreased odds of the imaging sign being present for a 1 point change in NIHSS score, a 1 year change in age, or a 1 h increase in time to randomisation. Each of the eight imaging variables was used separately in two logistic regressions: first on age and NIHSS score (both linear) to give the values in the first two pairs of columns; and second on age, NIHSS score, and time to randomisation (all linear) to give the values in the last pair of columns. IST-3=third International Stroke Trial. NIHSS=National Institutes of Health Stroke Scale.

c regressions: first on age and NIHSS score (both linear) to give the values in the first two pairs of columns; and second on age, NIHSS score, and time to randomisation (all linear) to give the values in the last pair of columns. IST-3=third International Stroke Trial. NIHSS=National Institutes of Health Stroke Scale. * Large lesion defined as a combination of large and very large on IST-3 score. Table 3 Logistic linear regression analysis of associations between individual imaging signs and primary and secondary outcomes, adjusted for age, NIHSS score, and time to randomisation

c regressions: first on age and NIHSS score (both linear) to give the values in the first two pairs of columns; and second on age, NIHSS score, and time to randomisation (all linear) to give the values in the last pair of columns. IST-3=third International Stroke Trial. NIHSS=National Institutes of Health Stroke Scale. * Large lesion defined as a combination of large and very large on IST-3 score. Table 3 Logistic linear regression analysis of associations between individual imaging signs and primary and secondary outcomes, adjusted for age, NIHSS score, and time to randomisation Symptomatic intracranial haemorrhage Death at 7 days or before Death at 6 months or before Alive and independent (OHS score 0–2) Favourable outcome (OHS score 0–1) Odds ratio (95% CI) p Odds ratio (95% CI) p Odds ratio (95% CI) p Odds ratio (95% CI) p Odds ratio (95% CI) p Early ischaemic signs Visible infarct 1·48 (1·00–2·19) 0·049 1·64 (1·24–2·16) 0·0004 1·39 (1·15–1·68) 0·0007 0·67 (0·55–0·81) <0·0001 0·63 (0·51–0·79) <0·0001 Hypoattenuation* 1·54 (1·04–2·27) 0·032 1·64 (1·25–2·16) 0·0004 1·39 (1·15–1·68) 0·0007 0·66 (0·55–0·81) <0·0001 0·62 (0·50–0·78) <0·0001 Severe hypoattenuation† 1·31 (0·67–2·56) 0·432 1·04 (0·63–1·74) 0·872 0·92 (0·63–1·33) 0·649 0·87 (0·60–1·27) 0·482 0·78 (0·51–1·19) 0·246 Large or very large lesion‡ 1·32 (0·85–2·05) 0·218 2·22 (1·67–2·96) <0·0001 2·07 (1·64–2·60) <0·0001 0·51 (0·38–0·68) <0·0001 0·40 (0·28–0·58) <0·0001 Very large lesion§ 1·51 (0·89–2·57) 0·131 3·20 (2·29–4·47) <0·0001 2·28 (1·68–3·11) <0·0001 0·29 (0·17–0·47) <0·0001 0·22 (0·10–0·46) <0·0001 Swelling 1·31 (0·87–1·97) 0·199 1·55 (1·17–2·06) 0·002 1·43 (1·16–1·77) 0·0008 0·59 (0·46–0·75) <0·0001 0·55 (0·41–0·73) <0·0001 Hyperattenuated arteries 1·54 (1·03–2·29) 0·034 1·44 (1·09–1·91) 0·009 1·41 (1·15–1·73) 0·001 0·59 (0·47–0·75) <0·0001 0·63 (0·48–0·83) 0·001 Pre-existing signs Any leukoaraiosis¶ 1·01 (0·68–1·50) 0·967 1·09 (0·82–1·45) 0·536 1·38 (1·14–1·67) 0·001 0·72 (0·59–0·87) 0·0007 0·62 (0·50–0·76) <0·0001 Severe leukoaraiosis‖ 1·15 (0·77–1·70) 0·499 1·17 (0·89– 1·54) 0·267 1·43 (1·18–1·72) 0·0002 0·66 (0·54–0·80) <0·0001 0·62 (0·50–0·78) <0·0001 Atrophy** 0·97 (0·58–1·64) 0·917 0·83 (0·57–1·20) 0·315 1·22 (0·93–1·60) 0·149 0·74 (0·59–0·94) 0·013 0·64 (0·50–0·82) 0·0004 Severe atrophy†† 1·02 (0·64–1·63) 0·923 0·87 (0·63–1·22) 0·422 1·28 (1·03–1·59) 0·026 0·79 (0·63–1·01) 0·057 0·75 (0·57–0·99) 0·040 Old infarct 1·72 (1·18–2·51) 0·005 0·94 (0·72–1·22) 0·622 1·05 (0·87–1·26) 0·603 0·88 (0·73–1·05) 0·149 0·79 (0·64–0·96) 0·017 Associations for visible infarct (the summary variable) are provided for completeness.

64–1·63) 0·923 0·87 (0·63–1·22) 0·422 1·28 (1·03–1·59) 0·026 0·79 (0·63–1·01) 0·057 0·75 (0·57–0·99) 0·040 Old infarct 1·72 (1·18–2·51) 0·005 0·94 (0·72–1·22) 0·622 1·05 (0·87–1·26) 0·603 0·88 (0·73–1·05) 0·149 0·79 (0·64–0·96) 0·017 Associations for visible infarct (the summary variable) are provided for completeness. The odds ratio is the estimated odds of an outcome happening when the imaging feature is present, divided by the odds of an outcome happening when the imaging feature is absent. NIHSS=National Institutes of Health Stroke Scale. OHS=Oxford Handicap Scale. * Mild or severe hypoattenuation versus none. † Severe hypoattenuation versus mild or none. ‡ Large or very large lesion versus no lesion or small or medium lesion. § Very large lesion versus no lesion or small, medium, or large lesion. ¶ Mild or severe leukoaraiosis versus none. ‖ Severe leukoaraiosis versus mild or none. ** Moderate or severe atrophy versus none. †† Severe atrophy versus moderate or none. Table 4 Full multivariate logistic regression models for symptomatic intracranial haemorrhage and functional outcome at 6 months

§ Very large lesion versus no lesion or small, medium, or large lesion. ¶ Mild or severe leukoaraiosis versus none. ‖ Severe leukoaraiosis versus mild or none. ** Moderate or severe atrophy versus none. †† Severe atrophy versus moderate or none. Table 4 Full multivariate logistic regression models for symptomatic intracranial haemorrhage and functional outcome at 6 months OHS score 0–2 versus OHS score 3–6 Symptomatic intracranial haemorrhage Odds ratio (95% CI) p Odds ratio (95% CI) p Age (years) 0·96 (0·96–0·97) <0·0001 1·00 (0·98–1·02) 0·911 NIHSS score 0·83 (0·82–0·85) <0·0001 1·06 (1·03–1·10) <0·0001 Time to randomisation (h) 1·04 (0·96–1·13) 0·303 0·98 (0·83–1·16) 0·814 Alteplase versus control 1·13 (0·94–1·35) 0·192 6·65 (3·89–11·35) <0·0001 Antiplatelets at the time of stroke versus none .. <0·0001 1·60 (1·07–2·38) 0·021 Large or very large lesion versus small, medium, or no lesion 0·69 (0·49–0·99) 0·043 0·97 (0·55–1·72) 0·919 Swelling 0·79 (0·56–1·11) 0·168 0·95 (0·54–1·69) 0·867 Hyperattenuated artery 0·70 (0·54–0·91) 0·007 1·45 (0·92–2·28) 0·114 Mild tissue hypoattenuation versus none 1·11 (0·52–2·38) 0·783 0·68 (0·17–2·78) 0·588 Severe tissue hypoattenuation versus none 1·40 (0·61–3·21) 0·432 0·71 (0·15–3·26) 0·658 Old infarcts 0·90 (0·74–1·09) 0·278 1·75 (1·17–2·63) 0·007 Mild leukoaraiosis versus none 0·81 (0·65–1·00) 0·051 1·01 (0·64–1·59) 0·975 Severe leukoaraiosis versus none 0·64 (0·48–0·85) 0·002 0·92 (0·51–1·66) 0·772 Mild atrophy versus none 0·87 (0·67–1·12) 0·273 0·83 (0·47–1·48) 0·532 Severe atrophy versus none 0·72 (0·52–1·00) 0·052 0·82 (0·40–1·67) 0·590 Numbers of patients in each category are shown in table 1. Odds ratios indicate the increased or decreased odds of the outcome being present for a 1 year increase in age, a 1 point increase in NIHSS score, or a 1 h increase in time to randomisation. Hosmer-Lemeshow tests for lack of fit: OHS outcome, p=0·59; symptomatic intracranial haemorrhage outcome, p=0·76. Additional time windows are shown in the appendix (pp 8–9). NIHSS=National Institutes of Health Stroke Scale. OHS=Oxford Handicap Scale.

