Richard Lehman’s journal review – 30 April 2012

Richard LehmanJAMA  25 Apr 2012  Vol 307
1717   Any budding young cardiology academic wishing to set up a publication of her own could do worse than start a Journal of Negative Stem Cell Trials in Heart Failure. There are enough of these to fill a volume every quarter-year or so, and editorials could reflect on all sorts of fascinating issues to do with how to wash bone-marrow cells, whether to pre-treat them with this or that, which bit of myocardium to put them in, whether tiny differences in this or that functional measure in various aggregated subgroups indicated that this treatment might actually work one day, etc, etc. This would save the rest of us from having the disappointment of bumping into these papers on a regular basis in the main medical journals. Ten years ago, they were really exciting, and we all took heart, so to speak; but the FOCUS-CCTRN published here is just another failure like the rest. The cells were autologous bone marrow mononuclear cells; they were introduced by transendocardial injection, mostly into male hearts damaged by ischaemia, and at six months there was no evidence that they were doing anything to any of 8 outcome measures.

1727   About 6% of infective endocarditis is associated with implantable cardiac devices, and the vast majority of the culprits are pacemaker batteries. So although the wires are in the heart, the germs are on the subcutaneous box and reach the heart valves through the bloodstream. The treatment is to get the device out as soon as possible: these infections carry a substantial mortality which increases with delay in removal.

1736   Recurrent severe migraine is a blight on anyone’s life, for which there are a number of prophylactic drug treatments, none of them infallible, and many of them unsatisfactory. So a simple mechanical cure would be a great breakthrough, were it to exist. Botulinum toxin injections to the forehead muscles do alleviate some types of headache, as this meta-analysis shows, but the effect is modest and is only detectable in chronic migraine and chronic tension headache. Botox does nothing for the prophylaxis of recurrent episodic migraine in randomized controlled trials.

NEJM  26 Apr 2012  Vol 366
1567   When the first case-series reports of bariatric surgery for type 2 diabetes came out, it was clear that something huge was afoot. In many patients, blood sugar levels dropped and stayed down immediately after surgery, before there was significant loss of weight. Now we have two randomized controlled trials of surgery versus optimal medical therapy in poorly controlled T2DM, and the conclusion of the first paper states that “12 months of medical therapy plus bariatric surgery achieved glycemic control in significantly more patients than medical therapy alone.” I don’t think I’ve ever read such a gross understatement in the conclusion of an abstract—which is such a strange feeling that I don’t know where to put it. “Significantly” here doesn’t mean a statistical trick to magnify an unimportant change in the surrogate end-point of glycaemia. In fact you don’t need statistics at all to describe the success of these treatments: gastric bypass and sleeve-gastrectomy cause massive weight loss (29kg and 25kg respectively) and would have eliminated the need for diabetes treatment in most patients had the HbA1c target not been set artificially low at 6%. The authors commendably advise caution until we have long-term outcomes, and I can hardly object to that. But this is a breakthrough, and will have profound consequences for the future management of T2DM.

1577   The Italian authors of the second paper are equally circumspect in their conclusion, but in the meat of their account of this trial they use the word remission. Biliopancreatic diversion achieved a remission rate of 95% at two years; gastric bypass achieved 75% in severely obese diabetic patients. Whoopee! But that isn’t all. “Preoperative BMI and [postoperative] weight loss did not predict the improvement in hyperglycemia after these procedures.” So these kinds of major surgery have some direct effect on beta-cell function which we don’t fully understand. “Type 2 diabetes”—a complex metabolic disorder, now perhaps curable—has plenty of surprises left up its sleeve. All eyes should be on the detail of what happens following biliopancreatic diversion, in the hope that a simpler form of surgery or some new kind of medical therapy could result. But patience will also be needed: as in all diabetes trials, the outcomes that really matter are cardiovascular events, limb loss, blindness, and renal failure. We still need to be certain we are doing more good than harm.

1596   Diagnosing appendicitis for me is a matter of simple rule-in and rule-out tests in the consulting room. The ones that haven’t made it into the textbooks are the Mars bar test (“if I gave you a Mars bar now, would you eat it?”), the speed bump test (“did you go over any speed bumps on the way here and did you hold your tummy?”), and the hopping test. If any of these are positive, there is a-priori evidence of an inflamed viscus, and the surgeons can work out the rest, especially if there are fever, RIF tenderness, rebound etc. Somebody needs to ascertain the diagnostic characteristics of these “frugal heuristics” which haven’t yet made it into JAMA’s Rational Clinical Examination series. Everything needs to be done to avoid recourse to abdominal CT scanning, which uses frightening amounts of radiation and really should be avoided in children. This study compared low-dose CT with standard- dose CT in 891 patients with suspected appendicitis in a single institution. “Low-dose” is a relative term, meaning about a quarter of standard dose: still very big. The negative surgery rate was the same in both groups, at just over 3%; and the perforation rate was the same too, at around 25%. To me, that suggests too much diagnostic delay: “if in doubt, whip it out,” would be my watchword if I were a surgeon. Perhaps just as well that I’m not.

Lancet  28 Apr 2012  Vol 379
1561   “Unprecedented momentum is gathering to put physics into the centre of global health policies,” declares The Lancet this week. Oops, sorry, that was last week. Let’s try again. “Unprecedented momentum is gathering to put adolescents into the centre of global health policies.” But no, this can’t be the way forward at all. Let’s put some grown-ups into the centre of global health policies. Otherwise we might have editors of international medical journals rushing about like teenagers from one conference to another, blogging crazily and spending all their time on Twitter while neglecting their homework. And that would be terrible.

