A multicenter trial of the efficacy of using tirilizad mesylate in cases of head injury

The European Brain Injury Consortium Survey of Head Injuries

“…It is slightly higher than the 15% reported from the recent International Tirilazed study [25] which considered only events in the ®rst four hours. In the current series, the reported incidence of hypoxia is very similar (27%) to the report from the Traumatic Coma Data Bank (26%) [5]; in the International Tirilazad trial the incidence in the ®rst four hours was again lower (15%) [25].…”

“…Time points include: on arrival at hospital [10], or at the neurosurgical department [3], after completion of`nonsurgical' resuscitation [23,27,46], within four hours of injury [25] [30]. Approaches used to allow for missing data have included allocating a`pseudo score' of 1 for the verbal portion of the GCS in an intubated patient [23], but this results in a loss of information in severe cases [17,33] and may be especially misleading in moderate injuries.…”

The sequential analysis of repeated binary responses: a score test for the case of three time points

“…where certain sums are over 1, 2 and * (missing). The maximum likelihood estimates of the q's areq (1) i; g = n i••; g =n •••; g ,q (2) ij; g = n ij•; g =n i••; g andq (3) ijk; g = n ijk; g =n ij•; g . The predicted counts, e ijk; g , can be expressed in terms of the q's as e ijk; g = (n ijk; g + n ij * ; g q (3) ijk; g + n i * * ; g q (2) ij; g q (3) ijk; g )…”

“…so that A h e g = (e h11; g e h12; g e h21; g e h22; g e •h1; g e •h2; g e ••h; g ) , and similarly for A h N g ; h, g = 1; 2. Now, let U g denote the e cient score vector for treatment group T g U g = (@'=@r (3) 111; g @'=@r (3) 112; g @'=@r (3) 121; g @'=@r (3) 122; g @'=@r (2) 11; g @'=@r (2) 12; g @'=@r (1) 1; g ) g = 1; 2. The derivatives of ' with respect to the r's are best found as sums of products of derivatives of ' with respect to the p's and derivatives of the p's with respect to the r's.…”

“…The primary endpoint of such studies is usually a measure of functional status three months after treatment [1]. In trials in severe head injury the minimum follow-up time is usually taken to be six months, and the Glasgow Outcome Scale six months after treatment is used to assess the patient's condition [2]. Despite the length of follow-up felt necessary for individual patients, there is a need to monitor the accumulating results of studies in conditions as serious as stroke or head injury with a view to stopping recruitment as soon as it is clear that the experimental treatment either does or does not work.…”