JAMA  23 Feb 2011  Vol 305
783   The highly prevalent custom of poisoning the osteoclasts of old ladies with bisphosphonates for years on end seems surprisingly harmless. This case-control study confirms that there is a tiny increase in the risk of subtrochanteric fracture after five years, heavily outweighed by the protective effects of these drugs in the first years of use. Nobody really knows how long to use them for.

800   The osteoporosis market is huge and lucrative and teems with competing products: the last thing that pharma companies want is a competitor that costs practically nothing. Now what might that be? Clues: you might get a headache or a Nobel prize if you can come up with the answer. You may find it in a stone quarry or smeared onto an anal fissure, or else popped under the tongue of an elderly man hurrying for a bus. Yes, this trial confirms that it really does increase bone mass density in osteopenic  post-menopausal women, used as an ointment for two years. Nobody won the Nobel prize but a lot of them dropped out due to headache. No need, I hope, to name this dynamite.

820    Two years ago, the third trial to show the futility of lowering HbA1c below 8% or so in stable type 2 diabetics was published, leading me to write that all diabetologists should now be ordered to give a lecture called “How I Learned to Eat My Words and Realize that Tight Glycaemic Control has been a Colossal Waste of Effort  and Resources.”  As a result I was asked to write a couple of polemical pieces for the BMJ, which became much more measured (though still polemical) editorials thanks to the help of Harlan Krumholz and later John Yudkin. Just before Christmas the three of us met up for the first time at Yale and we also met a number of brilliant youngsters (that means under 40 for the likes of us), notably Kasia Lipska and Joe Ross, who have now written this beautifully argued piece about the doubts surrounding the whole thiazolidinedione class of oral hypoglycaemic agents. If it ends in “glitazone”, it’s in the danger zone.

NEJM  24 Feb 2011  Vol 364
701   Ever since its first performance in 1882, Wagner’s PARSIFAL has drawn mixed responses – many agreeing with Nietzsche that there is some of the old magic in it but a lot that is frankly crap. Perhaps this is why the name has been given to the first of two German studies of the effect of barnyard microbes on childhood asthma. The first hint that the Holy Grail of asthma prevention might lie in exposure to poo from fellow-children or from farmyard animals came from East Germany. There was a time when many infant Socialist Pioneers enjoyed close proximity with each other and with the daily by products of collective cattle and pig farms: and these children of the People’s Democracy were later found to suffer less asthma and atopy. The cross-sectional PARSIFAL study shows that childhood asthma rates are lowest where there is a rich fungal and bacterial flora in the environment: the second study GABRIELA is even more interesting in that it shows that atopy in general is specifically prevented by exposure to gram-negative rods. If you love your children, embrace dirt, real dirt.

719   Beware the word Asian in the medical literature. In the USA, most Asians are of Chinese or Japanese descent, whereas in the UK they are mostly from the Indian subcontinent. Their genetic characteristics and epidemiology vary strikingly, and this applies even more when these “Asians” are living in their native subcontinents. Here, an intriguing study looks at the effect of body mass index on mortality in more than one million people living east of Iran. Go due east along the silk road and you find that Chinese, Korean, and Japanese adults display the expected U-shaped relationship between BMI and death from all causes. But go down along the spice roads to the south east, and the results are startlingly different: there is no increase in mortality with a rise in BMI. In fact if you exclude the effect of respiratory disease, which kills enormous numbers of Indians and Bangladeshis with a BMI under 22, the line for cancers and cardiovascular disease is straight, and level, to 35 and beyond. The confidence intervals, however, get pretty wide as you go along, since there are as yet so few fat Indians in India. The “world tsunami” of obesity is real, and dangerous, but it will hit different shores with varying force.

730   After this gargantuan study of non-communicable disease, it’s good to see some meticulous epidemiological work on the local scale – the equivalent of John Snow’s cholera map of the Broad Street area, brought crackingly up to date with the use of genomics and social-network analysis. Between 2006 and 2008, there was an outbreak of tuberculosis in British Columbia, which eventually affected 41 people. By studying the genomics of the mycobacteria and the social connections between the individuals, the investigators established there were two lineages of bacteria, most probably spread by the use of crack cocaine.

749   Some years ago, the front cover of the BMJ sported a doctor with a portable ultrasound machine slung round her neck in place of the traditional stethoscope. How I wish I could use one of these things! Suspected biliary colic, threatened miscarriage, possible breast abscess, suspected pneumothorax – I might see any or all of these things during my 7-hour out-of-hours primary care shift tomorrow, and there isn’t a single ultrasonographer in the local hospital at weekends. This brief account of point-of-care ultrasonography says that the latest portable mini-machines are good for all of these things (pneumothorax included) and perhaps I’ll look into setting up a local trial.