increase in NIHSS score, or a 1 h increase in time to randomisation. Hosmer-Lemeshow tests for lack of fit: OHS outcome, p=0·59; symptomatic intracranial haemorrhage outcome, p=0·76. Additional time windows are shown in the appendix (pp 8–9). NIHSS=National Institutes of Health Stroke Scale. OHS=Oxford Handicap Scale. Table 5 Multivariate logistic regression models selected by stepwise logistic regression for symptomatic intracranial haemorrhage and functional outcome at 6 months

increase in NIHSS score, or a 1 h increase in time to randomisation. Hosmer-Lemeshow tests for lack of fit: OHS outcome, p=0·59; symptomatic intracranial haemorrhage outcome, p=0·76. Additional time windows are shown in the appendix (pp 8–9). NIHSS=National Institutes of Health Stroke Scale. OHS=Oxford Handicap Scale. Table 5 Multivariate logistic regression models selected by stepwise logistic regression for symptomatic intracranial haemorrhage and functional outcome at 6 months OHS score 0–2 versus OHS score 3–6 Symptomatic intracranial haemorrhage Odds ratio (95% CI) p Odds ratio (95% CI) p Age (years) 0·96 (0·95–0·97) <0·0001 0·99 (0·98–1·01) 0·556 NIHSS score 0·83 (0·82–0·85) <0·0001 1·07 (1·04–1·10) 0·0001 Time to randomisation (h) 1·04 (0·96–1·13) 0·288 0·98 (0·83–1·15) 0·798 Alteplase versus control 1·12 (0·94–1·34) 0·196 6·71 (3·93–11·46) <0·0001 Antiplatelets at the time of stroke versus none .. .. 1·60 (1·07–2·38) 0·021 Small or medium lesion versus no lesion 0·85 (0·67–1·07) 0·173 .. .. Large or very large lesion versus no lesion 0·54 (0·39–0·74) 0·0001 .. .. Hyperattenuated artery versus none 0·71 (0·55–0·92) 0·009 1·61 (1·07–2·42) 0·023 Old infarct versus none .. .. 1·67 (1·13–2·46) 0·009 Mild leukoaraiosis versus none 0·76 (0·62–0·93) 0·009 .. .. Severe leukoaraiosis versus none 0·59 (0·45–0·79) 0·0003 .. .. Numbers of patients in each category are shown in table 1. Odds ratios indicate the increased or decreased odds of the clinical factor being present for a 1 year increase in age, a 1 point increase in NIHSS score, or a 1 h increase in time to randomisation. Models selected by stepwise logistic regression from full models shown in table 4. Additional time windows are shown in the appendix (pp 8–9). The first four variables were forced into all models. We used p≤0·05 as criteria for both forward and backward steps. Blank cells represent variables that were dropped as non-significant during stepwise selection. The final nominal p values take no account of model selection. Age, NIHSS score, and time to randomisation were entered into the model as continuous variables; thus the odds ratio for age represents the estimated change in odds of the outcome for a 1 year increase in age, with all other variables unchanged. The units for NIHSS score are points on a scale from 0 to 37 (maximum observed in this trial). Factors with three levels were either retained or excluded at each step; the method did not permit separate consideration of the individual 1 df contrasts comprising the three-level factors. NIHSS=National Institutes of Health Stroke Scale. OHS=Oxford Handicap Scale.

scale from 0 to 37 (maximum observed in this trial). Factors with three levels were either retained or excluded at each step; the method did not permit separate consideration of the individual 1 df contrasts comprising the three-level factors. NIHSS=National Institutes of Health Stroke Scale. OHS=Oxford Handicap Scale. Panel Research in context Systematic review

scale from 0 to 37 (maximum observed in this trial). Factors with three levels were either retained or excluded at each step; the method did not permit separate consideration of the individual 1 df contrasts comprising the three-level factors. NIHSS=National Institutes of Health Stroke Scale. OHS=Oxford Handicap Scale. Panel Research in context Systematic review We searched the Cochrane Stroke Trials Registry, Medline, and Embase from January, 1990, to March, 2015, to identify all previous randomised controlled trials of thrombolysis versus control in acute ischaemic stroke with data for CT brain imaging and any interaction with thrombolysis.31 Three trials provided data for the extent of early ischaemic signs on CT brain imaging and functional outcome, in patients treated by thrombolysis and in controls, dichotomised on the Alberta Stroke Programme Early CT Signs (ASPECTS) score 0–7 and 8–10. Two trials tested intravenous alteplase (0·9 mg/kg)8,32 and one tested intra-arterial pro-urokinase.33 No other trials had data for imaging signs and outcome by treatment allocation for meta-analysis. In the three previous trials,8,32,33 patients with large early ischaemic lesions (ASPECTS score 0–7, n=521) had similar odds of good outcome (odds ratio 1·41, 95% CI 0·95–2·09) to those with small or no lesions (ASPECTS score 8–10, n=1029; odds ratio 1·59, 95% CI 1·24–2·04) after thrombolysis (test for subgroup differences, p=0·61). In the third International Stroke Trial,12 patients with large lesions (ASPECTS score 0–7, n=729) had similar odds of good outcome (odds ratio 1·03, 95% CI 0·70–1·51) to those with small or no visible ischaemic lesions (ASPECTS score 8–10, n=2288; odds ratio 1·08, 95% CI 0·91–1·27). Combining all four trials (figure 4) shows that patients with large infarcts (ASPECTS 0–7, n=1250) had similar odds of good outcome (odds ratio 1·20, 95% CI 0·91–1·58) to patients with small or no visible infarct (ASPECTS 8–10, n=3317; odds ratio 1·21, 95% CI 1·06–1·39) after thrombolysis (test for subgroup difference, p=0·94). These findings accord with those of the Interventional Management of Stroke III trial (no interaction between ASPECTS score and thrombectomy)34 and trials in which the classification of early ischaemic signs on CT precluded meta-analysis with our data (no interaction between early ischaemic signs and streptokinase in 1292 patients within 6 h,35 or alteplase in 624 patients in the National Institute of Neurological Disorders and Stroke trial within 3 h of stroke)6 on functional outcome.

assification of early ischaemic signs on CT precluded meta-analysis with our data (no interaction between early ischaemic signs and streptokinase in 1292 patients within 6 h,35 or alteplase in 624 patients in the National Institute of Neurological Disorders and Stroke trial within 3 h of stroke)6 on functional outcome. No data were available in these trials for analysis of pre-existing CT signs and effects of alteplase, nor to adjust for clinical prognostic variables. Interpretation Absence of an interaction between early ischaemic signs and thrombolysis has been disregarded by many up until now, perhaps because data came (in part) from trials that were neutral or that used a drug perceived to be harmful (eg, streptokinase). All data combined show that the extent of early ischaemic signs on CT per se should not deter treatment with alteplase. A meta-analysis of CT findings from individual patients in all trials of alteplase is planned as part of the Stroke Thrombolysis Trialists' Collaboration, to test further the effect of early ischaemic and pre-existing structural signs and combinations of imaging signs for interactions with alteplase, adjusted for baseline prognosis.2 Future trials should not only assess effects of pre-existing and early ischaemic changes for interactions with acute stroke treatments but also consider use of imaging variables for minimisation in view of their strong independent prognostic value.

