Journal Watch - December 2009

JAMA  16 Dec 2009  Vol 302

2557    Tarenflurbil has died. He did not live long: just enough to get a phase 3 trial which showed no effect on cognitive decline in mild Alzheimer's disease. His was quite an attractive name, as silly names go, but one you don't need to remember, unless you are stuck for something to write on a pet's tombstone:
      Here lies my short-lived favourite gerbil
      Who shares his fate with tarenflurbil.

2565    But what of leptin? (Rhymes with slept in.) You may remember that there was a bit of interest in this brain hormone a few years ago when it looked as if it might be the key to appetite and weight control by its actions on the hypothalamus. It's turned out to be a good deal more complicated than that, and leptin secretagogues have yet to find a clinical use, but there is a definite link between leptin levels and incident Alzheimer disease. It's an inverse relationship discovered in a Framingham study subset of 750 people of mean age 79 years who had their plasma leptin levels measured and who have been followed up for a mean of 8.3 years. Lots of leptin is a good thing: it keeps you thin and it keeps you bright. The paper tries to find ways of explaining this, but many more investigators must leap in before we fully understand leptin.

2573   Cystic fibrosis is an autosomal recessive disease and testing for carrier status has been available for the last two decades, though screening policies differ widely, sometimes in adjacent regions of the same country. This study reports on a natural experiment which arose in north eastern Italy when Veneto adopted a conservative carrier testing policy while next door in the Padua area, much wider testing was carried out. In both regions babies are also tested for CF at birth, and in the Padua area there were half as many as in the Veneto area. In other words, couples do make reproductive choices based on their carrier status.

2580    For those of you who like to think about the way evidence informs our daily clinical practice, here's an important and thoughtful paper called Heterogeneity Is Not Always Noise. The examples here are complex interventions requiring behavioural change in an entire organisation "“ the kind of thing that NHS managers with an average tenure of 3 years frequently start up and leave someone else to sort out. This paper argues that such interventions can be usefully studied provided you use the right methodology. In the current political climate it may be hard to commit to the right blend of statistical sophistication and qualitative analysis, but somebody needs to do it if we are to know which interventions actually achieve health care improvement as opposed to Innovation Bid funding.

NEJM  17 Dec 2009  Vol 361

2405    We can take comfort in the fact that if a virulent form of pandemic influenza ever arises, the technology is in place to create an effective vaccine for survivors of the initial wave. In most other ways, it seems to me that creating an H1N1 vaccine has been wasted effort "“ a dummy vaccine for a dummy pandemic, because the disease itself is generally so mild and so difficult to distinguish from a range of other upper respiratory viruses. Vaccines have been developed rapidly and successfully in Australia (here), China (p.2414) and the UK (p.2424). They are associated with a high incidence of local reactions and headache and one woman in our practice has been in hospital with encephalitis. I have yet to have mine.

436    Iron is important, not just for your blood but for all the red bits inside you, such as your skeletal muscle, your heart, your kidneys and liver. For reasons we don't fully understand, people with progressive cardiac impairment tend to become progressively iron depleted, and this Europe-wide trial sought to find out whether intravenous iron (ferric carboxymaltose) might improve quality of life in mild-to-moderate systolic heart failure. The blinding procedures were worthy of a medical school pantomime "“ they used black syringes because the iron solution was black, but they couldn't get black tubing so "a curtain (or something similar) was used to shield the injection site from the patient's view." The result: those who, all unawares, got the black liquid felt better than those who didn't. Since the New England Journal does not recognise Christmas, this study gets a serious editorial on p.2472.

