Bad medicine: paediatric ear, nose, and throat surgery
BMJ 2010; 341 doi: https://doi.org/10.1136/bmj.c6560 (Published 17 November 2010) Cite this as: BMJ 2010;341:c6560
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
I wasn't going to post anymore more responses but this piece has had
more comebacks than Status Quo. I do regret that the pieces title suggests
I am having a go at all of Paediatric ENT surgery. I toyed with other
titles but they were too long. I do appreciate that there is much more to
the speciality than Grommets and Tonsillectomy ! I have had the privilege
of having patients with Cochlear implants whos lives have been
transformed.
I also regret the suggestion that the assessment of OM with effusion
was wholly subjective. It is in primary care but clearly not during the
period of assessment in ENT. This was poorly put.
On balance I still believe that Grommets and tonsillectomy are not
necessary.
As for my photos . I am weather beaten from all this years storms of
outrage caused by the "Bad Medicine" pieces. These are not personal but
naturally people take them personally. Perhaps I might be proved right or
perhaps wrong , only time will tell. And as promised and in spirit of
fairness first one next year is "Bad Medicine General Practice" and I am
currently working on the referencing.
Lastly , re my typos and spelling mistakes from undiagnosed Glue
ear and subsequent Dyslexia. I am too old to have a Dyslexia label and my
audiology was normal . The teachers thought me stupid and my wife thinks
me lazy . Either way, as much as I try I can't help them.
Competing interests: No competing interests
Oh dear. I've just returned from a couple of weeks away from my own
front doormat to find this article, already 3 weeks old, and a simmering
on-line row...
I should first declare at least one potential conflict of interest; I am
Editor, with Peter Bull and Glenis Scadding, of a text book, published in
2007 by Springer called 'Pediatric ENT'. (Don't blame us for the
transatlantic spelling, the publisher insisted).
It does give one quite a shock to come back innocently from holiday
and find that the product of one's main occupation in the years 2005 to
2007 is now declared a non-subject, and even 'Bad Medicine' in such a
respected journal as the BMJ! Having declared a conflict of interest I
assume it may perhaps be ethical to suggest Dr Spence might consider
buying a copy of this book in order to establish in his own mind the scope
of the clinical and academic field he has so roundly and inclusively
condemned in public. Sadly the soft cover version is currently reprinting,
but I'm sure the BMJ would get him a discount on the hard cover one.
Checking through the 47 chapter headings of our book, and the
contributions from the 65 authors, I confess that chapters 14 and 42 do
indeed deal with glue ear and tonsils and adenoids. However since Dr
Spence has insisted, in his responses on the 'rapid response' section of
the BMJ on line, that every word of his original article is correct, we
can assume this applies to the headline: 'Bad Medicine: Paediatric ENT
surgery' and the unequivocal remarks that 'Paediatric ear, nose, and
throat surgery is an example of the power of a medical belief system over
common sense and the evidence of harm. GPs should stop referring patients
and surgeons should stop doing the operations, for this is bad medicine'.
Apart from his evident allergy to glue ear and tonsils and adenoids,
subjects very ably covered by your previous correspondents, presenting
what I thought were a balanced selection of references, where does this
leave the handling of acquired and congenital tracheal stenosis? It looks
as if Dr Spence clearly and strongly believes that these children should
be left with life long intubation or tracheostomy. What about head and
neck masses and salivary tumours in children? Cochlear implants for the
profoundly deaf? Laryngeal webs? Tracheomalacia? Laryngeal papillomas?
Thyroid disease? Choanal atresia? Congenital atresia of the ear? Chronic
otitis media, with and without cholesteatoma (generally the result of the
same failure of middle ear ventilation of which glue ear is the first
stage)? All, it seems, bound for the dustbin!
How should we react? As I write the bars round here are shut and
we've run out of gin. I suppose that one could try applying to General
Practice Dr Spence's elegant strategy of binning the whole of a specialty
on the basis of his own dislike of parts of it. How would this look on
paper? Well, since anecdotes are in vogue, I know two people whose
symptoms of large bowel cancer was dismissed by separate GPs as wind,
piles or attention seeking and one person whose chest secondaries from
already treated breast cancer were attributed to a viral infection
('there's a lot of it about at the moment, dear'). No doubt we all know of
comparable cases, perhaps in our own extended families. From this, and in
the present climate - financial, not global warming - one might infer that
GPs are acting mainly as a barrier to effective medical management
(practicing really Bad Medicine in these cases) and so are doing their
best to save tax-payers' hard-earned cash, since palliative care is
clearly cheaper than long term aggressive treatment and, with the
persistence in the NHS of the 'Market', that untested and now in many ways
discredited Thatcherite (Thatcherwrong?) concept, Money Speaks. Why not,
therefore, get rid of our GP system and allow patients to go direct to
specialists, or to their local pharmacy? It would save the NHS a fortune
and allow us to continue funding the legions of people who administer the
Market by 'commissioning' and so forth.... Logical or wot?