Introduction Intravenous alteplase has been approved for treatment of acute ischaemic stroke in Europe for patients who are younger than 80 years and can be treated within 4·5 h. Such use is associated with improved functional outcome at 3 months after stroke,1 but whether treatment improves survival and sustains functional recovery in the long term is unclear. Of the 12 completed randomised controlled trials, ten reported outcomes at 90 days or less,1 two reported outcomes at 6 months,2,3 and one reported outcomes at 12 months,3 but none have reported effects at more than 1 year after stroke. Furthermore, the effect of thrombolysis on health-related quality of life—an important measure of the clinical and economic value of treatment—has not been reported to our knowledge.

ported outcomes at 6 months,2,3 and one reported outcomes at 12 months,3 but none have reported effects at more than 1 year after stroke. Furthermore, the effect of thrombolysis on health-related quality of life—an important measure of the clinical and economic value of treatment—has not been reported to our knowledge. The third International Stroke Trial (IST-3)2 recruited 3035 patients—half of whom were older than 80 years—to assess the effect of thrombolytic treatment with intravenous alteplase within 6 h of onset of acute ischaemic stroke. The results showed that although thrombolytic treatment was not associated with a significant difference in the proportion of patients who were alive and independent at 6 months, treatment did seem to improve functional outcome. A prespecified secondary ordinal analysis of Oxford handicap scale scores showed that treatment was associated with a favourable shift in the distribution of Oxford handicap scale scores (odds ratio [OR] 1·27, 95% CI 1·10–1·47; p=0·001).2 A secondary aim of IST-3 was to assess whether thrombolytic treatment improved outcomes more than 1 year after stroke, and sought to assess survival, functional outcome, health-related quality of life, overall functioning, and living circumstances at 18 months.4,5

res (odds ratio [OR] 1·27, 95% CI 1·10–1·47; p=0·001).2 A secondary aim of IST-3 was to assess whether thrombolytic treatment improved outcomes more than 1 year after stroke, and sought to assess survival, functional outcome, health-related quality of life, overall functioning, and living circumstances at 18 months.4,5 Methods Study design and participants The methods of the trial have been described in full previously.2,4–6 IST-3 was a randomised, open-label trial of intravenous alteplase (0·9 mg/kg) plus standard care compared with standard care alone (control). Eligibility criteria were: symptoms and signs of clinically definite acute stroke, known time of stroke onset, treatment could be started within 6 h of onset, and exclusion by CT or MRI of intracranial haemorrhage and structural brain lesions that could mimic stroke (eg, cerebral tumour). A patient could only be included in the trial if both they (or a proxy) and their clinician believed that the treatment was promising but unproven—ie, there was neither a clear indication for treatment, nor a clear contraindication against treatment. The effect that using this uncertainty principle approach as a key eligibility criterion had on the type of patients included and excluded from the trial has been described in detail elsewhere.2,6 Generally, patients who could be treated within licence were rarely enrolled, unless there was a specific reason that led the clinician or patient to be uncertain about whether to treat or not; as a result, 95% of enrolled patients did not meet the terms of the prevailing EU approval for treatment. All participants or proxies gave informed consent. The protocol was approved by the Multi-Centre Research Ethics Committee (Scotland) and by local ethics committees.

t to be uncertain about whether to treat or not; as a result, 95% of enrolled patients did not meet the terms of the prevailing EU approval for treatment. All participants or proxies gave informed consent. The protocol was approved by the Multi-Centre Research Ethics Committee (Scotland) and by local ethics committees. For the analysis presented here, we planned to assess outcome in patients who had follow-up at 6 months and 18 months. In seven countries (Austria, Belgium, Canada, Italy, Mexico, Poland, and UK) follow-up had to cease on Jan 30, 2012; therefore, we excluded any patients from these countries who were recruited after June 30, 2010, because they would not reach the 18-month follow-up point. In three countries (Australia, Norway, and Sweden), all recruited patients were to be followed up to 18 months, as part of a sub-study. Two countries (Portugal and Switzerland) followed up patients to 6 months only and were not included in this analysis.

2010, because they would not reach the 18-month follow-up point. In three countries (Australia, Norway, and Sweden), all recruited patients were to be followed up to 18 months, as part of a sub-study. Two countries (Portugal and Switzerland) followed up patients to 6 months only and were not included in this analysis. Randomisation After enrolment, patients were randomly assigned by a secure central telephone or web-based computer system, which recorded baseline data and generated the treatment allocation only after the baseline data had been checked for range and consistency. The system used a minimisation algorithm to balance for key prognostic factors: geographic region, age, National Institutes of Health stroke scale score, sex, time since onset of stroke, stroke clinical syndrome, and presence or absence of visible ischaemic change on the pre-enrolment brain scan.4,5 To avoid predictable alternation of treatment allocation, and thus potential loss of allocation concealment, patients were allocated with a probability of 0·80 to the treatment group that would minimise the difference between the groups for the key prognostic factors. Recruitment in the small double-blind phase (n=276) began in May, 2000, continued without interruption into the open-treatment phase (n=2759), and was completed in July, 2011.

llocated with a probability of 0·80 to the treatment group that would minimise the difference between the groups for the key prognostic factors. Recruitment in the small double-blind phase (n=276) began in May, 2000, continued without interruption into the open-treatment phase (n=2759), and was completed in July, 2011. Procedures In the ten countries participating in follow-up at 6 months and 18 months after enrolment (Australia, Austria, Belgium, Canada, Italy, Mexico, Norway, Poland, Sweden, and UK), if the patient was not known to have died, staff at each national coordinating centre contacted the patient's doctor (or hospital coordinator) to confirm that the patient was alive and that they might be approached for follow-up. In Austria and Italy, experienced stroke physicians, masked to treatment allocation, contacted all patients by telephone. In the other eight countries, IST-3 trial office staff posted a questionnaire to patients to assess outcome. Non-responders were sent a second questionnaire. If no questionnaire was returned, an experienced, masked clinician or stroke nurse assessed the patient by telephone interview. Telephone assessment of disability in stroke survivors is as valid as face-to-face interviews7 and postal questionnaires.8

s to assess outcome. Non-responders were sent a second questionnaire. If no questionnaire was returned, an experienced, masked clinician or stroke nurse assessed the patient by telephone interview. Telephone assessment of disability in stroke survivors is as valid as face-to-face interviews7 and postal questionnaires.8 The primary outcome of the trial was the proportion of patients alive and independent with an Oxford handicap scale9 score of 0–2 at 6 months (this outcome was chosen instead of survival alone because many people regard survival after a stroke in a disabled or dependent state as worse than death). The secondary endpoints at 18 months were: survival, Oxford handicap scale score, health-related quality of life, overall functioning, and living circumstances. The Oxford handicap scale is a six-point scale almost identical to the modified Rankin scale.10 In emergency care of acute ischaemic stroke, recording quality of life at baseline before randomisation was not possible; instead, quality of life was measured at 6 months and 18 months with the EuroQoL instrument,11 which assesses current self-rated health by a combination of questions about wellbeing and a visual analogue scale score. The questions are about the five dimensions of mobility, self-care, activity, pain or discomfort, and anxiety (the EQ-5D). Each dimension has three levels (no problems, some problems, severe problems), which can be presented individually. A unique health state is defined by combining one level from each of the five dimensions. Patients' responses can then be combined into an EQ utility index with scores ranging from −1 to +1 (where +1 represents perfect health, 0 represents a state equivalent to death, and −1 represents a state worse than death). Calculation of the EQ utility index requires valuations for all health states, and these have been estimated for the UK and other European populations.12 For the visual analogue scale, 100 represents the best imaginable health and 0 the worst imaginable health.

