NEJM 10 Mar 2011 Vol 364
907 “Microalbuminuria is an early predictor of diabetic nephropathy and premature cardiovascular disease.” That’s the opening message of the ROADMAP trial, and it will just about do. What it does not mean is that any agent that reduces the appearance of tiny amounts of albumin in the urine will therefore stop people with diabetes getting cardiovascular disease and renal failure. This trial of the angiotensin receptor blocker olmesartan is an important addition to the evidence base showing that reduction of microalbuminuria is in fact a totally useless surrogate marker for therapeutic benefit in diabetes. Even though it wasn’t powered to show a difference in cardiovascular death between groups, it managed to prove (p<0.01) that olmesartan is associated with more deaths from cardiovascular causes than placebo over a median of 3.2 years in this group of diabetics. Fifteen deaths v three is quite a result. It is reported as follows: “Olmesartan was associated with a delayed onset of microalbuminuria… The higher rate of fatal cardiovascular events with olmesartan among patients with preexisting coronary heart disease is of concern.” You betcha. And yet the NEJM’s deputy editor, Julie Ingelfinger, tries to allay these doubts in her editorial. ” The concept of preemptive renoprotective therapy has gained traction,” she writes, and concludes that ” These results provide hope that it may be possible to prevent progressive chronic kidney disease in the many persons worldwide who have type 2 diabetes mellitus.” It seems that many diabetologists are still stuck in the wishful mindset that puts surrogate markers ahead of actual outcomes – an attitude that has already done harm to patients and needs to lose traction at once.
918 Now let’s refresh our brains with some nice classical detective work in the Typhoid Mary tradition. Three years ago, there was a widespread outbreak of Salmonella enterica serotype Saintpaul across most of the United States, centred on Texas and New Mexico. Among the 1500 cases studied here, 2 died and 21% ended up in hospital. At first, tomatoes seemed to be the culprit (odds ratio 5) but tomato-tracing led nowhere. The association with jalapeño pepper initially seemed weaker (OR 3) but as the detectives fanned out from Mexican restaurants to home fridge containers, these pungent seed-pods became hot favourites, and were eventually caught red-handed. Eye-watering puns continue as the chili-hearted miscreants were found to have been peppered with bacteria from contaminated farm water in Texas and Mexico. ¡Ole!
928 And now, reluctantly, back to the efforts of drug manufacturers to find a new market for some angiotensin receptor blocker (aka angiotensin 2 inhibitor) that still has a few years left on its patent. In this trial, BristolMyers Squibb and Sanofi Aventis hoped to show that irbesartan would reduce cardiovascular events in people with atrial fibrillation. This was a reasonable hope, because the most frequent cardiovascular events in AF are stroke and heart failure, and you might expect a statistically significant result from any drug that blocks angiotensin and reduces blood pressure. But no, irbesartan did nothing. At least it didn’t kill people.
Lancet 12 Mar 2011 Vol 377
905 It’s pretty depressing to see the New England Journal including two company-sponsored ARB trials, but at least they were well-conducted and unlikely to generate a lot of money for the publisher from the sale of reprints for promotional use. But The Lancet doesn’t seem to share any such scruples. Axcan Pharma Inc decided to try bundling three off-patent drugs – tetracycline, metronidazole, and bismuth subcitrate potassium – into a single capsule and seeing if, in combination with omeprazole, it would eradicate Helicobacter pylori as effectively, or better, than the standard regime of clarithromycin and amoxicillin plus omeprazole. Nothing sophisticated: just an open-label randomised trial in 7 European countries with varying degrees of H pylori resistance to standard antibiotics. Sure enough, the Axcan capsule worked better and now it can be sold with the help of huge numbers of Lancet reprints if Axcan so choose. There are scores of other ways to address this issue using different drug combinations and proper blinded randomisation, but the British flagship of medical science has chosen to publish this one.
914 In fact, The Lancet seems to be giving notice this week that it considers that the double-blinded trial is no longer necessary for drug companies wishing to show the superiority of their latest product. The Esai company manufactures eribulin and would like us to believe that it represents a breakthrough in the therapy of advanced, multi-treated breast cancer. Sadly, this is not a rare condition, but Esai decided to collect fewer than 7 patients for each of the 135 centres around the world who took part in the EMBRACE trial, comparing open-label eribulin with whatever alternative last-ditch treatment their oncologist decided. Eribulin is a microtubule dynamics inhibitor, like the taxanes, and all these women had had previous taxane treatment. Those receiving open-label eribulin lived on average for about 3 months longer than the miscellaneous control group.
