This Article Full Text (PDF) All Versions of this Article: 36/2/230    most recent 01.STR.0000152953.71415.01v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mann, J. Articles by Christianson, T. J. H. Search for Related Content PubMed PubMed Citation Articles by Mann, J. Articles by Christianson, T. J. H.

(Stroke. 2005;36:230.)
© 2005 American Heart Association, Inc.

Letters to the Editor

NINDS Reanalysis Committee’s Reanalysis of the NINDS Trial

Jeffrey Mann, MD

Salt Lake City, Utah

To the Editor:

In their Special Report,¹ the authors stated, "there was no evidence that the imbalance in the distribution of baseline National Institutes of Health Stroke Scale (NIHSS) (stroke severity) between the treatment groups had either a statistically or clinically significant effect on the trial’s results."

First, note that the imbalance only existed in the 91- to 180-minute arm of the trial, and the National Institute of Neurological Disorders and Stroke (NINDS) Reanalysis Committee did not specifically address the issue as to whether the imbalance affected the correct interpretation of the results of the 91- to 180-minute arm. In their original report,² the NINDS Study Group Investigators reported the favorable stroke outcome results for the 91- to 180-minute arm of the trial as follows: absolute risk difference, 20%; OR, 2.4; RR, 1.8 (for a modified Rankin Scale 0,1 stroke outcome result). The 20% absolute risk difference (RD) is the most relevant measure because it allows one to demonstrate that the number needed to treat value is 5, and that the harm/benefit ratio is <1.0 (average risk of a symptomatic intracranial hemorrhage of 6%). After I received the NINDS raw data in October 2003, I wrote a critical article³ suggesting that the absolute RD value of 20% represents the "apparent" efficacy of tPA, and that the "true" efficacy RD value was more likely to be 8% to 11% (50% of the "apparent" efficacy of tPA), and that the "difference" was mainly caused by the imbalance in baseline stroke severity.

I suggest that the NINDS Reanalysis Committee did not uncover the extent to which the imbalance in baseline stroke severity affected the correct interpretation of the NINDS trial’s results, because they only approached the problem from one perspective. In a critique⁴ of the NINDS Reanalysis Committee’s Special Report, I have demonstrated how varying the absolute number of very mild stroke (baseline NIHSS, 0 to 5) and very severe stroke (baseline NIHSS >20) patients between the tPA and placebo groups significantly affects the final estimated RD values.

I believe that the stroke research community has not fully appreciated how stroke severity heterogeneity imbalances can affect the accurate interpretation of a tPA-for-stroke randomized control trial’s results, and I think that this problematic issue should be more thoroughly investigated and debated.

References

1. Ingall TJ, OFallon WM, Asplund K, Goldfrank LR, Hertzberg VS, Loius TA, Christianson TJ. Findings from the Reanalysis of the NINDS Tissue Plasminogen Activator for Acute Ischemic Stroke Treatment Trial. Stroke. 2004; 35: 2418–2424.[Abstract/Free Full Text]
2. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med. 1995; 333: 1581–1587.[Abstract/Free Full Text]
3. Mann J. A personal analysis of the NINDS trial using patient-level data. Available at http://www.homestead.com/emguidemaps/files/NINDSpersonalanalysis.html. Accessed December 16, 2004.
4. Mann J. A Critique of the Reanalysis of the NINDS trial. Available at http://www.homestead.com/emguidemaps/files/IngallCrtitique.htm. Accessed December 16, 2004.

Response

Timothy John Ingall, MB, BS, PhD

Department of Neurology, Mayo Clinic Scottsdale, Scottsdale, Ariz

William Michael O’Fallon, PhD

Division of Biostatistics, Mayo Clinic Rochester, Rochester, Minn

Kjell Asplund, MD, PhD

National Board of Health and Welfare, Stockholm, Sweden

Lewis Robert Goldfrank, MD

Department of Emergency Medicine, New York University School of Medicine, New York, NY

Vicki S. Hertzberg, PhD

Department of Biostatistics, Emory University, Atlanta, Ga

Thomas Arthur Louis, PhD

Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD

Teresa J. Hengy Christianson, BS

Division of Biostatistics, Mayo Clinic Rochester, Rochester, Minn

Dr Mann correctly points out, and in our article we acknowledge, that there was an imbalance in baseline stroke severity favoring the tissue plasminogen activator (tPA) group for patients randomized between 91 and 180 minutes after stroke in the National Institute of Neurological Disorders and Stroke (NINDS) tPA for acute ischemic stroke treatment trial.¹ As indicated on page 2421 in our article,² and detailed on page 53 in our report,³ using a wide variety of standard accepted statistical adjustment approaches, we found neither a clinically nor a statistically significant effect of this imbalance on the tPA–placebo comparison. Furthermore, subgroup analyses need to be interpreted cautiously, with the burden of proof resting on making the case that the treatment effect is different between subgroups. For example, it is good practice to estimate or test for a treatment by subgroup interaction. If this is not compelling, as was the case in our analysis, then concluding there are subgroup differences must rest on strong supporting evidence from other studies.

Although no statistical analysis can definitively rule out subgroup differences, passionate belief in a personal perspective will not change the biology. As stated by Brookes et al,⁴ subgroup analyses must be interpreted very cautiously because the erroneous identification of differential subgroup effects may lead to inappropriate provision, or withholding, of treatment. Additional data from an adequately powered new study are needed to assess possible subgroup effects. Until an adequately powered assessment of the effect of tPA in the 91- to 180-minute group stratified by baseline stroke severity is performed, we have no basis to believe that tPA is not equally effective among acute ischemic stroke patients with different baseline stroke severity treated at any time with intravenous tPA within 3 hours of stroke onset.

References

1. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med. 1995; 333: 1581–1587.[Abstract/Free Full Text]
2. Ingall TJ, O’Fallon WM, Asplund K, Goldfrank LR, Hertzberg VS, Louis TA, Christianson TJH. Findings from the reanalysis of the NINDS tissue plasminogen activator for acute ischemic stroke treatment trial. Stroke. 2004; 35: 2418–2424.[Abstract/Free Full Text]
3. O’Fallon WM, Asplund K, Goldfrank LR, Hertzberg VS, Ingall TJ, Louis TA. Report of the t-PA review committee. National Institute of Neurologic Diseases and Stroke; 2004. Available at: http://www.ninds.nih.gov/t-PA_review_committee. Accessed October 25, 2004.
4. Brookes ST, Whitely E, Egger M, Smith GD, Mulheran PA, Peters TJ. Subgroup analyses in randomized trials: risks of subgroup-specific analyses; power and sample size for the interaction test. J Clin Epidemiol. 2004; 57: 229–236.[CrossRef][Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				V. Hertzberg, T. Ingall, W. O'Fallon, K. Asplund, L. Goldfrank, T. Louis, and T. Christianson Methods and processes for the reanalysis of the NINDS tissue plasminogen activator for acute ischemic stroke treatment trial Clinical Trials, August 1, 2008; 5(4): 308 - 315. [Abstract] [PDF]

This Article Full Text (PDF) All Versions of this Article: 36/2/230    most recent 01.STR.0000152953.71415.01v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mann, J. Articles by Christianson, T. J. H. Search for Related Content PubMed PubMed Citation Articles by Mann, J. Articles by Christianson, T. J. H.