NEJM 26 Jun 2014 Vol 370
2478  Cryptogenic is a good word. It’s up there with “idiopathic” and “pleiotropic” and “diathesis” for covering gross ignorance with a smattering of Greek. “Cryptogenic” sounds as if it was first used to describe the odd symptoms that Superman experienced when exposed to kryptonite. However, its first use was recorded 30 years before the caped crusader first appeared in the skies above Metropolis in 1938. “Cryptogenic stroke” is a fairly recent term, covering 20-40% of incident stroke, and it challenges researchers to hunt around the garden looking for kryptonite hidden under stones. A patent foramen ovale! Ah yes, but it may be an innocent bystander. Atrial fibrillation (AF) then! Possibly, according to the EMBRACE study, but still unlikely to account for most unexplained ischaemic strokes. The Canadian researchers monitored 572 patients who had had an unexplained ischaemic stroke or transient ischaemic attack, half of them using standard 24 hour ECG, and half with a 30 day event monitor. Their mean age was 72, and the detection rate for AF was 3.2% versus 16.1% in the two groups.

2478  In another study—a multinational trial funded by Medtronic—the longer the investigators looked for AF after an unexplained ischaemic stroke, the more they detected. Their cohort was considerably younger, with a mean age of 61.5. Here the comparison was between 24 hour ECGs done at one, six, and 12 months and continuous monitoring for a whole year using an implantable event detector. There was a cumulative difference through the year, at the end of which the detection rate was 2% versus 12.4%. These data lead the editorialist to conclude: “The results of two studies published in this issue of the NEJM indicate that prolonged monitoring of heart rhythm should now become part of the standard care of patients with cryptogenic stroke.”

JAMA 25 Jun 2014 Vol 311
2508  Do you truly and solemnly believe in instrumental variable analysis? And I’m sorry, I’m not going to let you go away and look it up. You need to know the answer straight away if you are to make sense of the following statement: “Among adults in acute care hospitals in New York State undergoing hip repair, the use of regional anesthesia compared with general anesthesia was not associated with lower 30 day mortality, but was associated with a modestly shorter length of stay. These findings do not support a mortality benefit for regional anesthesia in this setting.” The people who wrote this conducted a massive database study of 56 729 patients in New York State acute hospitals who had surgery for hip fractures between 2004 and 2011. 15 904 (28%) received regional anesthesia and 40 825 (72%) received general anesthesia. Using hospital discharge coding, they matched these into pairs according to recorded factors that might affect hospital stay and mortality. I think this is what is meant by instrumental variable analysis (I haven’t looked it up either), and I am unconvinced that it can be more than a hypothesis generating exercise. If you really want to know whether or not regional anaesthesia for hip fracture is safer than general anaesthesia, do a randomised trial with properly blinded allocation.

NB: For those who find this topic endlessly fascinating, there is an almost identical paper using the same methods, but a different US database, which has just appeared on The BMJ’s website.

2518  “Chronic kidney disease (CKD) is a worldwide public health problem, with increasing prevalence, poor outcomes, and high treatment costs.” I don’t know if I believe a word of this. To me, the kidneys are complex appendages of the cardiovascular system, and most “CKD” is just an expression of cardiovascular disease. Its increasing prevalence is because people are living longer, and ditto its high costs. As the circulation fails, or diabetes gets into its final phase, people show a steep decline in glomerular filtration rate (GFR); this is not good news for these individuals, but it is hardly a public health problem. The more they are tinkered with, the higher their treatment costs become. Nobody likes to tell them how futile most of this treatment is, and how little renal physicians know about what works. I am sometimes called cynical for saying things like this, but am I missing something? This entirely nephrocentric paper paints a picture of gloom, based on the prognostic characteristics of a swiftly falling estimated GFR. We are indeed all doomed; but in my experience of generalist practice, to die of primary renal failure is actually quite rare.

JAMA Intern Med Jun 2014
OL  A London registry lists 41 688 patients who had percutaneous coronary intervention (PCI) in NHS hospitals between 2004 and 2011. During this time, clever cardiologists in our centres of teaching and excellence increasingly adopted sophisticated techniques such as intracoronary ultrasound and intracoronary pressure wire measurement of fractional flow reserve. This long term survival study shows that they made no difference to outcomes compared with old fashioned angiography guided PCI. This paper does not bear the “Less is More” label, but it could.

Lancet 28 Jun 2014 Vol 383
2232  Forgive me if you have heard this before: colchicine is a good treatment for acute pericarditis and for preventing recurrences. Colchicums abound on the hills of Northern Italy, where this study was conducted. Don’t eat any part of them or you might die an unpleasant death. However, if you are an adult weighing more than 70kg and happen to have pericarditis, you may take 0.5mg twice daily for six months and this will halve your chance of having recurrences.