ent to death, and −1 represents a state worse than death). Calculation of the EQ utility index requires valuations for all health states, and these have been estimated for the UK and other European populations.12 For the visual analogue scale, 100 represents the best imaginable health and 0 the worst imaginable health. We used the EuroQoL instrument because it is short and simple, and in patients with stroke it has been validated,13–17 is responsive to change,18 and is associated with higher response rates and fewer missing data than more complex instruments.16 Many patients who have had severe strokes might not be able to complete the questionnaire themselves and because responses from a proxy have reasonable validity,15,19 we therefore accepted responses submitted by a spouse, partner, close relative, or carer.

e rates and fewer missing data than more complex instruments.16 Many patients who have had severe strokes might not be able to complete the questionnaire themselves and because responses from a proxy have reasonable validity,15,19 we therefore accepted responses submitted by a spouse, partner, close relative, or carer. We also assessed binary (yes or no) answers to two questions, about global functioning: “Has the stroke left you with any problems?” and activities of daily living: “Do you need help from anybody with everyday activities (in washing, dressing, feeding, and going to the toilet)?” These questions have been validated17 and were used previously in a large trial.20 We also asked whether patients were living in their own home, a relative's home, a residential home, a nursing home, or were still in hospital. Finally, the questionnaire asked patients enrolled in the open-label treatment phase what treatments they recalled being given in hospital, including thrombolysis with alteplase. If the patient or proxy did not complete a specific item on a postal questionnaire, we did not re-contact them.

e, or were still in hospital. Finally, the questionnaire asked patients enrolled in the open-label treatment phase what treatments they recalled being given in hospital, including thrombolysis with alteplase. If the patient or proxy did not complete a specific item on a postal questionnaire, we did not re-contact them. Statistical analysis All randomly assigned patients who were due to be followed up at 18 months were included in the analysis of survival. We constructed Kaplan-Meier survival curves, and compared treatment groups with the log-rank test. Survival times were censored at 548 days after enrolment if patients died at a later date or returned an 18-month form at a later date. For patients from the Australia, Norway, Sweden, and UK, where reporting of deaths was prompt, if there was no known death date and no return of an 18-month form, patients were censored at 548 days. For patients from other countries who had no reported death date and no 18-month form, survival was censored at the date of return of the 6-month form or at the last date of contact, whichever was later. The justification for, and the methods for statistical adjustment of, the outcomes and the ordinal analyses of the Oxford handicap scale score at 18 months were specified in the statistical analysis plan and also described in the report of the primary outcomes.2,5 We divided the Oxford handicap scale into five levels: 0, 1, 2, and 3 were retained and 4, 5, and 6 were combined into a single level. The treatment OR between one level and the next was assumed to be constant, so a single parameter (a common OR) summarises the shift in outcome distribution between treatment and control groups.

d the Oxford handicap scale into five levels: 0, 1, 2, and 3 were retained and 4, 5, and 6 were combined into a single level. The treatment OR between one level and the next was assumed to be constant, so a single parameter (a common OR) summarises the shift in outcome distribution between treatment and control groups. In the main analysis, we report results without imputing missing data. In the sensitivity analysis, for patients with an unknown Oxford handicap scale score at 18 months, we imputed the value from their 6-month assessment (last observation carried forward). For the EuroQoL instrument, we analysed the three levels of each EQ-5D domain as ordered categories by ordinal logistic regression, calculated the mean overall difference in visual analogue scale score between treatment groups, and estimated the EQ-5D index—calculated with a set of valuations derived from a sample of the UK population with the time trade-off method and also the UK visual analogue scale and European visual analogue scale valuations.12 Analyses were adjusted for baseline prognostic factors (age, National Institutes of Health stroke scale score, delay between onset and enrolment, and presence of acute ischaemic change on the baseline scan). We did several sensitivity analyses to assess the effect of missing data for Oxford handicap scale score and EQ-5D, and we assessed the effect of setting utility to zero for patients who had died. We did subgroup analyses of the effect of treatment on Oxford handicap scale score (ordinal logistic regression, as in the study by Frank and colleagues21) and utility subdivided by age (>80 vs ≤80 years), time to randomisation (≤3·0, >3·0–4·5, >4·5–6·0 h), baseline National Institutes of Health stroke scale score (0–5, 6–15, 16–25, >25), phase of the trial (masked vs open label), and by the person completing the form (patient vs proxy). For National Institutes of Health stroke scale score, we also fitted a model with baseline severity as a linear regressor with treatment-specific slopes. Analyses were done with SAS (version 9.3).

–5, 6–15, 16–25, >25), phase of the trial (masked vs open label), and by the person completing the form (patient vs proxy). For National Institutes of Health stroke scale score, we also fitted a model with baseline severity as a linear regressor with treatment-specific slopes. Analyses were done with SAS (version 9.3). This study is registered with controlled-trials.com, number ISRCTN25765518. Role of the funding source The sponsors had no role in data collection, data storage, data analysis, preparation of this report, or the decision to publish. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. Results Of the 3035 patients enrolled by 156 hospitals in 12 countries, 2348 (77·4%) met the criteria for inclusion in the 18-month follow-up study—1169 assigned to alteplase, 1179 assigned to control (figure 1). The baseline characteristics of this subset were well balanced between groups (table 1) and were not much different from those who were ineligible for the 18-month follow-up analysis (appendix). Of the 2348 patients scheduled for 18-month follow-up, vital status and Oxford handicap scale score at 18 months were known for 2290 (97·5%). Survival at 18 months did not differ significantly between groups: 408 of 1169 (34·9%) participants allocated to alteplase versus 414 of 1179 (35·1%) allocated to control died (log-rank p=0·85; figure 2).

uled for 18-month follow-up, vital status and Oxford handicap scale score at 18 months were known for 2290 (97·5%). Survival at 18 months did not differ significantly between groups: 408 of 1169 (34·9%) participants allocated to alteplase versus 414 of 1179 (35·1%) allocated to control died (log-rank p=0·85; figure 2). At 18 months, of 2348 participants, vital status and disability were known for 2239 (95·3 %), vital status only was known for 51 (2·2%), and vital status and disability were unknown for 58 (2·5%). Oxford handicap scale scores were available for 1117 participants assigned to alteplase versus 1122 assigned to control. 391 (35·0%) patients allocated to alteplase versus 352 (31·4%) allocated to control were alive and independent (Oxford handicap scale score 0–2) at 18 months (adjusted odds ratio 1·28, 95% CI 1·03–1·57; p=0·024; unadjusted OR 1·18, 95% CI 0·99–1·40; p=0·068; table 2), with a favourable shift in Oxford handicap scale score (adjusted common OR 1·30, 95% CI 1·10–1·55; p=0·002). The size and statistical significance of the effect on Oxford handicap scale score at 18 months was robust to sensitivity analyses for missing data (data not shown). The appendix shows Oxford handicap scale score at 6 months in patients scheduled for 18-month follow-up who had data available at 6 months.

·55; p=0·002). The size and statistical significance of the effect on Oxford handicap scale score at 18 months was robust to sensitivity analyses for missing data (data not shown). The appendix shows Oxford handicap scale score at 6 months in patients scheduled for 18-month follow-up who had data available at 6 months. The EQ utility index could be calculated for 1341 (91·3%) of the 1468 patients who were alive at 18 months. 591 (44%) of these assessments were completed by patients themselves, 724 (54%) by a valid proxy, and 25 (2%) by a doctor. Treatment was associated with significant improvements in mobility, self-care, ability to do usual activities, and pain or discomfort, with no evidence of an effect on anxiety or depression (table 3). At 18 months, alteplase was associated with significantly fewer patients reporting being left with problems and needing help with everyday activities (table 3). Although treatment with alteplase was associated with a significantly higher EQ utility index in survivors (p=0·028; table 4), the mean adjusted difference in visual analogue scale score was not significant (p=0·072; table 4). These findings were robust in the sensitivity analyses (data not shown). The appendix shows EQ-5D, EQ utility index, and visual analogue scale score at 6 months and 18 months using different valuations. Of the participants who were still alive, the proportion who were resident at home did not differ significantly between groups (appendix).