'Are ye lookin at me Jimmie? Or is it yer amblyopic ee, that one the quack
didna recognise when ye was a wee bairn?'
Ho hum. Better get on to our publishers now and tell them not to
worry about that second print run.
As an afterthought and intended with absolutely no disrespect, I did
notice ten misprints in Dr Spence's various letters of reply on this
website. I know he's a busy man, as he doesn't have any more time to deal
with this correspondence, but the thought did occur to me that someone who
had such a traumatic episode during an attack of AOM, and who may most
reasonably have refused point blank to go near that hospital again, might
have had prolonged, undiagnosed glue ear afterwards and thus suffered a
form of acquired dyslexia because of this.... Also, in passing, if you
compare the youthful photo of Dr Spence that you can see next to his
weekly column with the rather more careworn one at the head of this
website you might concede that this whole episode may have subjected the
poor chap to a great deal of stress.
John Graham
Competing interests: Many: 1. Co-editor of 'Pediatric ENT'. 2. 1979-2005 member of multidisciplinary children's speech and hearing clinic (dealing mainly with non-middle class families living in Somers Town, Tower Hamlets and North and East Islington, with referrals almost entirely from Community doctors); on call permanently for acute laryngeal and hearing problems at UCH regional neonatal unit. 3. Editor of journal:Cochlear Implants International: more than 50% of content deals with paediatric ENT surgery in the form of cochlear implants. 4. Founding president of the British Association for Paediatric Otorhinolaryngology, past president and current general secretary of the European Society of Pediatric Otorhinolaryngology
In the section ' Views and Reviews' the BMJ has published another
badly written paper. Some of the errors and considerations are given
below.
The author is making the classical error of applying today's
standards of care retrospectively. It is the same as complaining that over
30 years ago the motor car was neither as reliable nor as economical as
the modern motor car. Over 30 years ago, when the author had his ear
suctioned, there was neither a Cochrane review nor SIGN guidelines and
doctors did what was the standard of care at the time. Unfortunately, the
author makes other errors too. His lecturing on the rates of tonsillectomy
and grommets shows how little he knows about past rates of the procedures
and current ones and how out of date he is. The rates have been going down
for some time (1) and this to large extent is due to guidelines issued by
various paediatric, ENT, GPs societies (2), NICE (3) and the work by
doctors to implement them. It would have helped had Dr Spence checked that
the even the journal he writes for does not recommend tonsillectomy
lightly and this was as far back as 1981 (3). But why should one allow
facts to get in the way of a 'good story', or non-story as it turns out?
The article is factually wrong. Dr Spence claims that 'the clinical
diagnosis is wholly subjective'. This is not so, hearing test and
tympanometry being mandatory (2,3). The author states 'tonsillectomy
doesn't prevent ...quinsy'(www.sing.ac.uk/pdf/sign117.pdf). This is
nonsense. The guideline says nothing of the sort, nor can it. One does not
need even to read the guideline or to be a doctor in order to know that .
Anyone armed with a dictionary and a modicum of common sense would be able
to reason that, as 'quinsy' is the lay term for peritonsilar abscess, it
would follow that if there were no tonsils there would be no quinsy.
I would advise latter-day Rip van Winkles to sniff the air before
issuing rallying cries for causes that have been fought and won long ago.
Otherwise they risk looking hopelessly out of date and attracting
ridicule. The plural is used here advisedly because it includes not just
the author but also the editors, who would have read the piece and
approved it for publishing.
This piece says nothing new, or original or witty. The author's
experience as a child is irrelevant to modern day ENT practice. Some of
his statements are unsupported by the evidence and some are factually
wrong. The question is: why has the BMJ published out of date and
factually wrong rubbish when the Daily Mail and Pulse do this much better?
