Arch Intern Med  23 May 2005  Vol 165

1095    We all know that warfarin interacts with all sorts of drugs and some foods, and this is a comprehensive overview of the evidence. It’s a bit alarming to realise how poor most of it is. Table 2 is worth downloading for reference, though it has its oddities: acetaminophen is down as a highly probable cause of moderate potentiation, whereas paracetamol (the same drug) is listed separately as a probable cause of serious potentiation. According to the table, alcohol is safe, and the drinks to avoid are grapefruit of cranberry juice; whereas the text warns against alcohol and doesn’t bother about the fruit juices.

1111   Will the coming of computers to hospitals reduce the rate of adverse drug events? In this study, a “highly computerized” US veteran’s hospital still managed to produce ADEs in half of its patients.

1140   Doctors who practised before1988 will remember the coming of the NNT, the number-needed-to-treat, which promised to make decisions about drug choice and rationing so much easier to explain to patients. Well, does it? This intriguing study was conducted in Norway, where the people are known for their hardiness, cleanliness and cool logic. When asked to make therapeutic choices, most of them ignored NNTs altogether.

1147   Of all the drugs we use, anti-hypertensives and lipid-lowering agents are among the commonest and also, incidentally, have the least favourable NNTs. This study looked at whether people in the USA actually take them. Full adherence rates for those on combined treatment were 44.7% at 3 months, and 35.9% at six months, after which it levelled off. The depressing conclusion is that people find it difficult to take even two medications consistently, at any rate when they have to pay for them.

1167   A small proportion of vitamin B12 will get absorbed whether or not you have any intrinsic factor, and many elderly people don’t. So why not give them big oral doses, instead of injections? This dose-finding study concluded that 500mcg of cobalamin is the right daily dose: you can buy it over the counter, but you can’t prescribe it, which is silly. And giving it weekly might be better.

1185   Patients with atrial fibrillation which is “cured” by cardioversion probably should not come off their warfarin: the AFFIRM study shows that they have a lower stroke risk if they stay on it, even in sinus rhythm.

1192    Having the metabolic syndrome not only increases your chances of myocardial infarction, it also worsens your outcome afterwards. This may be one group where bezafibrate is still worth prescribing, as it reduces the chances of a further infarct (see p.1154) in people with “MS”. Which, maddeningly, is the new abbreviation for Metabolic Syndrome.

J A M A  25 May 2005  Vol 293

2471    Coronary angiography requires time, equipment and cardiologists: multislice computed tomography requires less time, different equipment, and radiologists. In most patients, the imaging is as good, and if it helps to reduce the queuing time for cardiac investigations, it has to be a good thing. As for the argument that you can’t put stents in with a CT machine, see below for evidence that most patients would be better off with bypass grafting.

2479   Survival after a diagnosis of breast cancer depends on many variables, one of which seems to be exercise. This paper shows a dose-related effect in prolonging life in 2,987 patients who developed breast cancer during the Nurses Health Study, 1984-1998.

2496   In this issue of JAMA, the original proponents of a link between Epstein-Barr virus and multiple sclerosis admit that they got some of their data mixed up (p.2466). But that didn’t alter the validity of their conclusion, and it now gets a boost from another study looking at the timing of and level of EBV titre rises and the development of MS. Interesting, but maybe not definitive.

N E J M  26 May 2005  Vol 352

2163    A recent study convincingly showed that patients with mild chronic asthma had the same outcomes whether they took their steroid inhalers regularly or when they felt the need. This smallish (n=97) study from New Zealand attempted to use exhaled nitric oxide as the marker for airways inflammation to guide the dose of inhaled steroid. Patients randomised to NO measurement ended up using less inhaled steroid with the same outcomes as those following a conventional stepped-dose guideline. Experts always like to have something to monitor rather than letting the patient be the judge. I am not an expert; and even if I was, I suspect my patients would continue to do as they please.

2174    This study looks at the outcomes of nearly 50,000 patients who had coronary artery bypass grafting or stenting for multivessel coronary disease between 1997 and 2000. The CABG group showed better survival after adjustment for other factors, but as the authors point out, an updated study will have to compare off-pump total-arterial CABG with drug-eluting stents.

2184    Another reason to put statins in the water supply: they reduce the risk of colorectal cancer, according to this case-control study from Israel. The effect size in an unselected population could be as high as 50%, but we need better evidence.

2211   Do you know your pharmacodynamics from your pharmacokinetics? If not, you risk the ire of my distinguished correspondent Jeff Aronson - who would, I feel, have written a much more readable review of Drug Metabolism and Variability among Patients in Drug Response (and thought of a better title too). But more likely, he would wisely have refused to attempt compressing the whole of clinical pharmacology into ten pages, even for the NEJM. Read this, if you have the patience, for some insight into the role of different cytochrome P-450 enzymes.

