Journal Watch - May 2010

JAMA  26 May 2010  Vol 303

2035    We know that only about half of exacerbations of chronic obstructive pulmonary disease are infective, but the trouble is that we don't know which half, and we seldom look for other common causes, such as occult pulmonary embolism and cardiac ischaemia. So should we be using fewer antibiotics? God forbid. This huge retrospective study of nearly 85 000 patients admitted to US hospitals with exacerbations of COPD shows that we should be using earlier antibiotics. In fact the benefits from using antibiotics in the first two days are so clear that for once an extrapolation to primary from secondary care seems warranted: give anyone with an exacerbation of COPD an antibiotic at the first visit. I'd suggest that somebody needs to do a big observational primary care study at once, measuring D-dimer, troponins, CRP and procalcitonin at the same time. 
http://jama.ama-assn.org/cgi/content/abstract/303/20/2035

2043    We spend an awful lot of our time giving drugs to perfectly healthy people, and a good thing too. The Healthy People 2010 goal in the USA was to achieve blood pressure control on 50% of the population, and it was achieved with a surplus of 0.1%, according to NHANES. I'm sure that more than 50.1% of Britons have good blood pressures though I can't quote the latest figures, which would be derived from QOF, which encourages cheating. On the other hand, any slight tweaking downwards by payment-oriented GPs would be more than offset by millions of falsely high readings taken by nurses using wretchedly unreliable electronic machines instead of mercury sphygmomanometers and properly sized cuffs, as mandated in NHANES.
http://jama.ama-assn.org/cgi/content/abstract/303/20/2043

2051    "Don't worry, it's only a virus: Shaylee will soon be better." Or worse, as in the case of infants and children admitted with severe pneumonia in a rural Kenyan hospital. Over one half had evidence of viral infection, using comprehensive and sensitive molecular markers. Captain of these child-abusing viruses is respiratory syncytial virus followed a long way behind by coronavirus 229E, influenza A, parainfluenzae type 3, adenovirus and metapneumovirus. We urgently need a vaccine against RSV, which also caused the severest disease.
http://jama.ama-assn.org/cgi/content/abstract/303/20/2051

2058    Rawadawa. If you belong to the Munda tribe of peoples scattered around east India, Bangladesh and Nepal, you will know that this useful word means "the sensation of suddenly realising you can do something reprehensible because nobody can see you doing it." (cf. HL Mencken's definition of conscience - "the still small voice inside you that tells you somebody might be watching." ) This paper is a warning to those who want to spin the results of their non-significant interventional trial - Doug Altman is watching you! He and his fellow-detectives spot widespread spin and inconsistency in published trials with non-significant outcomes. Do not be tempted to enjoy rawadawa with your raw data.
http://jama.ama-assn.org/cgi/content/abstract/303/20/2058

2065   The Clinical Crossroads piece this week is titled "A 42-year-old man considering whether to drink alcohol for his health". At 42, you start realising that you really are going to die and that the process has already started. 18 years later, the process is well under way and you've got used to the thought  that all you can hope to do is use your brief opportunity to be useful and enjoy yourself - which often amount to the same thing. So the main reason to drink alcohol is not for the sake of your health but to be happy with others and yourself, and the main reason not to drink too much is not because you damage yourself physically but because you become an incapable bore. It just so happens that by drinking the right amount you also protect your heart and reduce your risk of Alzheimer's disease, though women increase their risk of breast cancer and both sexes their risk of oropharyngeal and oesophageal cancers. But then, as the Daily Mail long ago discovered, everything in life increases or decreases your risk of something else.
http://jama.ama-assn.org/cgi/content/abstract/303/20/2065

NEJM  27 May 2010  Vol 362

1959    I was last taught neonatology as a medical student,  when I remember a debate about how much oxygen should be given to extremely premature preterm infants, given the risk of retinopathy from high levels of oxygenation. Here, 36 years later, is a randomised factorial trial to settle the question. These tiny infants were given a target range of oxygenation between 85-89% or between 91-95%. There was no difference in retinopathy between the groups, but more babies in the lower concentration group died.
http://content.nejm.org/cgi/content/abstract/362/21/1959

1970   The other big issue all those years ago was how to make artificial surfactant to treat the respiratory distress syndrome of extreme prematurity. But with this study we have come full circle: it may well be that for most such babies, continuous positive airway pressure is just as good a treatment as surfactant with intubation.
http://content.nejm.org/cgi/content/abstract/362/21/1970

1193   Although I have always striven to maintain low standards of domestic hygiene in the interests of my children's immune systems, it has taken one of my sons until his early twenties to develop infectious mononucleosis. A virologist friend used to class universities by the proportion of students entering with positive antibodies to Epstein-Barr virus - Oxford came lowest. Which is rather odd since so many of the men in those days (and there were  also a few women, I seem to remember) had spent time in the dormitories of private schools, where practices took place that were highly favourable for the transmission of EBV, according to legend. Anyway, it's still best to get it out of the way while you're young, according to this excellent review. But the heterophile antibody test doesn't become reliable until after the age of about 10, and then only relatively: more than 10% of atypical lymphocytes is also a goodish test. As for "recurrent" glandular fever, it's a hard one to diagnose: once you have EBV, you have it for life, and it would take at least 11 years of antiviral treatment to rid you of it.
http://content.nejm.org/cgi/content/extract/362/21/1993
2001   Here is an article worth keeping handy: Genomic Medicine - An Updated Primer. It's a very well written and enthusiastic account, prefaced with a clinical vignette about a Jewish lady called Cathy who has breast cancer. "How can any clinician understand the diagnosis and treatment of breast cancer, much less explain it to a patient like Cathy, without a rudimentary understanding of genomic medicine?" ask the authors. Now Cathy belongs to an ethnic group with a raised risk of breast cancer, partly mediated by BRCA genes, which she pays to have tests for (they are still patented) which are negative, and she then has an alleged cancer detected by mammography, which is oestrogen-receptor positive and HER2 negative. So I can't really see what genomics has contributed to Cathy's diagnosis and management  except some extra expense. However, we must go forward in faith, and this piece has some very useful tables and figures. The NEJM is giving the full catechism away free:
http://content.nejm.org/cgi/content/full/362/21/2001