1591   Peter Rothwell continues his investigation of the long-term randomized trials of aspirin for clues about the possible value of this drug in the prevention and treatment of cancer. Here he generates the hypothesis that the observed short-term reduction in cancer seen in five British trials of aspirin (for cardiovascular protection) may be due to a suppressant action on the mechanism of metastasis, especially for adenocarcinomas, and especially in smokers. This is interesting, but speculative, and these subgroup effects will need to be confirmed by large and lengthy prospective trials.

1602   And then there is the more general question of whether aspirin has the ability to alter what is likely to appear on your death certificate, including the date. Frankly, this is a matter of indifference to me; I take daily low-dose aspirin, but only in the hope of avoiding migraine with aura.

Overwhelmingly, you die of something you can’t avoid, at a time not of your choosing, so it is a waste of time giving the matter any thought. Doctors in the last few decades, however, have taken it upon themselves to try and raise the average age of death in the population by any means possible, in the hope of seeming useful. In a widely discussed earlier study, the Rothwell team showed that daily low-dose aspirin has no effect on cardiovascular mortality in the general population but a detectable effect on cancer mortality. Here they concentrate on the short term reductions in cancer incidence, but get no further in proving any statistically significant overall prolongation of life from the general use of aspirin. The editorial puts this all into context, and also notes the omission of the two largest aspirin trials and other methodological flaws. So take aspirin if you have some reason to, but don’t count on it altering your death certificate.

1613   Kawasaki disease is every doctor’s and parent’s nightmare: cause unknown, so rare that most doctors never see a case, thus easily missed, and potentially fatal due to coronary arteritis. There are about 40 cases a year in the UK, but this Japanese trial managed to collect 298 children with severe Kawasaki’s and randomize them to receive intravenous immunoglobulin with or without prednisolone. The steroid-treated group were left with fewer coronary artery abnormalities.

BMJ  28 Apr 2012  Vol 344
Did you know that a lot of medicine can be done over the telephone? And that if you do that, you can save the NHS more than £1billion per year? No, I didn’t either, but the Health Secretary has evidence that he cannot reveal. As for the evidence we can access, here are two studies in which telephone support was added to routine asthma management, and to smoking cessation aided by nicotine replacement. It didn’t have the slightest effect. So perhaps we need to try harder and target the sickest patients with severe chronic disease. In an astonishing trial on the Archives website, which I shall come back to another time, this is just what they did. Mortality was nearly four times higher in the telemedicine intervention group. A great way to reduce healthcare costs, undoubtedly: do it to enough elderly patients, and you can probably save £3bn.

A useful observational study looks at risk factors for mortality from imported falciparum malaria in the UK over the past 20 years. I can hardly do better than quote the summary, in the hope that colleagues still in UK general practice will take note and keep their radar on: “Most travellers acquiring malaria are of African heritage visiting friends and relatives. In contrast the risks of dying from malaria once acquired are highest in the elderly, tourists, and those presenting in [should be from?] areas in which malaria is seldom seen.” So if they come back from their exotic holiday with “flu,” send them for a blood smear right away.

Arch Intern Med  23 Apr 2012  Vol 172
611   Most of my generalist readers are probably ready to pelt me with rotten eggs when I bring them another paper about stents, but I can’t help telling you about this individual patient data meta-analysis of drug-eluting vs bare-metal stents for primary angioplasty because it’s probably quite important. No, hold the ripe tomatoes too. All I’ll do is give you the conclusion:

“Among patients with STEMI undergoing primary percutaneous coronary intervention, sirolimus-eluting and paclitaxel-eluting stents compared with BMS are associated with a significant reduction in target-vessel revascularization at long-term follow-up. Although there were no differences in cumulative mortality, reinfarction, or stent thrombosis, the incidence of very late reinfarction and stent thrombosis was increased with these DES.”

Yes, read that again. To me that seems to say that these vastly more expensive stents, which also need a year of costly clopidogrel treatment afterwards, don’t actually perform any better than bare metal stents for most patient-important outcomes, and any possible advantages seem to be balanced by disadvantages. Or have I missed something?

623    Another meta-analysis, this time of trials of warfarin to prevent stroke in people with nonvalvular atrial fibrillation. What’s the biggest breakthrough here? Why, it’s making sure people are in the INR target range. Because if they are, they have fewer strokes and fewer bleeds. Medicine is sometimes, though sadly not often, perfectly logical.

Plant of the Week: Cornus florida

I’ve never had much luck trying to grow flowering dogwoods on limy clay and in a Limey climate, but over here in New England they are among the greatest sights of the spring. They are native trees, growing on forest margins, but I haven’t been able to get out and see them in their natural habitat. However, there is little need, as they have been planted by the hundred along the grander suburban streets of New Haven, where the older ones form magnificent mountains of flower in front of huge clapboard houses of various and exotic design.

Except that one should not say “flower” but “bracts.” The true sexual organs of the cornel family are paltry affairs, but they are surrounded by these stupendous quadriform codpieces and crinolines of white, pink or red, presumably to attract the emerging insects of spring. The reddest ones are the creation of human hybridists working with natural sports, and we must be grateful for their patient efforts towards securing a true deep red, most nearly perhaps in “Cherokee Chief.” There is also a purple-leaved form with purple-red bracts, called “Purple Glory,” which I haven’t seen.

I don’t know if there are places in the UK where these trees flourish and display their splendour to an East American degree, but I rather doubt it. For us, they will always bring back memories of wandering down quiet streets of exuberant New England mansions and their gardens in spring sunshine: one of the world’s truly delightful experiences.

  • Guest

    Ref Cornus florida: at National Trust’s Hidcote Manor, Gloucestershire. Yes, rare indeed, in the UK.