Lancet  19/26 Dec 2009  Vol 374

2055    This important American trial helps to define standard treatment for endocrine-responsive, node-positive breast cancer in postmenopausal women. Those given an initial course of chemotherapy with cyclophosphamide, doxorubicin and fluorouracil (CAF) survived better than those given tamoxifen alone; this is already widely reflected in standard management, though there may be a subgroup of women who do not benefit from chemo "“those with least tumour spread. Even for those with the highest number of involved lymph nodes, the benefit of this toxic chemo regime was not very large. The second question the trial addressed was whether tamoxifen works best given with or after CAF chemo. At a median of nearly 9 years' follow-up, there was no significant difference. The accompanying editorial looks critically at these data and calls for more precisely defined cohorts in future breast cancer trials.

2072    When I describe H1N1 influenza as a "dummy pandemic", that is not to say that it hasn't killed people or put a strain on health services "“ it just hasn't been nearly as bad as we expected from initial reports from Mexico. At 4 100 deaths worldwide to Sep 27, it seems rather more benign than seasonal influenza, whereas at its onset it killed several hundred Mexicans within a short time and a small area. Here is a retrospective analysis of the Mexican epidemic, with a number of odd features, such as the rapid falling off in lethality and the failure of a second wave to materialise when children returned to schools closed during the initial outbreak.

2115   All of which makes it very difficult to model the risk/benefit profile of H1N1 infuenza vaccines. Eighteen authors wade through evidence of variable relevance to reach a hesitant conclusion full of generalities and admonitions. I wonder how many of them have had the vaccine themselves.

BMJ  19 Dec 2009  Vol 339

It's a miserable business trying to read the Christmas issue of the BMJ online instead of sitting with it open by a roaring fire with drink in hand, slippers, faithful hound/inquisitive cat, fond relatives, new cardigan etc. Things just aren't witty in HTML on a monitor. I keenly await the arrival of my mustard coloured print copy with lots of red headlines in time for Christmas.

Arch Intern Med  14/28 Dec 2009  Vol 169

2053    Coffee prevents diabetes. So does tea, and so does decaffeinated coffee. Which is odd because you'd think that the only thing tea and coffee have in common is their caffeine. Anyway, take comfort in the fact that the evidence is strongest for proper coffee, according to this systematic review. No milk or sugar thanks.

2064    Say no to surrogate outcome measures, especially if they cause cancer. You could argue that all surrogate end-points are per se a cancer on the Body Medical, but this applies literally to annual measurement of coronary calcification in trials of preventive therapies. Ten trials are reviewed here, and they all used annual CT scanning, which is completely unethical (a) because it doesn't correspond to any hard end-points and (b) because it is bound to induce a substantial number of cancers (see paper below) in the trial subjects, whether they receive placebo or active treatment. This practice really does need to end immediately.

2078    So what exactly is the lifetime attributable risk of cancer from common CT examinations? This paper tries to give some answers, and also discovers that radiation doses are often higher than those generally quoted. Computerised tomography is not a harmless procedure. If, for example, you enrolled 270 women in their forties into a trial measuring coronary calcification each year, you would induce at least one cancer per year, with the figure increasing as the cumulative dose increased.

2087    I really like this study comparing prediction models for fractures in older women. It did the job properly: gathered 6252 women aged 65 or older and followed them up for 10 years. Then it compared the area under the ROC curve for a variety of fracture prediction models which include measurement of bone mineral density and those ("parsimonious") ones that don't. There is no difference. Save DEXA money through parsimony. There is an editorial (p.2094) asking if there should now be a fracas over the fracture score FRAX but I won't fag you with fragments of this fatuous and fractious affray.

2124    Here's an invited commentary, ostensibly about two rather weak papers but well worth reading as a summary of the benefits of physical activity "“ An Investment That Pays Multiple Health Dividends. These aren't just what you expect but include the prevention of recurrent colorectal cancer (p.2102) and the slowing of renal function decline (2096). About half of your fitness is genetically determined, but the rest is up to you. Says he, glued to a computer screen all weekend on your behalf.