Lancet  29 May 2010  Vol 375

1875   Fibrates have been around for 40 years, during which time we have been waiting for some convincing evidence that they improve hard outcomes. We know that they increase HDL cholesterol by acting as agonists at the α-receptor for perixosome proliferator, and they reduce levels of triglyceride. Otherwise, they don't seem to do very much except increase statin myopathy when co-prescribed. This large systematic review of the randomised trials shows that they have no effect on all-cause mortality, cardiovascular mortality or sudden death. They produce a fall of about 10% in "major cardiovascular events". Given this evidence, it is hard to see who should be given them, but the editorial gets excited by a putative subgroup who might benefit, subject to further trials, "thus providing a first step to personalised medicine." i.e. the theoretical use of drugs of dubious value in people who can be persuaded to take them on the basis of surrogate biochemical or genomic markers of risk.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60656-3/abstract

1885   It will be interesting to see how the new centre-right coalition in the UK addresses the problem of opioid drug addiction. If they followed the evidence they would decriminalise the possession of any kind of drug and allow the supervised use of injectable heroin or methadone as a maintenance treatment in those who do not respond to oral methadone. This randomised British trial confirms the findings of several others around the world.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60349-2/abstract

1920   Some years ago we saw the first progress for a century in diagnostic testing for tuberculosis, but this is a complex and difficult area which I think has now become the sole province of experts. At least that's my excuse for not trying to provide you with a synopsis of this lengthy and exhaustive paper, with its 218 references. The trouble is that Myocobacterium tuberculosis and its cousins are lazy old beasts which have been living with mammals and reptiles for many, many millions of years and can leave their signature everywhere without doing very much, while failing to alert us when they becomes active.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60359-5/abstract

BMJ  29 May 2010  Vol 340

1177   Some years ago I tried to master the clinical literature of vitamin B12 deficiency in the hope of joining Harold Hin in doing some primary care studies. We were particularly interested in the limitations of the commonly used assays and the putative superiority of holotranscobalamin II and methylmalonic acid. We needed to identify a few risk groups and one that stood out was patients who take metformin. This otherwise excellent drug drops your B12 level by about 20%, as this study shows, and in some people this may be clinically important. But the trial does not make a strong case for routine biochemical monitoring, as the editorial points out. And if you do replace it, for goodness sake use a high dose  oral preparation; the only real reason to use IM hydroxocobalamin is to give patients a two week period of euphoria, a common and well-attested phenomenon which has no medical literature that I know of.
http://www.bmj.com/cgi/content/full/340/may19_4/c2181

1178   Giving people the label of diabetes can have the most shattering effect on them, as you can see on healthtalkonline.
http://www.healthtalkonline.org/chronichealthissues/Diabetes_Type_2/Topic/2089/topicList

So why do we do this to asymptomatic people? The answer must be because we think that at this point they will benefit from weight loss, a low glycaemic index diet, more exercise, better blood pressure control and perhaps a bit of metformin. But doesn't that apply to about half the population anyway? This Chinese study takes the glucose tolerance test as its gold standard and derives a ROC curve comparing it with glycated haemoglobin (HbA1c). The level of divergence is 6.3%, so the authors suggest that we use this as the new definition of diabetes. Whatever.
http://www.bmj.com/cgi/content/full/340/may17_1/c2249

1179    This Spanish study of heterosexual couples where one partner has HIV-1 reassuringly finds that provided the infected partner keeps taking antiviral drugs, the risk of infection to the other partner is very low.
http://www.bmj.com/cgi/content/full/340/may14_1/c2205

1185   I hate winter. I become lethargic and irritable, I eat more, I want to sleep, I dread the increase in workload, and the deaths of patients I have become close to over many years. I become SAD, as in this review. And mad. And bad. Perhaps I should be sectioned and given light therapy next winter, but I have decided to try vitamin D and retirement instead.
http://www.bmj.com/cgi/content/extract/340/may21_1/c2135

Arch Intern Med  24 May 2010

880   Here is a really fascinating paper from primary care in the Netherlands which generates the hypothesis that long-term β-blockers may reduce mortality and risk of exacerbations in chronic obstructive pulmonary disease. And although this may surprise many doctors who habitually prescribe inhaled or even oral beta-agonists to these patients, it is in fact quite logical that beta-antagonists should be more beneficial, since the main mode of death in COPD is cardiovascular, and many exacerbations are hidden cardiovascular events. But this paper falls well short of proof. It uses a database from Dutch general practice to identify patients with COPD but without overt cardiovascular disease who have been prescribed β-blockers and compares their outcomes with a matched group. There is abundant opportunity for hidden confounding here, and we need a randomised controlled trial.
http://archinte.ama-assn.org/cgi/content/abstract/170/10/880

Plants of the Week: Tall Bearded Irises

The arrival of flowers on tall bearded irises marks the high point of the gardening season. Once they disappear, at the end of June, the best is over, so enjoy them while you may. Their form is miraculous, they are borne in generous succession, and you can choose more or less any colour you like. It would be silly to suggest one variety as there are several hundred; you can buy them from a specialist catalogue over the winter months or from a nursery at this time of year. This allows you the immeasurable advantage of being able to smell them: blues and yellows tend to fruit salad and pineapple, dark browns and blacks to chocolate and cloves. Oh for a boundless garden to grow them in huge drifts, and cut armloads for the house.

