JAMA  3 Dec 2008  Vol 300

2497    Here’s an interesting and methodologically rigorous decision analysis paper comparing outcomes from continued medical treatment versus surgery for pharmacoresistant temporal lobe epilepsy. “Pharmacoresistant” is a coinage I haven’t encountered before, but I think I know what it means: or do I? I had a patient identical to the one in this decision analysis: he had TLE from birth and he was getting full seizures every week with lots of partial ones between, despite all the drugs the National Centre for Epilepsy could think of. In his mid-thirties, he was on the verge of having surgery when a new agent appeared: levetiracetam (Keppra). Once he was up to full dosage, his fits more or less disappeared. He went from being highly pharmacoresistant to being highly pharmacoresponsive. And since this analysis was done mainly on the basis of studies carried out before levetiracetam appeared, I don’t think we can believe its conclusions. “Pharmacoresistant” now means something different from pharmacoresistant when its data were collected, let alone what it will mean in the future when more new drugs will have appeared. 
http://jama.ama-assn.org/cgi/content/abstract/300/21/2497

2506    No reasonable person could doubt that setting up something called a Rapid Response Team would improve outcomes in critical hospital emergencies, especially in Kansas City, where a sheriff’s posse of the fastest shooters was selected and given swift steeds to ride to the rescue the moment an outlaw or Injun appeared. The result? No reduction in hospital-wide code rates (crash calls) or mortality. Hand over that badge, Mr. Sheriff. 
http://jama.ama-assn.org/cgi/content/abstract/300/21/2506

2514   When I started in UK general practice thirty years ago, most drugs were prescribed by brand name, and prescribing generically made people suspect you of  being left-wing or academic (often both). It wasn’t popular with patients or with receptionists who tried to explain that the new white ones were the same as the old blue, even though they smelt different. What was a young chap of left-wing and academic leanings to do? I persevered heroically and now the rest of the world has caught up with me, though World Revolution is still some way off. Of course, wicked capitalist drug companies have done their best to spread rumours that generic drugs are not as good as the branded equivalent, but this systematic review shows that they are.
http://jama.ama-assn.org/cgi/content/abstract/300/21/2514

NEJM  4 Dec 2008  Vol 359

2417   I hope some mathematically inclined reader will write in to let me know the number of possible hypertension treatments based on currently available agents if the patient is on (a) two drugs or (b) three. Let’s assert for now that it’s an awful lot, and that it doesn’t much matter what they are so long as they reduce BP and they don’t harm the patient. One drug I bet you never choose is benazepril, a late-comer in the ACE inhibitor market which is full of excellent and effective generics. To ACCOMPLISH some good publicity, its manufacturers ran this trial in 548 centres around the world, combining it with amlodipine or hydrochlorthiazide in 11,506 high-risk hypertensive patients, counting events for a mean of three years. If you’d asked me to take bets on the basis of known data and basic physiology, I’d have predicted that the thiazide combination would have come out better: but in fact the amlodipine/ACEi combination was clearly superior. Another few thousand similar trials, and we might know which combination of all the drugs available is the best. So, mathematicians, how many patient-years of study would that involve?

http://content.nejm.org/cgi/content/abstract/359/23/2417

2442    Comparing drug treatments for hepatitis B is a piece of cake by comparison, except that you have to remember the names tenofovir disoproxil fumarate and adefovir dipivoxil. Randomise patients to one or the other and the compare their HBV DNA levels and liver histology at 48 months: tenefovir wins.

Tenofovir, adefovir –

there’ll soon be others too, I fear:

with disoproxil, dipivoxil,

maybe even smellysoxil.

http://content.nejm.org/cgi/content/abstract/359/23/2442

2456    When I started getting interested in heart failure from the perspective of primary care, I was shocked by echocardiography studies that seemed to show that many of the patients we were treating had a normal ejection fraction. The obvious conclusion – widely publicised – was that GPs were lousy at diagnosing heart failure. It has taken more than a decade for British cardiologists to concede what is patently obvious to their European and American colleagues - that heart failure with a preserved ejection fraction is common and important. The problem is that all the major heart failure trials of the 1980s did their selection on the basis of impaired EF, so we don’t know how to treat these people – mainly old ladies with high levels of co-morbidity. In this large trial, irbesartan had no benefit. But this may have been an unfortunate choice of drug class, since even in systolic heart failure, angiotensin II receptor blockers have had mixed success.  http://content.nejm.org/cgi/content/abstract/359/23/2456

