Richard Lehman’s journal review—29 April 2013

Richard LehmanJAMA  24 Apr 2013  Vol 309
1691  Last week I welcomed the imminent arrival of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) because it would classify every human being as insane, and so should provide the world with a good opportunity to step back and decide what psychiatry is really about. This open access piece about the new manual is written by DSM enthusiasts, so do read it and make up your own minds, because on my own admission there is a high probability that I am insane. And naturally, the same applies to you.

1696   One of the DSM categories which is changing in the new edition is autism, which becomes autistic spectrum disorder. Just how helpful this label really is I cannot decide, but in general it is considered to be something best prevented – after all, you wouldn’t want your child to grow up into a Newton or a Wittgenstein, would you? In a population study of children born to all Danish mothers who took valproate during pregnancy, there is a clear increase in diagnoses of autistic spectrum disorder, compared with the population rate and to mothers taking several other anticonvulsant drugs. The authors are wisely reluctant to make firm clinical recommendations.

1704   A great big retrospective study finds that taking a beta-blocker is associated with a small reduction in post-operative mortality and major cardiac events in some subclasses of patient undergoing non-vascular surgery. Alarm bells: how tightly were these groups pre-specified? How watertight was the propensity scoring and matching? The NNT to prevent one post-op death was 241: this could be massaged down by factoring in cardiac risk factors, to about 18 for the highest risk patients—but questions remain. The comparison was not between patients given beta-blockers short term for the peri-surgical period, but with people mostly taking them long-term. And high-risk patients undergoing vascular surgery showed no benefit. It doesn’t really add up: this is why we always need to do prospective randomized trials to decide the merits of treatment, rather than processing big bagfuls of old data.

1722   More than twenty years into JAMA’s excellent Rational Clinical Examination series, standards remain high, though topics have recently tended to become vaguer. You can’t say that of ectopic pregnancy, however: either you have it or you don’t. As usual, the evidence search is exemplary. History alone does not get you far, and short of laparoscopy, transvaginal ultrasound gets you furthest. No great surprises, then: some people also use quantitative serum hCG testing, but a definitive level has not been established.

NEJM  25 Apr 2013  Vol 368
1575   Trimethylamine-N-oxide (TMAO) is the new enemy within. We make it in our bowels, I am sorry to have to report, by microbial metabolism of the choline moiety in dietary phosphatidylcholine (lecithin). The fault lies with our microbiome, because as this study demonstrates, if you kill off the bacterial flora with metronidazole and ciprofloxacin, production of TMAO stops totally. The second part of the study goes on to associate TMAO levels with atherosclerosis, proving to its own satisfaction that if you produce shitloads (I use the word in its technical sense) of TMAO then you get major adverse cardiovascular events, independent of other known risk factors. I am naturally inclined to pooh-pooh this idea, but to do so would involve going into a level of detail that neither you nor I could quite face. There is lecithin in the kinds of food that cardiologists traditionally dislike—eggs, red meat, liver and pork—but there is also plenty in fish and most plants. I have just eaten a bagful of whelks—probably lots of lecithin there too, but you won’t stop me so easily. After all, “Lecithin may help to prevent the buildup of fats in the body and protect against cardiovascular disease by lowering cholesterol,” says the Nutritional Supplements Health Guide website.

1585   I keep a safe distance from interventional cardiologists, but hope that when my turn comes they will know what they are doing. If I had systolic heart failure, I would most definitely not want a cardioverter-defibrillator, but if I had atrioventricular block I would want to have whatever kind of pacing gave me the most benefit. In this life-shortening condition, I would place a high value on symptomatic improvement. Unfortunately, this Medtronic-funded trial of right ventricular pacing versus biventricular pacing did not measure quality of life at all: it went for the usual mixture of hard end-points (death, admission for IV treatment) and a surrogate—an end-systolic volume increase of 15%. Biventricular pacing proved superior on those counts, and we also know from other studies that it improves symptoms, so that’s what I’d go for if I had grade 1-3 systolic HF. My aim would be to live better and then die suddenly: so yes to BVP, and no, no, no to that ICD.

