Full text of DOI:10.1046/j.1365-2044.2003.03554.x

Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
Correspondence
Consent for anaesthesia
thesia with postoperative continuous
spinal or epidural analgesia.
Although the Association has pro-
Reference
1 Macdonald R. Problems with regional
anaesthesia: hazards or negligence?
Anaesthesia 1994; 73: 64–8.
The special article on consent for anaes-
thesia (White & Baldwin. Anaesthesia duced thoughtful guidelines, I would
2000; 58: 760–74) is timely, relevant and agree with White and Baldwin’s three
merits detailed attention.
criticisms and am persuaded that sep-
In 1994 when I wrote a paper on arate consent for anaesthesia should
‘Problems with regional anaesthesia: be taken. Obviously, for more simple
hazards or negligence?’ [1], I advocated procedures, e.g. anaesthesia for most
a moderate view regarding informed day surgery and investigative proce-
consent and obstetric regional anaesthe- dures, an information sheet carefully
sia. Indeed, on reflection, I ‘sat on the worded may suffice.
Academic anaesthesia
Academic anaesthesia is clearly at a cross
roads and alternative structures for the
provision of academic support for our
speciality need to be found. As Jackson
suggests (Jackson et al. Anaesthesia 2003;
58: 911–2), one possible solution is to
take academic anaesthesia away from
the medical school environment. Our
experience in the South-west of Eng-
land over the past few years is an
example of this.
In 1997, we had established an aca-
demic department of anaesthesia as part
of the Plymouth Postgraduate Medical
School. However, with the arrival of
the new undergraduate Peninsula Med-
ical School in 2001, things were soon to
change. The Peninsula Medical School,
for all of the reasons that have been
elucidated by Jackson, did not feel able
to continue to support an academic
department of anaesthesia. Despite
numerous attempts by anaesthetists in
the region and recurrent funding being
secured from the NHS, an academic
department of anaesthesia within the
Peninsula Medical School was not
established and our young academic
department was lost. We therefore took
fence’. The way in which medicine is
How should the actual process of
practised continually evolves and has to taking consent be implemented? First,
respond to the ever-changing values every department should discuss this
and philosophy of the society ⁄ commu- paper at a departmental meeting atten-
nity that it serves: our patients are better ded by consultants and trainees i.e. if
informed; their knowledge of both their such occasions still occur given the
illnesses and rights is continually being abundance of shift working. Second, a
reinforced by increasing use of the consensus should be reached, and thirdly
Internet, the media and various pressure pilot consent taking will be necessary
groups.
followed by audit and the cycle closed.
Early during my tenure as Post- Finally, a plan should be presented to the
graduate Dean, it became apparent that managers to implement since presuma-
it was inappropriate for preregistration bly that is one of their functions.
house officers to take consent for Such a development must improve
complex surgery about which they patient ⁄ doctor communication, a basic
had very little understanding. The tenet of a modernised Health Service.
educational contract was appropriately However, this will take time, which
modified. They also had little know- means sessional commitment, otherwise
ledge of anaesthetic techniques and patient throughput will be reduced. The
postoperative pain relief and so appro- College or the Association may wish to
priate tutorials were included in their coordinate this process with a view to
teaching sessions but we were left with achieving national consensus.
the problem of consent for anaesthesia,
R. Macdonald
particularly where increasingly com-
Leeds LS16 9JF, UK
plex techniques were being used e.g.
E-mail: rosemacuk@yahoo.co.uk
combined general and regional anaes-
All correspondence should be addressed to Dr David Bogod, Editor of Anaesthesia, Anaesthesia, Ist Floor, Maternity Unit, Nottingham
City Hospital, Hucknall Road, Nottingham NG5 1PB, UK.
Letters (two copies) must be typewritten on one side of the paper only and double spaced with wide margins. In addition, please
include your letter as a Word for Windows or .rtf document on disk or alternatively submit as an e-mail attachment addressed to
anaesthesia@nottingham.ac.uk. Alternatively, a response to a previously published article or letter can be submitted online at
www.anaesthesia correspondence.com. Copy should be prepared in the usual style and format of the Correspondence section. Authors must
follow the advice about references and other matters contained in the Notice to Contributors to Anaesthesia printed at the back of the
January and July issues. Author guidelines can also be found on www.blackwellpublishing.com/journals/ana/submiss.htm. The degree and
diploma of each author must be given in a covering letter personally signed by all the authors.
Correspondence presented in any other style or format may be the subject of considerable delay and may be returned to the author for
revision.
Ó 2003 Blackwell Publishing Ltd
1235

Correspondence Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
the only route open to us and estab- available for the safe completion of the study was greater than in their pilot,
lished an NHS based academic depart- flight and include non-command mem- such that they had a clinically significant
ment of Anaesthesia, Critical Care & bers being assertive when necessary. difference that was not statistically sig-
Pain Management under the auspices of Stress management and ensuring that nificant. Although the intubation times
the South-west School of Anaesthesia. an understanding of personal and work in the study by Marfin et al. were not
This provides a focus for academic related stresses would make colleagues normally distributed, the total ranges are
activity within the NHS and the School alert for any decline in performance.
of Anaesthesia. The academic lead is by Critique with the ability to analyse a tiple-use). This suggests a much smaller
NHS clinicians with a background in future, current or past plan of action. variability than the s.d. of 20 s used in
research and rotates on a three yearly
In anaesthesia we already use many the initial power ⁄ sample size calcula-
only 20 s (single-use) and 27 s (mul-
basis. This provides academic time for of the above components in our prac- tion, which in this case only serves to
NHS clinicians to allow us to continue tise but in the airline industry there is increase the power of the study. How-
to provide an academic environment formalised training in CRM and a ever, with median intubating times of
for trainees in the South-west School of requirement for pilots to adhere to it. I 22 and 26 s for the two bougies, how
Anaesthesia.
think we have much to learn from the on earth could the authors expect to
I believe that this is a constructive airline use of CRM to improve patient find a ‘clinically important difference
and pragmatic way forward in the safety. Maybe we should consider that between intubating times’ of 20 s?
presently hostile climate that anaesthesia such CRM type training relevant to Furthermore, let us consider the
finds itself in, both in established and, anaesthesia should be included in the one significant result in the study: the
from our experience, also within the FRCA syllabus?
new medical schools.
proportion of successful intubations.
Remember that the sample size was
calculated to detect a difference in
intubating times and no mention was
made of the number of successes. To
detect a reduction in the number of
successful intubations from 95% to 55%
(percentages changed slightly for ease of
calculation) with 80% certainty would
require about 30 patients in each group.
Or, to put it another way, the current
study had a 50% chance of detecting this
difference, if it really exists [2]. Were
the authors just lucky or is this a type I
or a error? (In fact, a reduction from
95% to 55% is an incredibly ambitious
target to anticipate from the outset.
A reduction from 95% to 75%, per-
haps more reasonable to expect, would
require about 60 patients in each group
for the same power).
The authors mention the possibility
of bias in their method of laryngoscopy,
but fail to mention the bias that might
be introduced by their past experiences
and preconceived ideas, clearly stated in
the introduction, that the single-use
bougie has ‘…very low rates of suc-
cess…’ and ‘…is intrinsically difficult to
use’. So, rather than showing a ‘striking
difference between multiple-use and
single-use bougies’ we have a barely
significant result (p ¼ 0.03) in an
underpowered and potentially biased
study. Don’t get me wrong, I’m not
an apologist for the single-use bougie,
but if we are to condemn it to a landfill
site let’s do it with good science.
A. Levison
J. A. Langton
Wexham Park Hospital,
Plymouth Hospitals NHS Trust,
Plymouth PL6 8DH, UK
E-mail: Jeremy.langton@phnt.swest.
nhs.uk
Slough SL2 4HL, UK
E-mail: andrewflyboy1@aol.com
Underpowered and overbiased?
Potentially unfair to the
single-use bougie
Patient safety
I read with interest the Editorial about I hope the authors (Marfin et al. Anaes-
patient safety (Jorm. Anaesthesia 2003; thesia 2003; 59: 852–5) will not feel
58: 833–4). Dr Jorm mentioned the offended if I use their article to illustrate
concept of ‘six sigma’, which has been some recent problems I have encoun-
achieved in the airline industry and has tered surrounding the question of
made commercial flying without doubt power estimation and sample size cal-
the safest form of transportation. Part of culation in clinical studies.
this greatly increased safety has been due
Whereas it is a welcome change to
to the introduction over the past now see referenced evidence of efforts
25 years of Crew Resource Manage- to calculate a sample size which pro-
ment (CRM) taking into account that vides a study with the recommended
over 80% of aviation accidents were power, we non-statisticians should not
caused by human error. This is defined be lulled into a false sense of security by
as ‘better recognition and utilisation of the apparent science. The authors used a
all available resources, information, standard deviation (s.d.) of 20 s from
equipment and people to achieve safe their preliminary trials to calculate the
and efficient flight operations! The sample size. If this is the best available
components of CRM include:
evidence then fair enough, but what
Communication ⁄ interpersonal skills should be done when the s.d. or vari-
including polite assertiveness, active ability of the studied parameter turns
listening and feedback.
out to be different in the actual study?
Situational awareness and the avoid- Should the authors perform a post hoc
ance of ‘tunnel vision’ in a crisis; problem power analysis to prove that the study
solving, decision making and judgement. still had adequate power to support their
Leadership with the pilot in com- conclusion? Evans et al. [1] fell foul of
mand being responsible for accessing this problem when the standard devi-
and managing all resources that are ation of morphine consumption in their
1236
Ó 2003 Blackwell Publishing Ltd

Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
Correspondence
One other thing whilst I’m having a two groups would be ꢀ40 s, yielding a clinically important difference between
bit of a power rant. Why is it that we post hoc power of > 95% and a statistically groups; (b) the expected standard devi-
accept a 5% (1 in 20) chance of finding a significant difference of p < 0.01 [2]. ation of the measure of interest.
true difference that doesn’t exist, but are However, we were careful to avoid such
The choice of (a) can be open to
happy to risk a 20% (1 in 5) chance of dramatic treatment of the statistics and much debate: why did we choose 20 s
missing an intervention or treatment favoured a more careful approach, which as the clinically important difference in
that may be beneficial to our patients?
minimised the risk of bias. Therefore, our study; why not 15 s; or why not
Dr Morris’ view of power is simplistic: 45 s? This (artificial) precision has a
careful handling of the data is more large impact upon the power, yet the
important than any aspect of power choice is often (and sometimes neces-
S. Morris
University Hospital of Wales,
Cardiff CF14 4XW, UK
E-mail: Stephen.morris@
cardiffandvale.wales.nhs.uk
analysis.
sarily) arbitrary. Where no prior data
Dr Morris suggests that sample size exists, both (a) and (b) can be ‘best
also influenced our statistically signifi- guesses’: this is self-evidently a poor (but
cant result, and in this assertion makes a clearly acceptable) means of calculating
common error. The purpose of a prior power [4]. Alternatively (a) and (b) may
power analysis is to be reasonably be estimated from a set of ‘pilot’
confident that a negative (i.e. non-signi- observations. In this case, great caution
ficant) result is meaningful. If, on the must be exercised: the pilot observa-
other hand, a positive (i.e. statistically tions are, by definition, less robust than
significant) result is obtained, the prior the results of the planned study. Pilot
power analysis is irrelevant [1,2]. One studies are not themselves subjected to a
simple illustration might help. Let us power analysis (and are rarely, if ever,
suppose that we wish to design a study subject to rigorous randomisation and
to test the hypothesis that ‘all swans are control) [5]. Thus, philosophically, it
white’, and prior power analysis sug- seems curious to lend equal or greater
gests a very large sample size of (e.g. weight to a small pilot study than to the
10 000 swans). However, I choose to larger substantive study. Perhaps the
look at only one swan (which is black). power analysis itself should be expressed
Thus, the hypothesis is disproved in its in terms of a confidence interval (e.g.
entirety on the basis of a sample size of ‘our power analysis indicated that we
one. Regarding our study [1], one might use between 20 and 60 subjects
comparison was ‘significant’ at the to be between 70 and 90% confident of
p < 0.03 level. This simply means that excluding a true difference between the
this result might have occurred by means of 10–50 s…’) [6].
References
1 Evans MJ, Harrison CA, Mapleson
WW. A pitfall in the use of the Altman
nomogram for power estimation.
British Journal of Anaesthesia 1996; 76:
334.
2 Miller DK, Homan SM. Graphical aid
for determining power of clinical trials
involving two groups. British Medical
Journal 1988; 297: 672–6.
A reply
Dr Morris criticises our study as being
underpowered. The points made are
mathematically interesting and require
us also to consider more generally the
limitations of power analysis.
Our study: power and significance
One smallerror in thetext might have led
to some misunderstanding. Our power
analysis was based on a reduction in total
time (not intubating time as written) of
20 s as being clinically important. The
post hoc power analysis thus shows a mean
difference in total time of ꢀ12 s with a
standard deviation of ꢀ15 s. This gives a
required sample size of ꢀ20 [1]. We used
16 subjects, so (post hoc) our study had
ꢀ65% power [2] to exclude a difference
in total times for intubation between the
two groups at the p < 0.05 level. This is
indeed slightly lower than conventional
levels. However, it is clear from our
chance no more than 3% of the time.
The crudeness of power analyses is
Dr Morris’ description of this result as reflected in the different sample sizes
‘barely significant’ is therefore impre- yielded by different methods. If we
cise: it was significant at the 3% level: assume that in a study the clinically
whether this is described as ‘barely’ or important difference is 1 (arbitrary units)
otherwise, is purely subjective. Note and the standard deviation is 0.8, then
that sample size is not relevant to this variouspowercalculationsgivethesample
debate. Generally, if there is concern sizeineachgroup(80%poweratp< 0.05)
about a false-positive result (a or type 1 to be: 10 [3], 11 [7], 13 [8], 18 [1].
error), the level at which a result is
Finally, I can answer the question in
accepted as ‘significant’ would need Dr Morris’ last paragraph (at least in part)
adjustment. We can see no compelling by reference to Lilford and Johnson [9].
need for this in our study.
Studies often compare two ‘treatments’,
one established and one new. Usually,
the new treatment carries greater ‘costs’
paper that we were intentionally very General limitations of power calculations
conservative in our handling of the data.
The total times for intubation we quoted
for the single-use bougie were only for
those attempts which were successful.
Most investigators would have included
the times at which failure of insertion
occurred (i.e. ꢀ120 s). If we had done
this, the actual difference between the
Just as experiments have their limita- (e.g. financial, clinical or organisational).
tions, so do power calculations. Blind Therefore, in general it is appropriate to
faith in power calculations, as shown by design studies with the slight risk of
Dr Morris, is misplaced. Power calcu- rejecting new treatments which might
lations are only crude estimates (with crude work. However, where ‘costs’ of two
emphasised in the title of some relevant treatments are thought to be equal, it is
papers [3]). Two important factors entirely reasonable to power the study to
contribute to power [1,2]: (a) the greater than 80% [10].
Ó 2003 Blackwell Publishing Ltd
1237

Correspondence Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
10 Bacchetti P. Peer review of statistics in
medical research: the other problem.
British Medical Journal 2002; 324:
1271–3.
graphic scanning (PET) for the utilisa-
tion of glucose as a marker for activity,
showed decreased activity in the areas of
the brain involved with attention and
higher order cognitive thinking [2].
Consider the case of Gary Hart, after
the Selby Rail Disaster he was sen-
tenced to 5 years in prison for causing
the deaths of 10 people; in summing up
Justice MacKay said: ‘In moral terms I
see little to choose between a driver
who sets off on a journey affected by
drink and a driver setting off in your
condition.’ [3]. Gary hart had had little
or no sleep the previous night.
Therefore, continuing to work when
you have the subjective feeling of
tiredness (with the support of objective
research evidence), could be considered
a breech of General Medical Council
(GMC) guidelines, on the grounds of
putting the patient at risk, when you
know it is unsafe to continue [4,5].
Should something happen, it is not
impossible that you could be proven
‘negligent’ on the grounds of ‘forsee-
ability’ [6]. Even if you are not phys-
ically working, just being ‘on call’ will
affect your sleep pattern and quality of
sleep [7], which may have similar effects
on your performance the next day if
you had been working, as pointed out
in the study. Appropriate management
of this problem will result in reduced
day time activity, but failure to act will
continue to create risk. The problem
will become more acute in August 2004
as the European Working Time Direct-
ive bites for junior doctors hours.
Summary
Bacchetti [10] has explicitly stated that
criticisms of power calculations should
only be made when unavoidable: that is,
when there is little else in a study to
criticise, or when a study states clearly
unrealistic goals, or fails to appreciate its
other experimental limitations. Five of
11 paragraphs in our discussion drew
attention to important experimental
problems and limitations of the study.
These potential criticisms (which we
were careful to discuss of our own
study) far outweigh in importance the
rather obscure statistical points made by
Dr Morris regarding power calculations.
Sleep deprivation and
performance
I read with interest the study on sleep
deprivation and performance (Murray
& Dodds. Anaesthesia 2003; 58: 520–5).
I have been following the research into
sleep deprivation and performance
since, when driving home post ‘on call’,
I found myself on the wrong side of the
road with a large truck bearing down on
me. This was in the bad old days of
72 h weekends, so things must be better
now. But are they?
While juniors are given the day off
post call, and many work shifts, consult-
ants are expected to continue with the
rest of the days work. Murray and Dodds
do not state when their subjects started
working in the morning, but 07:00–
08:00 would cover most anaesthetic
consultants. If you have then been
supervising juniors, covering the ICU
and ⁄ or maternity while on call, it is not
uncommon to be up until 02:00 the
next day before you can contemplate
sleep, which will then be disturbed again
4–5 h later to get up to go to work again,
or possibly sooner. You may well have
been up all night and are then expected
to continue to work the routine list the
following day, as our contracts do not
recognise ‘on call’ as work.
One study has equated the level of
performance decrement using hand eye
co-ordination after sustained wakeful-
ness, and alcohol consumption [1]. After
18 h of wakefulness, the performance
was equivalent to 0.05% blood alcohol
concentration, and after 24 h, to 0.1%
blood alcohol concentration. If the Trust
found you to have such levels of alcohol
in your blood while working, you would
expect suspension and probably erasure
from the medical register, and possibly
criminal charges. However, if you have
been working all night, it is considered
acceptable to have the same level of
performance, as if you had been drink-
ing, and you still have to drive home.
In subjects who were sleep deprived
for 24 h, Positron Emission Tomo-
J. J. Pandit
John Radcliffe Hospital,
Oxford OX3 9 DU, UK
E-mail: jaideep.pandit@physiol.
ox.ac.uk
References
1 Young MJ, Bresnitz EA, Strom BL.
Sample size nomograms for interpre-
ting negative clinical studies. Annals of
Internal Medicine 1983; 99: 248–51.
2 Altman DG. Statistics and ethics in
medical research. III. How large a
sample? British Medical Journal 1980;
281: 1336–8.
3 Lehr R. Sixteen s squared over
d squared: a relation for crude sample
size estimates. Statistics in Medicine
1992; 11: 1099–102.
4 Glantz SA. Primer of Biostatistics.
McGraw-Hill, New York 1992.
5 Horrobin DF. Rationale for requiring
power calculations is needed. British
Medical Journal 2002; 325: 491–2.
6 Grieve AP. Confidence intervals and
sample sizes. Biometrics 1991; 47:
1597–602.
Anaesthesia has led the way in
reducing patient risk and making life
bearable for its staff; surely the time has
now come for both the Association
and the College of Anaesthetists to get
together and support further research
on the topic of performance and sleep
deprivation?
In the meantime, there is enough
evidence to suggest that it is unsafe for
patient care, and the individual anaes-
thetist, to continue working the day after
a busy ‘on call’, regardless of the grade.
7 Campbell MJ, Julious SA, Altman DG.
Estimating sample sizes for binary,
ordered categorical, and continuous
outcomes in two group comparisons.
British Medical Journal 1995; 311:
1145–8.
8 Florey CV. Sample size for beginners.
British Medical Journal 1993; 306:
1181–4.
S. R. Price
9 Lilford RJ, Johnson N. The alpha and
beta errors in randomised trials. New
England Journal of Medicine 1990; 322:
780–1.
Burnley General Hospital,
Burnley BB10 2PQ, UK.
E-mail: srprice@doctors.org.uk
1238
Ó 2003 Blackwell Publishing Ltd

Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
Correspondence
I doubt whether this can have influ-
enced the patients taking part in the Excel 210) and the integrity of the carbon
survey.
The anaesthetic machine (Ohmeda
References
1 Reid K. Fatigue, alcohol and per-
formance impairment. Nature 1997;
388: 235.
dioxide absorber and circle breathing
system was checked for leakage prior to
use and was deemed intact. The gas and
anaesthetic agent monitor that was in use
was the Datex Ohmeda (5250 RGM).
This has a side-stream sampling port and
was calibrated before use. After induc-
tion of anaesthesia for the first case, we
noticed that the baseline carbon dioxide
detected on capnography was elevated at
about 4–7 mmHg. A check on the
carbon dioxide absorber showed that it
was neither due to tracking nor exhaus-
tion of soda lime. The expiratory valve
on initial assessment appeared intact.
However, a further investigation into
the possible causes revealed that the
knife-edge of the valve seat of the
expiratory valve was cracked and
chipped (Fig. 1). It was most probably
that the hole in the valve seat, despite the
disc sitting in a correct position, allowed
positive pressure ventilation to occur
through the expiratory limb of the circle
breathing system as well. This would
explain the raised baseline carbon diox-
ide concentration.
J. Palmer
2 Thomas M, Sing H, Belenky G, et al.
Neural basis of alertness and cognitive
performance impairments during slee-
piness.1. Effects of 24 hour sleep
deprivation on waking human regional
brain activity. Journal of Sleep Research
2000; 4: 335–52.
Hope Hospital
E-mail: james.palmer@srht.nhs.uk
References
1 Culver DH, Horan TC, Gaynes RP,
et al. Surgical wound infection rates by
wound class, operative procedure, and
patient risk index. National Noso-
comial Infections Surveillance System.
American Journal of Medicine 1991; 91:
152–7S.
3 Davis P. Are we a catastrophe waiting
to happen? Anaesthesia News 2002: 176.
4 Good Medical Practice. General
Medical Council 2001.
5 Personal communication. Medical
Protection Society.
2 You and Your Anaesthetic. Royal
college of Anaesthetists and Association
of Anaesthetists of Great Britain and
Ireland February 2003.
6 Hepworth v Kerr 1995 6 Medical LR,
139.
7 Torsvall L, Akerstedt T. Disturbed
sleep while being on call: an EEG study
of ships’ engineers. Sleep 1988; 11:
35–8.
3 Anaesthesia Explained. Royal College
of Anaesthetists and Association of
Anaesthetists of Great Britain and
Ireland February 2003.
Patients ideas of surgical risk
Faulty unidirectional expiratory
valve as a cause of rebreathing
The survey on information that patients
request (Moores & Pace. Anaesthesia
2003; 58: 703–6) was very timely, and
made interesting reading. The authors
state that the commonest question asked
of them by patients was that of the likely
duration of surgery. This puzzled me
slightly until it crossed my mind that
perhaps patients were using the length
of surgery as a signal of the level of risk
and severity of the procedure. This is an
analogous situation to that of the first
question asked of many patients after
surgery by relatives and friends, i.e.
‘How many stitches did you have’.
Patients may therefore be using surgical
duration as a self selected guide to their
particular level of risk, erroneously
basing their question on newspaper
reports and television documentaries
where length of surgery is frequently
discussed. It is easy to dismiss their
question as one based on ignorance, but
strangely enough, they may well be
right [1]. It is also of interest to note that
the question of surgical duration is not
one highlighted in the recently pub-
lished information booklet for patients,
‘You and Your Anaesthetic’ [2], but is
briefly mentioned, and linked to risk, in
We would like to report an incident
involving a faulty unidirectional expir-
atory valve of a circle system with
carbon dioxide absorber (Series 5 A
absorber, Ohmeda) that led to rebreath-
ing of carbon dioxide.
A quick check of other similar
Ohmeda series 5 absorbers showed that
irregularities such as minute cracks and
deformed edges were common. We
speculate that this may be due to wear
‘Anaesthesia Explained’ [3]. However, Figure 1
Ó 2003 Blackwell Publishing Ltd
1239

Correspondence Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
and tear, or may be damaged during concludes with a misleading statement. study, capnography was used solely to
refitting of the cage onto the valve seat The study demonstrates that needle detect ‘correct tracheal needle inser-
after cleaning. This demonstrates the placement within the trachea is equally tion’ implying that the tip of the
importance of proper maintenance of likely with either method.
I take needle was in the right place. How-
the expiratory unidirectional valve, exception to that being called ‘correct ever, we clearly state that needle
which may be a more common cause needle insertion’. As they state in their insertion remains blind and we make
of rebreathing of carbon dioxide than discussion, the bronchoscope ‘ensured no reference to capnography locating
reported. Vigilance of the attending midline needle insertion at an appro- correct needle position.
anaesthetist and adequacy of monitoring priate tracheal level’. The capnograph-
A. Mallick
are equally important as well.
ic method is a blind technique which
is unable to provide any information
about the proximity to the midline, or
the level of insertion. A cadeveric
study [1] demonstrated marked tra-
cheal stenosis in two patients following
oblique insertion of the cannula into
the trachea. This complication would
go completely unnoticed in a blind
technique placement of the needle and
tracheostomy. The other concern
when trying to fashion tracheal access
with a tracheal tube in situ is that the
tracheal tube will be punctured by the
needle insertion as happened in one
patient in the capnography group in
this paper. Again, this is another
example of the needle being in the
right place (the trachea), but not being
in the correct place (within the in-
dwelling tracheal tube).
S. C. Elliot
W. S . Ye w
Leeds General Infirmary, Leeds, UK
E-mail: abhiram.mallick@
leedsth.nhs.uk
N. C. Hwang
Singapore General Hospital,
Singapore
E-mail: ganyws@sgh.co.sg
A hidden safety feature in the
event of oxygen failure
Six smiths and safety
We would like to report a recent
incidental discovery. Whilst checking
an anaesthetic machine (Blease Front-
line Plus 560TM), it was discovered
that failure of the oxygen supply not
only shuts off the nitrous oxide and
sounds the oxygen failure alarm, but
also supplies compressed air to the air
flowmeter irrespective of the position
of the air ⁄ nitrous oxide interlock
switch.
No mention of this feature is made in
the company’s brochure advertising the
machine, although it is mentioned in
passing in the user manual. This is
achieved by means of a gas relay that, in
the event of an oxygen supply pressure
below 225 KPa, supplies air directly to
the air flowmeter bypassing the inter-
lock switch.
All the recent correspondence regarding
syringe labelling and avoidance of drug
errors reminds me of an occasion whilst
working as a urology house officer in
Bournemouth. At the time there were
six patients named Smith on the ward.
To my dismay, the sister in charge
moved all six Smiths into the same six-
bedded bay. However, her management
was of course brilliant. Any intervention
involving a Smith now demanded that
the patient details were checked and
checked again! I am certainly not advo-
cating that all drugs be packaged in an
identical manner, but I would agree that
familiarity with the packaging and label-
ling of drugs definitely breeds a degree
of complacency, as suggested by your
I think a more accurate conclusion
would be to say that the capnography
method detects placement of the needle
in the trachea, and in the absence of a
bronchoscope being available, this
should be considered. However, tra-
cheal placement alone does not mean
correct placement.
correspondents (Amoako
& Lavies.
Anaesthesia 2003; 58: 816). I am sure I
am not alone in taking extra precautions
whenever using 0.9% saline, water for
injections or 1% lidocaine, all manufac-
tured by Braun and presented in iden-
tical Mini-PlascoÒ vials.
We believe that this is an important
added safety feature, which ensures a
continued supply of air to the common
gas outlet in the event of an oxygen
failure whilst remedial action is taken.
However, the feature is only effective
if the air flowmeter valve is left open,
if it is not then no advantage is gained.
We feel therefore that it should be
recommended by the manufacturer
that the air flowmeter valve is left
routinely open even when oxygen and
nitrous oxide are the gases in use.
S. W. Benham
John Radcliffe Hospital,
Oxford OX3 9 DU, UK
E-mail: maguerite.scott@orh.nhs.uk
Reference
J. G. Hudsmith
1 Van Heurn LWE, Theunissen PHMH,
Ramsay G, Brink PRG. Pathologic
changes of the trachea after percutane-
ous dilatational tracheotomy. Chest
1996; 109: 1466–9.
Peterborough Hospitals NHS Trust,
Peterborough PE3 9GZ, UK
E-mail: jonhudsmith@doctors.org.uk
Capnography detects tracheal
placement, but is it ‘correct’?
A reply
We agree that there is a difference
between needle placement in the tra-
chea and the actual positioning of the
needle through the anterior tracheal
wall, with the latter only able to be
confirmed with bronchoscopy. In our
C. M. Hearn
The recent study (Mallick et al. Anaes-
thesia 2003; 58: 864–8) comparing cap-
nography and bronchoscopy to detect
correct needle insertion before per-
forming a percutaneous tracheostomy
K. Chandradeva
Queen Mary’s Hospital,
Sidcup DA14 6LT, UK
E-mail: chandra.chandradeva@
qms.nhs.uk
1240
Ó 2003 Blackwell Publishing Ltd

Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
Correspondence
Fatal cardiac perforation by
central venous catheter
dilators: does the length
matter?
tracheal ring after serial dilatational
percutaneous tracheostomy [1] and we
are not aware of this complication
having previously been reported after
the ‘Blue Rhino’ technique. The inci-
dence of tracheal ring fracture is, how-
ever, probably higher with the ‘Blue
Rhino’ technique than when serial
dilators are used. We are aware of at
least three previous studies that support
this observation [2–4].
It would seem likely that, in the
case we describe, the same tracheal
ring could have been transected dur-
ing both needle introductions. The
fact that the cartilage was only iden-
tified in the trachea after the trache-
ostomy tube had been sited, suggests
the cartilage could have been disloca-
ted during the insertion of the trach-
eostomy tube.
A reply
We thank Drs Hearn and Chandradeva
for drawing readers’ attention to this
feature. It is nice to know these hidden
benefits in our equipment are being
appreciated by anaesthetists.
Recently, the Medical Devices Agency
has written to all parties involved with
the insertion of central catheters. This
letter dated 18th July 2003 (reference
MDA ⁄ 2003 ⁄ 020) warned of the dan-
gers of over insertion of dilators that
come with various single and multi-
lumen central catheters. The solution
proposed was to increase education and
awareness of the issue.
Fatal cardiac or vascular perforation,
although fortunately not common, is
certainly not rare. There are many case
reports published and most anaesthetists
and intensivists will be aware of many
unreported cases. The dilator has been
implicated in both major vessel [1] and
cardiac penetration [2]. Sensibly, clini-
cians should be aware of, and trained in
the avoidance of, all the various
mechanical complications of central
venous access.
A. Ferguson
Blease Medical Equipment Ltd,
Chesham HP5 2WX, UK
Tracheal ring fracture –
dislodgement after Blue Rhino
percutaneous tracheostomy
Percutaneous dilatational tracheostomy
has become a standard procedure on
many ICUs and we have carried out
over 150 such procedures in the last
12 months. In spite of the frequency
with which the procedure is carried out,
it still carries the risk of serious compli-
cations. We would like to report a case
of tracheal ring fracture–dislodgement
following tracheostomy with a Cook
‘Blue Rhino’ percutaneous tracheo-
stomy kit.
The case illustrates the importance of
bronchoscopic control during percuta-
neous tracheostomy and of inspection
of the tracheal after the procedure has
been completed.
A. Thomas
In almost all Seldinger method kits,
the dilator is substantially longer than the
needle or cannula used to insert the
wire. This is unnecessary and we have
asked various manufacturers why the
dilators need to be so long. The dilator
in percutaneous tracheostomy sets is the
same length as the introducer cannula
and functions satisfactorily. We would
propose that dilators in central line kits
do not need to be as long as they are, and
I suspect that the reason is convenience
of manufacture and possibly also histor-
ical acceptance.
S. Subramani
A 36-year-old man had had a coiling
of a ruptured anterior cerebral artery
aneurysm and required a tracheostomy
as part of his postoperative care. Under
general anaesthesia and bronchoscopic
control, the trachea was punctured with
the introducing cannula. The first punc-
S. Mitra
Salford Royal Hospital,
Salford UK
E-mail: tony.thomas@srt.nhs.uk
References
ture was not in the midline and a second 1 Sun KO. Fracture and dislodgement of
puncture was required. The tracheo-
stomy was then carried out uneventfully
and a size 8 Portex tracheostomy tube
tracheal cartilage during percutaneous
tracheostomy. Anaesthesia 1995; 50:
370.
was placed using the 24 French gauge 2 Beiderlinden M, Karl Walz M, Sander
introducer. Unfortunately, during the
immediate post procedure broncho-
scopy, a foreign body was seen in the
mid trachea. This was removed with the
suction port of the bronchoscope and
A, Groeben H, Peters J. Complications
of bronchoscopically guided percuta-
neous dilational tracheostomy: beyond
the learning curve. Intensive Care
Medicine 2002; 28: 59–62.
W. C. Russell
J. Greiff
Leicester Royal Infirmary,
Leicester LE1 5WW, UK
E-mail: William.russel@uhl-tr.nhs.uk
was found to be a piece of cartilage 3 Byhahn C, Wilke HJ, Halbig S,
1.8 cm long and 4 mm wide. The
dimensions suggested that it formed part
of a tracheal cartilage. The patient made
a slow recovery from the subarachnoid
haemorrhage and the trachea was even-
tually decannulated after 8 weeks. The
Lischke V, Westphal K. Percutaneous
tracheostomy: Ciaglia blue rhino versus
the basic Ciaglia technique of percuta-
neous dilational tracheostomy.
Anesthesia and Analgesia 2000; 91:
882–6.
References
1 Robinson JF, Robinson WA, Cohn A,
Garg K, Armstrong JD. Perforation of
the great vessels during central venous
line placement. Archives of Internal
Medicine 1995; 155: 1225–8.
patient had no symptoms suggestive of 4 Edwards SM, Williams JC. Tracheal
tracheal narrowing and no change in
voice; a further bronchoscopy was
therefore not performed.
We are aware of only one previous
case report of fracture dislodgement of a
cartilage fracture with the Blue Rhino
Ciaglia percutaneous tracheostomy
system. European Journal of Anaesthesiol-
ogy 2001; 18: 487.
2 Porter JM, Page R, Wood AE, Phelan
D. Ventricular perforation associated
with central venous introducer-dilator
Ó 2003 Blackwell Publishing Ltd
1241

Correspondence Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
systems. Canadian Journal of Anaesthesia
1997; 44: 317–20.
face-mask, but the unusual noise con-
tinued. The patient was turned into the
left lateral position and subsequently
suffered a bout of coughing, expector-
ating a lump of white material. The
noise from her airway promptly ceased.
On closer inspection, the object proved
to be a piece of chewing gum. This had
not been seen at laryngoscopy. Either it
had been adherent to the teeth or gums,
or she had swallowed it and it had been
regurgitated into the pharynx. The
patient was eventually discharged home
with no problems.
This is not the first case report descri-
bing chewing gum obstructing the air-
way [1–4], although previously, the gum
has adhered to an airway device (tracheal
tube, etc.). We were not so fortunate.
Chewing gum can be a very real cause of
airway obstruction with potentially life-
threatening consequences. Where there
is a real chance of operative intervention,
the obstetric population being a prime
example, should chewing gum not be
forbidden? After our experience, we feel
that this should be the case, or at the very
least it should be something the anaes-
thetist should chew over.
The Negus slotted laryngoscope
We read with interest your recent letter
describing the use of the anterior com-
missure laryngoscope and we agree that
this style of laryngoscope has particular
advantages that have been described
very well by the authors (Davies &
Balachandran. Anaesthesia 2003; 58:
721–2). As they state, the closed barrel
style of laryngoscope may be of partic-
ular use in the patient with a difficult
airway and in whom fibreoptic laryng-
oscopy may not be possible, e.g. when
there is already, or is likely to be,
significant bleeding. However, we
would like to make a suggestion that
in fact there may be an even more
useful piece of equipment amongst the
ENT surgical instruments. In common
with the anterior commissure laryngo-
scope, the Negus slotted laryngoscope
also has an enclosed barrel, with all
the advantages previously described
(Fig. 2). The cords may be visualised
using the paraglossal approach from the
right side, and the bougie advanced
between the cords. However, the pos-
terior section of the barrel of the Negus
slotted laryngoscope may then be re-
moved completely by sliding it towards
the operator (Fig. 3). The tracheal tube
Figure 3 The Negus slotted laryngoscope with
posterior part of blade retracted
tracheal tube. Thus the trachea may be
intubated under direct visualisation, and
further local damage caused by blind
advancement of the tracheal tube may
be avoided.
Of course, it could be argued that
fibreoptic intubation should be the ‘plan
A’ in all patients in whom a pharyngeal,
laryngeal or lingual lesion is suspected,
but even then the use of the Negus
slotted laryngoscope would make a very
reasonable ‘plan B’.
A. Dering
R. Corbridge
A. Kitching
A. Afzaal
Royal Berkshire Hospital,
Reading RG1 5AN, UK
E-mail: andydering@onetel.net.uk
J. Kirk-Bayley
D. R. Uncles
Worthing Hospital,
Worthing BN11 2DH, UK
E-mail: azedafzaal@hotmail.com
may then be passed over the bougie and Something to chew over?
the laryngoscope removed completely
We would like to present an unusual
case of airway obstruction in a 30-year- 1 Bevacqua BK, Cleary WF. An unusual
References
without having to pass the laryngoscope
back over the tracheal tube, and the
tube connector may remain attached old-primigravida female with
a
case of endotracheal tube cuff dys-
function. Journal of Clinical Anesthesia
1993; 5: 237–9.
35-week twin pregnancy who opted
for vaginal delivery. An epidural was
originally incorporated to allow the sited without incident, and the first twin
throughout. This removable section of
the slotted Negus laryngoscope was
2 Wenke M, Akca O. Chewing gum on
a Laryngeal Mask Airway. Anesthesiol-
ogy 2002; 97: 1647–8.
was successfully delivered vaginally.
Unfortunately, the second twin suffered
advancement of a rigid bronchoscope,
but this also allows the passage of a
shoulder dystocia and fetal distress, so 3 Cupitt JM, Roye SE. A sticky situation.
the mother was taken for emergency
Anaesthesia 1999; 54: 1127–8.
Caesarean section. For expedience, the 4 Nowicki RW. Another sticky situ-
epidural was not used and a general
anaesthetic was administered with intu-
ation. Anaesthesia 2000; 55: 305.
bation of the airway. The operation was The ProSeal laryngeal mask
uneventful and the patient was extubat- airway: an easier and safer
ed at the end. Shortly afterwards, a approach to tracheal tube/
stertorous noise was heard, although the laryngeal mask exchange
patient continued to breath adequately
with good oxygen saturations. Pharyn- Exchanging the tracheal tube (TT) for
geal suctioning was repeatedly per- the classicTM laryngeal mask airway
formed and oxygen delivered with a (classic LMA) has been used following
Figure 2 The Negus slotted laryngoscope
1242
Ó 2003 Blackwell Publishing Ltd

Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
Correspondence
a variety of surgical procedures to hood that it will form an effective had been successfully treated with DC
facilitate a smooth emergence, and barrier to regurgitation. We have used cardioversion. A lateral neck X-ray
offers advantages over the face-mask in this technique in 18 patients, mostly demonstrated a radio-opaque foreign
terms of respiratory complications [1]. after thyroid surgery or suspension body in the oesophagus at the level of
Although the technique is effective and laryngoscopy, without any problems.
probably safe, there are two reports of
J. Brimacombe
C6 [Fig. 4]. This was thought most likely
to be a piece of bone. The patient was
listed for urgent rigid oesophagoscopy.
Examination revealed an obese,
123 kg gentleman in obvious discom-
fort. He was cardiovascularly stable
with heart rate 100 beat.min)1, BP
140 ⁄ 85 mmHg and oxygen saturations
in air of 97%. Airway examination was
classified as Mallampatti II. The patient
was noted to be exquisitely tender
over the anterior part of his neck at
the level of C6. The anaesthetist, con-
cerned about the use of cricoid pressure
during rapid sequence induction and
the risk of this causing oesophageal
perforation, consulted his ENT col-
league. The opinion of the ENT sur-
geon was that there was a small risk
associated with the use of cricoid pres-
sure but this was significantly less than
the risk of perforation if the patient was
left for 6 h to allow his stomach to
empty. On this basis it was decided to
proceed with the case. The patient was
premedicated with intravenous raniti-
dine 50 mg and metoclopramide10 mg.
Anaesthesia was induced in theatre with
alfentanil 1 mg, propofol 230 mg and
succinylcholine 150 mg. Cricoid pres-
sure was applied by a trained assistant
immediately following loss of con-
aspiration during the exchange [2,3]. In
N. Brimacombe
principle, the risk of aspiration can be
Cairns Base Hospital,
reduced by maintaining anaesthesia or
Cairns, Australia
muscle relaxation (to prevent airway
E-mail: jbrimaco@bigpond.net.au
protective reflex activation), passage of a
C. Keller
gastric tube (to empty the stomach) and
Leopold-Franzens University,
placement of the classic LMA before
Innsbruck, Austria
extubation (to form a barrier in the
hypopharynx). We describe a technique
involving the ProSealTM laryngeal mask
airway (ProSeal LMA) that we consider
easier and safer than using the classic
LMA.
References
1 Brimacombe J. Laryngeal Mask Anes-
thesia. Principles and Practice, 2nd edn.
London: Harcourt Brace in press.
2 Vogelsang H, Uhlig T, Schmucker P.
The technique involves the following
[Severe multiorgan failure caused by
steps, which should only take place with
the patient deeply anaesthetised (>1.3
MAC [4]), or paralysed:
(1) pharyngeal suction,
(2) placement of the ProSeal LMA
using the laryngoscope-guided, gum
elastic bougie-guided technique [5],
(3) partial inflation of the ProSeal
aspiration with laryngeal mask airway].
Anaesthesiologie Intensivmedizin Notfal-
medizin Schmerztherapie 2001; 36: 63–5.
3 Langer A, Hempel V, Ahlhelm T,
Heipertz W. [Experience with use of
the laryngeal mask in over 1900 general
anaesthetics]. Anaesthesiologie Intensiv-
medizin Notfalmedizin Schmerztherapie
LMA cuff,
(4) holding the ProSeal LMA at the
1993; 28: 156–60.
4 Takita K, Yamane M, Morimoto Y,
mouth to maintain the position of the
Kemmotsu O. The ED (95) of end-
distal cuff in the hypopharynx,
tidal sevoflurane concentration for the
(5) inserting a gastric tube via the
smooth exchange of the tracheal tube
drain tube and emptying the stomach,
for a laryngeal mask airway is 2.97%.
(6) disconnecting the anaesthesia
breathing system from the TT and
deflating the TT cuff;
(7) tracheal extubation while main-
taining the position of the ProSeal
LMA;
Canadian Journal of Anaesthesia 2003; 50:
184–8.
sciousness. Laryngoscopy revealed a
5 Howarth A, Brimacombe J, Keller C.
grade II view of the larynx and the
patient’s trachea intubated with an 8.0-
mm armoured tracheal tube. Anaesthe-
sia was maintained in an oxygen ⁄ air
mixture with 2% sevoflurane. Rigid
oesophagoscopy revealed a segment of
bone in the oesophagus with sur-
rounding tissue oedema. The bone
was removed and further examination
demonstrated no evidence of perfor-
Gum elastic bougie-guided insertion
of the ProSeal laryngeal mask airway. A
new technique. Anaesthesia and Intensive
(8) connecting the anaesthesia brea-
thing system to the ProSeal LMA;
Care 2002; 30: 624–7.
(9) fixing the ProSeal LMA in posi-
tion with tape;
(10) reversal of muscle relaxation, if Contraindications to cricoid
required, and discontinuation of anaes- pressure
thesia.
The advantages of the ProSeal LMA A 29-year-old man presented to the ation. The patient was extubated
over the classic LMA are easy access to A & E department of his local hospital uneventfully and a postoperative chest
the stomach and improved ventilatory complaining of the sudden onset of X-ray was normal.
capability; a disadvantage is that a painful dysphagia. This had occurred
In Sellick’s original paper, only vom-
smaller fibreoptic scope must be used immediately following the ingestion of iting is mentioned as a contraindication
if laryngoscopy is required. The advant- a piece of lamb shank as part of a three- to the use of cricoid pressure [1]. Since
age of the gum elastic bougie over the course meal. Past medical history was then, a number of clinical conditions
digital and introducer tool techniques is unremarkable apart from an episode of have been reported as possible contra-
that the distal cuff will be guided into atrial fibrillation following accidental indications to its use. These include
the hypopharynx, increasing the likeli- electrocution one year previously. This actual or suspected cervical spine injury,
Ó 2003 Blackwell Publishing Ltd
1243

Correspondence Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
grave consequences as a mismatched
transfusion.
Mr CC is a 73-year-old gentleman
who presented to the Accident and
Emergency Department with a month
long history of left hip pain. He could
not identify any traumatic precipitating
event. His admission clerking states that
his only past medical history was that of
multiple myeloma diagnosed 6 months
earlier for which he had received
three courses of chemotherapy. X-rays
revealed a fracture of the left neck of
femur with evidence to suggest that
the fracture was pathological. In the
margin of the clerking, the blood results
had been documented. Apart from a
low white cell (WCC) count of
2.2 · 109.l)1, the full blood count and
clotting were normal, as were his elec-
trolytes and renal function.
He was seen by an anaesthetist and
both patient and anaesthetist agreed on
spinal anaesthesia as the best course of
action. His operation was delayed until
the following day due to lack of theatre
time and so he was seen by a second
anaesthetist who questioned the pres-
ence of an isolated low white cell count
in a patient who may well have poorly
functioning bone marrow and other bio-
chemical abnormalities. Indeed, upon
checking the admission blood results
on the computer they were found to
Figure 4
be: WCC 2.2 · 109.l)1, haemoglobin
laryngeal trauma and the presence of a
foreign body in the upper airway.
Although intuitively it would seem
reasonable to assume that the presence
of a sharp foreign body (e.g. a bone) in
the oesophagus would contraindicate
cricoid pressure, a literature search
revealed no case reports of oesophageal
perforation in this context. A study by
Canter et al. in cadavers [2] demonstra-
ted no evidence of oesophageal trauma
when sharp foreign bodies were placed
in the cervical oesophagus and cricoid
pressure applied. From this study and
our literature search we would conclude
that cricoid pressure can safely applied
in this situation. We would welcome
other readers’ thoughts on the manage-
ment of this case.
References
11.5 g.dl)1, platelets 62 · 109.l)1, urea
1 Sellick BA. Cricoid pressure to control
regurgitation of stomach contents dur-
ing induction of anaesthesia. Lancet
1961; 2: 404–7.
7.8 mmol.l)1, creatinine 151 lmol.l)1
.
The origin of the incorrect blood
results remains unclear. Once the error
had been identified, the advice of the
haematologists was sought and the
operation went ahead under general
anaesthesia with broad spectrum anti-
biotic and platelet cover. The conse-
quences of proceeding with the
operation under spinal anaesthesia on
the basis of the erroneous blood results
2 Canter RJ, Gath AM, Harris D. Cricoid
pressure and sharp foreign bodies.
Clinical Otolaryngology 1985; 10:
343–4.
Incorrect blood results
Errors in blood transfusions have resul- may have been disastrous. Increased
ted in rigid protocols concerning the surgical bleeding may have been
drawing of blood transfusion samples, encountered and an epidural or spinal
their labelling, handling in the laborat- cord haematoma may have resulted
ory and subsequent checking of prod- from the choice of anaesthetic.
ucts for transfusion. However, a more
The question now is not who to
casual approach to the handling of other blame for this incident but how to
blood samples remains. I would like to prevent a recurrence. The manual tran-
present a case where relying on incor- scribing of results from computer to
rect blood results may have had as house officer notebook to patients’ case
S. Lewis
P. Magee
Royal United Hospital,
Bath BA1 3NG, UK
1244
Ó 2003 Blackwell Publishing Ltd

Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
Correspondence
notes is obviously unacceptable, but Thirty-six hours post surgery, an the use of factor VIIa in such individuals
laboratory generated results forms may increase in bleeding was noted and over is well founded.
take more than 24 h to return to the the next 18 h this continued despite two
J. L. Brown
ward. My preference would be compu- re-exploration thoracotomies, neither
S. Varma
ter generated printouts on sticky labels of which revealed an obvious bleeding
N. J. Morgan-Hughes
complete with full patient details and point. Non-operative treatment during
date of test that could be stuck in to the this period included the administration
patients’ notes. This particular incident of aprotinin 2000 000 units, desmo-
appears to have resulted from an error in pressin 16 lg, 17 units of packed cells,
transcription of results but perhaps it 2.5 L of fresh frozen plasma and 10 units
highlights the fact that all blood samples of platelets.
G. Rocco
Northern General Hospital,
Sheffield S5 7AU, UK
E-mail: n.hughes@sheffield.ac.uk
References
warrant the same degree of respect as
The bleeding continued and 6 h after
1 Jurlander B, Thim L, Klausen NK, et al.
Recombinant activated factor VII
(rFVIIa): characterization, manufac-
turing and clinical development.
Seminars in Thrombosis and Hemostasis
2001; 27: 373–84.
blood transfusion specimens, with three the second thoracotomy, he continued
forms of patient identification and sig- to lose an average of 500 ml.h)1 into
natures of the doctor, nurse or phlebot- the chest drain. The haematolo-
omist taking the sample on bottle and gists were consulted and factor V11a
form. It also reminds me that all test (0.1 mg.kg)1) was given. Within 30
results should be viewed in the context min of receiving factor V11a the bleed-
of the patient to whom they relate, and ing settled. Blood loss was only 250 ml
2 Worth LL, Hoots WK. Development
of a subdural vein thrombosis following
aggressive factor VII replacement for
post natal intracranial haemorrhage in
a homozygous factor VII-deficient
infant. Haemophilia 1998; 5: 757–61.
3 Heuer L, Blumberg D. Recombinant
factor VIIa – a new option in intract-
able bleeding? International Journal of
Intensive Care 2003; 10: 73–80.
causes for any anomaly sought.
in the subsequent 6 h after which, all
bleeding ceased. He required only one
further unit of packed cells.
E. Combeer
East Surrey Hospital,
Redhill RH1 5RH, UK
The patient’s AV fistula was docu-
mented to have had a thrill between
the two re-exploratory thoracotomies.
Eight hours later and 30 min after the
factor VIIa was commenced, the renal
team reviewed the patient. At this time,
a thrill was initially absent although
returned after the fistula was massaged.
The next morning, 14 h after the factor
VIIa, the fistula had no thrill and was
presumed clotted. The patient recov-
Thrombosis related to
emergency factor VIIa
treatment
Factor VIIa is emerging as a treatment
for a variety of coagulation disorders
characterised by impaired thrombin
generation and life-threatening bleed-
ing. It initiates the coagulation cascade
by binding to exposed tissue factor at
the site of endothelial injury and causes
the formation of thrombin to trigger
coagulation [1]. Based on its mechanism
of action, thrombosis is a potential risk,
although this has only rarely been
reported [2]. We report a case in which
treatment with factor VIIa was life
saving; however, the treatment was
complicated by the clotting of an arte-
riovenous (AV) fistula.
Malignant hyperthermia –
patients don’t always read the
textbook
ered and was subsequently discharged to An individual anaesthetist may never see
the renal unit.
Factor VIIa is a highly effective hae-
a patient with malignant hyperthermia
during their career; they are even less
mostatic agent that has been used in the likely to see an unusual presentation.
prevention and treatment of bleeding in
patients with coagulopathies and in the
setting of massive haemorrhage. It’s
The Association of Anaesthetists of
Great Britain and Ireland has issued
guidelines on the presentation and man-
mechanism of action, dependent on the agement of malignant hyperthermia
interaction with exposed tissue factor,
localises its effect [3]. However, patients
with other areas of abnormal endothe- often sudden, other features can be
(MH). While tachycardia and hypercap-
nea are invariably present and their onset
lium may be at risk of thrombosis at these
sites, for example those with coronary
atheroma. Although we cannot be abso-
lutely certain that the Factor VIIa initi-
ated thrombosis of this patient’s AV
variable and the presentation insidious.
A recent case illustrated this to us.
A 50-year-old man, who had not
previously been anaesthetised, presented
for re-implantation of his left forearm
A 58-year-old man with diabetes
mellitus, end-stage renal failure and a
right forearm AV fistula was scheduled
for an emergency thoracotomy for a
right-sided empyema. The pre-operative
coagulation profile was normal. The
surgery was uneventful and the patient
was transferred to the Cardiac Intensive
Care Unit (CICU) postoperatively. The
patient underwent haemofiltration and
the AV fistula was kept warm and
examined regularly for a thrill.
fistula, the timing is suggestive. High following traumatic amputation. He
flows through the fistula may have led to received an intravenous induction with
vessel wall damage, predisposing it to
thrombosis once the factor VIIa was
given. This case suggests that thrombotic gen, air and isoflurane. He remained
fentanyl, propofol and rocuronium and
anaesthesia was maintained with oxy-
events in patients with abnormal vascu-
lature are more than just a theoretical
concern and that a cautious approach to incremental metaraminol to a total dose
haemodynamically stable for 3 h but
then became hypotensive requiring
Ó 2003 Blackwell Publishing Ltd
1245