JAMA  19 May 2010  Vol 303

1921   This week's JAMA is devoted to mental health issues, as we now call them. When I began my medical career, we would have called these psychiatric problems. But now we have moved away from the model of the fretful psychiatrist taking on the personal problems of a multitude of individuals to a model of team care where the main aim seems to be to prevent any personal bonding between patients and professionals. This, plus the unwieldy nature of the interventions, the huge time delays and the constant changes of service provision, all mean that patients turn to GPs for the timely personal involvement that they really need. They may make our hearts sink, but that is the nature of the job. Anyway, that's enough from me: I am retiring and saying a genuinely fond farewell to many of them. On to the delivery of evidence-based treatment for multiple anxiety disorders in primary care. This is American mental health care at its most up to date, i.e. indistinguishable from British mental health care, and a world away from the bearded man sitting at the end of the patient's couch who still appears regularly in New Yorker cartoons. Instead the anxious patient (or should I say client?) sits in front of a computer doing CBT. If this is deemed insufficient, then serotonin reuptake inhibitors are allowed, following a web-based optimization schedule. Then, if all else fails, give them some diazepam. This is known as CALM (Coordinated Anxiety Learning and Management) and it proved superior to Usual Care in 17 US primary care clinics.
http://jama.ama-assn.org/cgi/content/abstract/303/19/1921

1938   Traumatic brain injury does not do wonders for your mood. Over half the patients in this cohort of brain-injured patients from Seattle were deemed to have major depressive disorder. This in turn was a predictor for poorer health-related quality of life. The figures were based on a recruitment rate of 52% and telephone interviews using PHQ-9, so the figures aren't the most robust. I'd also like to know how many of these patients have undetected pituitary dysfunction. Also we don't know what interventions might work best for these people; and from watching Frasier I have concerns about the quality of psychiatric services in Seattle.
http://jama.ama-assn.org/cgi/content/abstract/303/19/1938

1954   It's been known for years that depression is one of the worst single predictive factors for death and hospitalization in end-stage diseases such as heart failure and advanced pulmonary disease. Here the same is found for chronic kidney disease. Some of this may be to do with the will to live; some of it is also no doubt due to all the feel-bad chemicals (such a cytokines) that accompany the dying process in these advanced disease states.
http://jama.ama-assn.org/cgi/content/abstract/303/19/1946

1961   Are men allowed to have postnatal depression? I think the answer from this meta-analysis of 43 studies is: not really. The men who became depressed after the birth of a child were generally the men who had depression before and who were reacting to a new stressor, in the form of the baby or its depressed mother. Maybe this is what Freud meant when he told Mahler that he was much too neurotic to father children.
http://jama.ama-assn.org/cgi/content/abstract/303/19/1961

1970   Progress in psychiatry in my working lifetime has meant a fall in the suicide rate and an increase in the number of people taking long-term drug treatment. Some drugs are slightly better than the ones I started out with; most are just a lot more expensive. Psychopharmacology works in the most mysterious ways its wonders to perform. It seems that you can make people feel better either by inhibiting serotonin reuptake or promoting it. Serotonin reuptake inhibitors aren't meant to be addictive but most people feel worse when they stop them. The newer ("atypical") antipsychotics may be more toxic than the phenothiazines they replaced. Benzodiazepines have become an absolute taboo, but I don't see in what ways they are worse than the drugs that we are supposed to prescribe instead. The way out of this morass is going to come with better understanding of genomics, according to this interesting short editorial, Rethinking Mental Illness. Or rather, because genomics is proving useless at uncovering most of the determinants of mental illness, the answer will lie in epigenomics - how life exposures influence the expression of genes. The authors find "strong bases for hope that insights emerging from genetics and neuroscience will be translated into rational development of robust and personalized treatments." Lift up your hearts. R: We lift them up unto the genes, O Lord.
http://jama.ama-assn.org/cgi/content/extract/303/19/1970

NEJM  20 May 2010  Vol 362

1863   British studies of endovascular versus open repair of abdominal aortic aneurysm dominate this week's New England Journal: there are two of them, or three, if like Andrew Marvell you regard the Netherlands as just a few bits of England that have washed up on the other side of the North Sea:
HOLLAND, that scarce deserves the name of Land,
As but th'Off-scouring of the Brittish Sand;
And so much Earth as was contributed
By English Pilots when they heav'd the Lead;
Or what by th' Oceans slow alluvion fell,
Of shipwrackt Cockle and the Muscle-shell;
This indigested vomit of the Sea
Fell to the Dutch by just Propriety.
The Character of Holland (?1653)
In the interests of friendly relations between our two kingdoms I should hasten to add that the British and the Dutch studies here concur and that renewed naval conflict is unlikely to be required at this time. The British data come from two studies called EVAR, with a median follow-up time of 6 years from randomisation to either open surgery or endovascular repair in EVAR-1. Although there was an initial mortality benefit from endovascular repair, this had disappeared by two years and there was a higher rate (30%) of additional procedures in this group. Unfortunately the rate of major operative complications of open repair (e.g. wound hernia) is not given, but the editorial (p.1930) uses US registry data to suggest that these are common and important and probably mean that the two procedures are equally good and equally hazardous.
http://content.nejm.org/cgi/content/abstract/362/20/1863

1872   EVAR-2 looks at the effect of endovascular repair or no treatment in patients with too much comorbidity to be considered for open repair. Those operated on die less from their aneurysms but overall die just as quickly as those left alone.
http://content.nejm.org/cgi/content/abstract/362/20/1872

1881   The Dutch Randomized Endovascular Aneurysm Repair (DREAM) study was no mere off-scouring of the British ones but an entirely independent large trial which reached exactly the same conclusion - initial benefit from endovascular repair, but long-term mortality equivalence and more complications.
http://content.nejm.org/cgi/content/abstract/362/20/1881

1901   Although I am forever deriding the contribution that genomics has so far made to the practical business of medicine, I am at heart an honest and nobly scientific creature whose greatest delight will be to be proved wrong in the fullness of time. So I commend to your attention this Brief Report (8 pages) on L-histidine decarboxylase and Tourette's syndrome. The investigators examined a unique kindred (family to you and me) with autosomal dominant heritability of Tourette's, which is otherwise a sporadic disorder with 10-15% recurrence in first-degree relatives. They were able to identify a genetic locus for Tourette's in the HDC gene which encodes for L-histidine decarboxylase. This may soon translate into a pharmacological treatment for Tourette's, because work has already been going on for years on the pharmacology of the presynaptic autoreceptor H3R and its role in Tourette's. So although genomics cannot take the credit for starting this search, at least it supports its logic.
http://content.nejm.org/cgi/content/abstract/362/20/1901