Lancet  26 Feb 2011  Vol 377
693   I have just been reading through the personal narrative of someone with systemic lupus erythematosus, which might be summed up as “BLISS is it not this day to be alive, And to be young is very hell.” Like so many people with SLE, she longs for better public understanding, and for a breakthrough in treatment. She would also like better access to the medical literature, and I’d happily send her the BLISS-52 study which shows promising results from belimumab, a fully human monoclonal antibody against B-lymphocyte stimulator. But despite a very positive editorial, I’m not convinced that this agent really is the beginning of a new dawn in SLE treatment. As usual, The Lancet fails to edit out a summary conclusion which smacks of salesmanship rather than science.

732    My father had heart failure and angina and developed a hatred for motor exhaust fumes, especially around bus stations, which we all put down to him “being Polish.” But there is now a lot of good epidemiology suggesting that atmospheric pollution is a potent source of cardiac distress, quite unconnected with ethnic emotional stereotypes. In this study, a sort of meta-analysis with bells and whistles, moderate atmospheric pollution is identified as a potent trigger for myocardial infarction on a level with “other well accepted triggers such as physical exertion, alcohol, and coffee.” Blimey, I’ll be extra careful next time I have a beer and a coffee after carrying luggage into a Polish bus station.

741   Just in case you might think I had missed it, I shall here mention the SCAST study of reducing blood pressure during acute stroke using candesartan. Do not reduce BP during stroke by this or any other means: it is probably harmful.

751   Influenza pandemics only happen every twenty to thirty years, so if H1N1 counts, we’ve been very lucky and we can sit back complacently for another couple of decades. There should be no sudden rush to produce a new vaccine quickly during that time, but if and when we do, Vero cell culture may be the way to do it, following this successful trial of a trivalent vaccine. Or perhaps human influenza will be a thing of the past by then, eradicated by an effective omnivalent vaccine.

BMJ  26 Feb 2011  Vol 342
479   The association between alcohol consumption and coronary heart disease is very simple: make sure you drink at least 10 units a week. As far as I can tell from this meta-analysis and other papers, there is a threshold dose but no upper limit (a fact deeply hidden in the text). For stroke, however, the maximum dose for benefit is less than 10 units a week, and there is definite extra risk above about 60u a week. Total abstinence conveys risk for all cardiovascular outcomes. It may, on the whole, be better to drink no more than a bottle of wine a day.

484    The generally excellent Clinical Review section maintains top quality with a piece on the treatment of breast infection as carried out in the Western General Hospital in Edinburgh. From a primary care point of view, the key message is to treat early with standard doses of flucloxacillin or similar for ten days, and to refer urgently anything that recurs or fails to resolve. Needle drainage of abscesses may need daily follow up with ultrasonography. Mastitis in non-lactating women is almost confined to smokers. Sub-optimally managed breast infection can land you in court as it can lead to lasting disfigurement. All useful points.

Plant of the Week: Corylus avellana “Fusco-rubra“

This is the time of year when hazel catkins decorate every wood and roadside hedge, and you can try to imagine the hazel-dominated England which would have greeted European hunter-gatherers as they moved in after the last Ice Age. Everywhere there would have been forests of oak or beech or lime, with alder dominating the many marshes, and the understory in all these woods would have been hazel.

But although C avellana is the most abundant of native British shrubs, it has sported very few variants. The one I celebrate this week bears strikingly attractive red catkins: at least it does if it’s the small tree I recently spotted in a nearby village. It also bears red leaves later on, and still later I imagine the bracts of its nuts will be red. It is a more attractive plant than the widely popular purple hazel, C maxima “Purpurea”, according to Bean, whose word is law.

This then is a large subject which is attractive in flower, leaf, and fruit, but only rarely seen. I’m afraid I have only just discovered it, and long ago I planted its coarse purple cousin instead. However, the filberts of maxima are a little bigger than the cob nuts of avellana. Both are of course a great attraction to squirrels, and can provide a fine garnish to a dish of Squirrel Nutkin, though I have not perfected this to my own satisfaction. Supplied from time to time with freshly killed young squirrels by the exertions of our fearless lion-coloured Burmese cat, Spencer, I have got as far as this attempt:

Top and tail your squirrel(s), then skin. Dispose of the entrails including the liver, which is very bitter. The legs alone are substantial, but the rest of the carcase, suitably jointed, can be included in the stew. The first step is to brown the pieces in hot walnut oil with a little chopped onion. Then add white wine and water and a few sections of chopped leek with a couple of stems of thyme, and bring to the boil. Simmer gently for 30-40 minutes. Meanwhile chop some filberts or cobnuts and place on an oven tray rubbed with walnut oil. Heat these in a fairly fast oven so that they brown. Take the squirrel pieces out of their cooking liquid and put them on a separate tray in the same oven. Reduce some of the liquid to form a sauce. Remove the squirrel from the oven after a few minutes and decorate with the roasted nut pieces. Flame with brandy or nocino (a North Italian nut liqueur). Pour a little of the cooking sauce on heated plates and serve the squirrel. Depending on the size of the animal, allow half to one squirrel per person, though most people prefer none.