OL  There is an immense backlog of papers on the Lancet website. This is good for lean weeks like the present, but it’s hard to pick any outright winners for a generalist readership, especially as most of you cannot penetrate the Elsevier paywall. If you can, you will certainly be interested to browse an immense study on global overweight and obesity, funded by the Gates Foundation. Over one third of the world’s population now has a BMI over 25. There is no country in the world where people are getting thinner. In less developed countries, women tend to outrun men in the race towards obesity. In developed countries, we suffer from a syndrome which could be called RBS: rushed but sedentary. Our race towards obesity, however, is generally slowing down. To accompany this survey there is a superb commentary piece by Klim McPherson: but again behind the paywall.

OL  Among the dud trials on the website, there is one of a needle free injector for influenza vaccination. The primary endpoint was a non-inferior immune response compared with needle and syringe administration. But it is more likely to cause local itching and pain, which can last for several days. If you are needle-phobic, you are better off just to look away while you have the needle.

The BMJ 28 Jun 2014 Vol 348
Here’s a demonstration of true grit. Toby Leslie from the London School of Hygiene and Tropical Medicine organised a study, which recruited 5794 patients of all ages with suspected malaria and involved 80 clinicians in 22 clinics. Imagine doing that in rural Afghanistan! Not only that, but he was in effect challenging the usual practice of these scattered clinic doctors. In the low malaria prevalence areas, only half had microscopes, and so they often missed the one malaria case in every 100 febrile patients they saw. In the higher prevalence areas, everybody with a fever would tend to be prescribed antimalarials just in case. To see if this could be changed, the LSHTM team tried introducing rapid diagnostic testing kits capable of diagnosing malaria reliably, and also of distinguishing between Plasmodium vivax and Plasmodium falciparum. The strategy showed a very useful degree of success.

“Attention-deficit/hyperactivity disorder (ADHD) is a common neurodevelopmental disorder that can affect people across the lifespan.” I wonder how much ADHD is diagnosed in Afghanistan. However, this study was carried out in Sweden, where people with ADHD are often prescribed drug treatment with stimulant drugs or atomoxetine. The investigators worked really hard to examine the precise correlation between suicide and ADHD treatment using within patient analysis. A model piece of work, which concludes: “This study found no evidence for a positive association between the use of drug treatments for ADHD and the risk of concomitant suicidal behaviour among patients with ADHD. If anything, the results pointed to a potential protective effect of drugs for ADHD on suicidal behaviour, particularly for stimulant drugs.”

The news that statins can “cause diabetes” broke about two years ago. This new Canadian study looks at several large primary care databases and discovers that: “In the first two years of regular statin use, we observed a significant increase in the risk of new onset diabetes with higher potency statins compared with lower potency agents (rate ratio 1.15, 95% confidence interval 1.05 to 1.26). The risk increase seemed to be highest in the first four months of use (rate ratio 1.26, 1.07 to 1.47).” So statins cause small elevations in blood glucose that correlate with their potency, i.e. their degree of LDL-C lowering. The usual reason to take a statin is not to lower LDL-C but to lower cardiovascular risk, and we do not really know whether risk lowering is directly linked with “potency”. We have only one individual patient data meta-analysis to go by, and its data are not available to researchers outside the Oxford Clinical Trials Support Unit. As for the “diabetes” bit, it should not matter to any individual whether they are classed as lying slightly below or slightly above the arbitrary definition line. What matters is the long term risk of harm associated with that particular level of blood sugar and their likelihood of benefit from treatment. And playing those odds, they would be better off staying on their statin, whichever it may be. But the choice is always a personal one.

Plant of the Week: Tropaeolum tuberosum var. lineamaculatum “Ken Aslet”

Beautiful June, when everything is fresh and lovely, is coming to an end. The bright leaved trees are turning dark Constable green. There will be no more bearded irises. From now until November, we gardeners have to be a bit more creative to keep the show going. Our greatest friends are the later flowering climbers.

Every year we buy a number of clematis hybrids, if possible with some viticella blood in them to make them late flowering and wilt resistant. I shall no doubt find myself writing about some of these later in the summer. They and the roses are undoubtedly the best of the ornamental sprawlers and climbers, but they need quite a lot of training and attention.

The climber I most wish I had planted earlier in the year is Ken Aslet’s tropaeolum. It is an edible tuber that sprouts a mass of upward scrambling stems, bearing expanses of flaming yellow-orange flowers through the summer. A must for any expanse of ivy (such as we have), yew, or other dark evergreen hedging.

I have always longed to possess Ken’s potato, as we call it, in sufficient quantity to use it as a food crop. But it does not seem to like our heavy limy clay, and it seldom lasts a winter. I must pay more attention to its welfare if we are ever to serve it with the Christmas goose.