[ccm-l] SAFE TBI in today's NEJM

[Ian Seppelt]

Thanks, Stephen. See below
Ian

>>> "Stephen Streat" <StephenS at adhb.govt.nz> 30/08/2007 2:15pm >>>
Dear Ian

Interesting stuff for sure. 

As one who created (with Colin McArthur) the draft criteria for the
pre-determined subsets in SAFE I would point out a few things --
correct
me if I'm wrong -- I've been known to be wrong before ...

1) Trauma was a certainly a pre-determined subgroup (along with sepsis
and ARDS) -- see attached SAFE study original report.

--- correct

2) However, was "trauma including TBI" a pre-determined subgroup or
did
this arise later -- when the SAFE-TBI study group started looking at
the
reason for the interesting difference in the trauma patient group? 

--- TBI was predetermined but not stratified. The study was stratified
for sepsis, ARDS and trauma. The original case report form had a tick
box for TBI [defined as presenting GCS 13 or less with abnormal CT
scan]. For the SAFE TBI 24 month followup the accuracy of these
classifications were checked, misclassifications corrected and a few
other patients found

3) Todays NEJM paper reads as if TBI was a pre-determined subgroup in
SAFE.

--- see above

Now a couple of comments --

1) The albumin group in the TBI paper are a bit older, a bit sicker
(APACHE-II) and had slightly higher injury severity at baseline --
none
of these individual issues being significant in their own right
(Injury
severity is mistakenly reported as "AIS" in the paper and not picked
up
by the reviewers -- what are presented in Table 1 are surely ISS, not
AIS -- incidentally while I'm going there -- ISS is a non-linear
ordinal
variable with values between 1 and 75 comprised of the sum of three
squares of integers between 1 and 5. It should not be treated
parametrically, again something that the NEJM reviewers overlooked --
as
do many people .. Sigh ...)

--- the SAFE TBI II post hoc that is currently underway will look at
this inb more detail

2) Unfortunately "the number of patients who became brain dead was not
recorded". I'm surprised that this information was not easily
obtainable
post-hoc by case review. It would help -- especially with the next
point. Similarly, I would bet real money that every one of the
Australian and NZ units could tell you about which patients died after
treatment was withdrawn because of TBI ... Short of brain death. Was
that information about "mode of dying" sought?

--- not available on the original CRF but will certainly be there on
the SAFE TBI II analysis

3) The proportion of deaths (by 28 days) in the albumin group in which
the "primary cause of death" was "traumatic brain injury" was similar
--
(slightly less in fact) -- to the proportion in the saline group. Does
this mean that albumin caused more MOF deaths in these patients with
TBI
(unlike its performance in sepsis -- where if anything it was better
than saline?).

--- don't know

More questions than answers --

My interpretation --

Useful and interesting data. Hard to justify using albumin routinely
over saline in any patient, although both were "SAFE" overall in
all-comers. In that context, where one (like me) had already made a
shift to saline, these data don't change my practice. They do not
examine the use of albumin (along with diuretics) as "oncotherapy" for
intracranial hypertension -- see word document attached. That is, this
NEJM study arose out of a study of fluids used for circulatory
resuscitation. "Absolutely contraindicated" Ian -- no, I disagree.
Should not be used as "resuscitation fluid of choice" maybe. As the
authors of todays paper wrote (SAFE executive members included perhaps
Ian ?) wrote -- "It remains possible that our results represent a
chance
subgroup finding."

--- fair comment

;-)

S

Stephen Streat FRACP
Intensivist                                                  
Clinical Director, Organ Donation New Zealand
P : +64 9 630 9812
F : +64 9 630 6490
stephens at adhb.govt.nz 





  

 

 

-----Original Message-----
From: ccm-l-bounces at ccm-l.org [mailto:ccm-l-bounces at ccm-l.org] On
Behalf
Of Ian Seppelt
Sent: Thursday, 30 August 2007 14:34
To: ccm-l at ccm-l.org; trauma-list at trauma.org 
Subject: [ccm-l] SAFE TBI in today's NEJM

Today's NEJM has the 2 year followup of the traumatic brain injury
cohort from SAFE which clearly shows worse outcomes in the albumin
group. Based on this it is reasonable to say that albumin is
absolutely
contraindicated in TBI, and by extrapolation possibly in ANY trauma
[as
there is no suggestion of any benefit from albumin in trauma, mild -
moderate TBI can be missed acutely in the context of other injuries,
plus the cost of albumin].

The million dollar question is WHY albumin is deleterious. A post hoc
analysis of these patients is underway looking at whether there is any
subtle difference in baseline (eg differences in ICP or management of
intracranial hypertension, etc etc) that might account for it.

[CAVEAT: I was an investigator in SAFE and am an executive member of
the ANZICS CLinical Trials Group. Interpret any implied bias how you
will!!!!]

Cheers, Ian

Ian Seppelt FANZCA FJFICM
Senior Staff Specialist
Dept of Intensive Care Medicine
The Nepean Hospital, PO Box 63 Penrith NSW 2751
Director of Clinical Research, Sydney West AHS
Clinical Lecturer, University of Sydney