2128    If you are interested in the potential harm of antidepressants prescribed to women (and you jolly well should be, because you prescribe enough of them), then here is a paper to look out and spend some time over. It is a careful compilation and stratification of data from women newly prescribed antidepressants in the Women's Health Initiative: a case-control study within a much bigger study. There is some comfort in the fact that cardiovascular harms do not differ between tricyclic and SSRI antidepressants in this study "“ though that may not be true of some TCAs. On the other hand, cerebrovascular and total mortality show a definite increase in the treated group.

2148   I shouldn't need to keep pointing this out, but the biggest favour you can do your smoking patients is to help them stop smoking. This big, cheerfully pragmatic primary care study allocated 7128 smokers to five different pharmacotherapies and measured abstinence at six months. The best result (29.9%) came from bupropion plus nicotine lozenges. Nearly as good at 26.9% was the combination of nicotine patches and lozenges. I would suggest we forget rationed monotherapy and go all out to save these people from premature death through addiction.

Ann Intern Med  15 Dec 2009  Vol 151

829    I've already given you my views on vaccination against H1N1 influenza a couple of times: for another analysis deeply informed by mathematical modelling and suchlike, you can turn to this paper and the one after (p.840) and then make up your mind whether to roll up your sleeve, assuming you haven't already.

854    When you look at the relationship between glycaemic control and cardiovascular outcomes in type 2 diabetes, you observe a striking dissonance between the observational data "“ showing a linear relationship "“ and the data from interventional trials, showing a complete lack of benefit from HbA1c lowering below about 8.5, let alone 7. I won't go into the detail, but this Italian study lies primarily in the observational category and does show a lower overall level of CV events in patients whose HbA1c stayed below 7. These are the fitter patients with less co-morbidity. If you stratify patients by Total Illness Burden Index, you find that the ones with most co-morbidity benefit least (i.e. not at all) from better glycaemic control. There are lucky patients with diabetes, who stay well controlled and get little illness, and unlucky ones, who stay unlucky whatever drugs you use to control their blood sugar. Control more important things instead.

861    Ever since the HOPE trial (2000) showed that ramipril improves outcomes in ischaemic heart disease, we have been ramming prils down all our IHD patients. Some have coughed back at us and we've moved them on to angiotensin II receptor blockers. For people with reduced systolic function, we know that ARBs have equivalent benefits to ACE inhibitors, but is this true of all patients with IHD? The answer from this systematic review is that we don't know for sure. ARBs probably aren't quite as good for secondary cardiac protection as ACEis.

Plant of the Week: Ilex aquifolium

For your Christmas walk, go to a wood with holly trees in it. In northern England, such places traditionally bear the name "hollins" and were once used by medieval peasants for their shivering animals to graze upon during winter. Hollinsdale, Hollinscough: it's grim up North.

In the effete south, holly trees will often grow to heights of 10m and more in the competitive environment of a deciduous wood, producing smooth grey trunks which are particularly lovely in snow and frost. If you have enough room for a holly in your garden, plant "Silver King" for its variegated foliage and all-year handsomeness. Mind you, you won't live to see it at its best, as this takes 100 years or so. However, it will bear you red berries, because this particular king is female.

JAMA  9 Dec 2009  Vol 302

2437    According to one theory, humans developed big brains because they could cook. This meant that they got much more nutrition from the food they hunted and gathered and could cease to jump from branch to branch having constant diarrhoea and settle down to walking about and developing new forms of consciousness and firm stools. Plants and seeds that were poisonous when eaten raw could be valuable food sources when cooked - a good example being the annual legume Glycine max or soy. The soy bean is a small unattractive object which contains a lot of isoflavones, sometimes known as phyto-oestrogens. The more you boil it, the more edible it becomes, and the more the isoflavones get broken down. The soy protein products that are used to bulk up foods in the West contain hardly any oestrogenic chemicals, whereas the kinds of soy which can form a substantial part of the diet in Shanghai contain a lot more. The surprising thing about these oestrogens is that they seem to improve survival in women with breast cancer according to a study of soy intake carried out in Shanghai. It's unlikely that anyone would want to consume soy at the top quartile level of this study, but for those who were worried that a bit of the stuff might cause breast cancer to recur, this study is soy reassuring.

