JAMA 3 Feb 2010  Vol 303

423     Everybody dies; quite a lot of people get myocardial infarction; few people get renal failure. This study of 920,985 people in Alberta ran for 35 months, so only 3% of them died. Proteinuria was more predictive of these outcomes than estimated glomerular filtration rate, and a combination of the two was especially predictive. But it's pointless to give the risk estimates because they were unadjusted for anything else. The tables are not much help, and you could spend a long time poring over this paper trying to work out how this might impact on clinical practice. Instead you should spend these hours writing to all your patients with an eGFR of less than 60 asking them to bring along a urine sample. In this way you can cause them stacks of anxiety for no known benefit, waste a lot of nurse time, and collect several thousand pounds of QOF income.
http://jama.ama-assn.org/cgi/content/abstract/303/5/423

438    "Evidence-based Medicine Requires Appropriate Clinical Context" declares the thoughtful editorial on this meta-analysis of deep vein thrombosis following a single negative whole-leg ultrasound. By lumping all the studies and patients from the community with those who have cancer or have had major surgery, the reviewers come up with a figure of 0.57% for VTE in the three months following negative ultrasound. But as the editorial points out, this is not the way we should work in real life. Systematic reviewers are always lumpers, but clinicians should always be splitters.
http://jama.ama-assn.org/cgi/content/abstract/303/5/438

448    If I ever get round to compiling the Good Death Cookbook from the recipes appended to these reviews, I shall have to confront the evidence linking dietary sodium with cardiovascular disease outcomes. All of it is observational; and according to this article, the studies are in equipoise. That's right: there are some studies showing cardiovascular harm from lowered salt intake; most are neutral; some show benefit. But there has never been a prospective randomised trial.
http://jama.ama-assn.org/cgi/content/extract/303/5/448

NEJM  4 Feb 2010  Vol 362

387    We have now definitely entered the age of effective oral treatment for relapsing multiple sclerosis. Effective means clinically effective , resulting in a clear reduction in disability over two years in this placebo-controlled RCT of fingolimod, a sphingosine-1-phosphate-receptor modulator that prevents the egress of lymphocytes from lymph nodes. Since we often need our lymphocytes need to achieve egress from their little nodes, you might expect a lot of infection-related adverse events; but in fact the biggest problems seem to have been bradycardia and macular oedema - both uncommon. Naturally, the safety jury will be out for some time yet.
http://content.nejm.org/cgi/content/abstract/362/5/387

402    The next trial of fingolimod was only a year long but this time it was compared with intramuscular interferon in a double-dummy double blinded RCT. And in this study two subjects receiving the higher dose of fingolimod died of infective complications - generalised herpes zoster in one case and herpes simplex encephalitis in the other. There were also six cases of macular oedema in the orally treated groups, all of which resolved, but it looks as if regular eye examinations are going to add to the cost of this drug.
http://content.nejm.org/cgi/content/abstract/362/5/402

416    Another promising oral drug for relapsing MS is cladribine. This is a fairly basic chemical (metabolised to 2-chlorodeoxyadenosine triphosphate) which interferes with DNA synthesis and repair. Surprisingly it is quite specific in its effects and knocks out CD4+ and CD8+ lymphocytes preferentially. Over the best part of two years, it was clinically beneficial in reducing the number and time course of relapses, compared with placebo. Inevitably it caused lymphocytopenia and there were 20 instances of herpes activation, none of them fatal, and one instance of TB activation, which was fatal. Expect many more studies over the next few years, including head-on efficacy and safety comparisons between Cladry Bean and Fingummybob.
http://content.nejm.org/cgi/content/abstract/362/5/416

440    In general, I can be fairly smug about my carbon footprint, and I read this article on jet lag with purely academic interest. Of course, if anyone wants to pay for this to change, they should contact me immediately. My lecturing fees, over and above first-class air travel and accommodation, are quite reasonable. I shall buy my own melatonin and try to follow the excellent advice summarised in Table 2 of this paper.
http://content.nejm.org/cgi/content/extract/362/5/440

Lancet  6 Feb 2010  Vol 375

481    Just over a year ago, I wrote about VADT, the third interventional trial in type 2 diabetes to show that lowering glycated haemoglobin (HbA1c) below 7 had no meaningful benefit. I suggested that all diabetologists should forthwith give a lecture eating their previous words on this subject. Fiona Godlee spotted my rantings and invited me to write a BMJ editorial on the subject, which I duly did with marvellous assistance from Harlan Krumholz. We ended with the suggestion that the QOF incentive to lower HbA1c to 6.5 might cause harm and should be abandoned forthwith. It wasn't, of course. Its proponents riposted feebly citing observational evidence of a linear relationship between all adverse events and HbA1c at every level, which we did not dispute. It's what happens to real patients when you treat them that counts. Well, here are the observational data from UK primary care, where I work. Turn to figure 1, and you will see that in diabetic individuals treated with combined oral therapy, all cause mortality varies little between HbA1c 7 to 9, being lowest at 7.5. Go as low as 6.4 and it jumps to higher than at 9.4. In patients treated with insulin, the 7.5 target is even more important; if you go any lower, mortality gets higher, even at 7.2, which is worse than being at 9. These data are from 28,000 patients observed after an intensification of treatment. So eat your words, diabetologists, and bin your target, QOF.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)61969-3/abstract