2468    The mainstay of treatment for Parkinson’s disease for the last fifty years has been levodopa – a simple and effective drug, but a difficult one to use in daily practice once the on/off effects and dyskinesias have started to cut in (at 5 years from commencement of treatment on average). A jack-of-all-trades GP should know how to use this drug, and this is an excellent guide; but in practice it is a huge advantage to have the services of a specialist Parkinson’s nurse to help patients and their carers through the fine-tuning of drug dosing, which may need to be a often as every two hours.  http://content.nejm.org/cgi/content/extract/359/23/2468

Lancet  6 Dec 2008  Vol 372

1938   Traditional Western medicine (up to about 1850) was a collection of time-hallowed nonsense based on wildly mistaken theories of disease and usually did more harm than good. Is traditional Chinese medicine any better? Although written with great politeness by three Chinese authors, this editorial essentially gives the answer no. It is just a load of old Yin-Yang and bear’s spleen.

1962   From time to time, I’ve mentioned the issue of perioperative β-blockade for patients having non-cardiac surgery, in particular the POISE trial which showed an increase in overall mortality, surprisingly mostly from stroke. Nonetheless, American guidelines apparently continue to recommend perioperative beta-blockers for patients with evidence of cardiovascular disease. This meta-analysis shows that too many poor-quality trials with high levels of potential bias led to the adoption of this practice: pooling reliable data shows no evidence of benefit and a slight tendency to harm.  http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(08)61560-3/abstract

BMJ  6 Dec 2008  Vol 337

1328    This is a 30 year study of outcomes following admission for suicide attempt in Sweden: it is followed by a study of self harm following discharge from British inpatient psychiatric care. Trying to compare the two is a maddening exercise, likely to result in self harm. The Swedish data stretch back to the time when Sweden was the suicide capital of the world, and the biggest risk for completed suicide that emerges is psychosis and schizophrenia, followed by “bipolar and unipolar disorder”. The British study, on the other hand, takes schizophrenia and psychosis as a sort of control group, against which all other disorders are compared by hazard ratio, and they all carry a higher risk of self harm, especially a category called “other”. Unipolar and bipolar disorder are no longer lumped together, but depression and anxiety are, and carry a HR of 3.7. The only conclusion the two papers share is that the highest risk is in the period shortly after discharge. The increasing role of Ikea superstores in both countries is nowhere discussed.  http://www.bmj.com/cgi/content/abstract/337/nov18_3/a2205
http://www.bmj.com/cgi/content/abstract/337/nov18_3/a2278

1334  Caffeine has been blamed for lots of things over the years, but usually exonerated later. Now, however, the caffeine police seem to have found a charge that will stick: drinking the average amount during pregnancy (mostly as tea) may result in a smaller baby. Ban Tea for Bigger British Babies! Cut Coffee and Create Colossal Kids!   http://www.bmj.com/cgi/content/abstract/337/nov03_2/a2332

1341  The introduction of a national colon cancer screening programme in Finland created the opportunity to nest within it a randomised trial of faecal occult blood testing. This unappealing test misses half of colorectal cancers, as this two-year study shows. That’s actually better than most previous studies, and Finland may be on target to reduce its colorectal cancer rate by 20% eventually. The most worrying aspect to me is that three people presented with cancer within two years of a “normal” colonoscopy, but I guess that’s bound to happen when colonoscopists get an enormous workload of mostly normal examinations.  http://www.bmj.com/cgi/content/abstract/337/nov20_2/a2261

Ann Intern Med  2 Dec 2008  Vol 149

787    Ages ago I described mitral valve prolapse as “darling of the show-off medical registrar”, but I meet so few medical registrars these days that I don’t know if this is still true. This paper describes outcomes in an enormous collection of patients (over 8,000) with floppy mitral valves seen at the Mayo Clinic between 1989 and 1998. This was no mean exercise: they went through thousands of old echocardiograms and then traced the patients. Most of it is mild and confers no added mortality risk. For the details of the rest, consult the paper, especially if you are a medical registrar and like showing off.  http://www.annals.org/cgi/content/abstract/149/11/787

816   The little bits of research I have done in my GP career have involved case-finding in primary care using new diagnostic tests – endomysial antibody for coeliac disease, B-type natriuretic peptide for heart failure and holotransocobalamin  for functional B12 deficiency (though others did the real work). This brought home to me the difficulties of comparing the new test with the current reference standard, because changing the test can mean radically changing our picture of the disease: coeliac patients are no longer withered children but often asymptomatic adults. So how do you compare and validate the new test against the old? If there really was a “gold standard” we would all be using it anyway. Oddly enough, this ground-breaking paper suggests that it is possible to compare two silver level tests by using a bronze-level “fair umpire” test. If you want to know how, you must read this paper, written by three of the clearest thinkers in the field, Paul Glasziou, Les Irwig and Jon Deeks. It is beautifully written but needs twenty uninterrupted minutes, a clear head and a cup of coffee (unless you are pregnant): it is destined to become a classic of the diagnostic literature.  http://www.annals.org/cgi/content/abstract/149/11/816

