Low Body Temperature Does Not Compromise the Treatment Effect of Alteplase
Background and Purpose—Hypothermia is neuroprotective in ischemic stroke models. The influence of baseline body temperature on outcomes after thrombolytic therapy is unclear. We examined outcomes after alteplase treatment across baseline body temperature for patients with ischemic stroke in data held within the Virtual International Stroke Trials Archive (VISTA; 1998 to 2007).
Methods—We collated data on age, baseline severity (National Institutes of Health Stroke Scale), and 90-day modified Rankin Scale score on patients presenting with acute ischemic stroke. We compared 90-day modified Rankin Scale score between thrombolyzed and nonthrombolyzed comparators across baseline body temperature. We report age and baseline National Institutes of Health Stroke Scale-adjusted Cochran-Mantel-Haenszel probability value and proportional OR with 95% CI for improved modified Rankin Scale distribution. We report temperature profiles over 72 hours after stroke by treatment group.
Results—Rankin data were available for 5586 patients with acute ischemic stroke in VISTA (1980 received alteplase). Age and baseline severity were similar (age 68.0±13.0 years versus 69.9±12.3 years, National Institutes of Health Stroke Scale 14.2±5.2 versus 13.0±5.6). Alteplase was associated with improved outcome (OR, 1.49; 95% CI, 1.35 to 1.65, P<0.0001). Alteplase treatment effect was not associated with baseline temperature (P=0.14). Point estimates showed benefit of alteplase treatment across 35.5°C to 37.5°C but showed a negative trend >37.5°C. Alteplase did not influence temperature profiles over 72 hours after stroke.
Conclusions—There is no evidence of influence of body temperature on alteplase treatment response. These results are reassuring that low temperatures across a physiological range do not compromise therapeutic effect of alteplase.
Alteplase is the only proven therapy available for patients with ischemic stroke presenting within 4.5 hours of symptom onset. Body temperature is an important factor that influences clinical outcome.1 Current recommendations suggest use of antipyretics for febrile patients with stroke.2 Ongoing clinical trials are examining the influence of induced hypothermia on neuroprotective mechanisms and prolongation of the therapeutic window.3 The influence of hypothermia on outcomes after acute stroke is of immense interest, but interpretation is confounded by contradictory data from experimental models and human studies.4,5 To examine the influence of temperature on outcomes after thrombolytic therapy, we analyzed clinical data held within the Virtual International Stroke Trials Archive (VISTA).6 We report outcomes for various levels of baseline body temperature by comparing patients who received alteplase with those who did not.
Data Source and Patients
We collated data from neuroprotection trials conducted from 1998 to 2007 lodged within VISTA (www.vistacollaboration.org).6 Eligibility criteria required demography, body temperature, and complete outcome data on patients who had an ischemic stroke, who were offered neuroprotective agents, and on whom there was no influence of neuroprotective agents on the outcomes as reported in their respective clinical trials report.7
We performed nonrandomized adjusted comparison of outcomes between patients who received alteplase and patients who did not (control subjects) for various levels of baseline body temperature (32.8 to 35.5°C, 35.51 to 36.0°C, 36.01 to 36.5°C, 36.51 to 37.0°C, 37.01 to 37.5°C, 37.51 to 38.0°C, 38.0 to 40.5°C).
For each contrast, we compared the distribution of modified Rankin Scale at Day 90 between the 2 groups using the full range of scores, adjusted for age and baseline National Institutes of Health Stroke Scale. We used Cochran-Mantel-Haenszel statistic and then proportional odds logistic regression analysis to test for an association of outcome distribution with the use of alteplase.
We compared body temperature over the first 72 hours between thrombolyzed and control patients. Analyses were undertaken using SAS 9.2.
We obtained data on 9665 patients. Of the excluded patients, 571 had hemorrhagic stroke, 36 had stroke of unknown etiology, 3767 had no recorded temperature data, 65 had ambiguous unit of temperature, and 386 had missing outcome data. Included patients totalled 5586. Baseline characteristics are shown in the Table.
Outcomes were better among all patients who received alteplase (P<0.001; OR, 1.49; 95% CI, 1.35 to 1.64; N=2097). Using proportional odds logistic regression, we found no overall interaction of baseline temperature with outcomes after thrombolytic therapy (P=0.14). No interaction was found when the test was conducted within each temperature bracket (P>0.05 for all groups). Figure 1 illustrates the odds for improved outcomes after alteplase treatment within various categories of baseline body temperature.
We found no difference in means of body temperature between treatment and control groups from 2 to 72 hours after symptom onset (P<0.0001 at baseline, 1, 2 hours; 8 hours P=0.7; 24 hours P=0.2; 72 hours P=0.1; Figure 2).
We found no interaction between baseline temperature and outcomes after alteplase treatment in patients with acute ischemic stroke. We observed significantly improved outcomes across temperatures 35.51°C to 37.5°C. Significance was lost at temperatures <35.5°C and >37.5°C, although point estimates suggest that hyperthermia attenuates thrombolysis-associated benefit. Our data support an association between hyperthermia and poorer outcome1,4 and support existing guidelines that recommend treatment of pyrexia.2
In vitro evidence suggests that lower temperatures reduce the rate of fibrinolysis by alteplase,7 whereas hyperthermic stress may reduce endogenous fibrinolytic capacity.8 The overall efficacy of thrombolysis will depend on temperature effects on these individual components. Our included patients had temperatures mainly ranging from 35.5°C to 38°C. Induced hypothermia is likely to lower temperature to approximately 33°C.9 We cannot draw firm conclusions regarding effect of temperature <35.5°C, but our data are reassuring that low temperatures across a physiological range do not compromise alteplase activity.
This is an analysis of a large clinical trial data set. Included trials were closely monitored with verification of source data and end points. We used nonrandomized data, which we accept as a limitation of our analysis. Owing to the retrospective nature of our data, we cannot control for any selection bias. This underlines the desirability of confirming our findings in a prospectively designed study.
Source of Funding
Supported by a Chest Heart and Stroke Scotland research grant.
P.D.L. reports grant income from the National Institutes of Health, Veteran's Affairs Department, American Heart Association, and is a consultant to Photothera Inc, CoAxia Inc, Benechil Inc, Mitsubishi Inc, and ZZ Biotech LLC. A.S. has received honorarium and is on Advisory boards of Sanofi-BMS, BI, Pfizer, Bayer, and Roche.
Costantino Iadecola, MD, was the Consulting Editor for this paper.
The abstract was presented as a poster at the International Stroke Conference, Los Angeles, 2011.
- Received December 29, 2010.
- Revision received March 7, 2011.
- Accepted March 23, 2011.
- © 2011 American Heart Association, Inc.
- Saini M,
- Saqqur M,
- Kamruzzaman A,
- Lees KR,
- Shuaib A
Guidelines for Management of Ischaemic Stroke and Transient Ischaemic Attack 2008. European Stroke Organisation web site. Available at: www.eso-stroke.org/pdf/ESO08_Guidelines_Original_english.pdf. Accessed December 1, 2010.
- Hemmen TM,
- Raman R,
- Guluma KZ,
- Meyer BC,
- Gomes JA,
- Cruz-Flores A,
- et al
- Mishra NK,
- Diener HC,
- Lyden PD,
- Bluhmki E,
- Lees KR