Suggested answer: because the BMJ is not very much read, not even by its
editors. Suggested remedy: 1. split the BMJ into BMJ medical and BMJ
banality, claptrap and infotainment, and let doctors choose which
section(s) they want to subscribe to 2. Tighten up the editorial process
1. Bisset AF. Glue ear surgery in Scottish children 1990-1994: still
plenty of ENT and public health challanges. Clin Otolaryngol Allied Sci
1997; 22:233-238
2. American Academy of Family Physicians, American Academy of
Otolaryngology - Head and Neck Surgery, and American Academy of Pediatrics
Subcommittee on Otitis Media with Effusion. Pediatrics 2004;113:1412-1429
3. NICE. National Collaborating Centre for Women's and Children's
Health. Surgical management of otitis media with effusion in children.
Clinical Guideline 2008
3. Ludman H. ABC of ENT. BMJ 1981; 282:628-631
Competing interests: No competing interests
Dr Spence asks, "But what is tonsillitis?" I would ask, "What is
recurrent tonsillitis?" Given that by no means all episodes of sore throat
are brought to the doctor's surgery, it is difficult to know the incidence
in the population, but let's assume that the average five year old gets
two episodes in a year. By statistical chance alone, some children will
have no sore throat in a given year; some will have half a dozen episodes.
If the unfortunate child who has had six episodes undergoes tonsillectomy,
and if by chance (as is overwhealmingly likely) he or she has fewer
episodes in the following year, the operation will be declared a success.
I believe that Dr Spence is right to question tonsillectomy for sore
throats, because far from being due to diseased tonsils, in reality
recurrent sore throats may be explained by statistical chance alone and
nothing more.
Competing interests: No competing interests
Dear Sir,
Thank you for your reply. I have taken your invitation to review the
references in the Golz paper re :cholesteatoma. I think as per all
scientific review we can pick and choose the papers that best fit our
clinical views. Padgam et al (JLO 1984) in another 20 year (1966-1986)
series (this time from Dundee)reported that despite a sixty fold increase
in grommet insertion their mastoidectomy rate varied between 0.9-
1.88/10,000 population. The mean incidence of cholesteatoma was
1.32/10,000, leading to their conclusion that despite a massive increase
in grommet insertion there was no rise in cholesteatoma.
Either way and whatever our viewpoints my honest belief is that we
should present our patients with balanced information and a range of
options. There is "no one size fits all" choice for glue ear or
tonsillitis.
Finally for all the academic debate that has preceeded, not at one
point throughout this entire stream has anyone suggested that the
operations discussed are without risk for this would be disingenious and
wrong. Therefore whatever your clinical views, on your last point I
contend , Dr Spence , you are factually incorrect.
Yours Sincerely,
Samit Ghosh
Competing interests: I am a surgeon with a paediatric practice in a DGH. I hopefully practice evidence based and pragmatic medicine (if these two approaches can co-exist!)
I would genuinely like to thank you all for all your responses. I
spent this week responding as I can. I have read the various guidelines
and maintain that my article and responses are factually correct. Whether
you accept my interpretation is another question.
I will need to sign off on this one currently as I have other work.
A last thought. In the guidelines there is little by way of the
complication rate and severity. This is important as both the conditions
are self limiting.
In respect to cholesteatoma. There are other papers and I would
suggest looking at the reference section of the paper given. I understand
the population risk is about 0.1% therefore having had surgery for a
transient and non life threatening condition the associated rate is 10
time higher than this. Sometimes association is causation. Also what about
the risk of chronic supportive otitis media ?
To suggest these operations are without risk is not true.
Thanks again and hope that the BMJ will publish a fuller response to
allow a proper right of reply.
Competing interests: No competing interests
Would I ever recommend Tonsillectomy?
Most sore throat are caused by viruses not bacteria, tonsillitis is
caused by bacteria. But most pressure for tonsillectomy comes from parents
who believe in tonsillectomy will cure sore throat, often having had this
operation themselves. Also attending a doctor does not depend on severity
of illness but in the main health seeking behavior. For every one sore
throat we see in the surgery there are nearly 20 we don't see. Therefore
the reporting of sore throat is defunct and in truth bacteria tonsillitis
infection must be relatively rare.
So with sore throat I try not to prescribe antibiotics as this only
reinforces illness behavior. I encourage the parents to send their
children to school with simple analgesia - as missing school is the main
parental concern. Remember kids not going to school has little to do with
the severity of the illness and much to do with parental anxiety , so
insisting they go to school is fundamental. With this policy I can't ever
remember making a referral to ENT for tonsillectomy. Clearly if parents
want a referral , of course I would do this. In my medical opinion
tonsillectomy carries risks unacceptable for a self limiting condition
like sore throat .