obust in the sensitivity analyses (data not shown). The appendix shows EQ-5D, EQ utility index, and visual analogue scale score at 6 months and 18 months using different valuations. Of the participants who were still alive, the proportion who were resident at home did not differ significantly between groups (appendix). For the ordinal subgroup analysis of Oxford handicap scale score at 18 months, significant interactions existed between baseline variables and treatment effect. Greater differences in favour of alteplase were reported for age older than 80 years (p=0·032) and high National Institutes of Health stroke scale score (p=0·021), but not for time to treatment, respondent (patient vs proxy), or masking of assessment of outcome (double blind vs open label; appendix). When age, delay, and National Institutes of Health stroke scale score were treated as continuous variables, the interaction of ordinal Oxford handicap scale score with age became non-significant, delay remained non-significant, and for National Institutes of Health stroke scale score the p value for a trend was 0·004 (appendix). For EQ utility index, when subgroups were in discrete categories, none of the interactions were statistically significant (appendix). However, when the National Institutes of Health stroke scale score was treated as continuous, every five-point increase in score reduced the EQ utility index by 0·12 in the alteplase group versus 0·15 in the control group (adjusted estimates; p=0·008 for difference in slopes). For delay in enrolment time and age there was no trend in EQ utility index, irrespective of whether the variables were grouped or entered into models as a linear trend (data not shown).

utility index by 0·12 in the alteplase group versus 0·15 in the control group (adjusted estimates; p=0·008 for difference in slopes). For delay in enrolment time and age there was no trend in EQ utility index, irrespective of whether the variables were grouped or entered into models as a linear trend (data not shown). Of the 1468 patients who were alive at 18 months, 1260 were asked to recall if they had been given thrombolytic treatment (appendix); 273 in the alteplase group versus 156 in the control group correctly recalled whether or not they had received thrombolytic treatment. In both treatment groups, the ability to recall treatment correctly was associated with better outcome; patients with correct recall were more likely to have an Oxford handicap scale score of 0–2 than were those who remembered incorrectly or did not know (62·5% vs 49·3%; 0·0001). Of patients with correct recall, those treated with alteplase were more likely to have an Oxford handicap scale score of 0–2 than were those in the control group (66·7% vs 55·1%; 0·018), whereas of those who did not remember correctly, outcomes did not differ significantly between groups (OHS 0–2 48·6% vs 49·9%; 0·714); a significant interaction existed between recall status and treatment (p<0·0001).

ave an Oxford handicap scale score of 0–2 than were those in the control group (66·7% vs 55·1%; 0·018), whereas of those who did not remember correctly, outcomes did not differ significantly between groups (OHS 0–2 48·6% vs 49·9%; 0·714); a significant interaction existed between recall status and treatment (p<0·0001). Discussion We have shown that, for treatment of acute ischaemic stroke, thrombolysis with intravenous alteplase seems to provide a benefit at 18 months. Treatment had no effect on survival, but was associated with a significant increase in the likelihood of being alive and independent. However, the unadjusted absolute difference in the number of patients alive and independent at 18 months was not significant, so judgment on whether or not the results are clinically significant rests on the quality of the data and the overall patterns of effect seen across all measures. The ordinal estimates of effect at 6 months and 18 months were similar and significant. Treatment was also associated with a gain in health-related quality of life that was significant for four of the five dimensions of the EQ-5D and the overall EQ utility index (though not for visual analogue scale score). Living circumstances did not differ significantly between groups.

ths were similar and significant. Treatment was also associated with a gain in health-related quality of life that was significant for four of the five dimensions of the EQ-5D and the overall EQ utility index (though not for visual analogue scale score). Living circumstances did not differ significantly between groups. Strengths of this study are the large number of patients and the completeness of follow-up. Of the patients scheduled for 18-month follow-up, a small proportion were missing data for both vital and functional outcome status. We estimated the EQ utility index in more than 91% of survivors (a similar proportion to that in a trial23 of younger and less impaired patients with coronary artery disease) and our sensitivity analyses also showed that the estimates of overall health-related quality of life with the EQ utility index were robust to various assumptions about missing data. Although thrombolytic treatment was associated in survivors with less functional impairment, better health-related quality of life, and less likelihood of being left with problems and needing help with daily activities after stroke, it did not translate into a higher proportion of patients living at home at 18 months, perhaps because living circumstances are affected by social and financial factors that are not influenced by treatment. We believe that the direction and size of the effects are clinically significant and will inform health economic assessments of thrombolytic treatment. For example, in 2002, the estimated cost of long-term care of an independent survivor of stroke was £876 per year and that of a dependent survivor was £11 292 per year,24 so even a small difference in the proportion of patients who survive and are independent will have substantial economic impact.

rombolytic treatment. For example, in 2002, the estimated cost of long-term care of an independent survivor of stroke was £876 per year and that of a dependent survivor was £11 292 per year,24 so even a small difference in the proportion of patients who survive and are independent will have substantial economic impact. Lyden25 has identified limitations of IST-3, chiefly that treatment was not masked. Patient-reported outcomes—eg, health-related quality of life—are subjective,26 and recall of thrombolytic treatment could affect patient responses. Only 30% of survivors correctly recalled whether or not they had received thrombolytic treatment. As expected, accurate recall was associated with better outcome in both treatment groups. Thus, recall bias might have affected our findings. However, the analysis of recall was based on a variable measured in a subset of survivors after randomisation and so could itself be biased. The effects of treatment on the Oxford handicap scale score and EQ utility index were much the same in the masked and open-label parts of the study (appendix). Assessment of health-related quality of life is limited because many patients who have had a stroke are unable to complete the form themselves. The high proportion of forms completed by a proxy in IST-3 is a result of the severity of stroke in the patients included in the trial. Although the use of surrogates is a potential weakness, it did enable us to achieve satisfactory response rates; however, because proxies tend to assign worse health status than do patients,15 we were reassured that there was no interaction between the person responding and the effect of treatment on utility or Oxford handicap scale score. Not all enrolled patients were scheduled to be followed up for 18 months, but the selection criteria for the longer follow-up cohort did not seem to introduce relevant imbalances at baseline, nor were the characteristics of the cohort substantially different from those not included in long-term follow-up. We therefore believe the 18-month follow-up cohort is representative of the trial as a whole.

e selection criteria for the longer follow-up cohort did not seem to introduce relevant imbalances at baseline, nor were the characteristics of the cohort substantially different from those not included in long-term follow-up. We therefore believe the 18-month follow-up cohort is representative of the trial as a whole. Another weakness is that the trial was under-powered, so the subgroup analyses of the effects of baseline age, stroke severity, and delay to enrolment on the Oxford handicap scale score and health-related quality of life should be treated with caution. These are secondary analyses of a secondary outcome, and the apparent lack of effect of time to treatment might be due to chance. Furthermore, a more appropriate assessment of the complex interactions between age, stroke severity, and time to treatment will be available from a meta-analysis of individual patient data by the Stroke Thrombolysis Trialists.27 In conclusion, IST-3 adds to the evidence from previous trials (panel) and shows that although thrombolysis for acute ischaemic stroke with intravenous alteplase does not improve survival, there is evidence of improvement in several measures of function and quality of life in survivors of all ages for up to 18 months after treatment. Correspondence to: Prof Peter Sandercock, Division of Clinical Neurosciences, University of Edinburgh, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK peter.sandercock@ed.ac.uk Supplementary Material Supplementary appendix

