Richard Lehman reviews the latest research in the top medical journals

NEJM  27 July 2017  Vol 377
Tocilizumab in giant cell arteritis
This week’s opening mab is tocilizumab, a humanized monoclonal antibody that targets the interleukin-6 receptor. As mab names go, this one is not too cili (I’m only doing this so you will remember it). In the UK, it costs about £12 000 per year. Here it is compared with placebo in a trial of steroid withdrawal in giant cell arteritis (GCA): not an enormous market, one would imagine, but then this is a very expensive drug in search of an indication. What this trial tells us is that when you taper off prednisone (costing almost nothing) very slowly over two years, GCA will relapse in about 50% of people, whereas if you give them weekly subcutaneous tocilizumab, the figure is more like 20%. We are all familiar with the long term harms of corticosteroids (they include respiratory infection and elevation of blood pressure and cholesterol, for example), but few of us know the long term harms of tocilizumab.

JAMA  25 July 2017  Vol 318
Football on the brain
“Punch drunk” didn’t sound medical enough in the 1920s, so it became dementia pugilistica: now it’s been extended to brain damage from other contact sports under the name of chronic traumatic encephalopathy (CTE). American football is, I am told, a contact sport. I’ve tried to watch two matches on television with the help of American friends, but I was entirely unable to work out why people were running around in the directions they did, and generally failed to spot either the ball or (more importantly) the collisions of great importance that were taking place during the brief periods of play. Collisions between players lead to the presence of phosphorylated tau aggregates in brain cells: they have been found in 99% of deceased National Football League players who donated their brains to science. Such insoluble bits of material do not help brain function and are diagnostic of CTE.

It’s possible that the same thing happens on a lesser scale among people who play proper football (soccer), since concussion in World Cup games is common and, according to a research letter, poorly handled or ignored in 63% of cases.

JAMA Intern Med  July 2017
Dizziness and one minute hypotension
I’ve recently had an experience of postural hypotension, and can confirm that a 50mm Hg systolic drop from sitting to standing causes symptoms. In my youth, we were taught to measure it using a long rubber tube connected to a mercury sphygmomanometer, hopefully within one minute, and ideally from a supine position. For some reason, the practice of measuring it after three minutes became favoured and is now hallowed by various guidelines. This is wrong. In a cohort of 11 429 middle-aged Americans followed up over 20 years, the one minute drop rather than the three minute drop was most closely associated with dizziness on standing, falls, fractures, syncope, car crashes, and death.

Sharing knowledge—must try harder
I’ve set myself up for an uncomfortable old age thinking about the shared understanding of medicine. First we need to understand it ourselves. Then we need to share it with the people we treat, in ways that make sense and don’t overburden or confuse them. Also we need to keep firefighting the simplistic or delusory ideas that are spread about, not just by commercial interests but also by experts who think they know best. And we need to think about how this could be done much better and more systematically, while acknowledging the limits of the possible. That’s where the sharing comes in. Tammy Hoffmann and Sharon Straus continue the Sharing Medicine series with a succinct and excellent overview of Sharing Knowledge for Health Care. At the moment it can be a grindingly slow and almost random process.

Ann Intern Med  25 July 2017 
Resuming statins and cardiovascular events
Statins are a classic case of how hard it is to achieve a shared understanding of medicine. Primary prevention with drugs is a relatively recent and problematic idea, and what doctors think people should do may not accord with what healthy feeling people wish to do. The evidence from double blinded randomised trials is pretty straightforward, while the observational evidence is often confusing. In this cohort study from China, based on electronic medical records, most people who were recorded as having had an adverse reaction to statins continued to receive prescriptions for a statin, including over 80% who were given a different statin and then had a further reaction to it. Over a mean of four years, people who continued their statins did slightly better than those who did not, judged by a composite outcome of myocardial infarction, stroke, or death. But the quality of the data is questionable, and this 1.7% difference could well be artefactual.