2451     Cluster headache is the rather prosaic name given to an agonising form of migraine which prostrates patients with unilateral head pain and autonomic eye swelling, watering and nasal congestion. It comes in bouts (clusters), usually every few years but sometimes much more frequently. Over the years I've treated it with opioids and injections of sumatriptan, but recently high-flow oxygen has been suggested as the treatment of choice. This British trial shows that it works better than a placebo of high-flow air, though there was a huge drop-out rate. It seems reasonable to ensure that everyone who suffers from this excruciating condition should have an oxygen supply handy, and perhaps a nasal triptan spray and a bottle of Oramorph for good measure.

NEJM  10 Dec 2009  Vol 361

2309    Cangrelor is like a name from an Arthurian legend. The castle of Cangrelor was mightily girt, and surrounded by a loathly mire that no man might pass, save that he knew the secret way. Wherefore Sir Perceval was wont to return by another path, until he spied a gleam that came from the chapel of that Castle, and bethought himself that this might perchance be the Holy Grail. So Sir Perceval knelt upon the ground and signed himself with the sign of the cross, purposing to dare the loathly mire and attain upon Cangrelor. But lo! at the same sign the Castle of Cangelor sank into the mire and was seen no more and Sir Perceval said unto himself that this was surely the work of Merlin and beshrove himself and thanked God for such a deliverance. Which is roughly speaking what has happened to the real cangrelor in this week's New England Journal. It was billed to be the Holy Grail of perfect platelet inhibition, replacing clopidogrel which depends on CYP metabolism, but in two trials it proved itself no better when given to patients undergoing percutaneous coronary intervention. The Quest goes on.

2343     But in the Quest for the Perfect Anticoagulant, a Grail is declared! Let all true knights fall before dagibatran, humbly giving thanks for a direct thrombin inhibitor which will replace warfarin and make INR testing a thing of the past. The trials so far have reported equivalence or superiority in prophylaxis for lower limb surgery, but this one (RE-COVER) went for a direct comparison in the treatment of acute venous thromboembolism. A fixed dose of dagibatran was compared with INR-guided warfarin over six months following VTE. There was no significant difference in bleeding or in recurrent VTE. As I've said before, Boehringer-Ingelheim should do the world a favour by pricing this drug at exactly the equivalent cost to warfarin and the cost of its monitoring. At a stroke great number of doctors, nurses and lab technicians would be able to do something more valuable and less boring than monitor anticoagulation. B-I could pocket billions and have a good conscience at the same time. The Holy Grail indeed.

Lancet  12 Dec 2009  Vol 374

1967    It's clear that the more you inhibit clotting, the more you will encourage bleeding. Perhaps that's even the origin of the term "bleeding obvious" - and yet we don't really know to what extent. The Danish nationwide register is the place to look, because it records which antiplatelet agents and vitamin K antagonists were given to which patients following myocardial infarction, and how many of them were readmitted to hospital with major bleeding. The really dangerous combination is warfarin with clopidogrel: don't use it unless you really have to. In ten years we may know the exact risk from dagibatran, but I don't expect much will be heard of cangrelor.

1975    It's time we took cervical cancer off the world map. To see that map, turn to page 1949, with the poorest countries showing the highest risk: and the means to abolish it is an effective human papillomavirus vaccine. The Glaxo-Smith-Kline HPV-16/18 ASO4-adjuvanted vaccine will probably do nicely, according to this 6.4 year study of its immunogenicity. As the editorial title asks, Global access to HPV vaccination: what are we waiting for?

BMJ  12 Dec 2009  Vol 339

Rarely in the field of medical research have so many people worked for so long to so little effect. There are several candidates for this accolade, but I think the prize must go to pre-eclampsia. We know an almost infinite amount about it, except what causes it and how to prevent it. The latest thing to emerge is that it might predict intra-partum thyroid insufficiency and the later development of the same, permanently. This may be mediated by soluble fms-like tyrosine kinase-1. Yes, I know you may not care; but we have to encourage researchers in PET or they might give up.