Fungus of the Week: Tricholoma saeva
There’s something rather thrilling in finding a good-sized handsome mushroom growing on a patient’s lawn amid the fog and ice of early December, and this prejudices me this week in favour of the wood blewit, as it is commonly known. It can indeed haunt woods, where it likes a lot of leaf litter, and often hides itself by fallen trunks; but equally you can find it on open grass, as I did, or by the side of muddy footpaths and road verges in early winter. Its cap varies from purple to brown but there can be no mistaking its purple gills and blue-tinged stem.

It is now cultivated commercially around the year in small quantities, and commands high prices, like so many “wild” fungi popular with restaurateurs. But our last few trials of it have been disappointing: its texture turns easily towards sliminess and its taste really isn’t as good as the ordinary cultivated mushroom. And its magnificent purple colours disappear on cooking, unless you augment them with beetroot as in the French soupe mauve.

JAMA  10 Dec 2008  Vol 300 

2621    In the folklore of British otorhinolaryngology, there was once an open-topped London bus which used to run between the primary schools of central London and the Charing Cross Hospital, where the little Cockneys would all have their tonsils out and go back to school in the afternoon, covering the omnibus sides with blood as they leaned over to be sick after their anaesthetic. This crude but still occasionally necessary procedure is apparently covered by a perioperative dose of dexamethasone in many centres, on the basis that this cheap intervention reduces postoperative pain. And so it does, according to this Swiss randomised trial; but this is at the expense of more bleeding, and so probably shouldn’t be done.  http://jama.ama-assn.org/cgi/content/abstract/300/22/2621

2631    Almost anyone can lose weight for four months if they really try, whether for a wedding, an operation, or a cash prize, as in the case of this cunningly designed trial, in which the prize could vary from nothing to $252 per month. But maintaining lower weight over many years is the challenge, and I suspect that even cash can’t achieve that. Global hunger, war or effective pharmacology will have to provide the answer: not much of a choice.  http://jama.ama-assn.org/cgi/content/abstract/300/22/2631

2647   The treatment of pressure ulcers is a topic most doctors fight shy of, leaving the choice of dressings and techniques to nursing staff in hospitals and nursing homes. And some of them are astonishingly good at it, despite the fact that science holds but a flickering light over the broken down skin of the sacrum, buttocks and heels. “Take the weight off and put something over it” seems to be about as far as this systematic review takes us. There are plenty of trials, though 80% don’t reveal potential conflicts of interest and most are not very large and not very good. This includes most trials of nutritional interventions and even trials of human platelet-derived growth factor and nerve growth factor, which may provide a step forward if they ever become affordable.  http://jama.ama-assn.org/cgi/content/abstract/300/22/2647

NEJM  11 Dec 2008  Vol 359

2521   An effective malaria vaccine is one of the biggest prizes still to be won in the field of medicine, and two trials in this week’s NEJM show that we now have a vaccine (RTS,S) that produces good levels of antibody to circumsporozoite protein. The sporozoite is the infective agent before it enters the red blood cells. Once it’s inside those, or sleeping in the liver (as the hypnozoite), it’s hard to get at. But in fact protection against malaria – of around 60% - in these two children’s trials bore no relation to levels of antibody against circumsporozoite. The vaccine works, to an incomplete but useful degree, possibly by other means; and compared with rabies vaccine or Hep B vaccine, it has a low incidence of adverse effects in children. A worthy step, but we’re not there yet.  http://content.nejm.org/cgi/content/abstract/359/24/2521  http://content.nejm.org/cgi/content/abstract/359/24/2533

2579    There’s a paradox in this useful review of the prevention and treatment of seasonal influenza: rates of flu vaccination in over-65s in the USA have gone up from 32% in 1989 to 67% in 1997, but flu-related hospital admissions in this age group have gone up over the same period, though there has been no overall increase in influenza. You might be tempted to give it up as a bad job, though the author here prefers to advocate universal vaccination. Antiviral drugs get rather short shrift: it is very difficult to use them effectively in practice, and in fact the article begins with the real case of a fifteen year old Texan girl who died from necrotising pneumonia despite early administration of osteltamivir. If flu vaccination has a role, it must include people who look after the ill and elderly: so go on, raid the nurses’ fridge and give yourself one. It doesn’t hurt, honestly. Well, not much.  http://content.nejm.org/cgi/content/full/359/24/2579 