In conclusion, IST-3 adds to the evidence from previous trials (panel) and shows that although thrombolysis for acute ischaemic stroke with intravenous alteplase does not improve survival, there is evidence of improvement in several measures of function and quality of life in survivors of all ages for up to 18 months after treatment. Correspondence to: Prof Peter Sandercock, Division of Clinical Neurosciences, University of Edinburgh, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK peter.sandercock@ed.ac.uk Supplementary Material Supplementary appendix Acknowledgments We thank all the patients who participated in the study, and the many individuals not specifically mentioned in the report who have supported the study. We also thank NIHR Stroke Research Network, NHS Research Scotland, and the National Institute for Social Care and Health Research Clinical Research Centre. IST-3 is an investigator-led trial. The University of Edinburgh and the NHS Lothian Health Board are cosponsors. The double-blind start-up phase was supported by the Stroke Association (UK). The expansion phase was funded by the Health Foundation UK. The main phase of the trial was funded by: UK Medical Research Council and managed by the National Institutes of Health Research on behalf of the MRC-NIHR partnership, the Research Council of Norway, Arbetsmarknadens Partners Forsakringsbolag Insurances Sweden, the Swedish Heart Lung Fund, The Foundation of Marianne and Marcus Wallenberg, Stockholm County Council, Karolinska Institutet, Polish Ministry of Science and Education, Australian Heart Foundation, Australian National Health and Medical Research Council, Swiss National Research Foundation, Swiss Heart Foundation, Foundation for Health and Cardio-/Neurovascular Research (Basel, Switzerland), Assessorato alla Sanita (Regione dell'Umbria, Italy), and Danube University (Krems, Austria). Boehringer Ingelheim GmbH donated drug and placebo for the double-blind phase, but thereafter had no role in the trial. The UK Stroke Research Network adopted the trial on May 1, 2006, supported the initiation of new UK sites, and in some centres, data were collected by staff funded by the Network or working for associated NHS organisations. Central imaging was done at the Brain Imaging Research Centre (University of Edinburgh), partly funded by the Scottish Funding Council and the Chief Scientist Office of the Scottish Executive. Additional support was received from Chest Heart and Stroke Scotland, DesAcc, University of Edinburgh, Danderyd Hospital R&D Department, Karolinska Institutet, Oslo University Hospital, and the Dalhousie University Internal Medicine Research Fund.

unding Council and the Chief Scientist Office of the Scottish Executive. Additional support was received from Chest Heart and Stroke Scotland, DesAcc, University of Edinburgh, Danderyd Hospital R&D Department, Karolinska Institutet, Oslo University Hospital, and the Dalhousie University Internal Medicine Research Fund. Contributors The study was conceived by the cochief investigators—PS, RIL, and JW. JW led the imaging. The study was designed by PS, RIL, and JW, with input from all others who were coordinators of the trial in their own country. PS, RIL, JW, MD, and KI wrote the protocol. KI is the study coordinator. GC is the study statistician who prepared the analyses. GM advised on statistical aspects. PS, RIL, MD, WW, GV, AC, AK, EB, KBS, VM, AP, GH, KM, SR, GG, SP, AA, MC, and PL recruited patients. GV, AC, AK, EB, KBS, VM, AP, GH, KM, SR, GG, SP, AA, MC, and PL were national coordinators. PS wrote the first draft and all authors commented on subsequent drafts and approved the final version.

on statistical aspects. PS, RIL, MD, WW, GV, AC, AK, EB, KBS, VM, AP, GH, KM, SR, GG, SP, AA, MC, and PL recruited patients. GV, AC, AK, EB, KBS, VM, AP, GH, KM, SR, GG, SP, AA, MC, and PL were national coordinators. PS wrote the first draft and all authors commented on subsequent drafts and approved the final version. Writing committee University of Edinburgh (Edinburgh, UK)—Peter Sandercock, Joanna M Wardlaw, Martin Dennis, Geoff Cohen, Gordon Murray, Karen Innes, Will Whiteley; Sydney Medical School—Westmead Hospital and The George Institute for Global Health (University of Sydney, Sydney, Australia)—Richard I Lindley; Sheffield Teaching Hospitals NHS Foundation Trust (Sheffield, UK)—Graham Venables; Institute of Psychiatry and Neurology and Medical University of Warsaw (Warsaw, Poland)—Anna Czlonkowska; Institute of Psychiatry and Neurology (Warsaw, Poland)—Adam Kobayashi; Department of Neurology (Ospedale, Citta' di Castello, Italy)—Stefano Ricci; Karolinska Institutet (Stockholm, Sweden)—Veronica Murray; Oslo University Hospital (Oslo, Norway)—Eivind Berge, Karsten Bruins Slot; School of Medicine and Pharmacology (The University of Western Australia, Perth, Australia) and Royal Perth Hospital (Perth, Australia)—Graeme J Hankey; Hospital Geral de Santo Antonio (Porto, Portugal)—Manuel Correia; Cliniques Universitaires Saint-Luc (Bruxelles, Belgium)—Andre Peeters; Landesklinikum Donauregion Tulln (Tulln, Austria)—Karl Matz; University Hospital Basel (Basel, Switzerland)—Phillippe Lyrer; Dalhousie University and Queen Elizabeth II Health Sciences Centre (Halifax, Canada)—Gord Gubitz, Stephen J Phillips; and Instituto Nacional de Neurologia (Mexico City, Mexico)—Antonio Arauz.

rs; Landesklinikum Donauregion Tulln (Tulln, Austria)—Karl Matz; University Hospital Basel (Basel, Switzerland)—Phillippe Lyrer; Dalhousie University and Queen Elizabeth II Health Sciences Centre (Halifax, Canada)—Gord Gubitz, Stephen J Phillips; and Instituto Nacional de Neurologia (Mexico City, Mexico)—Antonio Arauz. Conflicts of interest EB and AC have received honoraria and travel costs from Boehringer Ingelheim. GB has received honoraria and speaker fees from Boehringer Ingelheim, Sanofi Synthelabo Aventis, Hoffman La Roche, and Novo Nordisk. AK has received lecture fees and conference travel costs from Boehringer Ingelheim. RIL has been paid for his role as a member of a conference scientific committee and for lectures by Boehringer Ingelheim and has attended national stroke meetings organised and funded by Boehringer Ingelheim. PS has received lecture fees (paid to the Division of Clinical Neurosciences, University of Edinburgh) and travel expenses from Boehringer Ingelheim, was a member of the independent data and safety monitoring board of the RE-LY trial funded by Boehringer Ingelheim for which attendance fees and travel expenses were paid (to the Division of Clinical Neurosciences, University of Edinburgh). KBS has received an honorarium for a lecture from Boehringer Ingelheim and had costs for participating in scientific meetings reimbursed; is a member of the European Medicines Agency's Committee for Medicinal Products for Human Use and the Cardiovascular Working Party. The views expressed in this Article are the personal views of KBS and should not be understood or quoted as being made on behalf of or reflecting the position of the European Medicines Agency or one of its committees or working parties. VM has received an unrestricted educational grant from Boehringer Ingelheim for a meeting on thrombolysis in stroke at which IST-3 was discussed.

KBS and should not be understood or quoted as being made on behalf of or reflecting the position of the European Medicines Agency or one of its committees or working parties. VM has received an unrestricted educational grant from Boehringer Ingelheim for a meeting on thrombolysis in stroke at which IST-3 was discussed. JMW received funding to the Division of Clinical Neurosciences, University of Edinburgh for reading CT scans for ECASS III from Boehringer Ingelheim, is the contact reviewer for Cochrane systematic reviews of thrombolytic treatment for acute stroke, has attended meetings held by Boehringer Ingelheim as an unpaid independent adviser during the licensing of alteplase, but was refunded her travel expenses and the time away from work, has attended and spoken at meetings organised and funded by Boehringer Ingelheim for which she received honoraria and travel expenses, and is director of the Brain Research Imaging Centre for Scotland, which has received some funding supplemented by grants and donations from Novartis, Schering, General Electric, and Boehringer Ingelheim. All other members of the writing committee declare that they have no conflicts of interest. Figure 1 Trial profile EQ=EuroQoL. *Of the patients who were known to be alive at 18 months, 24 in the alteplase group versus 27 in the control group had a known date of death more than 18 months after enrolment, but their disability status at 18 months was unknown. Figure 2 Kaplan-Meier survival curves Table 1 Baseline characteristics of patients included in 18-month follow-up

