Richard Lehman reviews the latest research in the top medical journals

NEJM  4 Jan 2018  Vol 378

Swapping cervical nerves

Starting out as a medical student, I got interested in what the biceps and brachioradialis reflexes might tell you about the location of a lesion. Their  innervation was said to be C6/7, but as I dug deeper I came to realise one of the most important lessons of medicine: textbooks will tell you all sorts of things, but almost never the thing you want to know. Now, 45 years later, I’ve found it in the preamble to this Chinese trial:

“Among the spinal nerves, the five that give rise to the brachial plexus (C5, C6, C7, C8, and T1) together contain approximately 40,000 to 69,000 nerve fibers and innervate the entire upper extremity. The C7 nerve accounts for approximately 20% of these fibers. Because the motor function of the C7 nerve largely overlaps with that of the other four nerves that give rise to the brachial plexus, severing this nerve usually results in only transient weakness and numbness in the ipsilateral upper extremity.” So it seems that C7 is a sort of back-up nerve, which is why the textbooks so often throw it in with C6. Chinese neurosurgeons spotted an opportunity here: 36 young patients from Shanghai had suffered brain injury manifesting mainly as spasticity and weakness in the upper extremity contralateral to the cerebral lesion, which had failed to improve over at least 5 years. So in half of them, they tried implanting the affected arm with the C7 nerve root connected to the undamaged ipsilateral brain and waited to see what happened. There was immediate decrease in spasticity, and some return of muscle power over the following year, but it’s unclear whether this is going to be a worthwhile long-term option for this kind of neurological injury.

JAMA Intern Med Jan 2018

Puffing it and snuffing it

Unless pharmacology has changed since I last read any, beta-adrenergic receptors react to constant stimulation by down-regulation. So it should come as little surprise that inhaled long-acting β2 agonists are at their most dangerous in the first weeks of use, when they cause most of their cardiovascular overdrive. Whole population data from Taiwan show that the same applies to long-acting muscarinics. “New initiation of LABAs or LAMAs in patients with COPD is associated with an approximate 1.5-fold increased severe cardiovascular risk, irrespective of prior CVD status and history of exacerbations.” I’ve never been quite sure why we use them so much. After a month, the CV risk fades, but does any meaningful benefit accrue? How can any individual with COPD judge whether they are getting fewer exacerbations than they otherwise would have?

Gabapentinoids: Less is (probably) More

The widespread use of gabapentin as an analgesic is a classic tale of drug mission creep. Some of it was good. When it appeared as an anticonvulsant drug in the early 1990s, gabapentin offered little advantage over existing drugs for epilepsy. But gradually it was adopted for use in neuropathic pain, where its adverse effect profile and effectiveness gave it the edge over carbamazepine. GPs needed little encouragement to give it a try. But in the UK, I don’t think use really shot up until the 2000s. In the US, this research letter tells us that prescribing of gabapentin and pregabalin went up threefold between 2002 and 2015. People prescribed the “gabapentinoids” tend to be older and already on opioids and benzodiazepines. From my own experience, I suspect this is in part due to pain clinics and palliative care teams wanting to help patients with residual pain, by reclassifying it as “neuropathic”. We need to think of ways to reverse this trend. For every patient to benefit, there are probably ten who end up fuddled and immobile. Every time you see one of these drugs on a chart (usually somewhere in the middle, as drugs accumulate geologically), think “don’t I need to check this with the patient?”.  

Ann Intern Med  2 Jan 2018  Vol 168

Checking out checkpoint inhibitors

Ipilumab, nivolumab, pembrolizumab, atezolizumab, durvalumab, avelumab. What do they have in common? Silly names. Yes. Exorbitant prices. Of course. Severe adverse effects related to inflammatory upregulation? Sadly, yes. Although they work by slightly different mechanisms they are all called checkpoint inhibitors (don’t ask me, look it up on Wikipedia), and they do increase overall survival in a number of cancers. Next list: psoriasis, ulcerative colitis, Crohn’s disease, multiple sclerosis, myasthenia gravis, autoimmune thyroid disease and sarcoidosis. What do these have in common? Yes, they are all nasty and incompletely understood. Yes, they have autoimmunity as part of the disease process. So what happens if you give a checkpoint inhibitor to cancer patients with these diseases in order to prolong their overall survival? According to a mixture of observational evidence, their autoimmune diseases get worse, but they live a bit longer, especially if they have severe adverse effects. These effects can in turn be alleviated somewhat with anti-immune drugs (mostly steroids). This is a seriously tricky area for shared decision making, but all the more necessary for that reason. Particularly in the USA where the cost of treatment is likely to clear out your bank account and could lose you your home.