Lancet  22 May 2010  Vol 375

The Lancet has gone nobly global this week, mostly on the subject of tuberculosis. This is what the editor Richard Horton does best, and it won't benefit from the comments from a middle-English GP who rarely sees any TB. You're best to go straight to the papers of interest - about early treatment outcomes and HIV status in extensively resistant TB in South Africa (p.1798)
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60492-8/abstract

and the wide-ranging, magisterial surveys on p.1814 - tuberculosis control and elimination 2010-50
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60483-7/abstract

 - and on p.1830 - multi-resistant and extensively drug-resistant tuberculosis.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60410-2/abstract

For the general reader in relatively TB-free countries, Horton's "Offline" page (p.1766) continues to provide a pleasant diversion full of surprising insights and unexpected modesty. The "Art of Medicine" section on TB (p.1774) however strikes me as a great opportunity missed. We get a picture of TB in Australia in the past to compare with HIV now. What we might have had  is a discussion of the effect of TB on literature and music: imagine Keats growing old writing verse plays to rival Shakespeare, and epics of the soul complete and luminous, of the sort he started to sketch with Hyperion. Schiller likewise in Germany would have created magnificent new texts set to music by Weber (and by Schubert too, had syphilis been conquered). A golden age of gentle humanity would have settled over central Europe as these men and Chopin continued to write masterpieces until their deaths in the 1870s and 1880s. Upstarts like Wagner would have found no place and Prussian militarism would never have taken root with deadly effect. Chekhov in Russia would have continued to celebrate and transform the human spirit right up to the 1940s, a decade of universal peace and plenty. Ah me, I fondly dream. Mycobacterium tuberculosis got there first.

BMJ  22 May 2010  Vol 340

1120   I started prescribing antibiotics in primary care in the mid-1970s, when the commonest ones were amoxicillin, erythromycin, cefalexin and co-trimoxazole. They usually worked, though we were told that in a few years they might become useless due to bacterial resistance thanks to overprescribing. Nowadays I prescribe the same antibiotics (though trimethoprim alone rather than as co-trimoxazole); they still usually work, though here is another paper telling us they might breed resistance in individuals and hence in the community. And that is true: the people for whom they work least well are the ones who most need them - those with recurrent UTIs, COPD and bronchiectasis, who get repeated courses. But nobody is suggesting we should stop giving them to these people. Whether we should give fewer antibiotics to healthier people is a moot point, about which I have no strong convictions. I don't think it greatly matters, as they will get them much less frequently than every 12 months. That seems to me the message of this immensely diligent meta-analysis, though it's not the conclusion of the authors, who take the Orthodox Calvinist position.
http://www.bmj.com/cgi/content/full/340/may18_2/c2096

1122    Accelerated recovery from ankle sprain is the one subject I have contemplated the most in the last month. I hobbled home after I had done it and observed the increasing swelling with bemusement. I went to bed but woke up with the pain and found myself unable to bear weight, so I went to A&E. It was only after a clear X-ray that I had enough sense to put some ice on it and keep it up properly. Blood breakdown products tracked around in the most interesting fashion over the next few days, causing a stinging inflammation wherever they emerged, remote from the point of injury. This is my patient narrative and I hope it has increased your empathy. Now for the business: does early therapeutic exercise in the first week produce a better functional outcome than standard PRICE? I can't really answer the question personally as I got very bored of lying with my leg in the air and reading a learned work about Akhenaten, and besides I was on my own and needed to eat. This randomised trial found some functional advantage for active rehabilitation after one week, little difference thereafter. I suspect it doesn't make much difference in the real world - ankle sprains get better whatever you do.
http://www.bmj.com/cgi/content/full/340/may10_1/c1964

1124    Multiple sclerosis is a disease of sun-deficient countries, which has given rise to the hypothesis that low maternal vitamin D during pregnancy may play a role in its pathogenesis. Support comes here from an historico-metereological study from a MS register which gives the dates of birth of Australian MS patients. Examine this and then estimate the amount of ambient ultraviolet radiation during gestation from weather records and you get the expected inverse link with MS, particularly if you look at the first trimester. Ingenious.
http://www.bmj.com/cgi/content/full/340/apr29_1/c1640

1125   When I first conceived a BMJ series to be called "Commoner Than You Think" 8 years ago, high on the list was antiphospholipid syndrome. But when the series finally emerged as Easily Missed, I was told that the syndrome was already earmarked for a Clinical Review, and here it is. This is a much better format for going into detail about this intriguing and distressing range of conditions, which can be anything from recurrent miscarriage to multiple arterial infarcts, first identified 27 years ago. This article from Holland continues the high standards of these BMJ disease reviews. But I gawped at the advice given in the final sentence:
"We advise that any patient with a suspected antiphospholipid syndrome should be seen by a multidisciplinary team of specialists that ideally includes a rheumatologist, haematologist, neurologist, nephrologist, and obstetrician for diagnosis, treatment, and education." Any patient with suspected antiphospholipid syndrome? Surely this list should include a porcine aviation facilitator.
http://www.bmj.com/cgi/content/extract/340/may14_1/c2541

Ann Intern Med  18 May 2010  Vol 152

621   Deprive a chap of androgen, and his bones begin to thin. So at what point should you be thinking of using a bisphosphonate? Here's a cost-effectiveness modelling exercise from the USA, based on the case of a 70-year-old man with prostate cancer starting anti-androgen therapy. The advice that emerges from the Markov state-transition model is to do bone density measurement first and only give alendronate to men who are osteoporotic already, or at high risk of hip fracture. But to apply this to a UK context, you would have to factor in an entirely different set of costings, which somebody now should do.
http://www.annals.org/content/152/10/621.abstract