The Lancet  29 July 2017  Vol 390
C1 esterase inhibitor for deficiency
What might be the best treatment for hereditary C1 esterase inhibitor deficiency? The ideal agent might be C1 esterase inhibitor itself, you’d have thought. Pharming Technologies have now produced a recombinant form of the human inhibitor and tested it on 35 people, 26 of whom completed the study. This did reduce their attacks of hereditary angio-oedema, and it’s nice to see someone addressing this rather rare condition. I can’t see it coming cheap, though.

Romosozumab
In the title of this paper about romosozumab, the Lancet kindly adds “sclerostin monoclonal antibody,” to let us know what it does. It’s not as arcane as you think: inhibiting sclerostin should increase bone production and reduce bone resorption. If a woman with postmenopausal osteoporosis can’t tolerate or doesn’t respond to a bisphosphonate, this might be a choice for her, if it is safe and effective over a long period of time. In this open label trial funded by Amgen, Astellas, and UCB Pharma, romosozumab was compared with teriperatide in such women. The mab increased hip bone density over one year, whereas teriperatide did not. So here’s an unblinded, short term trial of two very expensive long term agents using a surrogate endpoint. Not my favourite kind of evidence.

The BMJ  29 July 2017  Vol 358
Inhaling the nicotine without the toxins
There’s a reluctance in most of us to accept addictive substances being sold to the public for profit, even if they are relatively harmless. In the US (and often in The BMJ) this aversion has extended to e-cigarettes, so it’s good to see this American observational study appear in The BMJ to redress the balance. The point of course is that most nicotine addicted people currently breathe in products of combustion that damage their lungs and circulation and shorten their lives, so a relatively harmless, pleasurable substitute may be a very good thing. In the US, greater use of vaping has coincided with a steeper drop in smoking, and many vapers say that they are using e-cigarettes as a means of smoking cessation.

Pharma: openness and closedness
Just six years ago, the idea that pharmaceutical companies would share data from all their trials seemed aspirational, if not ludicrous. Now a survey by Ben Goldacre and colleagues finds that 22 out of 23 top companies are committed to sharing all summary data prospectively and 17 are committed to doing it retrospectively. But there is a wide spread once you drill down to detail, especially about individual participant data. Massive progress, but still some way to go.

How long is a course of antibiotics?
I’ve occasionally taken antibiotics, but I don’t think I’ve ever finished a “course” of them. As soon as I’ve felt myself get better, I’ve stopped. I suspect most doctors are the same, and yet every pack we prescribe sternly warns the patient to complete the whole lot. I am glad to see the science of this unpicked so neatly in this article.

Although it has 10 authors, it reads like the product of a single person who can order thoughts well and convey them with clarity. I particularly like the division between “professional pathogens,” which contrive active harm to their hosts, and opportunist pathogens, which usually sit around contentedly eating snot, skin debris, or whatever it is they fancy and only get dangerous by unhappy chance.

Plant of the Week: Clematis “Kaiu”
The name of this clematis sounds Japanese, but is actually Estonian, being the name of the village where it was first noticed. The mother plant was C viorna, a modest native of the southeastern US whose flowers hang bell-like for a few weeks with recurved petals but never fully open. The seed parent is unknown, but I would guess it might have been the European thug called vitalba, or Old Man’s Beard, a great coverer of trees.

When C kaiu was first brought over to the UK in 2006, it was thought to grow to a maximum of about two metres and was applauded for flowering abundantly from July to August. But over coming years these plants never stopped growing, and developed a habit of producing flowers from late June right until the first frosts. Ours was bought to hide a domestic oil tank, but it could actually hide the whole garage if we didn’t cut it back ruthlessly during and after flowering. It doesn’t in the least mind our dry alkaline clay soil. It is lavishly covered with nodding bell-flowers of white flushed with pale purple near the base, curved prettily but without scent.

Here then is a great and imperturbable climber to send up a tree or wherever you might otherwise think of planting a Russian vine. As the years go by, be bold: it will come up perfectly well if you cut its rough liana trunk back to within a foot of the ground in winter.