1625   Mention the word “itch,” and you will scratch yourself: usually on the head, but I’ve just noticed this place on my back… So it is kind of the NEJM to use the title “Chronic Pruritus” for this review of the subject, because saying “pruritus” never made anyone scratch, since Roman times. People who live in a state of perpetual itchiness have my heartfelt sympathy: I know they have already tried most of the drugs I will suggest to them, and they will go straight to the bottom of the non-urgent dermatology waiting list. When they finally get seen, they will be told that they should carry on using whatever it is that they are using and that skin tests will not help. It seems from this review that this is a largely evidence-free area. The options are clearly laid out, and there are enough of them to go on trying for most of a lifetime, with little indication of what might work for whom.

Lancet  27 Apr 2013  Vol 381
1461  I still don’t quite understand what the net effect of the H1N1flu epidemic of 2009 actually was. From the crude figures, it seems it was followed by a marked reduction in death from seasonal flu, and was therefore a good thing. On the negative side, it’s certain that many children with serious febrile illness had their diagnosis delayed by the instructions issued by the UK government that summer, and vast sums were spent on antivirals for which there is still no clear evidence of benefit. Developed countries also embarked on large vaccination programmes, and this study shows that this probably resulted in a few extra cases of Guillain-Barré syndrome. There must be lessons to be learnt from the balance sheet of this pandemic, but so far I haven’t come across a good review.

1469   Last week I had a bit of fun at the expense of Richard Horton’s Offline column, but to his great credit, the Lancet continues to be the most globally focussed of all the main journals, and actively encourages research on important clinical questions in the developing world. Here’s a good example: the DEVTA (Deworming and Enhanced Vitamin A) studies, which do what the acronym promises. In north India, vitamin A deficiency (retinol <0•70 μmol/L) is common in pre-school children and 2-3% die at ages 1-6 years. Previous studies have shown that this mortality can be reduced by giving supplements of vitamin A to all children: but DEVTA, surprisingly, does not. However, with unusual and becoming modesty, the Oxford team sets aside its own findings and concludes that “Meta-analysis of DEVTA plus eight previous randomised trials of supplementation (in various different populations) yielded a weighted average mortality reduction of 11% (95% CI 5—16, p=0•00015), reliably contradicting the hypothesis of no effect.” Hmm. “Reliably” depends on the quality and context of the other trials: but let that pass.

1478   And now for the D in DEVTA: Deworming. Once again, one million North Indian children got cluster-randomized, some to albendazole every six months, others to nothing. And once again, statistical significance was not reached: “regular deworming had little effect on mortality in this lightly infected pre-school population.”

So DEVTA was a logistical triumph, but a disappointment in terms of improving child health in Uttar Pradesh. Still: many congratulations to all concerned for model trials that need replicating in other locations of high child mortality.

1499   Two million children die each year from diarrhoea and pneumonia, and the Lancet has been running a series on this for a couple of weeks. This too is admirable, but the prose that results from such extensive coverage is less so. “Increasing of awareness of the size of the problem; strengthening of leadership, intersectoral collaboration, and resource mobilisation; and increasing of efficiency through the selection of the optimum mix of a growing set of cost-effective interventions depending on local contexts are the priority actions needed to achieve the goal.” Very true. This is indeed the solution to all the world’s problems: of achieving basic education, or of reducing carbon emissions, or of sharing water resources, or of improving labour relations. But in fact the sentence ends “of ending preventable deaths from pneumonia and diarrhoea by 2025.” True of that too, no doubt. It’s a great sentence for sounding important about anything global. One I like better is: “Scaling up of existing interventions against the two diseases to 80% and immunisation to 90% would eliminate more than two-thirds of deaths from these two diseases at a cost of US$6•715 billion by 2025.” That figure is uncannily close to 1% of the US defense budget for a single year.