It's hard to believe, but less than a year ago the manufacturers of Tamiflu and Relenza were wringing their hands in the expectation that all their stock would soon become worthless as new variants of epidemic flu became resistant to existing neuraminidase inhibitors. Then with the arrival of H1N1 pandemic influenza every fruit machine in the entire gaming arcade spilt out money at once. I'm told that Sir Iain Chalmers, founder of the Cochrane Collaboration, still wants to believe in the concept of evidence-based politics. But I don't think he'll be able to take much comfort from this saga of panic-purchased antivirals - which the Cochrane review reveals to be a clear case of we're In The Thick of It, never mind the evidence, let's be seen to do something. Politics as usual, in other words. It began as a long tale of drug companies playing by inadequate rules, and ended with them hitting a jackpot that they hardly deserved. I haven't time to take you through it here, but it is very well told by Fiona Godlee in her editorials and in various pieces throughout this week's BMJ. Fiona calls for access to the raw data but in a Rapid Response Michael Power goes one step further and suggests that "Individual patient data (appropriately anonymized) from all trials of all health interventions should be readily available for scrutiny by scientists, investigative journalists, and others working for the public interest." Which might even include governments, in an ideal world.

Hepatocellular carcinoma for the non-specialist is what it says on the tin. And don't think you don't need to open this tin because it's relatively rare in the UK: its incidence here has more than doubled in the last 30 years. It almost always occurs against a background of hepatic cirrhosis and infection with hepatitis B or C and thus presents some chance of early detection by ultrasound screening of high-risk groups. The outlook for treated small tumours is actually quite good, but the world-wide burden of advanced disease remains pretty horrific.

Plant of the Week: Rudbeckia laciniata

These are the golden-headed marigold-like flowers that sometimes go by the name of Black-Eyed Susan, though her name is sometimes given to other plants as well. The Rudbeckias are sprawling perennials with abundant cheerful flowers of the most untasteful kind, bred widely in Germany and found in town markets rather than in posh nurseries and garden centres that do bronze piglets.

I got ours for about GBP1.50 in the local market and it has done splendidly near our kitchen window, clashing violently with a purple penstemon nearby. It is coarse. It is gay. It is delightful. And it is still in flower.

JAMA  2 Dec 2009  Vol 302

2345    Rarely has an issue of JAMA contained so little worth commenting on. I only point your attention to this article on the assessment of claims of improved prediction beyond the Framingham score because our overlords and paymasters sometimes direct us to use such scores, and perhaps some of you actually do. Personally, I seldom bother. Still less do I care about the veracity of claims that additional factors improve the score: and in fact this paper shows that there is scant evidence that they do. The authors don't enter into the Q-RISK debate arena because that's about a separate issue of generalisability to different populations. The fact is that in the NHS every major cardiac risk factor is subject to its separate irrational dictates. Treat all blood pressure if it is above a certain limit. Treat cholesterol only if it part of a risk score. Treat obesity with interventions which have been shown not to work, but not with bariatric surgery, which does work, and usually cures diabetes to boot. Treat smoking with small supplies of nicotine replacement rather than with drugs that have been shown to work much better. And so on. A percentage point or two in some score is never going to bring reason into this chaos.

NEJM  3 Dec 2009  Vol 361

2241    This week's New England Journal is no diamond mine either, but this study of advanced heart failure treated with a continuous-flow left ventricular assist device may be a pointer for the future. At the moment, if one of your patients with advanced HF becomes pulseless, you can predict with considerable certainty that he is dead. But increasingly you will encounter pulseless HF patients walking around alive, each of them carrying a briefcase. This does not contain native soil to allow daytime rest in a coffin, but heavy duty batteries to keep a continuous-flow pump going in the left ventricle. I know one such patient who had his briefcase pinched, but managed to survive. LVADs are still very expensive and fraught with problems, but this study shows that they do help patients with very poor LV function to live on and that they are much better than pulsatile devices.