630   From time to time I have tried to fill you in with the emerging literature of non-invasive testing for coronary artery disease, and in particular computed tomography coronary angiography. This may grow in importance as stress testing fades away under the new NICE guidelines. But like all tests, its predictive value depends on the pre-test probability, in this case of coronary artery disease. Once again we go to Marvell's Holland to get a good thorough Bayesian analysis of the value of CTCA based on their patients in Rotterdam. They conclude that this kind of imaging is most useful in the intermediate probability range. Those with low pre-test probability can be given the all clear after a negative stress test; those with a high probability should get a proper angiogram.
http://www.annals.org/content/152/10/630.abstract

Plant of the Week: Erodium chrysanthum

The erodiums are small members of the geranium family, with pretty and usually abundant flowers over a long period. They make ideal fillers for every small space of bare earth you may happen to have. Most of them have colours in the usual hardy geranium range from white through pink to streaky purple. But this Greek erodium has flowers of pale primrose, and to my mind it is the best.

I once found it on a market stall and then lost it during a hard winter. I looked for another but a specialist alpine nurseryman assured me he knew of no such plant. Then I found it last year in some garden centre, and its wonderful cut silvery-green foliage has shown no fear during the worst winter for three decades: it even started flowering in late March, and it will continue until autumn. I suspect it comes with varying degrees of vigour and hardiness. Anyway, look out for it and buy it as soon as you see it. Despite my experiences, the Plant Finder classes it as "widely available".

JAMA  12 May 2010  Vol 303

1815    We all have our deficiencies, and in the case of people living in the British Isles, these generally include vitamin D. Correcting vitamin D deficiency is therefore a matter of national importance, but another British deficiency is the lack of a National Formulary with any useful doses of vitamin D. One way our local endocrinologists get round this is to give people an annual overdose of 500 000 units of cholecalciferol which boosts their vitamin D levels all the year round. However, this Australian study of nursing home residents suggest that this is not a good idea: those randomised to the overdose showed an increase in falls and fractures especially in the first 3 months. In an accompanying editorial, various reasons are suggested, both biochemical and behavioural - for example, that the vitamin made the residents friskier and riskier. However, I think the urgent question flagged up is: how else should we take the extra vitamin D that most of us need?
http://jama.ama-assn.org/cgi/content/abstract/303/18/1815

1841    Mortality from coronary heart disease is falling throughout the developed world, which includes Canada. Here a group of Ontarians examines why that should be, by looking at aggregated data for Ontario between 1994 and 2005 (adults aged 25 to 84). As in previous studies, a model called IMPACT explains the great majority of the decrease by a fall in risk factors - except obesity and diabetes, which move in the opposite direction - and an improvement in treatment. We are getting most of this right, then going home and eating too much and doing too little.
http://jama.ama-assn.org/cgi/content/abstract/303/18/1841

1848   For many years our practice was blighted by the proximity of a private allergy clinic from which worried middle class patients would emerge asking us for more RAST tests and treatments for something called candeeder in their bowels. A Clinical Review of the diagnosis and management of common food allergies confirms that this is still a largely science-free area. What science there is and is not is carefully teased out in this systematic review, and nicely tabulated. A few interventions for babies and children have been studied with tolerable rigour, but, to quote the Abstract, "In high-risk infants, hydrolyzed formulas may prevent cow's milk allergy but standardized definitions of high risk and hydrolyzed formula do not exist. Conclusion: The evidence for the prevalence and management of food allergy is greatly limited by a lack of uniformity for criteria for making a diagnosis."
http://jama.ama-assn.org/cgi/content/abstract/303/18/1848

NEJM  13 May 2010  Vol 362

1761    Bisphosphonates are good drugs and undoubtedly prevent many fractures in the osteoporotic, but their amazing effectiveness at poisoning osteoclasts is a bit worrying. If osteoclasts are best poisoned, why do we have them? Two rare but worrying adverse effects of this drug class have emerged: osteonecrosis of the jaw (associated with high doses and poor dental hygeine) and more recently fractures of the shaft of femur. The latter can occur without any obvious trauma, as one did in a patient of mine recently. This study is a secondary analysis of three bisphosphonate trials aimed at determining the size of this risk: "very rare" is their conclusion. But worth being aware of.
http://content.nejm.org/cgi/content/abstract/362/19/1761

1795    Some studies of screening colonoscopy have shown surprisingly high rates of interval cancer especially in the left colon. Colonoscopists also vary quite widely in the time they take over the procedure and also as to whether they reach the caecum and how many adenomatous lesions they remove on average. This study used a regression model on data from 186 Polish colonoscopists performing 45 000 colonoscopies to determine how the last two characteristics affect the rate of interval cancer. Reaching the caecum doesn't seem to matter, but removing adenomas does, as you might expect.
http://content.nejm.org/cgi/content/abstract/362/19/1795

Lancet  15 May 2010  Vol

1695    What is an ecass and what are ninds?  Clues: ecass is a word from the lost language of Atlantis, and ninds is an epithet. Possibly. Anyway, set a lot of very eminent researchers to work on ECASS, ATLANTIS, NINDS and EPITHET and you get a good idea of how effective alteplase is when given for stroke. As you might expect, time is of the essence: high odds of benefit (2.55) when given within  90 minutes but no benefit and possible harm when given beyond 270 min.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60491-6/abstract

1704   Timor mortis conturbat me. In a few weeks I shall be sixty and nobody will care if I die. Premature adult death is defined in this marvellous global overview as death between the ages of 15 and 59, and to avoid it, you should migrate to Australia and avoid Africa and Russia. A good paper to ponder over:
The stait of man dois change and vary,
Now sound, now seik, now blith, now sary,
Now dansand mery, now like to dee;
Timor mortis conturbat me.
from Lament for the Makaris by William Dunbar c1500
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60517-X/abstract