EQ=EuroQoL. *Of the patients who were known to be alive at 18 months, 24 in the alteplase group versus 27 in the control group had a known date of death more than 18 months after enrolment, but their disability status at 18 months was unknown. Figure 2 Kaplan-Meier survival curves Table 1 Baseline characteristics of patients included in 18-month follow-up Alteplase group (n=1169) Control group (n=1179) Region Americas (Canada, Mexico) 5 (<1%) 6 (1%) Australia 89 (8%) 90 (8%) Eastern Europe (Poland) 158 (14%) 159 (13%) Northwest Europe (UK, Austria, Belgium) 550 (47%) 556 (47%) Scandinavia (Norway, Sweden) 251 (21%) 250 (21%) Southern Europe (Italy) 116 (10%) 118 (10%) Age 18–50 years 49 (4%) 57 (5%) 51–60 years 83 (7%) 81 (7%) 61–70 years 153 (13%) 158 (13%) 71–80 years 291 (25%) 304 (26%) 81–90 years 523 (45%) 512 (43%) >90 years 70 (6%) 67 (6%) Women 592 (51%) 596 (51%) National Institutes of Health stroke scale score 0–5 235 (20%) 236 (20%) 6–10 323 (28%) 330 (28%) 11–15 244 (21%) 235 (20%) 16–20 207 (18%) 219 (19%) >20 160 (14%) 159 (13%) Delay in enrolment ≤3·0 h 320 (27%) 307 (26%) >3·0–4·5 h 471 (40%) 481 (41%) >4·5–6·0 h 378 (32%) 389 (33%) >6·0 h 0 (0%) 2 (<1%) Atrial fibrillation 347 (30%) 331 (28%) Systolic blood pressure ≤143 mm Hg 380 (33%) 380 (32%) 144–164 mm Hg 379 (32%) 405 (34%) ≥165 mm Hg 410 (35%) 394 (33%) Diastolic blood pressure ≤74 mm Hg 342 (29%) 343 (29%) 75–89 mm Hg 409 (35%) 448 (38%) ≥90 mm Hg 406 (35%) 381 (32%) Blood glucose concentration* ≤5 mmol/L 202 (20%) 207 (20%) 6–7 mmol/L 501 (49%) 485 (47%) ≥8 mmol/L 324 (32%) 347 (33%) Treatment with antiplatelet drugs in previous 48 h 599 (51%) 610 (52%) Assessment of acute ischaemic change Scan normal 99 (8%) 102 (9%) Scan not normal but no sign of acute change 551 (47%) 579 (49%) Signs of acute change 511 (44%) 490 (42%) Predicted probability of poor outcome at 6 months† <40% 633 (54%) 640 (54%) ≥40–<50% 130 (11%) 113 (10%) ≥50–<75% 275 (24%) 304 (26%) ≥75% 131 (11%) 122 (10%) Stroke syndrome TACI 491 (42%) 509 (43%) PACI 460 (39%) 430 (36%) LACI 137 (12%) 133 (11%) POCI 79 (7%) 104 (9%) Other 2 (<1%) 3 (<1%) Data are n (%). TACI=total anterior circulation infarct. PACI=partial anterior circulation infarct. LACI=lacunar infarct. POCI=posterior circulation infarct.

%) ≥75% 131 (11%) 122 (10%) Stroke syndrome TACI 491 (42%) 509 (43%) PACI 460 (39%) 430 (36%) LACI 137 (12%) 133 (11%) POCI 79 (7%) 104 (9%) Other 2 (<1%) 3 (<1%) Data are n (%). TACI=total anterior circulation infarct. PACI=partial anterior circulation infarct. LACI=lacunar infarct. POCI=posterior circulation infarct. * Baseline glucose concentration was not recorded for the first 282 patients recruited; thus, glucose measurements were available for 2066 of 2348 participants (88%; 1027 allocated to alteplase and 1039 allocated to control). † Calculated from a model based on age and baseline National Institutes of Health stroke scale score.22 Table 2 Oxford handicap scale scores at 18 months

* Baseline glucose concentration was not recorded for the first 282 patients recruited; thus, glucose measurements were available for 2066 of 2348 participants (88%; 1027 allocated to alteplase and 1039 allocated to control). † Calculated from a model based on age and baseline National Institutes of Health stroke scale score.22 Table 2 Oxford handicap scale scores at 18 months Alteplase group Control group Adjusted anaylsis* Unadjusted analysis† Difference per 1000 patients†(95% CI) OR (95% CI) p value OR (95% CI) p value Planned 18-month follow-up 1169 1179 .. .. .. .. .. Missing OHS data at 18 months‡ 52 (4%) 57 (5%) .. .. .. .. .. Number analysed (both vital and OHS status known) 1117 (96%) 1122 (95%) .. .. .. .. .. OHS score at 18 months§ 0 119 (11%) 83 (7%) .. .. .. .. .. 1 135 (12%) 141 (13%) .. .. .. .. .. 2 137 (12%) 128 (11%) .. .. .. .. .. 3 132 (12%) 138 (12%) .. .. .. .. .. 4 81 (7%) 107 (10%) .. .. .. .. .. 5 105 (9%) 111 (10%) .. .. .. .. .. Died before 18 months§¶ 408 (37%) 414 (37%) 0·95 (0·78 to 1·16) 0·628 0·98 (0·83 to 1·17) 0·855 4 (−36 to 44) Alive and independent (OHS score 0–2)§ 391 (35%) 352 (31%) 1·28 (1·03 to 1·57) 0·024 1·18 (0·99 to 1·40) 0·068 −36 (−75 to 3) Alive and had favourable outcome (OHS score 0 or 1)§ 254 (23%) 224 (20%) 1·23 (0·98 to 1·55) 0·076 1·18 (0·96 to 1·44) 0·109 −28 (−62 to 6) Data are n (%) unless stated otherwise. OHS=Oxford handicap score. * Logistic regression of outcome on treatment group, adjusted for age, National Institutes of Health stroke scale score, and delay (all linear) and visible infarct on baseline scan.

Alteplase group Control group Adjusted anaylsis* Unadjusted analysis† Difference per 1000 patients†(95% CI) OR (95% CI) p value OR (95% CI) p value Planned 18-month follow-up 1169 1179 .. .. .. .. .. Missing OHS data at 18 months‡ 52 (4%) 57 (5%) .. .. .. .. .. Number analysed (both vital and OHS status known) 1117 (96%) 1122 (95%) .. .. .. .. .. OHS score at 18 months§ 0 119 (11%) 83 (7%) .. .. .. .. .. 1 135 (12%) 141 (13%) .. .. .. .. .. 2 137 (12%) 128 (11%) .. .. .. .. .. 3 132 (12%) 138 (12%) .. .. .. .. .. 4 81 (7%) 107 (10%) .. .. .. .. .. 5 105 (9%) 111 (10%) .. .. .. .. .. Died before 18 months§¶ 408 (37%) 414 (37%) 0·95 (0·78 to 1·16) 0·628 0·98 (0·83 to 1·17) 0·855 4 (−36 to 44) Alive and independent (OHS score 0–2)§ 391 (35%) 352 (31%) 1·28 (1·03 to 1·57) 0·024 1·18 (0·99 to 1·40) 0·068 −36 (−75 to 3) Alive and had favourable outcome (OHS score 0 or 1)§ 254 (23%) 224 (20%) 1·23 (0·98 to 1·55) 0·076 1·18 (0·96 to 1·44) 0·109 −28 (−62 to 6) Data are n (%) unless stated otherwise. OHS=Oxford handicap score. * Logistic regression of outcome on treatment group, adjusted for age, National Institutes of Health stroke scale score, and delay (all linear) and visible infarct on baseline scan. † Standard binomial test with normal approximation. ‡ Includes patients who did not return an 18-month form but died more than 18 months after enrolment (figure 1). § Percentages based on number analysed for OHS. For one participant, OHS was imputed on the basis of responses to EQ-5D.