Lastly , I find the reasoning and the evidence behind tonsillectomy
for recurrent sore throat conflicting and counter intuitive if most
infections are viral. This based on anatomy, bacteriology, virology and
epidemiology. It just couldn't work.
The issue of obstructive sleep apnea is different and I would
certainly refer.
Hope this helps.
Competing interests: No competing interests
Dear Sir,
May I congratulate Dr Spence on inviting superb debate on a current
topic and in doing so recieving structured replies from eminent authors
within a one week deadline! Surely a commissioning editors dream. I will
be recommending this stream to all ENT registrars preparing for the
upcoming FRCS!
Dr Spence is surprised that there are no ENT surgeons who agree with
him, I have some sympathy with his views, however it appears that he is
writing from a period in time when anecdotally the indication for a
tonsillectomy was " the presence of tonsils" . These days are long gone.
The current evidence base , as has been said is not arbitrary and based on
significant trial data (Paradise 1984, NESTACC 2010). Paradise reported In
the first year after randomisation, there was a 2.5-fold reduction among
surgical subjects in the mean number of throat infection episodes overall
(47 episodes in 38 surgical subjects versus 108 episodes in 35 control
subjects), but a 4-fold reduction in the proportion of surgical subjects
who had three or more episodes of any type (13% [5 of 38] versus 54% [19
of 35]) and a 14-fold reduction in the mean number of episodes rated
clinically as moderate or severe (3 episodes in 38 surgical subjects
versus 41 episodes in 35 control subjects). I agree that surgery cannot be
blinded for, but using this argument damns all trials researching quality
of life conditions. To claim a trial is ineffective with a generalised
criticism of "biased trial methodology" is too simplistic.
To present a slightly less conservative view , Paradise, (Pediatrics
Vol 110:1; July 2002,7-15) further reported a three way trial of 328
children randomised to adenotonsillectomy, tonsillectomy and a non
surgical control group. It found that using a less stringent criteria of 3
episodes of infection in one year, there was a modest reduction in
infection episodes in the surgical group compared to the conservative
group but could not recommend surgical intervention in these modestly
affected children. Therefore with regards to tonsillectomy in children
(for infection) the effectiveness is proportionate to the stringency of
indications.
Dr Spence further refutes grommets due to the risk of
cholesteatoma.The study quoted was a retrospective analysis of 2829
children(5575 ears) from 1978-1997. the follow up was anything was 1 -20
years, with no data stating time from grommet to cholesteatoma. The
decision that choleateatoma was due to the grommet was if the authors felt
that the cholesteatoma was near the site of insertion or behind an intact
ear drum. Unless these cholesteatomas were localised single pearls ( not
documented) this is a very difficult call. There is no control group in
this study ie how many cholesteatomas presented in 2829 children who never
had grommets? I agree with Dr Spence that there is "association" but let
us not confuse this with cause.
Finally and potentially more pertinantly , given Dr Spence's
admission that his referrals are wholly subjective,and he was directing
this to (our)primary care colleagues, is this a peek into the post "white
paper" future of GP commissioning!!
Yours sincerely
Samit Ghosh
Competing interests: No competing interests
Madam
As an ENT surgeon, I have been following the stream of responses to
Dr Spence's article with equal measures of outrage and bewilderment. On
the one hand, the ENT respondents have eloquently displayed the clinical
and scientific evidence. On the other hand, Dr Spence has elegantly dodged
the scientific debate with the ultimate suggestion that all surgical
trials are fundamentally biased as operations can not be blinded among
other extraordinary statements.
Ultimately, I do not believe there will be an end to the scientific
debate in this section as scientifically, it is a "no brainer".
Furthermore, I can not articulate the arguments any more clearly than they
already have been. However, the last letter opens perhaps a much greater
concern altogether on GP commissioning.
For patients not to be denied appropriate treatment or even the
opportunity to discuss such treatment options with a specialist due to the
prejudice of their GP, should GPs perhaps publish a list of conditions on
their website/bulletin board which they consider to be "aberrant health
seeking behaviour" ? Perhaps such an announcement could be accompanied by
a disclaimer saying that "This doctor believes that evidence based
medicine is widely flawed and the results of trials of the efficacy of
operations is fundamentally biased."
