NEJM  30 May 2013  Vol 368
2059   Don’t read this paper, but rejoice that it exists. It’s proof that cancer genomics is the best kind of science—incomplete, dynamic, complex, and full of hope. It is also open to all who can make use of it: “We identified at least one potential driver mutation in nearly all (adult) acute myeloid leukemia samples and found that a complex interplay of genetic events contributes to AML pathogenesis in individual patients. The databases from this study are widely available to serve as a foundation for further investigations of AML pathogenesis, classification, and risk stratification.” I think I am falling in love with the Cancer Genome Atlas Research Network.

2075   “Is there a doctor on the plane?” My knowledge of in-flight medical emergencies comes from watching Airplane twice, and once being called to see a hyperventilating lady on a flight to Poland. Alas, my ministrations were coldly received, because she was on her way home to see a proper Polish doctor since English doctors like me did not understand her condition and were trying to kill her. Medical emergencies on commercial flights are mostly of this kind, fortunately. If she had had a tension pneumothorax I’m not sure I would have been able to summon up the courage to improvise with a ball-point pen. Here’s a paper which collects the outcomes of events serious enough to lead airline crew to seek advice from a physician-staffed communications centre. These occur at the rate of one per 604 flights: fainting is the commonest event, followed by breathing problems, and vomiting. Some poor doctor on board rendered assistance in half of these cases. Death in flight is rare, but when it happens the passengers tend to be youngish: mean age 60.

2084   A neat study from Canada seeks to discover whether patients who need to stay on anticoagulants should continue warfarin or change to heparin while having a pacemaker or implantable cardioverter-defibrillator fitted. The main outcome measure was clinically significant device-pocket haematoma. The heparin group had more than four times as many pocket bleeds as the continued-warfarin group. There’s more on the general issue of antithrombotic treatment during invasive procedures in a clinical review article on 2113.

2093   The trouble with evidence is that it can change in the course of a few days. Last week you learnt that very premature babies do just as well with target arterial oxygen saturations between 85-89% as between 91-95%: it said so in JAMA. Now here in the NEJM are the results of three multinational randomized trials addressing the same subject, and they show that sats below 90% are associated with worse outcomes.

Lancet  1 June 2013  Vol 381
1905  By the time a youngster gets type 1 diabetes, most of the ß-cells are dead and those that remain are under attack from the immune system, led by interleukin-1. Small, unblinded single-centre studies have shown promising results from canakinumab, an anti-interleukin-1 antibody, and anakinra, a human interleukin-1 receptor antagonist. Can you guess what is coming next? Yes, they did proper large scale blinded randomized trials on these two agents in kids and adults with new-onset type 1 diabetes. And these showed no effect at all, even on the surrogate measure C-peptide, which measures ß-cell function. A good clear editorial explains why, and ponders the next moves.

1916   A few weeks ago, I decided that I would not direct your attention to anything about H79N influenza virus or novel coronaviruses. Very soon, they may cut a huge swathe through the human race and leave our towns and cities empty: vast cemeteries made hideous with the howling of wolves, the screams of carrion-feeding birds, and a constant scurry of rats. On the other hand, these viruses may just fizzle out. If you want to worry about such things, the Lancet offers two showcase papers by Chinese virologists and genomic scientists describing the outbreak and the structure of the H79N virus(es). But you would do better to go back a couple of weeks to a clinical account of the disease outbreak in the NEJM. As for me, I shall try to remain fatalistic and indifferent. Take no thought for the morrow. Sufficient unto the day is the evil thereof. Don’t watch Contagion.

1933   A likeable seminar discusses gastro-oesophageal reflux disease, as we have come to call it. GORD knows why, because it is not really a disease at all: “Reflux of gastric contents to the oesophagus is a physiological event: a healthy person typically has reflux episodes. Gastro-oesophageal reflux disease is defined as reflux that causes troublesome symptoms, mucosal injury in the oesophagus, or both of these.” Like many things, it becomes a disease when doctors get to see and treat it. Our first approach is with dietary advice, which has no evidence base, as these authors point out: “… the effectiveness of dietary recommendations has not been shown, and in view of this absence of evidence, limitation of dietary advice seems wise.” A sound principle across the whole of medicine.

1956   “Is type 2 diabetes a category error?” asks Edwin Gale in a wonderful Viewpoint essay. “Carl von Noorden regretfully noted in 1907 that diabetes can only be defined in terms of glucose. A century has passed since then in largely fruitless attempts to escape from this circular definition.” “A scientific definition needs an external point of reference, and a problem that cannot be defined in scientific terms cannot have a scientific solution.” “In my youth, we solemnly advised one another not to adjust our minds; there was a fault in reality. With age, we acquired more respect for reality. When a century of scientific endeavour brings us round to the conclusion that we cannot define what we are talking about, it might be time to consider adjusting our minds.” Thank you Edwin—reading this is the closest thing to bliss anyone will find in a medical journal.