Correspondence Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
surgical patients. Journal of Neurosurgical
Anesthesiology 1997; 9: 324–8.
of 9 mg over 45 min Shortly after the
After institutional approval and
onset of hypotension, his heart rate and informed consent, 57 patients, aged
end-tidal CO2 started to increase slowly, 19–87 years, undergoing non-neuro-
reaching peaks of 150 beat.min)1 and surgical procedure under spinal anaes-
94 mmHg, respectively. Naso-pharyn- thesia were enrolled in this study. Two
geal temperature also increased slowly to ml of arterial blood and 1 ml of CSF
a maximum of 39.5 ꢀC by which time were obtained prior to the surgical
the diagnosis had been made and intra- procedure for determination of plasma
venous dantrolene started. After 100 mg and CSF iMg concentration with an
of dantrolene had been infused, his heart electrolyte analyser CRT8TM (NOVA,
rate, temperature and end-tidal CO2 Waltham, MA, USA). The mean iMg
started to return to normal. The re- concentrations in plasma (0.49 SD
implantation was successfully completed 0.05 mmol.l)1, n ¼ 57) were lower
and the patient made a full recovery. He than those in CSF (0.77 SD
has since undergone a muscle biopsy, 0.04 mmol.l)1, n ¼ 57). There was no
and in-vitro contraction testing has correlation between plasma and CSF
confirmed his susceptibility to malignant iMg concentrations (Regression equa-
A novel role for magnesium?
We would like to highlight a recent
example of the use of intravenous
magnesium in a lady with severe asthma
undergoing elective lower segment
Caesarean section (LSCS) under com-
bined spinal-epidural anaesthesia. Des-
pite the increasing use of magnesium in
a wide range of clinical settings, there
have been no publications describing its
use in this scenario.
A 30-year-old Caucasian lady was
admitted for an elective LSCS for the
delivery of her second child. She had
been asthmatic since the age of 3 years,
requiring numerous hospital admissions.
On her previous admission for an acute
exacerbation, she had responded well to
intravenous magnesium administered in
the emergency department. Her asthma
was normally poorly controlled on a
combination of long-term high-dose
steroids (> 30 mg prednisolone daily),
theophyllines, nebulised b agonists and
nebulised antimuscarinics. Despite this
therapy, she had an average peak exp-
iratory flow rate (PEFR) of only
250 l.min)1 and required home oxygen,
especially at night. She had a body mass
index of 36 kg.m)2 and past medical
history included a thrombophilic ten-
dency having suffered from two previ-
ous deep vein thromboses. Her first
child had been delivered by LSCS
18 months previously under a combined
spinal-epidural anaesthetic. Throughout
the whole procedure she had felt
extremely short of breath and distressed
despite nebulised b agonists.
hyperthermia.
tion: y ¼ 0.745 + 0.058x, r ¼ 0.06; p ¼
This case illustrates two important 0.65, n ¼ 57), suggesting that CSF iMg
points. Firstly, there can be a delay concentration tends to remain stable
between exposure to trigger agents and regardless of variation in plasma iMg
the appearance of signs of MH. This may concentration in patients with an intact
be particularly important in shorter cases BBB. However, Fuchs-Buder et al. [3]
where the onset may be in the recovery reported that an intravenous adminis-
area where neither end-tidal CO2, nor tration of magnesium sulphate led to a
temperature is monitored. Second, des- significant increase in CSF iMg con-
pite a hypermetabolic state, the patient centration in neurosurgical patients and
can be hypotensive rather than hyper- that it took 90 min to raise CSF iMg
tensive. In addition, it is worth noting concentration following the injection.
that susceptible patients will not always The present results cannot rule out the
manifest the signs of MH on their first possibility that long-term loading or
exposure to trigger agents.
depletion of plasma iMg might affect
the CSF iMg concentration.
S. Wimbush
J. Homewood
R. Sasaki
Frenchay Hospital,
Bristol BS16 1LE, UK
E-mail: wimbush@ukgateway.net
K. Hirota
M. Yamazaki
Toyama Medical and Pharmaceutical
University, Toyama,
930–0194, Japan
E-mail: koki@ms.toyama-mpu.ac.jp
Ionised magnesium
concentrations in
non-neurosurgical patients
undergoing spinal anaesthesia
References
1 Kroin JS, McCarthy RJ, Roenn NV,
Schwab B, Tuman KJ, Ivankovich AD.
Magnesium sulfate potentiates mor-
phine antinociception at the spinal
Prior to her second LSCS she received
intravenous magnesium sulphate 2 g in
10 ml normal saline over 10 min. A
It has been reported that changes
in cerebrospinal fluid (CSF) ionised
magnesium (iMg) concentration can
modify analgesic effects in vivo [1] and
alter the general anaesthetic actions
in vitro [2]. Since it is important to
know whether measurements of the
plasma iMg concentration predict the
alteration in CSF iMg concentration,
we investigated the relationships be-
tween plasma and CSF iMg concentra-
tions in non-neurosurgical patients with
an intact blood–brain barrier (BBB).
level. Anesthesia and Analgesia 2000; 90: combined spinal-epidural was then per-
913–17.
formed in the sitting position and 0.5%
heavy bupivacaine 2.5 ml and diamor-
Yamazaki M. Extracellular magnesium phine 0.3 mg were injected into the
2 Sasaki R, Hirota K, Roth SH,
ion modifies the actions of volatile
anesthetics in area CA1 of rat hippo-
campus in vitro. Anesthesiology 2002;
96: 681–7.
subarachnoid space. She was placed in
the supine position with left tilt and
received nebulised b agonists with oxy-
gen 4 l.min)1 once the operation started.
3 Fuchs-Buder T, Tramer MR, Tassonyi There were no signs of tachypnoea or
E. Cerebrospinal fluid passage of intra- respiratory distress and her oxygen satu-
venous magnesium sulfate in neuro-
rations remained greater than 98%
1246
Ó 2003 Blackwell Publishing Ltd

Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
Correspondence
an adjunct in the treatment of acute
asthma. Chest 1995; 107: 1576–81.
6 Skobeloff EM, Kim D, Spivey WH.
Magnesium sulphate for the treatment
of bronchospasm complicating acute
bronchitis in a four months pregnant
woman. Annals of Emergency Medicine
1993; 22: 157–9.
throughout the duration of surgery
Magnesium may be a useful adjunct
(50 min). Her PEFR increased from in the brittle asthmatic undergoing
200 to 275 l.min)1 following adminis- Caesarean section and we would be
tration of the magnesium. She reported interested to hear if others have used it
no adverse side-effects and the surgeon in this situation.
reported no increase in blood loss.
J. Bromilow
Magnesium plays a vital homeostatic
A. McCormick
role within the body via transmembrane
Poole Hospital NHS Trust,
and intracellular modulation of electri-
Poole, UK
7 Stanton MF, Lowenstein FW. Serum
magnesium in women during preg-
nancy, whilst taking contraceptives and
after the menopause. Journal of the
American College of Nutrition 1987;
6: 313–19.
cal cellular activity. It causes smooth
E-mail: brom@doctors.org.uk
muscle relaxation by altering extracel-
lular calcium influx and intracellular
phosphorylation reactions. These prop-
erties are utilised in a range of pathol-
ogies including pre-eclampsia, cardiac
arrhythmias, ischaemic heart disease,
migraine and alcohol withdrawal [1].
References
1 McLean RM. Magnesium and its
therapeutic uses. A Review. Ameri-
8 Zalogd GP. Interpretation of serum
magnesium levels. Chest 1989; 95:
257–8.
can Journal of Medicine 1994; 96: 63–76.
2 Haury VG. Blood serum magnesium in
bronchial asthma and its treatment by
Magnesium was first reported anec-
the administration of magnesium sul-
dotally for the treatment of asthma
50 years ago [2]. Its use has been further
supported in recent years following
several published randomised clinical
trials. Skobeloff et al. [3] reported im-
phate. Journal of Laboratory Clinical
Diprifusor – things are not
always what they seem
Medicine 1940; 26: 340–4.
3 Skobeloff EM, Spivey WH,
McNamara RM. Intravenous magnes-
Computerised equipment is frequently
ium sulphate for the treatment of acute used in modern anaesthesia. In clin-
asthma in the emergency department. ical practice, we do not necessarily
Journal of the American Medical Society pay attention to their engineering
proved PEFR and discharge rates from
emergency departments in adults with
severe asthma. Two further trials [4,5]
have highlighted its effectiveness in
moderate to severe asthma in both
paediatrics and adults at a dose of 20–
25 mg.kg)1 and 2 g, respectively. Its
use in pregnancy for the treatment of
acute bronchospasm has also been
documented in a case report [6].
Pregnancy is often associated with
exacerbations of asthma especially as
gestation increases. This is multifactorial
due to an increase in prostaglandin
levels, raised upper airway resistance,
decreased expiratory reserve volume and
a decrease in cell-mediated immunity
resulting in an increased susceptibility to
upper respiratory tract infections and
inflammation. These factors, together
with a splinted diaphragm, anxiety and
the supine position, make it likely that
asthmatic symptoms are exacerbated
during Caesarean section. The use of
magnesium in this case is further sup-
ported because pregnancy can signifi-
cantly decrease serum magnesium levels,
especially in white patients [7].
Although our patient’s measured serum
magnesium was only just outside the
normal range (0.69 mmol.l)1; normal
range 0.70–1.10 mmol.l)1), serum mag-
nesium levels do not necessarily corre-
late with total body stores [8].
1989; 262: 1210–13.
details. ‘Diprifusor’ a Target Controlled
Infusion (TCI) pump for use with
propofol is a perfect example of this.
We discovered that by operating a latch
on the right hand side, the pump could
be flipped open as seen in Figs 5 and 6.
We performed a medline search to
resolve the mystery of what looked like
two pumps.
4 Ciarallo L, Sauer AH, Shannon MW.
Intravenous magnesium therapy for
moderate to severe pediatric asthma:
results of a randomised, placebo-
controlled trial. Journal of Pediatrics
1996; 129: 809–14.
5 Bloch H, Silverman R, Mancherje N.
Intravenous magnesium sulphate as
Figure 5 The ‘Diprifusor’ pump fac¸ade
Ó 2003 Blackwell Publishing Ltd
1247