2607   Tucked away at the back between the editorials and the letters is a nice crisp interactive discussion of the options for a relatively young man (aged 63) with early prostate cancer. Watch and wait, says the steady Dutchman, though this involves a rectal examination and PSA every 3 months; put in some nice little radioactive seeds, says the radium man from San Francisco, who prefers this to radical external beam radiotherapy; or find yourself a good radical prostatectomist, advises a urologist from Sloane-Kettering. You can even contribute your own opinion by using the link.  http://content.nejm.org/cgi/content/full/359/24/2605

Lancet  13 Dec 2008  Vol 372

The idea of universal human rights is a product of the European Enlightenment, first expressed in luminous prose by Tom Paine, Benjamin Franklin, Thomas Jefferson, William Godwin and the French Assembly of 1789. In the still fairly luminous prose of the International Declaration of 1948, it came to include the right to health, a concept that has been debated ever since; reaching a nadir of self-indulgent obscurity, as usual, in the works of Michel Foucault and his disciples. Can this lofty ideal even be defined, or are we, as the BMJ editorial (p.1361) complains, forever doomed to debate what “health” means on wikis, blogs and online social networks? Is the “right” to health, like other “human rights”, actually always contingent on participative citizenship, as Hannah Arendt argued? It’s a relief to turn away from theory and rhetoric and look at where the world stands, country by country, as judged by a survey using 72 fairly objective indicators in 194 countries. This stupendous overview (p.2047) takes up most of this week‘s Lancet, and is well worth mulling over in front of a roaring fire with a warming glass in hand. Most of the more rhetorical parts can be consigned to the flames, though Amartya Sen’s piece on p.2010 is a definite exception.  http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(08)61781-X/abstract

BMJ  13 Dec 2008  Vol 337

a2313   Here’s a great big systematic review of fibre, antispasmodics and peppermint oil for irritable bowel syndrome which is worth looking at on the website for its full version. Surprises abound: mebeverine is shown to make IBS worse, whereas the best antispasmodics are ones that never appered in the UK or disappeared ages ago, like cimetropium and otilonium. Fibre is pretty useless whereas peppermint oil may be the most effective agent of all.  http://www.bmj.com/cgi/content/full/337/nov13_2/a2313

A2299   For 99% of jobbing doctors, reading papers mostly involves skimming through the abstract and looking at a chart or two to get the general gist. Now and again we may sit down with the full text and read it from start to finish. But nobody under normal circumstances would compare the sample size calculations and statistical methods with the original protocol, which is just as well since they are often (51 times out of 62) either impossible to match or explicitly discrepant. Thank goodness there are policemen out there like Peter Gøtzsche and Doug Altman to do detective work of this kind. To err is human but to cheat is wrong.  http://www.bmj.com/cgi/content/full/337/dec04_1/a2299

Arch Intern Med  8/22 Dec 2008

2401   I took to reviewing the Archives some years ago after I picked up an issue casually in a library and discovered that this journal regularly contains more research papers of general interest than any other. This is a credit to its editor, Philip Greenland, who is now stepping down and being replaced by Rita Redberg. Best wishes to Philip, who has done a marvellous job in publishing masses of good straight-down-the-middle practical medical research without any frills. It’s boring work but somebody needs to do it. Please Rita, don’t change the Archives at all.

2405   If push came to shove during an influenza epidemic, could we make our stocks of vaccine go twice as far by using a half instead of a full dose of trivalent  inactivated influenza vaccine? The answer is almost certainly yes, at least in young people with a good immune response. They produced protective levels of antibody with either amount. in this nice simple large trial.   http://archinte.ama-assn.org/cgi/content/abstract/168/22/2405

2415   Heart failure continues to kill elderly patients, with little improvement in prognosis over the last few years (see the outcome study on p.2481). For systolic heart failure, three β-adrenergic blockers have been shown to improve prognosis – carvedilol, metoprolol and bisoprolol, whereas a lot of heart failure patients are on other, cheaper β-blockers, especially atenolol. Does this make any difference to mortality? A study of over 11,000 adults based on pharmacy records and mortality registers seems to show that atenolol treated patients actually fare slightly better, and a study of a similar number of elderly patients in North Carolina shows level pegging, but with higher rehospitalisation rates in those receiving the approved β-blockers.  http://archinte.ama-assn.org/cgi/content/abstract/168/22/2415  http://archinte.ama-assn.org/cgi/content/abstract/168/22/2422 
http://archinte.ama-assn.org/cgi/content/abstract/168/22/2481