* Logistic regression of outcome on treatment group, adjusted for age, National Institutes of Health stroke scale score, and delay (all linear) and visible infarct on baseline scan. † Standard binomial test with normal approximation. ‡ Includes patients who did not return an 18-month form but died more than 18 months after enrolment (figure 1). § Percentages based on number analysed for OHS. For one participant, OHS was imputed on the basis of responses to EQ-5D. ¶ If all patients known to be alive are included in the denominators, the percentage dead at 18 months are 35·8% in the alteplase group and 36·0% in the control group. Table 3 EQ-5D and other assessments of function at 18 months

§ Percentages based on number analysed for OHS. For one participant, OHS was imputed on the basis of responses to EQ-5D. ¶ If all patients known to be alive are included in the denominators, the percentage dead at 18 months are 35·8% in the alteplase group and 36·0% in the control group. Table 3 EQ-5D and other assessments of function at 18 months Alteplase group Control group Odds ratio (95% CI)* p value Difference per 1000 patients†(95% CI) EQ-5D Mobility 702 692 .. .. .. No problems walking 283 (40%) 259 (37%) 1·30 (1·05 to 1·61) 0·017 −29 (−80 to 22) Some problems walking 343 (49%) 346 (50%) .. .. 11 (−41 to 64) Confined to bed 76 (11%) 87 (13%) .. .. 17 (−16 to 51) Self-care 695 689 .. .. .. No problems with self-care 372 (54%) 328 (48%) 1·43 (1·16 to 1·78) 0·001 −59 (−112 to −7) Some problems washing or dressing 176 (25%) 191 (28%) .. .. 24 (−23 to 70) Unable to wash or dress 147 (21%) 170 (25%) .. .. 35 (−9 to 79) Usual activities 699 694 .. .. .. No problems with usual activities 235 (34%) 209 (30%) 1·32 (1·07 to 1·62) 0·008 −35 (−84 to 14) Some problems with usual activities 258 (37%) 256 (37%) .. .. 0 (−51 to 50) Unable to do usual activities 206 (29%) 229 (33%) .. .. 35 (−13 to 84) Pain or discomfort 698 694 .. .. .. No pain or discomfort 344 (49%) 304 (44%) 1·26 (1·02 to 1·56) 0·029 −55 (−107 to −2) Moderate pain or discomfort 316 (45%) 355 (51%) .. .. 59 (6 to 111) Extreme pain or discomfort 38 (5%) 35 (5%) .. .. −4 (−27 to 19) Anxiety or depression 693 690 .. .. .. Not anxious or depressed 353 (51%) 349 (51%) 1·05 (0·85 to 1·29) 0·668 −4 (−56 to 49) Moderately anxious or depressed 292 (42%) 290 (42%) .. .. −1 (−53 to 51) Extremely anxious or depressed 48 (7%) 51 (7%) .. .. 5 (−23 to 32) Additional questions about overall function Stroke left patient with problems 484/700 (69%) 542/699 (78%) 1·67 (1·30 to 2·17) <0·0001 84 (38 to 130) Needs help with everyday activities 298/696 (43%) 350/692 (51%) 1·59 (1·25 to 2·00) <0·0001 78 (25 to 130) Data are n (%) unless stated otherwise.

(7%) .. .. 5 (−23 to 32) Additional questions about overall function Stroke left patient with problems 484/700 (69%) 542/699 (78%) 1·67 (1·30 to 2·17) <0·0001 84 (38 to 130) Needs help with everyday activities 298/696 (43%) 350/692 (51%) 1·59 (1·25 to 2·00) <0·0001 78 (25 to 130) Data are n (%) unless stated otherwise. * Logistic regression of outcome on treatment group, adjusted for age, National Institutes of Health stroke scale score, and delay (all linear) and visible infarct on baseline scan. † Standard binomial test with normal approximation. Table 4 EQ utility index and visual analogue scale score assessment of overall health at 18 months Alteplase group Control group Adjusted analysis* Unadjusted analysis† n Mean (SE) n Mean (SE) Mean difference (SE) p value Mean difference (SE) p value Visual analogue scale score 653 62·07 (0·90) 648 60·57 (0·91) 2·18 (1·21) 0·072 1·49 (1·28) 0·244 EQ utility index 674 0·550 (0·015) 667 0·502 (0·016) 0·062 (0·020) 0·002 0·049 (0·022) 0·028 * Adjusted for age, National Institutes of Health Stroke Scale score, delay from onset to enrolment, and presence of visible ischaemia on the baseline scan. † Significance based on t test. Utility based on UK time trade-off valuations on a scale of −1 to +1. Panel Research in context Systematic review

Alteplase group Control group Adjusted analysis* Unadjusted analysis† n Mean (SE) n Mean (SE) Mean difference (SE) p value Mean difference (SE) p value Visual analogue scale score 653 62·07 (0·90) 648 60·57 (0·91) 2·18 (1·21) 0·072 1·49 (1·28) 0·244 EQ utility index 674 0·550 (0·015) 667 0·502 (0·016) 0·062 (0·020) 0·002 0·049 (0·022) 0·028 * Adjusted for age, National Institutes of Health Stroke Scale score, delay from onset to enrolment, and presence of visible ischaemia on the baseline scan. † Significance based on t test. Utility based on UK time trade-off valuations on a scale of −1 to +1. Panel Research in context Systematic review The primary results of IST-32 included a systematic review of randomised controlled trials of alteplase in acute stroke.1 To accompany this review we searched up to April 30, 2013, for additional randomised trials of intravenous alteplase versus control within 6 h of onset of acute stroke in the Cochrane Stroke Trials Registry, Internet Stroke Trials Centre, and reference lists in review articles and conference abstracts. For the Cochrane Stroke Trials Registry we searched for interventions with thrombolytic drugs in acute ischaemic stroke added since the last update of the Cochrane review. For the Internet Stroke Center, we searched for “acute ischemic stroke”, “acute ischaemic stroke”, “thrombolysis”, “thrombolytic therapy”, “alteplase”, and “recombinant tissue plasminogen activator”. For each trial, we checked the primary trial publication, and when available, the trial protocol, to determine if it was planned to collect long-term clinical outcome data (ie, more than 90 days after enrolment) or health-related quality-of-life data, as assessed by a valid instrument such as EQ-5D or Short Form 36.

”. For each trial, we checked the primary trial publication, and when available, the trial protocol, to determine if it was planned to collect long-term clinical outcome data (ie, more than 90 days after enrolment) or health-related quality-of-life data, as assessed by a valid instrument such as EQ-5D or Short Form 36. Of the 12 completed randomised controlled trials, ten reported outcome at 90 days or less,1 two reported clinical outcome at 6 months2,3 and one at 12 months,3 but none reported effects more than 12 months after stroke. The Second European Collaborative Acute Stroke Study collected data on health-related quality of life at 90 days with the SF-36, but has yet to report those data. In the NINDS Trial,3 mortality at 12 months did not differ significantly between alteplase and placebo groups (24% vs 28%; p=0·29). The primary outcome was favourable outcome, defined as minimal or no disability as measured by the Barthel index, the modified Rankin scale, and the Glasgow outcome scale, and the treatment effect was assessed with a global statistic. The global statistic favoured the alteplase group at 6 months (OR for a favourable outcome 1·7, 95% CI 1·3–2·3) and at 12 months (1·7, 1·2–2·3). Interpretation

Of the 12 completed randomised controlled trials, ten reported outcome at 90 days or less,1 two reported clinical outcome at 6 months2,3 and one at 12 months,3 but none reported effects more than 12 months after stroke. The Second European Collaborative Acute Stroke Study collected data on health-related quality of life at 90 days with the SF-36, but has yet to report those data. In the NINDS Trial,3 mortality at 12 months did not differ significantly between alteplase and placebo groups (24% vs 28%; p=0·29). The primary outcome was favourable outcome, defined as minimal or no disability as measured by the Barthel index, the modified Rankin scale, and the Glasgow outcome scale, and the treatment effect was assessed with a global statistic. The global statistic favoured the alteplase group at 6 months (OR for a favourable outcome 1·7, 95% CI 1·3–2·3) and at 12 months (1·7, 1·2–2·3). Interpretation IST-3 confirms the evidence from previous trials on the neutral effect of thrombolysis with alteplase on survival after stroke in a much larger sample, and adds to the evidence that improvements in function reported at earlier timepoints are evident at 18 months. IST-3 also provides the first validated estimates of the effect of thrombolysis with alteplase on health-related quality of life.