1721   Sooner or later, you'll send off a blood test and get back a report of monoclonal gammopathy of uncertain significance, and you will have the unenviable task of explaining this to the patient, or sending her/him off to a haematologist to do it for you. More than 3% of people over the age of 50 have MGUS, and while its significance is truly uncertain for the great majority, a few of those with light chain gammopathy do go on to develop multiple myeloma (0.3% per 100 person-years). A great many more of them - 23% - have impaired renal function.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60482-5/abstract

1729   Coming up to Oxford as a callow youth of 18, my first encounter with its fabled tutorial system was to sit with a few others and be tossed bits of bone to identify by a rotund and choleric anatomy tutor. As soon as I could, I changed to English literature. My next encounter with bone was in the orthopaedic operating theatre, after I had changed back to medicine. Here one learnt that bone is hard and bloody: after that, it's just something that shows up on X-rays. Unfortunately I suspect that most doctors have a similarly sophisticated understanding of this complex tissue, and we have certainly been lured into supposing that dual energy X-ray penetration of trabecular bone (the inside, holey bit) is a good measure of bone strength. This Australian study of intracortical remodelling and porosity in the distal radius and post-mortem femurs of women shows that we are missing a lot by failing to take into account the bony cortex. As age progresses, the cortex becomes thinner and less well organised, in men just as much as women. Most of this is missed with current methods of bone density measurement.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60320-0/abstract

1737   As I seem to have slipped into valetudinarian reminiscence mode, let me tell you that when I were a lad in Lincolnshire, you never saw a fat kid. Those who were accused of fatness by their peers then would pass for average now. We all played in the streets from dawn to dusk, there was no such thing as fast food (or edible food, I seem to remember), paedophiles were all safely confined to schools, choirs and scout huts, television was a flickering grey screen with people talking in posh voices, et cetera. In fact this sensible but naturally inconclusive review of childhood obesity goes much further than the usual grumbles about television and inactivity (though these are important) and is particularly useful for its Figure 1, which gives a simplified model of the leptin pathway which is probably the most important in controlling appetite and weight gain.

BMJ  15 May 2010  Vol 340

1070   If an elderly person falls at home, a paramedic is generally called in to pick them up and check them over, and the GP gets sent an A3 sized proforma diligently scribbled on which is then scanned into the computer, somehow. What we then need to do is refer these people for community falls prevention. Ideally this should include an assessment of the home for hazards and the provision of necessary aids, as in this study. Together with a medication review, medical assessment and training sessions, this will reduce the risk of future falls by more than half. The authors suggest that such a service should be offered to all elderly people who call paramedics following a fall, and how can one disagree?
http://www.bmj.com/cgi/content/full/340/may11_1/c2102

1071   This Norwegian study shows that there is a strong genetic element to hyperemesis gravidarum. If your mum had it, you are three times as likely to get it yourself.
http://www.bmj.com/cgi/content/full/340/apr29_1/c2050

1072   It's a pity that this weak study from Kaiser Permanente is published in the British Medical Journal, where it will be used, no doubt, to demonstrate that the removal of financial incentives in the UK will result in a fall in clinical quality indicators. It proves no such thing. It simply demonstrates that when something is an established part of good clinical care, it should be routinely provided without reference to doctors and their pay, as illustrated by screening for diabetic retinopathy. In a properly organised health system, interventions of proven benefit should happen automatically, and doctors be left free to exercise decision-making skills which lie beyond scrabbling for money.  
http://www.bmj.com/cgi/content/full/340/may11_1/c1898

1073    Male, lower social class medical students who fail exams are becoming a great rarity these days, from my limited experience. I think we need more of them. In the present system, these characteristics apparently predict later professional misconduct, and the thought police are hovering to eliminate unconventional medical students with the help of the GMC (see p.1064). On second thoughts, medical schools should only select ladies from good homes.
http://www.bmj.com/cgi/content/full/340/apr27_1/c2040

1074   Percutaneous endoscopic gastrostomy (PEG) has become a common procedure these days, though not without considerable hazards, as tabulated in Box 2 of this review. This is a clear and thoughtful account of the technique itself but above all of the often difficult decision-making which is involved in PEG insertion, especially if the patient does not have the capacity for consent.
http://www.bmj.com/cgi/content/extract/340/may07_2/c2414

Arch Intern Med  10 May 2010  Vol 170

747   I don't suppose that many readers actually read the Archives of Internal Medicine, or if they do, it's for the odd article of interest, among a sort of B-list for JAMA. I've long had a soft spot for this journal and I'm hoping it will grow in stature and importance with a new series called LESS IS MORE. As the name implies, this will question our over-use of many interventions. This week's target is proton pump inhibition. It's rather wonderful that by inhibiting the movement of two up-quarks glued to a down quark we can stop ourselves getting heartburn at night. But such tinkering on the quantum level is not without its hazards. Women taking PPIs show a small but significant increase in fractures of the spine, forearm and wrist, according to this analysis of data from the Women's Health Initiative.
http://archinte.ama-assn.org/cgi/content/abstract/170/9/765

772   A much larger hazard ratio applies to the use of proton pump inhibitors and the risk of recurrent Clostridium difficile infection. This retrospective study looked at the New England Veterans Healthcare database and puts the additional risk around 40% - higher if you are very old. It is followed (p.784) by another study which shows a dose-related increase in hospital-acquired C difficilefor patients given PPIs. At the very least we should try and avoid their use in people who have had C diff or who are at high risk due to antibiotic therapy.
http://archinte.ama-assn.org/cgi/content/abstract/170/9/772
http://archinte.ama-assn.org/cgi/content/abstract/170/9/784

804   Stroke medicine is an invention of the last two decades, spearheaded by neurologists in large centres who strove to prove that stroke units provide the best outcomes. It has taken a long time to tease out the processes of care which lead to better survival and better functional outcomes in stroke care, and this paper identifies a new one - detection and treatment of hypoxia. The other two factors which affect survival are attention to swallowing and DVT prophylaxis. It should be perfectly possible for any small hospital with CT scanning, alteplase, and a properly run ward with a good physio service to do as well as a stroke unit.
http://archinte.ama-assn.org/cgi/content/abstract/170/9/804