2489   “I’ve never has flu like this before, doctor” they say, coughing demonstratively and sinking into the consulting room chair. The well-oiled machinery of your brain glides noiselessly into action and rapidly produces words of sympathy, a perfunctory chest examination, a sick note and an unnecessary course of antibiotics, about which you hear yourself saying “in case it gets worse”, to salve your conscience. Achily putting on his overcoat, the patient asks, “So do you think it is flu, doctor?” to which you sagely reply “It certainly looks like it”, or something similar. If you were having a really bad morning you could snap, “How should I know? It might be any of a whole number of viruses, perhaps respiratory syncytial virus, or maybe human metapneumovirus, which is the latest fashion. They all cause much the same range of symptoms, varying from a slight sniffle and cough to death. Chances are you’ll be better in a couple of days.”  In fact over a third of flu-like illnesses were due to HMPV in this 3-cohort prospective study conducted between 1999 and 2003 in New Jersey.  http://archinte.ama-assn.org/cgi/content/abstract/168/22/2489 

Plant of the Week: Hedera helix  variegated forms

When the odd ray of sunshine breaks the gloom of mid-December, a few plants can still look pleasing and even cheerful. These are often the backdrop plants with evergreen foliage which you ignored in happier seasons of the year. Plain ivy itself cannot really look cheerful, and if you are so minded you can even enjoy an intensified gloom by planting sorts like “Vimii” which turn dark purple in winter. Better on the whole though to steal or buy bits of gay variegated ivy when you come across them and stick them by your fences and bare walls. They will become a nuisance when they are full grown, but at least they will provide attractive cover the whole year round.

JAMA  17 Dec 2008  Vol 359 

2742    Just in time for Christmas, here is another study showing a short-term reduction in glycosylated haemoglobin in a group of people with type 2 diabetes randomised to a low glycaemic index diet as opposed to a high cereal fibre diet for six months. So if you have type 2 diabetes, or a borderline fasting glucose, what’s it to be on December 25^th?

Smoked eel or salmon – yes
Oysters – yes
Brown bread – no
Cream and freshly dug horseradish – yes
Champagne – yes
Goose – yes
Turkey – only if there’s no goose
Sausages and bacon – always with turkey, never with goose
Prune, brandy and pâté stuffing – yes
Roast potatoes – no
Brussels sprouts – oh alright
Purple sprouting – lots please
Claret – is there another bottle, it’s lovely
Cheeses – yes, from cows, goats and sheep
Christmas pudding – only the sixpences
Brandy butter – only the brandy and the butter
Port – always unwise
Chocolates – what the hell, it’s Christmas and I’ve drunk too much.
http://jama.ama-assn.org/cgi/content/abstract/300/23/2742

2765    In case you are tempted to round off Christmas dinner with a cigar, here is a great big systematic review of all the studies which have explored a link between smoking and bowel cancer. You guessed it – smokers get more colorectal cancer, by about 25%.
http://jama.ama-assn.org/cgi/content/abstract/300/23/2765

2779    Over the years, I’ve praised The Rational Clinical Examination as the best series in any medical journal, little knowing that my comments were heard from afar by the series editor David Simel, who very kindly sent me a copy of the new book in which past contributions are compiled and expanded. This book is a must for all doctors who want to improve their own practice and teach others. This new paper in the series, “Has this prepubertal girl been sexually abused?” is one of the best and most important, because it sifts the evidence rigorously and concludes that there are no clinical features which of themselves can establish a diagnosis of sexual abuse beyond doubt. This may not be what courts, prosecutors and tabloid journalists want to hear, but it needs to be widely understood in order to prevent debacles like the California and Tyneside child abuse allegations which were based on spuriously specific physical signs.
http://jama.ama-assn.org/cgi/content/abstract/300/23/2779 

NEJM  18 Dec 2008  Vol 359

2641   Dicer and Drosha sound like the rival handlers in some John le Carré spy novel, but they actually belong to the rarified science of RNA interference and hence of cancer research. This study of specimens of excised ovarian cancer finds that high levels of Dicer and Drosha expression are actually a good thing for prognosis. If you really must know why, you will have to read the paper and the nice little editorial on p.2720. In the novel, Dicer has an affair with the lovely but mysterious Drosha in Prague in 1946, when both of them are working for their respective intelligence agencies. You can make up the rest. There’s just time to get the DVD for Christmas.
http://content.nejm.org/cgi/content/abstract/359/25/2641