Patients should have the information due to them before choosing
their GP and therefore gate-keeper (sorry, commissioner) for any further
specialist opinion.
Michael Kuo
Competing interests: I am a Children's ENT Surgeon
Otitis media with effusion: aiming the right therapy at the right patient
J.A.deRu@umcutrecht.nl
Considering the 'storms of outrage' caused by the article: 'pediatric
ENT, bad medicine', it seems that the discussion on the preferred therapy
for otitis media with effusion (OME) is far from closed. It was suggested
before that OME could function in defence of the body during the first
childhood years and that usually no treatment is needed. Promoting this
point of view can be helpful in the current state-of-war.
Our first statement is that OME is not a singular 'disease', but
merely a symptom. The second suggestion is that resolving the glue itself
is not a prerequisite, yet the cause for the necessity of glue formation
should be taken care of.
We think that the glue is formed in order to protect the middle ear
from further damage when other components of the human defence system have
failed. Various levels of resistance could be implicated, such as
inadequate antibody formation, craniofacial deformities, or Kartagener
syndrome, but it can also develop during a more common (viral) upper
respiratory tract infection. All of these are completely different causes
that consequentially would require different treatment options. The need
for this symptom to develop might theoretically be enhanced due to other
local factors that cause damage to the mucosa such as allergy and reflux.
Ventilation tubes improve the annoying hearing loss and can be
considered in patients with a definite need of improving it. In the past,
groups of patients that might benefit from grommets have been
'identified'. These mostly consist of patients with language and learning
problems. We do agree that a short-term benefit, theoretically, can be
desirable. However, even in these special groups only a marginal benefit
has been established.
Furthermore, over time grommets cause more hearing loss than an
observational policy, and although most eardrum pathology associated with
OME is selflimiting, myringosclerosis which is associated with the
insertion of grommets, shows the least tendency towards recovery. And, we
must not forget that grommet insertion is associated with the development
of otorrhoea. So, perhaps hearing aids might be preferable.
It was postulated that a long-term therapy of antibiotics might be
appropriate for specific patient populations. However, the benefits appear
to be limited, leaving the question whether the slight advantage outweighs
the risk of developing resistance to antibiotics? Furthermore, prescribing
antibiotics leads to more visits to the doctor's office. In light of the
above, we propose a significant decrease in antibiotic therapy, especially
because for most children 'watchful waiting' appears to be justified.
Bio-films are mentioned to explain why active bacteria are found in a
middle ear secretion that otherwise yields a negative culture. However,
the existence of a bio-film isn't of much significance in itself,
especially not, when we consider that this bio-film is not capable of
causing serious infections. This may reverse our thinking: it could be
that the body is capable of building such a strong defence, that bacteria
can only survive in a 'bio-film'.
Recently, it was mentioned that Alloiococcus otitidis might play a part in
OME. Yet, this in no way diminishes the conclusion that it is not possible
to establish any clinically important infection caused by these bacteria.
Reflux may be a contributing factor, based on the presence of bile
salts in the middle ear. This finding does lend more credence to the
premise that once children start spending more time in an upright
position, OME is self-limiting. One wonders whether treatment with acid
inhibitors is an option. Positive results with this kind of treatment are
scarce.
What do we suggest as treatment? There might be valid reasons to take
action against OME, see also the NICE-guidelines on this subject. If so,
adenoidectomy is in many cases probably the best form of treatment to
start with. This will reduce the number of infections as well as its
source.
Corticosteroids in nasal spray were successfully administered in one
study, however these results could not be duplicated in others. This could
be due to differences in the inclusion criteria. Patients in the study
that reported successful treatment were scheduled for grommets and
adenoidectomy, which suggests that these patients had more rhinological
(allergic?) symptoms and were more likely to benefit from said treatment.
The patients in the study with a negative result were a small subgroup
with non-resolving, bilateral OME.
Including every patient with OME in every study does not make sense.
Sub-groups possibly benefiting from the different treatment modalities
need to be defined.
In a recent study a nasal spray with Streptococcus sanguinis
constitutes an interesting new therapy. This therapy offers a boost to the
immune system. Its effectiveness needs further confirmation, but we like
the thought of improving the immune system.
In conclusion, the current surgical or medical therapy of OME often
is a form of 'over' treatment. When confronted with OME in children the
focus, more than ever, is a matter of aiming the right therapy at each
individual child.
Competing interests: No competing interests