500    Pulmonary embolism in pregnancy is a good illustration of the triad described by Rudolf Virchow - hypercoagulability, venous stasis, and vascular damage. At this point I usually go into a panegyric about Virchow, a wonderful scientist and visionary social progressive; but to redress the balance I must also point out that he was a stubborn German professor who blocked the careers of anyone supporting the germ theory of disease. This however has nothing to do with PE in pregnancy, which may be accompanied by negative leg vein scans as clots can arise from the pelvic veins. It's also tedious to treat, and if this is something you are involved with, here is all you need to know, with 135 references.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)60996-X/abstract

BMJ  6 Feb 2010  Vol 340

302    The dear young BMJ is always an interesting fireside read these days unless you want some research to get your teeth into. I decided to look online at this paper on social variations in access to hospital care in three common cancers because as usual I couldn't even understand the question it purports to address from the one-page condensation. So far as I can tell, the authors aren't claiming that people in the lower social categories have less access (in the normal sense) to hospital services for cancer, but that they present later to such services, and that this hasn't changed from 1999 to 2006. I am only telling you this because there is so little else to tell you this week.
http://www.bmj.com/cgi/content/full/340/jan14_1/b5479

303   It's a convenient belief, supported by some systematic reviews of randomised trials, that all blood pressure lowering regimens are equally beneficial in proportion to the  degree to which they succeed in reducing BP. This population based case-control study seeks to dispute that, and in particular to blacken the name of calcium-channel blockers compared to ACE inhibitors and ARBs. Again, you won't learn much from the one-page version. In the full on-line article, you can see the confidence intervals in all their unconvincing glory. There may be some differences, but we need better evidence than this.
http://www.bmj.com/cgi/content/full/340/jan25_2/c103

314   One of the reasons I proposed the Easily Missed series was to find out what I personally had been missing these last thirty-five years. Long QT syndrome is a definite case in point. If you have a young patient who has fainted during exertion or on being woken by a loud noise, get an ECG at once and make sure it is looked at carefully: the next episode may be sudden death.
http://www.bmj.com/cgi/content/extract/340/jan08_1/b4815

Ann Intern Med  2 Feb 2010  Vol 152

144    Just as you wouldn't give up and blame the patient if their blood pressure remained at 186/112 despite a short course of treatment, so you mustn't give up treating nicotine addiction until people no longer run the awful cardiovascular and pulmonary risks of smoking. Give them nicotine replacement therapy for as long as it takes, and bin any guidelines which instruct you to do otherwise on grounds of cost. This study unsurprisingly found that  a nicotine patches are more effective prescribed for 24 weeks than for 8. Many smokers won't need this length of treatment, others will need more.
http://www.annals.org/content/152/3/144.abstract

167    Non-invasive coronary angiography sounds like a great idea, but there are problems. Magnetic resonance imaging would be ideal if it worked, because it doesn't involve ionizing radiation. But this head on comparison with computed X-ray tomography shows that it is not nearly as accurate, according to the published studies. This may change as techniques develop, of course. The problem with CT is that it uses big doses of radiation and needs iodine-base contrast material; and so does the gold standard of coronary angiography, which the patient will then have to undergo if the CT shows a lesion requiring intervention. The real-life radiation dosage studies are worrying, though every article predicts that doses will fall in the future.
http://www.annals.org/content/152/3/167.abstract

Dish of the Week: Crab

Although the life of early humans may have been precarious, our ancestors must  also have enjoyed many hours of pure bliss as they wandered the shores gathering molluscs and crustaceans and spearing the odd fish.

The most highly prized crustaceans are fat lobsters and crayfish, but there is nothing to beat a good crab. And it seems unlikely that the human population will ever outstrip the supply of crabs, so they are cheap in relation to their merits, except in posh restaurants.

It is a mistake to eat crab in a restaurant in England. It is unlikely to be fresh, and you never get given enough. You need to buy your crab freshly boiled and directly from a fisherman you can trust. It can be as big as you dare, because big crabs do not get fibrous flesh, like big lobsters. You must set aside enough time to smash it and retrieve its contents, and to make a mayonnaise. This should be made with the best eggs you can find, hen's or duck's, with a mild olive oil and quite a bias towards lemon. However its crowning glory should be a few drops of truffle vinegar. The Italian firm called Elfos make one with Tuber melanosporum, which is excellent; but if you can find one made with white truffles, do let me know.

This is not an extravagant dish. A bottle of truffle vinegar will last you for months and a crab costing about £6 from a peripatetic fish seller will feed two people royally. However, if the heavenly nature of crab mayonnaise leads you to open a bottle of your best white Burgundy, the cost might soar. But it will be worth it.

 

 

 

 

 

 

Subscribe to the RSS Feed for this page

 

 

 

 

Page last edited: 08 February 2010

• Login

Designed and updated by Cache
Hosted and content managed by Minervation