Feeding statins to the herd

Speaking of shared decision making, how about statins for primary prevention of cardiovascular disease? Well, it’s pretty clear. There are two opposing camps who begin from entirely different premises (or premisses). Premiss (philosoph.) One: If everybody took statins, there would be less cardiovascular disease in the population. Premiss Two: It is up to every individual to decide whether to take medication for life, based on their individual risk and informed preference.

If you put people randomly in a room to discuss statins for half an hour, the space will get hotter, its pressure will increase, and the two factions will end up at opposite sides of the room. It’s like Boyle’s Law of Gases in reverse.

Now throw in a guideline. The gases will explode, but then the same negative entropy will reassert itself and within half an hour every particle will be back in its original position. You can do this five times and the same thing will occur each time. Here’s the proof. Danish investigators looked at 5 European guidelines recommending that statins should be offered to anything between 44% (max) and 13% (min) of the adult population. They then modelled the likely incidence of CV events based on Danish whole-population statistics if statins were taken by these percentages of the adult population. Naturally, the more Danes taking statins, the fewer CV events predicted in Denmark. But what if the population would not deign (sorry) to take them?

Now here’s the natural experiment I would like to see carried out, in Denmark or anywhere else. Offer statins free of charge to the whole population, together with a CV risk assessment tool and a multi-layered decision aid. Keep doctors out of it. Observe the pattern of uptake and adherence. Introduce an element of proactive pharmacovigilance to detect hidden adverse effects, and test these with n-of-one trials in individuals. Publish observational data every 5 years.

The Lancet 6 Jan 2018  Vol 391

Surgical mortality in Africa

This survey of short-term outcomes following surgery in 25 African countries does not read comfortably. The patients are mostly younger and fitter than in other parts of the world, but most of the surgery is done as an emergency by a desperately small and under-equipped workforce. Result: people die within a week of surgery in Africa twice more often than elsewhere.

The BMJ 6 Jan 2018  Vol 360

BP lowering drugs work incrementally for high-risk people

Here’s an important secondary analysis based on data from the SPRINT trial. If you are interested in managing high blood pressure, you need to pore over this paper which shows that the benefits of additional treatment accrue in proportion to total cardiovascular risk. I won’t say much more, because the issues for decision making are very simple in principle, and nearly identical to those for statin prescribing. But the devil is in the detail, and if you venture further, a whole Hieronymus Bosch landscape will open up before you, complete with pink torture-kettles, strangely toothed fish and large green buttocks.

Plant of the Week: Daphne bholua

Sixty years ago Major Spring-Smyth was walking on a ridge of the Nepalese Himalaya among the melting snows of March, with a trusty gurkha for companion. He caught sight of a shrub covered with starry pink flowers, breathing a fragrance that made his moustache bristle. Could this be the fabled Daphne bholua, described a century before but never properly introduced into European cultivation?

“Gurkha”, he said to his trusty companion, “be a good man, and dig up that bush and run down to the coast with it, would you? When you get there—it shouldn’t take many weeks—see that it gets on the next ship to England. Water it daily and make sure it doesn’t get too hot. If all goes well and it grows in English gardens, I’ll make sure that your name is remembered for this.”

It came to pass as he had said and the plant thrived and became known as Gurkha.

Although valued by a few connoisseurs for its wonderful pervasive sweet scent in winter, D bholua “Gurkha” never really took off as an English garden plant, because it presents an increasingly scraggy appearance as winter progresses and the leaves fall off. Then in 1982, Alan Postill at Hilliers Nursery raised a seedling of Ghurkha which proved to be hardy, evergreen and very free-flowering. He named it for his wife, and ever since then, D bholua “Jacqueline Postill” has been the one to have.

We used to have a mature specimen over 2m high which barely survived one exceptionally cold winter, and died after the next one. Since then, we’ve tried a few others—”Darjeeling” and a couple of white forms—none of which did much during their short lives. So we’ve gone back to Jacqueline Postill, ignoring the disdain of the specialist nurseryman who tried to sell us his most recent variety. For all his faint praise, mild eye-rolling and display of left nostril hair, this remains the best form of the best winter plant there is. Buy a well-grown specimen about a metre high (never mind the price tag) and place it by your front door, as I’ve told you many times. It will even make you want to go out in January.