2651   Out-of-hospital cardiac arrest means the same as dropping dead, unless somebody happens to be nearby with a defibrillator that knows what to do within two minutes (yes, that’s intentional: it’s the defibrillator that needs to know what to do). This European trial achieved the unlikely feat of collecting over a thousand patients with witnessed out-of-hospital arrest and giving them thrombolysis with tenecteplase during resuscitation. Then it was abandoned due to futility.
http://content.nejm.org/cgi/content/abstract/359/25/2651

2663   Another randomised trial that achieved nothing was this one of the alpha-adrenergic blocker alfuzosin for the symptoms of “chronic prostatitis” or male chronic pelvic pain syndrome. In this placebo-controlled study, half of both groups improved substantially within 3 months irrespective of what they were given.  http://content.nejm.org/cgi/content/abstract/359/25/2663

2685   PHD mutation and congenital erythrocytosis with paraganglioma. A must for all self-respecting medical Christmas quizzes.  http://content.nejm.org/cgi/content/abstract/359/25/2685

2693    As you groan through the next few days of compulsory over-eating, my wish for you is that you remain completely unaware of your eating apparatus. To become aware of the teeth, the tongue or the temporomandibular joint is a sign of trouble, and in the case of the TMJ, trouble that nobody really understands and that tends to grumble on for months or years. Dental luminaries from Tufts and Harvard here tell us all that is known about the subject and what can be done about it – not all that much on either count. Reassurance, rest, heat, NSAIDs, benzos, TCAs, jaw appliances, behavioural therapy, manipulation. If you try each one for three months, you might at least keep the patient busy for a couple of years.  http://content.nejm.org/cgi/content/extract/359/25/2693

Lancet  20 Dec 2008  Vol 372

2115   As a fairly frequent migraineur living in the same house as a teenager who is in almost permanent status migranosus, I warmly welcome the arrival of a whole new class of drugs for acute migraine, the pants. They are bound to take off. They work as well as the triptans but without the vasoconstrictive side-effects and with even more beautiful names like telcagepant and olcegepant. They were designed as antagonists of the calcitonin gene-related peptide receptor, though whether they really work by this central mechanism is hard to tell (see editorial, p.2089).
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(08)61626-8/abstract

2132   Carotid endarterectomy has become a very common operation, which can now be performed under local anaesthesia. This might offer the advantage of the patient being able to tell you if they are having a stroke, assuming they don’t become aphasic right away. But in fact this randomised trial (GALA) didn’t establish any superiority between local and general anaesthesia.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(08)61699-2/abstract

2143    A major improvement in neonatal survival followed the discovery that a single dose of antenatal corticosteroid would help babies cope with preterm birth. This blinded placebo-controlled Canadian trial went on to explore whether you could compound this benefit by giving multiple courses of antenatal corticosteroids to mothers likely to deliver very prematurely. Quite the opposite: you end up with smaller babies and no benefit. http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(08)61929-7/abstract

2124   If you are reading this, you were born lucky: in a prosperous country at a time before the world’s resources collapse and while medicine provides a well-paid and interesting job. It was not always thus, and may not be so for ever. If you live in Europe and are getting oldish, Great Britain is not a bad place to live by comparison with all European countries: for example, I can look forward to another 12 years of reasonable health (as an average kind of bloke) and maybe survive another ten years beyond that. Help! There is so much I would like to do before I die. But at least I don’t live in Estonia, where I might be dead tomorrow. Or in Cyprus, where over half of elderly people live in poverty. This “cross-national meta-regression analysis” offers lots of interesting browsing over the holiday period. Italy, as ever, is the best place to be. http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(08)61594-9/abstract

2152   When NICE was set up, those of a paranoid tendency such as myself feared it would become a government rationing tool and descend into political and scientific confusion. It has had a few bad moments, but overwhelmingly it has succeeded in a very difficult task, and much credit for this must belong to its director, Michael Rawlings. His paper here shows how sound his intellectual foundations are: it’s a magisterial survey of the types of evidence and how they should be used to decide the use of therapeutic interventions. Don’t be put off by the fact that it is called a Harveian Oration and that its title begins in Latin. It is beautifully written and completely practical in tone and worth reading by every jobbing doctor as well as every academic.