BMJ  1 June 2013  Vol 346
For the second week running the BMJ has the best research paper—and once again it comes from Australia. I’ve often expressed anxiety in these reviews about the increasing use of CT scanning in all age groups, especially children. But it is very hard to quantify the risk from lowish doses of ionizing radiation—it has been traditional to extrapolate from the exposures at Hiroshima and Nagasaki, but this has obvious limitations. Now a team of Australian scientists has gone through the data available for nearly 11 million people, identifying 680,000 who had CT scans between the ages of 1 and 19, with a mean follow-up of 9.5 years. It shows that those exposed to CT radiation, mostly at doses higher than are used now, have a 24% increase in cancer risk over this period; there is good evidence that this is cumulatively dose-related and will continue through life.

A Dutch systematic review usefully confirms that the cut-off level of D-dimer for detecting venous thromboembolism in people at low and medium risk needs to be adjusted for age. “The application of age adjusted cut-off values for D-dimer tests substantially increases specificity without modifying sensitivity, thereby improving the clinical utility of D-dimer testing in patients aged 50 or more with a non-high clinical probability.”

We inherited a large old Rover 75 which goes wrong every few weeks. “Must have been a Friday afternoon job” say the local mechanics, as they sigh and tot up the bill. Here is the already-famous paper which shows that surgical deaths within 30 days display the same pattern. The team from Imperial didn’t really expect to find this but were looking for the weekend effect already documented in other areas, e.g. cardiac survival in the USA found to dip by the Yale team. Weekends are certainly very dangerous times to have elective surgery in England, with an 80% higher mortality than on Monday: but the gradual increase in risk through the week is the most startling finding. By Friday afternoon you’re almost 50% more likely to leave hospital in a box.

JAMA Intern Med  27 May 2013  Vol 173
859   Implantable cardioverter-defibrillators stop some people from dying suddenly. The only other things they are capable of doing are (a) going wrong and/or (b) delivering inappropriate shocks. People who have these things fitted need to know their individual odds, as far as that is possible. I know several researchers who are trying to bring shared decision making into this difficult arena, and this descriptive review of the literature about outcomes by Fred Masoudi and colleagues will be useful to them. It should also be read by all interventional cardiologists. “Implantable cardioverter-defibrillators reduce the risk of sudden cardiac death and prolong life in selected populations; however, many patients will receive an ICD shock, either appropriate or inappropriate. It is imperative that patients be counseled regarding this risk and adverse outcomes associated with shocks. Reduction of ICD shock should be individualized to ensure that patients receiving these devices experience the maximal benefits of therapy while minimizing the adverse consequences.”

866   This study (marked Less is More) looked at nearly 300,000 men aged 65 and over in the US Veteran’s system and found that 25,000 had PSA levels over 4 ng/mL. Of these, 33% went on to have one or more prostate biopsies, and of these more than half had “prostate cancer”—at which point they were nearly all treated for this condition, regardless of age and co-morbidity. I find comfort in the fact that for the other two thirds, nothing further seems to have been done. So why take the blood in the first place?

Plant of the Week: Paeonia rockii “Rock’s Variety”

This was once one of the most sought-after plants in England, and over 20 years ago I spent silly money ordering one bare-rooted example from a nursery in Oregon. This grew and flowered and gave some pleasure for a few days: its scent was not all that was hoped for, a mixture of rose and dishcloth, and its papery flowers quickly fell apart in the rain. Plants which derive from Rock’s original introduction to the Arnold Arboretum in 1925 continue to command high prices: you can get one for about £100 from a specialist nursery in the UK if you are lucky.

On the other hand, if you are content to take a chance you can get a species “tree” for less than £5. The moutan is apparently an abundant plant on some mountainsides in Western China, where it is used for firewood. And the flowers, while they last, can be spectacularly beautiful with crinkled double white petals and a maroon or dark purple splash in the centre. Now that these are cheap, everyone should have a go: far from being real trees, “tree peonies” are medium sized shrubs and most gardens have enough room for two or three.

As for the eponymous Joseph Rock, he was even more exotic than his plants. He was born in Austria in 1884 and died in Honolulu in 1962. Wikipedia tells us that “He began by hunting the Chaulmoogra tree / in Burma, Thailand and Assam” which sounds like the beginning of a poem by Edward Lear. “Rock was cherished for his eccentricities, as well as his knowledge of botany and of ethnic minorities. He always travelled with a complete set of silverware, which was laid out for him at mealtimes. He also travelled with an Abercrombie and Fitch canvas bathtub, which his servants filled with hot water so that he could enjoy that most European of luxuries: a good soak in the bath.”