2261    Mitral valve prolapse is commoner than you think. If Framingham Offspring data are generalisable, then your personal list of 2,000 patients will include no less than 50 with the condition, and of these about 5-7 will eventually develop significant mitral regurgitation. The traditional cure is mitral valve replacement, but this clinical review cites good evidence that mitral valve repair is a better option. There are nice illustrations, as indeed there are in the LVAD paper. It's all rather awe-inspiring and a definite incentive to become a heart surgeon in your next life.

Lancet  5 Dec 2009  Vol 374

1889    How would you like to retire in France? Bliss might it be that day to be alive, And to be old like very heaven. And that, basically, is that, according to this study. People in France who retire gain about 8-10 years of restored health (self-rated) and maintain it for at least 7 years. So I'm off into retirement next summer, though I shall stay in England. Peut-etre...

1897   Photons are a pain in the neck. They annoy us by being both particles and waves and by being in two places at once. Fifty years ago we learnt to control some of their wayward behaviour by means of lasers, which still have a magical sound to much of the population, myself included. And anything that sounds magical can be used to sell therapeutic devices. There is no known mechanism of physics or physiology by which a spot of red light can alleviate pain, but this careful systematic review of all the properly conducted studies of low energy laser treatment of neck pain is forced to conclude that it works. Shine the laser at the tender points in your neck and the pain diminishes and stays diminished for months. Nobody has the faintest idea why.

BMJ  5 Dec 2009  Vol 339

As our lives (i.e. our computers) grow more dominated with an agenda which has nothing to do with what patients come to see us about, we need a whole new way of working. This will include self-generating risk scores. The indefatigable Julia Hippisley-Cox is the British leader here with her system of QResearch practices and her infinite patience with the data they generate. She has identified and validated a long check-list of factors which account for about 64% of the variance in risk for osteoporotic fracture in women and slightly less in men. As our older patients check in to see us a central computer should tot them all up, go bing! and say "Are you aware that you may be at risk of fractures? Take vitamin D and calcium/see the practice nurse/get a DEXA scan/order a wheelchair immediately". Then it could pour out a mass of advice leaflets, X-ray request forms, and packets of pills. This would then leave patients free to come in and tell us the reason that brought them in the first place.

I continue to follow the literature about salt and cardiovascular disease with a furrowed brow. This is not quite science as I understand it. The interventional studies are weak. The whole-population observational studies are painstaking but their measurement methods are poor and there is endless room for confounding. Proponents of the salt hypothesis (see editorial) sometimes resort to quasi-religious discourse about WHO targets and the Neolithic diet, i.e. a mixture of authority and ignorance. We might do better to state simply that salt is a modifiable factor which we would probably benefit from modifying, whether the central theory is true or not, since we have no way of knowing. A bit like climate change, which The Lancet discusses this week.

In the Zoroastrian scheme of things, Ahura Mazda, the God of truth and light, is engaged (through man and the Bountiful Spirits) in a constant battle against Ahriman, who embodies the Lie and works through fallen spirits (daevas) and evil creatures. Each Zoroastrian pledges her/himself five times a day to side with Ahura Mazda by maintaining asha, or integrity, the highest good. Somehow we have come to accept a similar dualism in the world of medical research. This is strikingly illustrated in the debate between Ben Goldacre and Vincent Lawton on conflicts of interest in drug company research in this week's BMJ, and by this paper outlining good publication practice for communicating company sponsored medical research. It is assumed that the pharma companies will try to spin their trials to promote their products, and that the scientific community should accept money to collude with this, while at the same time insisting on publication guidelines and employing whole teams of meta-analysts to cast doubts on the results. A bizarre way of doing things, more Manichaean than Zoroastrian. Guidelines in these circumstances are like a pact with the devil and should at least give the devil some hard work to do, which these do not. Ahriman and his daevas will be laughing.