JAMA  24 Dec 2008  Vol 300

2859    To create one of the many US cohort studies which report painstakingly from time to time, nearly 500 people aged 18 to 30 were plucked from the streets of Chicago between 1985 and 1986 with the idea of following them up for life. As suggested by the name CARDIA (an approximate acronym for Coronary Artery Risk Development in Young Adults), this was yet another search for hidden cardiac risk factors. Little could they know how soporific this subject would become after 20 years. Or perhaps they guessed, because they went to great lengths to measure sleep duration in their cohort. Since their oldest subjects are still only 53, hard cardiovascular events are few, but coronary calcification is beginning to creep in, and this shows a definite relation to short sleep duration. Now you have one more thing to worry about as you lie awake at night.
http://jama.ama-assn.org/cgi/content/abstract/300/24/2859

2886   There are a number of things you can do about not getting type 2 diabetes, and I am forming a New Year’s resolution to do some of them; but changing one’s birth weight is not among them. What you register on the midwife’s scales definitely counts as a non-modifiable risk factor for the rest of your life. In the world literature on the connection between birth weight and type 2 diabetes there are 327 papers, of which 31 were deemed worthy of analysis in this systematic review. Oddly enough, the two papers which carry the highest statistical weight counterbalance each other almost exactly, one showing an association with low birth weight and the other with high birth weight. Overall, however, the low connection wins. The paper concludes with a lengthy comment section in which it seems that all 35 authors get the chance to introduce their favourite red herrings. Or, in this case, Red Indians.
http://jama.ama-assn.org/cgi/content/abstract/300/24/2886

2898   For a long time now, JAMA has been running a rather uneven series called Perspectives on Care at the Close of Life. This one reminds us that most people die in a state of delirium, classed as either hyperactive or hypoactive. Even at this late stage, it is worth looking for reversible factors in an effort to avert distress: dehydration, hypercalcaemia and the wrong medication are common causes. The right medication, as far as we can tell from a patchy evidence base, is a phenothiazine, usually haloperidol in British practice, though chlorpromazine is an option too; new generation (“atypical”) antipsychotics on the other hand are probably best avoided.
http://jama.ama-assn.org/cgi/content/abstract/300/24/2898

NEJM  25 Dec 2008  Vol 359 

2753   Perhaps recollecting that the celebration of Christmas was outlawed by the governors of Boston, Mass. between 1659 and 1681, the New England Journal appears in plain Puritan garb on Dec 25^th with sober contributions on topics like shared and distinct genetic variants in type 1 diabetes and coeliac disease (p.2767) and inhibition of the Bcl-x_L deamidation pathway in myeloproliferative disorders. A bit more seasonal is this paper on childhood anxiety, a common condition in the days leading up to Christmas. Should we be treating it with sertraline, cognitive behavioural therapy, or both? If it’s a serious problem – and it often is, Father Christmas aside – then the answer seems to be both the SSRI and the CBT, though I do retain worries about tinkering with the neurochemistry of the developing brain.  http://content.nejm.org/cgi/content/abstract/359/26/2753

2790    ANCA-associated vasculitis encompasses Wegener’s granulomatosis and microscopic polyangitis and would often be fatal before the introduction of treatment with corticosteroids and toxic immunosuppressants. These regimes seem nasty and old-fashioned but they have not yet been superseded, and this French study looks whether azathioprine or methotrexate is the safer drug for maintenance treatment. Although methotrexate is more widely used, the figures in this study favour azathioprine as the safer drug, though the difference is not statistically significant.  http://content.nejm.org/cgi/content/abstract/359/26/2790

2804   The lung is a potent filter, not just of atmospheric rubbish but also of stuff that floats about in the venous circulation. Someone ought to write an essay about this: I would wager that for every cancer metastasis or pulmonary embolism that becomes clinically evident, there are a hundred that the lung has quietly got rid of without anyone noticing. When the sieve blocks, however, the consequences can be dire, and venous thromboembolism is responsible for about 15% of hospital deaths and up to twice that percentage of deaths associated with pregnancy and delivery. This single author clinical review is a clear guide to diagnosis and what is known about treatment, all the way from heparin to embolectomy. As I’ve recently pointed out, we still don’t know how best to select patients for thrombolysis. The rate of bleeding complications in the PE trials is far above that in the myocardial infarction and stroke trials, presumably because patients have first received anticoagulation.  http://content.nejm.org/cgi/content/extract/359/26/2804

BMJ Journals  Dec 2008 

ADC  1017   Tuberculosis in children in England and Wales is uncommon, but not vanishingly rare: about 500 cases a year between 1999 and 2006. As you might expect, the rate is far higher among children born outside the UK. It was not always so – in the sixteenth century it could even kill young kings of England, to the great detriment of English Protestantism; though Edward VI did also suffer the misfortunes of measles and smallpox in the year that TB finally carried him off. Nowadays practically every child with TB survives, despite a fairly high proportion of drug-resistant infection.  http://adc.bmj.com/cgi/content/abstract/93/12/1017