Correspondence Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
mean? Second, we have to evaluate
the effects of fentanyl on the bronchial
smooth muscles. If we could accept
that this effect is similar for each
inhalation agent, then this paper could
be of value. If not, what are the
conclusions? Considering the half time
of fentanyl, we have to keep in mind
the effects of this drug in this study.
Last but not least, the authors con-
clude that the effect of desflurane at 2
MAC results in an increase in airway
resistance. If we look at the results, the
respiratory resistance base-line value of
desflurane is 19.9 (4.0) and at 2 MAC
19.7 (3.4), 19.7 being statistically not
significant compared to the baseline.
In mathematics (19.9–19.7) ⁄ 19.9 ¼ 0.01
or only 1%. Is it clinically significant?
Please also take into consideration the
variability of the method of measur-
ing the respiratory resistance. Can we
conclude that we really have broncho-
constriction? What is its clinical signifi-
cance? All we can say is: if you go
abruptly from 6% to 12% desflurane for
only 5 min, there is no bronchodilation.
But it doesn’t necessarily mean bron-
choconstriction. This questions whether
the well known pungency of the drug
can explain this and according to the
paper of Ebert [1] about the cardiovas-
cular effects, if this effect is equally short
lived ( 7 min) or not? Finally how can
they write that at T 2 the PIP decreased
6.7% (13.8) with isoflurane, when the
reported values are 20.7 (5.9) and 18.8
(4.2)? In mathematics (20.7–18.8) ⁄
20.7 ¼ 0.09 or 9% and not 6.7%. If
you look at the other results you will
have the same surprises. If you consider
not the absolute mean value but the
dispersion around it, there could be
questions about the validity of the results
and the conclusion. Why should we
compare T3 with T2 and not with the
baseline T1?
Figure 6 Behind the fac¸ade of the ‘Diprifusor’ pump is a syringe driver.
British Journal of Anaesthesia 1991;
It became clear that behind the mask
of the TCI (Fig. 5) technology is the face
of a simple syringe-driver (Fig. 6). The
TCI software enables the syringe-driver
to deliver a desired target concentration,
the pharmacodynamic response being
influenced by pharmacokinetic param-
eters incorporated into the system.
The pharmacokinetic programme
selected for the ‘Diprifusor’ consists
of a three-compartment model with
modification for distribution to the
central compartment as described by
Marsh et al. This model tends to
produce a 16% greater blood concen-
tration than predicted by the system.
This improves predictive performance
of the system to provide acceptable
clinical anaesthesia.
67: 41–8.
3 White M. Evaluation of a computerised
propofol infusion system. Anesthesiology
1989; 71: A278.
Pulmonary mechanics and
volatile anaesthesia
The paper about the pulmonary
mechanics during isoflurane, sevoflura-
ne and desflurane anaesthesia (Dikmen
et al. Anaesthesia 2003; 58: 745–8) inves-
tigates an interesting topic namely the
bronchodilator effects of the different
inhalation anaesthetics. However, I have
some questions about the methodology,
results and conclusions of the authors?
The authors use the concept of
MAC, but in the methods we read
two important facts: First, they used
fentanyl. As we know, the MAC value
is defined for anaesthesia without mor-
phine derivates. The MAC with mor-
phine derivates is always lower than
the well-known and used MAC value.
On the other hand, the authors used
the term ‘equivalent doses… a con-
centration of approximately one
MAC’. What does approximately
means? Knowing that the pharmaco-
kinetics of the different inhalation
anaesthetics are different, can we prove
that effects were measured at a stable
steady state value of inspiratory fraction
(Fi) or expiratory fraction (Fe). What
does approximately in ‘steady state’
We think that for the casual TIVA
users like us this is an interesting
finding. We realise that anaesthetists
with a special interest in TIVA may be
already aware of this.
M. Vanarase
G. Biswas
Wycombe Hospital,
High Wycome HP11 2TT, UK
E-mail: hgargi@aol.com
J.-L. Demeere
Clinique Ge´ne´rale St Jean,
1000 Brussels, Belgium
References
Reference
1 Glen JB. Development of ‘Diprifusor’:
a TCI system for propofol. Anaesthesia
1998; 53 (Suppl. 1): 13–21.
1 Ebert TJ, Muzi M.Sympathetic hyper-
activity during desflurane anaesthesia in
healthy volunteers. A comparison with
isoflurane. Anesthesiology 1993; 79:
444–53
2 Marsh B. Pharmacokinetic model
driven infusion of propofol in children.
1248
Ó 2003 Blackwell Publishing Ltd

Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
Correspondence
A reply
constant). Also the increase in respirat- with the diagnosis and management of
Dr Demeere has pointed out some ory resistance after switching to 2 MAC ‘fetal distress’. The tool with which this
interesting points. First is the comment is statistically significant when compared diagnosis is often commonly made,
on MAC and fentanyl. It is widely to 1 MAC, which would indicate an namely the cardiotocograph (CTG), is
accepted that morphine derivatives may increase in bronchial tone. It is true that imprecise; therefore, it is not surprising
decrease the need for inhalation agents; it returns to baseline and it is not that the diagnosis would often be incor-
but it is impossible to find good evi- clinically significant. There would be rect. A normal outcome following
dence how these three agents are influ- only one explanation for this increase, fetal distress is a retrospective finding;
enced by the fentanyl dose used in our and that is bronchoconstriction. This therefore, it is understandable why obste-
study. The MAC values we used were may be explained by the pungency of tricians, midwives and obstetric anaes-
the concentrations in oxygen ⁄ air mix- desflurane, when you increase the con- thetists become ‘distressed’ on account of
ture. We believe that giving fentanyl centration, there may be an increased the fetus in the face of unsatisfactory
1 lg.kg)1 on induction should have a reaction to this agent.
minimal effect and should be in the same
When comparing the effects of dif- indicate fetal hypoxia, the severity of
proportion for all agents. The same ferent agents on different variables, sim- which may be impossible to estimate.
inhalation doses of isoflurane, sevoflu- ple mathematics can’t be applied. You
There are attempts by obstetricians to
CTG patterns given that is could possibly
rane and desflurane with fentanyl had either compare the individual values improve this situation by the presence
been used in other studies, without the obtained from every patient or you of senior obstetric cover on the Deliv-
need to discuss this issue [1]. The use of calculate the difference (i.e. percentage ery Suite as well as a new classification
MAC gives the opportunity to compare change) induced by the application (i.e. system for Caesarean section. Despite
the effects of different inhalation agents, increasing the concentration of agent) in these, the need to deliver babies at very
accepting that these concentrations are each patient. To my knowledge, it is not short notice will remain. What is
equipotent. The application of inhala- applicable to compare the difference in required is an agreed protocol for
tion doses for 5 min was considered means in this context.
sufficient to reach a steady state and to
We compared T3 with T2 and also
response in these situations.
There is evidence that the 30-minute
measure the effects of these agents on with T1 with the use of ANOVA for interval for delivery of a baby that may
bronchial tone, although longer periods repeated measures and Tukey tests. This be compromised is arbitrary and unsci-
such as 10 min were used in other is because we aimed to see that whether entific [1]. It is also illogical that this
studies for different end points [1]. As increasing doses of inhalation agents same time interval holds in both large
all agents are highly lipid soluble and causes an increased effect. We observed and small units where staffing issues are
have a potent action, and as they directly a continuing bronchodilating effect with clearly not the same.
interact with bronchial mucosa to exert 2 MAC sevoflurane, no change with
their effects, 5 min should suffice.
Another point is the effect of fentanyl resistance with desflurane.
Until we are better able to decide
isoflurane and an increase in respiratory whether fetus is compromised or not,
agreed protocols for rapid response
by all concerned, communication and
teamwork are the ways forward.
on bronchial tone and its duration of
Y. Dikmen
action. Fentanyl is a potent morphine
Medical School of Cerrahpasa
derivative with fast onset and interme-
34303 Istanbul, Turkey
U. I. Esen
diate duration of action (0.5–1 h) [2]. It
South Tyneside District Hospital,
South Shields NE34 0PL, UK
may cause rigidity of the muscles of the
thorax, but this effect may be inhibited
by the use of muscle relaxants. Even if it
References
1 Sponheim S, Skraastad O, Helseth E,
Due-Tonnesen B, Aamodt G, Breivik
had an effect on bronchial tone, this
Reference
H. Effects of 0.5 and 1.0 MAC isoflu- 1 James D. Caesarean Section for Fetal
would be the same for all patients as the
same dose was used in all. That’s why we
don’t consider the use of fentanyl as a
confounding factor.
When you compare the respiratory
resistance values at T1 and T3, namely
without inhalation anaesthetic and at 2
MAC inhalation, the values are quite
similar with no statistical difference. But
the decrease from base line at 1 MAC
rane, sevoflurane and desflurane on
intracranial and cerebral perfusion
pressures in children. Acta Anaesthesio-
logica Scandinavica 2003; 47: 932–8.
2 Barr J, Donner A.Optimal intra-
venous dosing strategies for sedatives
and analgesics in the intensive care unit.
Critical Care Clinics 1995; 11: 827–47.
Distress. British Medical Journal 2001;
322: 1316–17.
Syntocinon and ‘epidurals’ in
labour – which comes first?
Numerous studies have been published
that show an association between the
inhalation is significant and it would Fetal distress and the 30-minute use of regional analgesia and the need
for augmentation of labour [1–3]. We
have found that great emphasis is given
definitely indicate a decrease in bron- rule
chial tone (given the other variables
which effect respiratory resistance, such The Editorial (Yentis. Anaesthesia 2003; to this when non-anaesthetists discuss
as flow rate, tracheal tube size, are held 58: 732–3) highlighted the problems regional analgesia with women. How-
Ó 2003 Blackwell Publishing Ltd
1249