813   In my part of the world, and in Switzerland where this study was carried out, the great majority of patients with hip fractures are vitamin D deficient. Here 173 patients were randomised to receive either 2,000iu of vitamin D daily or 800iu, with or without extended physiotherapy. Extended physio reduced the number of falls, but not hospital admissions; whereas higher dose vitamin D reduced hospital admissions by 39%. I shall start taking twice as much immediately, to keep my hips intact and myself out of hospital.
http://archinte.ama-assn.org/cgi/content/abstract/170/9/813

Plant of the Week: Hyacinthoides non-scripta

There is no lovelier thing than an English bluebell wood in dappled sunshine, scented and buzzing softly with gratified insects.

Don't be tempted to introduce these plants into the ordered areas of your garden, where they will grow randomly and show their worst features, such as invasiveness and floppy leaves. Just go walking in the woods at this time of year.

JAMA  5 May 2010  Vol 303

1699   Pneumococcal vaccination does not reduce the risk of myocardial infarction or stroke in men. I didn't find this very interesting, because it had never entered my head that it might. And now this theory can leave my head once more (storage space being at a premium), thanks to a follow-up study of 84 170 men signed up with Kaiser Permanente in California. But the fact that should stay in my head and yours is that there is evidence from several studies that influenza vaccine protects the arteries of the heart and brain, according to the first sentence of this paper.
http://jama.ama-assn.org/cgi/content/abstract/303/17/1699

1707   Life expectancy in cystic fibrosis is increasing steadily due to incremental improvements in management, especially the use of antibiotics. The colonising bacterium most feared in CF is Pseudomonas aeruginosa, for which azithromycin is usually the antibiotic of choice. This randomised trial looked at the effect of 24 weeks of azithromycin in CF patients (mean age 10.7) withoutchronic Pseudomonas infection. Used in such children, the antibiotic seemed to have no benefit in terms of lung function or episodes of infection, though there was a small improvement in cough.
http://jama.ama-assn.org/cgi/content/abstract/303/17/1707

1723   Gastric cancer has gone out of fashion in the West, having been a major killer in past generations, as it still is in other parts of the world. We tend to assume that it will continue to decline, but data from the USA show that there is one group that bucks the trend. In white Americans aged 25 to 39, the incidence of (non-cardia) gastric cancer has actually increased.
http://jama.ama-assn.org/cgi/content/abstract/303/17/1723

1738   I shall shortly be retiring from clinical practice but will continue to live on the northern foothills of the Cotswolds, well known for their odd place-names. Chipping Ignorance or Much Dabbling would be the sort of names to suit me. Amongst the things I've dabbled in is coeliac disease, leaving Harold Hin to do the serious, ignorance-chipping work which forms reference 8 in this systematic review of diagnostic testing for coeliac disease in patients with abdominal symptoms. The radioimmune assay for endomysial antibody (IgA) that he used was derived from monkey oesophagus and proved pretty infallible. It is still the best test, as validated by small bowel biopsy. Next best is the anti-tissue transglutaminase (IgG) antibody, while antigliadin is unreliable in primary care or other relatively unselected populations.
http://jama.ama-assn.org/cgi/content/abstract/303/17/1738

 NEJM  6 May 2010  Vol 362

1663   O no, not another Stent Fight! This time it's between everolimus and paclitaxel. The Lancet would have us believe that poor old Paclitaxel has already been knocked out of the ring by Sir Olimus. Now it's the turn of his son, Ever. And there's another son waiting in the wings, Zotar. Why can't the Olimus clan leave poor old Paclitaxel alone and fight each other? Leaving the rest of us to read about better things than drug-eluting stents.
http://content.nejm.org/cgi/content/abstract/362/18/1663

1675   Not that non-alcoholic steatohepatitis (NASH) is a better thing. The best piece I've read about it was titled "NASH may be Trash" and nothing I have read subsequently has altered that impression. A small percentaqe of patients with NASH allegedly progress to cirrhosis and eventually to liver failure; a somewhat larger percentage progress to diabetes. Now the only way you can really tell what is happening in the liver is to biopsy it, which is an uncomfortable and potentially hazardous procedure. The alternative is to wait for liver failure to occur, which happened to only one of the 247 subjects in the 96 weeks of this study. That patient was in the group randomised to vitamin E, which nonetheless seemed to bring about the most histological improvement in the others. Another group received pioglitazone which had little effect on their liver biopsies but seems possibly to have prevented some diabetes at the expense of weight gain. Personally, I would advise anyone with fatty changes on their liver ultrasound to avoid hepatologists and clinical trials.
http://content.nejm.org/cgi/content/abstract/362/18/1675

1708   So, that was it: we had an influenza pandemic, and if it hadn't happened in the summer with all the world watching, we might hardly have noticed. I guess we would have been puzzled that a few children and pregnant women got so seriously ill and died, but if it happened in winter and we didn't have serotyping to identify H1N1 we might have called it a mild season. This excellent review sums up all that we have learnt about this virus, and identifies the main gaps. The most significant is the lack of a simple, cheap and highly accurate test for this subtype as it continues to spread around communities in coming years, causing no symptoms in a third of cases and mild ones in most of the rest.
http://content.nejm.org/cgi/content/full/362/18/1708

Lancet  8 May 2010  Vol 375

1609   Although I'm generally against the use of abbreviations in medicine as in most other areas, there are some that are just part of life and that everybody understands. Take MMR for instance. Would you give it to your child? I hope not, because it stands for maternal mortality ratio, if you happen to be an epidemiologist. This is a fascinating international survey of maternal death, a very sensitive indicator of the quality of maternity services. The worst figure (per 100,000 live births) comes from Afghanistan, at 1575. The best comes from Italy at 4, and Sweden and Australia at 5, which we can take as a sort of gold standard. The UK comes in at 8, flat for the last 20 years, while the USA stands at 17, up from 12 in 1980. There are lessons to be learnt here: we need to be more like Italy and less like the USA.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60518-1/abstract