Heart  1594   Here is a comparison between the Republic of Ireland and the UK province of Northern Ireland in the secondary prevention of cardiovascular disease. With their primary care system galvanised by the Quality and Outcomes Framework, Ulstermen enjoy lower levels of blood pressure and cholesterol: they are, however, less active, more miserable, and less likely to eat fruit and vegetables.  http://heart.bmj.com/cgi/content/abstract/94/12/1594

Heart  1656   Short of putting statins in the water supply, I am all for pushing the limits of established indications wherever there is a bit of evidence to support this: after all, no group has yet been identified in which these drugs do harm, and they undoubtedly reduce cardiovascular risk (and probably cognitive decline) in the vast majority of people, whatever their cholesterol or their initial level of risk. This Dutch paper reviews the place of statins in controversial subgroups, particularly ischaemic heart failure, chronic kidney disease, and the very elderly. The “Holy Grail Effect” – actual regression of atheromatous plaque – is also discussed. Watch these spaces for better evidence to come.  http://heart.bmj.com/cgi/content/extract/94/12/1656

Thorax  1103   The days we take off around Christmas provide a brief respite from the stream of babies with crackly chests we see at this time of year. Anxious to be seen to do something, we’re tempted into prescribing bronchodilators and steroids for acute bronchiolitis, though the evidence tells us very clearly that these do no good at all. In Switzerland, guidelines discouraging their use were promulgated in 2004 to all doctors in primary and secondary care, and this survey shows that rates of prescribing have dropped markedly between 2001 and 2006. I get the impression that UK GPs are somewhat lagging in this respect.  http://thorax.bmj.com/cgi/content/abstract/63/12/1103

Ann Intern Med  16 Dec 2008  Vol 149

845  My diagnosis rate for spinal stenosis with or without degenerative spondylolisthesis shot up a couple of years ago, when GPs locally first got direct access to lumbar spinal MRI. I duly sent most of them off to the surgeons (mainly neurosurgeons) and most of them duly got operated on. I guess about half of them wish they never had been, though I try to comfort them with the idea that if they hadn’t they might have got worse. The main source of our evidence about this is the SPORT study, which had a huge rate of cross-over, mainly from those randomised to conservative treatment. So we are still a bit in the dark as to whether operative treatment really helps and still more about whether it is cost-effective.  http://www.annals.org/cgi/content/abstract/149/12/845

861    There is a big literature about hope, truth and preparing for death in various contexts, but this paper breaks new ground by discussing these issues in connection with relatives or “surrogates” who may have to make decisions for those too ill to make their own. It is an excellent study of its kind, conducted in a San Francisco intensive care unit where no fewer than179 surrogate decision makers were interviewed. This analysis of the responses is too complex for an easy summary, but essentially it shows that most surrogates need preparing for their decision making, and that they rarely take the view that withholding a prognosis is a good way to maintain hope.  http://www.annals.org/cgi/content/abstract/149/12/861

889    Alas, I cannot expect you to share my enthusiasm for this paper on systematic reviews of diagnostic test accuracy. It is indeed rather chewy reading even for those who have spent years trying to explore this subject. Better, perhaps, to turn straight to the editorial on p.904 which emphasises the fact that beyond the world of specificity and sensitivity, ROC curves and fixed-effects models lies the plain fact that the predictive value of each test depends on the clinical context. We really need a whole new literature of diagnosis adapted to primary care. It may start appearing in 2009.  http://www.annals.org/cgi/content/abstract/149/12/889

New Year Projects 2009-2109: a request for help

I would like to pursue two projects in the New Year and I would greatly appreciate help from readers:

(a) what medical conditions do you think are Easily Missed? These need to be clinically important and to have clear diagnostic criteria, and there needs to be evidence that they are underdiagnosed, or missed at first presentation. A BMJ series of this name will begin to appear some time in mid-2009 and Anthony Harnden and I, the series advisers, would welcome your ideas.

(b) if you were a patient – and you will be one day, if not now – what kind of integrated care would you like to receive and in what setting? The immediate trigger for this question is a comprehensive review of medical services in my small town, but this needs to link in with a much grander vision of an academic Centre for Integrative Medicine if it is to succeed. This might begin simply with an amalgamation of certain acute primary and secondary care services on a single site, with shared working and training based on following individual patients through and beyond the location. But my hope is that it would link up with an international network of such centres in the context of widely differing health systems. Each of these would explore new models of care and teach a full range of health care professionals, and would exchange ideas and people. We would follow the complete lives of individuals across the whole spectrum of physical and mental experience, from genomics and epigenetics to narrative medicine. We would redefine medical generalism in ways that we cannot even conceptualise at present. That’s why I don’t see the project as ending until 2109 at the earliest. Give me your thoughts.