Correspondence Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
Table 1 Syntocinon requirements. Values are number (proportion).
retained products of conception but
finally ending in delivering a baby. We
would like to comment on one of the
statements made in the article about
obtaining consent. The authors men-
tioned that the patient’s mother could not
be traced for consent. This implies the
team had alternatively sought to seek
consent from the mother, which is not in-
line with the General Medical Council’s
(GMC) advice statement. The GMC
document ‘Seeking patients’ consent:
the ethical considerations’ [1] states that:
‘At age 16 a young person can be
treated as an adult and can be presumed
to have capacity to decide’.
Total
Spontaneous
Induced
n = 500
labour n = 326
labour n = 174
No syntocinon
Syntocinon before analgesia
Syntocinon after analgesia
198 (40%)
187 (37%)
115 (23%)
167 (51%)
70 (22%)
89 (27%)
31 (18%)
117 (67%)
26 (15%)
ever, little mention is made of work that ever, in the majority of cases (62%), this
suggests that this association may not be was either planned or already in progress
causal [2,3]. We had the impression that before the regional analgesia was com-
regional analgesia is often requested menced. Therefore, regional analgesia
when contractions become painful as a cannot be held solely responsible for high
result of syntocinon- being started, or the rates of augmentation of labour. A policy
woman is advised to ‘have an epidural for active management of labour is likely
before the drip which makes the con- to be associated with a higher demand for
No one may make decisions on
behalf of a competent adult.
tractions painful’.
epidurals [2].
We carried out a simple audit to
determine which came first, augmenta-
tion (or the perceived need for aug-
mentation) of labour or the use of
regional analgesia. We recorded whe-
ther a syntocinon- infusion was already
in progress, or had been planned, at the
time of requesting regional analgesia. At
the postnatal follow-up, we noted if a
syntocinon-infusion had been used prior
to delivery. Data were collected on 500
consecutive women requesting regional
analgesia. Of these, 61% were nulli-
parous and 65% had a spontaneous onset
of labour. Median gestation was
40 weeks (IQR 39–41 weeks) and the
median cervical dilatation at the time of
request was 3 cm (IQR 3–5 cm).
Of the women who received regional
analgesia, 302 (60%) also had intravenous
syntocinon to augment labour. Of these,
however, 187 (62%) were already receiv-
ing syntocinon or its use had been
planned by the time regional analgesia
was requested. In total, 115 women
(23%) required augmentation of labour
following regional analgesia (Table 1).
Of the 326 women in spontaneous
labour, 89 (27%) required augmentation
following regional analgesia (Table 1).
This compares with an overall augmen-
tation rate of 22% for spontaneous
labours in our unit over the same time
period. A survey of intrapartum care for
low risk primigravidas in the UK [4]
found that 28% of spontaneous labours
(range 13–46%) were augmented with
syntocinon. In conclusion, in our unit,
60% of women with regional analgesia
required augmentation of labour. How-
The patient mentioned in the article
was 16-year-old and according to the
GMC statement [1], can be treated like
an adult and no one can make decisions
on behalf of a competent adult.
N. L. Lewis
F. Plaat
A. M. Qureshi
Queen Charlotte’s and Chelsea
Hospital, London W12 0HS, UK
E-mail: FPlaat@hhnt.org
Under general anaesthesia, if a proce-
dure other than that for which the
patient has consented needs to be per-
formed, it may be done only in situa-
tions where the treatment is intended to
save life or avoid significant deterior-
ation in the patient’s health [1]. My
question to the author is whether they
would have done the Caesarean section
if they were able to get the parental
consent. We think it would have been
inappropriate to perform Caesarean sec-
tion on this patient even after parental
consent for the following reasons:
The clinical diagnosis was incorrect.
Proceeding to performing a Caesarean
section, even with parental consent, is
inappropriate since the patient was
not informed of the misdiagnosis and
offered an option.
Medical consent to perform Caesar-
ean section in this instance would be
inappropriate because that was not the
only option available for the patient, the
other being normal delivery as it had
happened.
The GMC ‘best interest’ principle
also states that in a patient who lacks
capacity to decide, one may take views
of the patient from a third party (like a
parent). However, one should also
consider the options that least restricts
the patient’s future choices where more
than one option (including non-treat-
References
1 Howell CJ. Epidural versus non-epidu-
ral analgesia for pain relief in labour. In:
The Cochrane Library 2003: (1).
2 Mayberry LJ, De Clemmens DA. Epi-
dural analgesia side effects, co-inter-
ventions, and care of women during
childbirth: a systematic review. Ameri-
can Journal of Obstetrics and Gynecology
2002; 186: S81–S93.
3 Dickinson JE, Godfrey M, Evans SF,
Newnham JP. Factors influencing the
selection of analgesia in spontaneously
labouring nulliparous women at term.
Australian and New Zealand Journal of
Obstetrics and Gynaecology 1997; 37:
289–93.
4 Williams FLR, Florey C du V, Ogston
SA, Patel NB, Howie PW, Tindall VR.
UK study of intrapartum care for low
risk primigravidas. a survey of inter-
ventions. Journal of Epidemiology and
Community Health 1998; 52: 494–500.
Ethical considerations
in obtaining consent under
anaesthesia
We read with interest the report (Kuehn
et al. Anaesthesia 2003; 58: 725–6), des-
cribing the management of a woman who
was actually planned for evacuation of
1250
Ó 2003 Blackwell Publishing Ltd

Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
Correspondence
ment) seems reasonable in the patient’s may be due to unrecognised air leak failed epidural following a positive
best interests. Hence, it would have secondary to poor connection of the Whoosh Test ) 2.
been inappropriate to do a Caesarean syringe to the hub of the needle, leak
R. M. Khan
section anyway.
between the barrel and the piston, and at
times leakage between the track made
by the larger introducer and the epidural
needle. Since the LOR syringe is usually
attached after the epidural needle has
been introduced to a depth of 2–3 cm,
any false positive LOR raises the appre-
hension of proceeding any further. To
overcome this dilemma, when suspicion
of false positive LOR exists, we perform
the Whoosh Test-2 originally suggested
for identifying the caudal epidural space
[2]. We now follow a protocol wherein
an injection of 1 ml saline with 0.5 ml
air is made through the epidural nee-
dle. An assistant auscultates for the
Whoosh – 2 sound two dermatomes
above the injection site. To elicit an
audible Whoosh – 2 sound, this injec-
tion of saline and air should be made
rapidly as the Tuohy needle is of a wide
bore and its spread is in either direction
unlike the small gauge needle used for
caudal injection where the spread is
J. Chabra
M. Ranganathan
M. T. Alam
G. Raghuraman
M. Ashraf
Birmingham Heartlands Hospital,
Birmingham B9 5SS, UK
E-mail: Raghulatha@aol.com
D. Jain
J.N. Medical College, A.M.U.,
Aligarh – 202002 [U.P.], India
E-mail: seeras_alig@rediffmail.com
Reference
1 General Medical Council Seeking
patients’ consent the ethical considera-
tions. London: 1998, GMC.
References
1 Atkenhead AR, Rowbotham DJ,
Smith G. Local anaesthetic techniques.
In: Textbook of Anaesthesia, 4th edn.
London: Churchill Livingstone 2001,
555–75.
A reply
Thank you for the opportunity to res-
pond to the comments made by Drs Ran-
ganathan and Raghuraman onourreport.
The consent for Caesarean section
from the mother was sought in the
consultant gynaecologist presence’s. This
was thought to be in the best interest of
patient and her baby under the circum-
stances (she was fully dilated and about to
go into the second stage of labour). With
hindsight and after a normal delivery, we
agree it might not have been the best
possible option for the patient.
2 Puddy BR The ‘Whoosh’ test.
Anaesthesia 1999; 54: 614.
Diagnostic tattoo in a
parturient with ‘ecstasy’ use
A 17-year-old woman at 38.5 weeks
gestation was in labour and requested
analgesia. When interviewed by the
anaesthetist, she denied any medical
and ⁄ or social problems (including sub-
stance abuse). However, preanaesthetic
physical examination of her lower
back revealed an unexpected ‘diagnostic’
tattoo in the sacral area (Fig. 7). Upon
predominantly upwards.
A negative
Whoosh – 2 sound clearly identifies
the false positive LOR and the operator
either proceeds afresh in an adjacent
epidural space and ⁄ or with a new LOR
syringe. To date, we have never had a
As regards to the misdiagnosis, another
fact came to light following the delivery.
For social reasons, she had deliberately
misled the medical staff by concealing the
date of her last menstrual period.
All we now can say is that perhaps
the inability to find the mother peri-
operatively was a blessing in disguise.
H. N. Saxena
B. Kuehn
G. Menon
Mayday Hospital,
Croyden, UK
E-mail: harshnsaxena@hotmail.com
Whoosh test 2 and confirmation
of lumbar epidural space
Loss-of-resistance [LOR] to injection of
air or saline as the tip of the epidural
needle penetrates the ligamentum fla-
vum is a popular means of identifying
the epidural space [1]. Operators use
either a glass syringe or a disposable
plastic syringe to elicit LOR. We have
observed that the LOR syringe, especi-
ally a glass syringe, occasionally elicits a
false positive LOR to air injection. This Figure 7 Lower back ‘Ecstasy’ (misspelled) tattoo encountered in our patient
Ó 2003 Blackwell Publishing Ltd
1251

Correspondence Anaesthesia, 2003, 58, pages 1235–1252
.....................................................................................................................................................................................................................
further questioning, the patient admitted unprotected sexual intercourse, and
that she had her ‘Ecstasy’ tattoo inscribed domestic violence [2].
References
1 Stephens MB. Behavioral risks associ-
ated with tattooing. Family Medicine
2003; 35: 52–4.
a few weeks after engaging into recre-
ational 3,4-methylenedioxymetham- the presence of a tattoo noted during
phetamine (MDMA, i.e. ecstasy) use. physical examination of young
In summary, in the clinical setting,
a
2 Roberts TA, Ryan SA. Tattooing and
high-risk behaviour in adolescents.
Pediatrics 2002; 110: 1058–63.
Tattoos are an increasingly prevalent parturient can alert obstetricians and
form of self-expression, especially for anaesthetist to the possibility of other
adolescents and young adults [1]. risk-taking behaviours.
Roberts and Ryan studied the beha-
K. M. Kuczkowski
vioural risk factors associated with tat-
UCSD Medical Center,
tooing in adolescents and young adults
San Diego, CA 92103–8770, USA
and concluded that permanent tattoos
E-mail: kkuczkowski@ucsd.edu
are strongly associated with high-risk
behaviours including substance use,
Announcement
The journal would like to inform its readers that the ‘Deaths of members reported to Council October 2002 to September
2003’ will appear in the published January 2004 issue.
1252
Ó 2003 Blackwell Publishing Ltd