1624   I do wish this study had appeared before the UK started its new policy of bowel cancer screening using faecal blood testing. Unlike the pilot studies for that, it shows a really convincing reduction in mortality from colorectal cancer - 43% in those that had the test. It seems to me that this large randomised trial (170 432 men and women in 14 UK centres) demonstrates that once-only flexible sigmoidoscopy, between the ages of 55 and 64, is the best option short of full colonoscopy. The present screening choice is bloody crap, frankly.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60551-X/abstract
To see what patients think of the faecal blood screening programme, see:
http://www.healthtalkonline.org/Cancer/Bowel_screening

1634   Seeing 101 patients might be tiring and depressing, but at least you would see some tangible results from time to time. Compare this with analysing 101 studies to assess the -1131T(rs662799) promoter polymorphism of the apolipoprotein A5 ((APOA5) gene in relation to triglyceride concentration, several other risk factors and the risk of coronary heart disease. You could be one of about a hundred international academics exchanging e-mails in your office between strolls out for cups of coffee. Your name could appear in tiny print at the end of a paper in The Lancet. This paper hints that there might be a weak genetic linkage between triglyceride-mediated pathways and coronary disease. How satisfying.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60545-4/abstract

BMJ  8 May 2010  Vol 340

1014   The place of testing for human papillomavirus in women invited for cervical screening is a topic that is bound to keep coming up in the future, but I'm not your best guide to it. A randomised study nested in the Finnish cervical screening programme discovered more CIN III lesions when HPV testing was incorporated for women under the age of 35. The net benefit of this is hard to judge, and you certainly won't be able to find it in the printed pico version. In fact reading the whole paper in PDF, I don't think anyone quite knows.
http://www.bmj.com/cgi/content/full/340/apr27_1/c1804

1015   This pico business had me piqued right from the start, but now the BMJ editors have kindly responded to my groans by inviting me to contribute to their revision process, so I should show some courtesy in return. However, I can't avoid recommending that you ignore the short version of this and the subsequent paper altogether and go for the real thing. The diagnosis of serious bacterial infection in young febrile children is one of the most important jobs we do. If we are to do it better we need to learn lessons from large, well-conducted studies, of which this is a fine example. But few of us work in a tertiary Children's Hospital with 220 beds, like Westmead in New South Wales, with access to immediate investigations. We diagnose pneumonia by listening to the chest, not necessarily by getting a chest X-ray. We probably tend to overprescribe antibiotics, rather than under-prescribe them like the well-behaved juniors in the Westmead emergency room. But there are still many lessons to be learned, and they have devised a diagnostic tool that we need to try out in our own setting. So, as the BMJ won't do it for you, print off the PDF of this article, and mull over it during your next out-of-hours session triaging poorly toddlers.
http://www.bmj.com/cgi/content/full/340/apr19_2/c1594

1017   For qualitative research, you need to get a feel of the quality, and again you can't do that from a pico version. You may respond by saying that you're a proper doctor and don't have time for page after page of ums and ers and sociological jargon, which is yet another reason you should print off the PDF of this paper. I defy you to read it without getting some new insights into the way people cope with the horrific event of a sudden bereavement due to trauma, violence or suicide. And you'll also be filled with admiration for Alison Chapple who spent several hours each with 80 people who had gone through such experiences  to find out the effect of viewing the body afterwards. You may also want to see the interviews themselves on www.healthtalkonline.org - a fantastic site which I now have the privilege of working on myself. Take that as a competing interest statement if you like, but do use this site for learning, teaching and helping your patients.
http://www.bmj.com/cgi/content/full/340/apr30_2/c2032
http://www.healthtalkonline.org/Dying_and_bereavement/Bereavement_due_to_traumatic_death

Ann Intern Med  4 May 2010  Vol 152

578   It's often a surprise to doctors entering general practice that shingles is common at all ages. However, it rarely causes more than transient discomfort in immunocompetent people under the age of around 60. In the USA, the use of a herpes zoster vaccine is now being encouraged to prevent the uncommon but nasty condition of persistent post-herpetic neuralgia. However, there is likely to be a certain amount of pain for this possible gain: in a placebo-controlled trial, 48% of active drug recipients experienced a significant local reaction. Serious reported harms, however, did not differ from placebo.
http://www.annals.org/content/152/9/545.abstract

578    Now one thing you can be quite sure of is the benefit of warfarin for recurrent venous thromboembolism. Wrong: this big systematic review of the randomised trials shows that in the first three months of treatment, those given warfarin stand the same risk of a major bleed as the untreated do of another thromboembolism. It isn't very straightforward at all. If you want to use anticoagulation, as I'm sure you do, make sure you monitor it very carefully.
http://www.annals.org/content/152/9/578.abstract

Plant of the Week: Clematis montana  

The season of miracles is now well upon us, spoilt only by overcast skies and biting north-easterly winds. In towns and villages cherry blossom is going brown and tulips are shedding their petals, but across walls of stone or brick great wisterias hang with scented blossom, and a few fortunate trees and roofs are covered with blankets of clematis blossom, ranging from pure white through pink to nearly purple.

I wouldn't really recommend letting a montana clematis take over your roof unless it is thatched. In fact I don't think these big sprawling climbers should really be trained up house walls, but rather be draped over garden walls and fences and trees and shrubs you don't much like. There are lots of good cultivars, all of them vigorous and hardy, so take your own pick. But be sure to do so at a time they are in flower, so that you can check them by nose. What you need from this clematis when fully grown is a great sheet of vanilla scent, dusty and almost choking, like inhaling sugar.

 

 

 

 

 

 

 

 

RSS Subscribe to the RSS Feed for this page

 

 

 

 

Page last edited: 08 June 2010