Arch Intern Med  9 May 2005  Vol 165

978    Is there any such thing as a safe non-steroidal anti-inflammatory drug? Basically no; if this case-control study is to be believed. It trawled through data from various counties of Denmark, matching hospital admissions for myocardial infarction with prior prescriptions of non-aspirin NSAIDs. Adjusting for known risk factors, users of rofecoxib fared worst, with an adjusted relative risk of 1.8: NSAIDs overall carried an ARR of 1.68; and celecoxib has an ARR of 1.25. Given that it is less likely to eat holes in your GI tract than most NSAIDs, I think it’s the one I’d prefer to take.

986   A systematic review of brief alcohol interventions in primary care. They actually work.

997    More evidence that dairy consumption reduces the risk of type 2 diabetes, this time in male health professionals. It isn’t a huge effect, but it applies at all levels of weight, and it does have a physiological rationale – see The Milk Debate on p.975.

1005   Men should take more physical exercise if they want to reduce their risk of prostate cancer.

1048   And to reduce their risk of renal dysfunction, men should drink more than 7 units of alcohol a week. I’m off for some wine and cheese, followed by a run.

Ann Intern Med  17 May 2005  Vol 142

805   A Mexican study shows that rifamixin, a non-absorbed antibiotic, really does prevent travellers’ diarrhoea. So don’t let Montezuma take his revenge: nor let Delhi harm your belly.

836   Lots of my patients travel to the Indian subcontinent without taking antimalarials, thinking that they are immune from being born there. So I see a steady stream of imported malaria, which is easily missed. Send any suspected patient for a peripheral blood film, but if your lab is closed, a rapid test for parasite derived chemicals might be the answer, according to this systematic review. I haven’t yet bought a kit.

813   A huge body of evidence has been accumulating over the last decade linking impaired glucose tolerance with adverse cardiovascular events. This study swims against the tide by finding no such association in postmenopausal women with existing coronary artery disease. Odd.

832   “Adults are whooping, but are internists listening?” asks this perspective piece. Pertussis is one of the commonest causes of cough lasting more than 2 weeks – 20-30% of adults, and in fact children too. There’s not a lot you can do, though this paper argues for antimicrobial prophylaxis for close contacts. Vaccination is clearly not as completely protective as we once believed: though it’s ages since I heard a true whoop, or had a red-faced coughing child be sick over my consulting room carpet.

Lancet  28 May 2005  Vol 365

1849    “O for a dragon that would lay me golden eggs!” cried the king. Forthwith there was fetched to him a golden egg, and it cracked open before him and brought forth a tiny dragon. The king ordered that the shell of the egg, being of purest gold, should be sent to his treasury, and that the infant dragon should be reared with utmost care, that it might bring forth further eggs. And lo!  The dragon grew and laid eggs, and belched forth sweet smoke, that the people flocked to breathe it. But they that breathed it did in after times choke and die. And the king said, “My dragon hath brought forth eggs, and my treasury fills with gold, even as the land fills with dragons: therefore it is my decree that no man in the land shall slay a dragon.” But the king’s physician cried, “Sire, but the dragons’ breath doth smite the people with cankers and choking, so that thy kingdom is laid waste.” And the king said, “What care I for that? I shall be the richest king of the earth, and perchance I shall of my pleasure give to the people a portion of the wealth which the dragon hath given me”.  And the physician was very sorry for the king’s words, and made of herbs a patch which did avail against the dragon’s smoke, and gave it to the people: and the king was sore displeased. And this did happen in the city of New York in the year of grace 2005. Free nicotine patches helped people stop smoking.

1877     The therapy of heart failure has accumulated in geological fashion, with one treatment being added on top of another in hospitalised patients with systolic dysfunction. Two leading practitioners of this approach here summarise the whole area; reasonably well, if such a strategy appeals to you. But heart failure in the community is a different animal, which has yet to be properly characterised and managed.

B M J  28 May 2005  Vol 330

1233    This MRC study compared surgical stabilisation of the lumbar spine with an intensive rehabilitation programme in people with chronic low back pain. The actual interventions were pretty heterogeneous, as were the patients, and the results were essentially the same in both groups. No great surprise, then; and given that most surgeons wisely avoid treating back pain, and that most patients don’t persevere with intensive rehab, what are we left to do for them?

1247   Thirty two people, originally from India or Pakistan with type 2 diabetes, were interviewed in Punjabi and/or English, to look for possible reasons for non-compliance with oral treatment. Trish Greenhalgh’s commentary (p.1250) is trenchant and sensible in pointing out that the main obstacles to what she calls “concordance” are not culture-specific but are shared with native British patients.

1251    With countless Foundation Year doctors coming and going in various directions (including general practice) every four months, who is going to assess these young sprats, and how? The Sheffield peer review assessment tool, SPRAT, described here, looks useful beyond the confines of South Yorkshire paediatrics. As for GPs and consultants in permanent posts, why not let all members of one group simply give a score for each of the other, as I suggested (a bit flippantly) in a BMJ letter a few years ago?