Foot complications are very prevalent in people with diabetes, and will remain so, however carefully we check their feet and control their blood sugar. Osteomyelitis complicates a good half of diabetic foot ulcers. Be generous with antibiotics and keep this excellent review article handy for your diabetic clinics.

Ann Intern Med  1 Dec 2009  Vol 151

775    Ping! goes the check-in computer: "Do you realise that you are at high risk for diabetes? Go on a diet/have fasting blood done/join exercise programme/see doctor/change your will." Out rattles a mixture of metformin and advice leaflets and gym subscriptions. "If you would still like to see the doctor about what you originally booked for, please proceed to waiting room." For those who would like to develop this computer programme, here is some essential reading - the development and validation of a patient self-assessment score for diabetes risk based on NHANES data. The necessary data about age, sex, history of high BP and obesity should already be on the computer and all that's needed is a score for physical activity.

812    Practices generally contain at least one GP who is a stern "most sore throats are viral" preacher plus at least one GP who prescribes the antibiotics two days later. It would be dangerous if it were otherwise, since the sore throats that are not viral can sometimes be life-threatening due to Lemierre's syndrome. Lemierre described a syndrome in adolescents and young adults which proceeds from pharyngitis to tonsillitis to suppurative thrombophlebitis of the internal jugular vein and metastatic spread of infection to the lungs. The mortality in his day (1936) was 90%. This syndrome is uncommon but not vanishingly rare, and it follows infection with Fusibacterium necrophorum in young people, which is as common as streptococcal throat infection, at around 10% of sore throats. I bet you didn't know that. Treat with a penicillin, cefalexin or clindamycin: add metronidazole if there is neck swelling.

784   Ping! goes the check-in computer: "Do you realise you may be depressed? Take just half a minute to complete this simple questionnaire." "Thank you. You are depressed. Pull yourself together/see a counsellor/do on-line CBT/try St John's wort/get a proper antidepressant/top yourself." Out splurges some self-help literature, advice on MIND courses, herbal cures, waiting times for CBT, citalopram. "If you still have the will to live and would like to see a doctor, proceed to the waiting room." Now if this is what you might like to trial in your practice, the US Preventive Services Task Force will not discourage you. Nor will it encourage you. It will sit on the fence about screening for depression in adults, in line with the large systematic review which it sponsored to seek evidence on the matter. There is very little to suggest that it affects hard outcomes.

Fungus of the Week: Pleurotus ostreatus

Once the frosts have come, few fungi are to be found, but most of those are edible. This is the largest, and grows on trees. It is the true oyster fungus and looks like the stuff you can buy in supermarkets, only bigger and more handsome with a cap of slate blue-grey.

I wish I could tell you that it tastes better than the cultivated supermarket varieties, but this I cannot promise. In my experience its flavour varies somewhat with the host tree and those that grow on a dead sycamore near us are not particularly pleasant. However, they can be impressively large and form a plate on which to serve the meats that go well with them. These include snails (excellent) and indeed oysters, though it's rather a waste of the latter, as they have to be cooked. For a very cheap and delicious dish you can use turkey mince. This tastes of nothing much but you can enliven it thus:

Trim off the fibrous stipe of your oyster mushroom, which should be at least 12 cm across. Soften it in a little oil and butter mixture with a scrap of garlic. In another pan fry the turkey mince mixed with a chopped leek, a little onion, cooked bacon finely chopped, and quite a lot of fresh thyme. Transfer the oyster fungus on to a heated dish and cover it with the fried mixture, adding a few drops of old Modena balsamic vinegar (no 12 if you can get it). Cover lightly with a mixture of grated parmesan and breadcrumbs and finish under the grill.





RSS Subscribe to the RSS Feed for this page





Page last edited: 03 January 2010