924 So what’s the subject of the next open label trial with the sort of positive message which might please a manufacturer sponsor? Why, it’s nothing less than a new insulin from Novo Nordisk, at present coming under scrutiny in the BMJ and elsewhere for the way it has marketed its analogue insulins while reducing the availability of cheaper human insulins. To keep the profits flowing, new insulins have to keep appearing and doing something to justify their ever increasing costs: the latest wheeze is to make them ultra-long acting, as in this latest insulin degludec. In this trial, it was non-inferior to Novo Nordisk’s existing insulin detemir and can be given less often. A 12-week open-label trial to assess a drug which will be given for decades? What is The Lancet coming to?
942 What a relief to turn to a world class clinical review of amyotrophic lateral sclerosis – or motor neurone disease as we still mostly call it in Britain. “Let us keep looking, in spite of everything. Let us keep searching. It is indeed the best method of finding, and perhaps thanks to our efforts, the verdict we will give such a patient tomorrow will not be the same we must give this man today.” said Charcot in 1889 as he demonstrated a case of the disease he had first described. Alas, not so: Charcot would be astounded by the science in this paper, but we remain in deep ignorance about the pathogenesis of this disease and how to alter its course. Palliative care gets a sentence at the end of the paper, though it is the most important subject from the patient’s point of view, with the possible exception of assisted dying, which gets no mention at all.
BMJ 12 Mar 2011 Vol 342
586 In cystic fibrosis, a fall in predicted FEV1 below 30% predicted used to signal a median life expectancy of less than two years. This improved dramatically in the late 1990s, rising to over 5 years, but sadly there has been no gain since. An interesting cohort study, with reflections on changes in therapy which might account for this, notably nebulised recombinant human DNase.
588 During a working lifetime as a jobbing GP, I have always hankered after a primary care research project that would change practice. My small and intermittent attempts can’t make such a claim, though at least I encouraged my work partner Harold Hin to succeed with his ground-breaking study of coeliac disease. Looking at this week’s tranch of dreadful secondary care studies, I don’t think primary care research is notably more futile than any other kind, but sadly it impinges little on the lives of most GPs. The judge of academic success is not the usefulness index but the citation index, and this bibliometric study shows that the UK does very well by this criterion, with only the Netherlands ahead when weighted for money spent. Like all old men, I reflect that we need less research but better. Ditto medical journals: on this week’s showing, about two would suffice. Once a year.
598 Roentgen had hardly published his work on X-rays before the great Russian physicist Nikola Tesla was warning of their possible harmful effects. That was in 1895, and we have learnt some lessons since then, but uncertainties remain. Procedures like abdominal CT and CT coronary angiography are attractive for diagnosticians but pretty hazardous for patients, carrying the dose of 400 and 800 chest X rays respectively. This clinical review is worth keeping for its useful tables.
Arch Intern Med 28 Feb 2011 Vol 171
Apologies for any inconvenience caused by the late arrival of this train.
286 Nick Wald gleaned little fame from teaching me medicine nearly 40 years ago, but his subsequent work on the prenatal detection of neural tube defects earned him a knighthood and FRS; and when he published a paper suggesting that everybody over a certain age should take a Polypill, the editor of the BMJ suggested it might be the most important medical development for 50 years. But one by one his favoured ingredients fell into disrepute, and this latest paper from him suggests that he may be disenchanted with the whole idea. The point of the Polypill was to save us bothering with the idea of individualised cardiovascular risk: the point of this paper is to discuss the use of risk factors as assessment tools. The Risk-Screening Converter looks like a really good tool, particularly valuable for instilling the rudiments of the subject into intractable students, as poor Nick once tried to do with me all those years ago.
333 I’ve heard it claimed that smoking is protective against dementia, but it is not. A cohort from Kaiser Permanente in California is here followed up for 20 years or more. Those who smoked heavily back then have more than twice as much Alzheimer’s disease and vascular dementia now.
Plant of the Week: Hepatica x media “Ballardii”
No decent gardener, with the possible exception of the late Christopher Lloyd, ever tired of trying to achieve combinations of soft yellow and clear blue. In early spring, this can best be done with primroses and well-chosen hepaticas. No garden can ever have enough of these plants.
Hepaticas are adorably shy and love shady spots. You can get them in whites and reds and purples, but blue is always best. They are slow and hate disturbance; they flower at a time when few people buy plants; nobody has made a fortune, or probably even a profit, selling hepaticas. Apparently you can grow them by scattering seed, but then you never know what you’re going to get. If they like your garden, they can form big clumps covered in flowers. Ours remain demure to the point of sulkiness. But we still yearn for them every late winter, and when they shine a blue eye at us, we well up with gratitude.
Primroses of course are much simpler. Throw away any that are not of the purest native kind. Pull the leaves off the chosen ones and shove their roots roughly into the ground. They will thrive on such treatment and sing beside your hepaticas.