BMJ  27 Apr 2013  Vol 346
An excellent surgical trial—acronym REFLUX (guess what it’s for)—gets an even better editorial from Peter McCulloch, putting it into context and drawing out the lessons for all surgical trials. This is good stuff, especially for British GPs, since it reflects the results which can be expected from UK hospitals with experienced surgeons performing minimal access surgery. The surgeons chose the kind of fundoplication they knew best, and they had to have done 50 to enter the trial. At five years, the fundoplication group had better relief of GORD symptoms than the medical treatment group. This trial also reflected the rough-and-tumble of real life in that over a third of the group offered fundoplication didn’t go ahead. It is evidence in the raw, but of a kind which bears interrogation: not like the ridiculous short term trial of a banding device with a surrogate end-point recently published in JAMA.

While you are snoozing, mowing the lawn, or watching your young play team sports, my weekends are usually spent writing these reviews, and doing out of hours primary care sessions. Theoretically, that leaves some weekdays free, though it never seems to happen. Anyway, I’m just telling you that I still see sick kids: a lot of them. And it always worries me. So I should welcome a clinical prediction model to aid emergency doctors managing febrile children at risk of serious bacterial infections, carefully developed and road-tested by a group of Dutch academics with advice from a GP colleague in Oxford. And yet I’m not really convinced that I would use this tool in my own out-of-hours primary care setting. It incorporates near-patient C reactive protein testing and has the capacity in some cases to rule out serious infection and the use of antibiotics. But for kids whose parents are really worried, and who look ill, I think I shall still call the paediatricians, and let them take over the worrying.

Just what is NICE for? And what should it be for? It has recently survived a nearly fatal attack by Lansley in his closing days as health secretary and is now under the direction of one of Britain’s most admirable GPs, David Haslam. Here is the article you need to read in order to understand what NICE is and where it might be going. It could yet be a great force for good in UK medicine. But it has many enemies in high and well-funded places.

If you are interested in end-of-life care, there is quite a lot to ponder over in this week’s BMJ. If you are not interested in end-of-life care, then you must be indifferent to your own inevitable fate and that of your loved ones. Should everybody with a terminal illness have the fact spelt out to them, in the name of informed decision-making? There’s a good head-to-head about that, and an excellent guide to the management of patients dying in hospital for the “Competent Novice”—worth reading by everyone, if only because all of us stand a worse-than-even chance of dying in hospital.

JAMA Intern Med  22 Apr 2013  Vol 173
614   Living with human immunodeficiency virus doubles your chance of having a myocardial infarction, according to this Veteran’s cohort study. It sound like everyone on antiretroviral treatment should be taking a statin.

632  Neat: a Boston group have come up with the HOSPITAL score for risk of readmission: “h emoglobin at discharge, discharge from an o ncology service, s odium level at discharge, p rocedure during the index admission, i ndex t ype of admission, number of a dmissions during the last 12 months, and l ength of stay.”

639   The debate about dietary calcium intake and cardiovascular disease continues. This study examines the relation in a whopping cohort of 388 229 men and women aged 50 to 71 years from the National Institutes of Health–AARP Diet and Health Study. The chalk-eating men seem to get more CVD whereas the women don’t.

664   There are far too many cardiovascular risk factors. Most of the studies are bad. Even when they are of reasonable quality, they tend to be of doubtful clinical significance. This paper written under the supervision of that invaluable sceptic, John Ioannidis, concludes that “Selective reporting biases may be common in the evidence on emerging cardiovascular biomarkers. Most of the proposed associations of these biomarkers may be inflated.”

Plant of the Week: Michelia doltsopa “Silver Cloud”

Exhausted by the long gloomy winter, we rushed off to Cornwall for a couple of days in the hope of experiencing spring. Hah! Most of England—including our village—enjoyed its first warmth, while Cornwall was cold and drizzly.

Still, there are quite a few plants you can only see close to the warming shores of Cornwall, and this is one of them. In Trebah garden it has been reclassified as a magnolia and planted in a bed of screeching orange and red dwarf azaleas. But there is no mistaking its class. It has slightly corrugated dark evergreen leaves, and is covered in large scented white magnolia flowers. Although it is quite a tender tree, it seemed to have fared better than many of the immense Himalayan magnolias which had come into flower and leaf in early March and suffered the consequences, looking lifeless and covered in brown remnants after the frosts.