Richard Lehman’s journal review—16 November 2015

richard_lehmanNEJM 12 Nov 2015 Vol 373
SPRINTing to conclusions
OL There’s no denying that the research event of the week has been the online publication of the SPRINT hypertension trial. So here I am departing from the usual pattern of these reviews, because it would seem odd not to begin with it. I’m going to be very brief. The trial shows that in people of mean age 68 years with elevated cardiovascular risk, but not diabetes, aiming for a systolic blood pressure target of 120 rather than 140 mmHg produced a significant reduction in death and cardiovascular events over a median period of three years and three months. The patient characteristics, benefits, and harms in this trial are very well described. If you are consulted by an individual who fits its criteria, you could have an informed conversation about their risk lowering options, of which tighter BP control might be one.

When “non-inferior” means “not as good”
1905 My guess is that you are not in the least interested in the comparative outcomes of a trial of everolimus-eluting bioresorbable vascular (Absorb) scaffold versus an everolimus-eluting cobalt–chromium (Xience) stent. But take the trouble to click on the link to Figure 1 of this article. It’s very clear that right from the start, the failure rate with the bioresorbable device is higher than with the metal stent. Which of course is not what the manufacturers hoped to show. So they report that its performance “was within the prespecified margin for noninferiority with respect to target-lesion failure at 1 year.” Various readers have complained over the years when I’ve used the inelegant term “non-inferior,” which means “not worse according to the arbitrary statistical margin used in the trial (almost always 5%).” It does not mean “same as,” and as this report shows, it doesn’t even mean “no worse than” as we usually understand it. It just means “non-inferior.” I hope that is clear.

The road to melanoma, cell by cell
1926 Mind, prepare to boggle. For more than a century, histopathologists have used much the same methods to slice and stain and peer down at bits of tissue. Rudolf Virchow, who in 1855 proclaimed the principle of omnis cellula de cellula (everything a cell that comes from a cell), would until recently have been able to settle down comfortably in a modern histology lab and just carry on doing the sort of science he first invented. But now we have genomics, and it is time to get out the old tissue blocks and look at them again. “A total of 37 formalin-fixed, paraffin-embedded (FFPE) melanocytic neoplasms, enriched for melanomas with histologically distinct precursors, were retrieved from the archives of the dermatopathology sections of the University of California, San Francisco; St. John’s Hospital in London; and the University Hospital of Zurich. In total, 150 distinct areas were manually microdissected for sequencing.” I can’t possibly tell you all that they found, but amidst the complexity, order emerged. “Our study defined the succession of genetic alterations during melanoma progression, showing distinct evolutionary trajectories for different melanoma subtypes. It identified an intermediate category of melanocytic neoplasia, characterized by the presence of more than one pathogenic genetic alteration and distinctive histopathological features.” Bravo. Virchow would be proud of you.

Carry on living with COURAGE
1937 The COURAGE trial was nearly ruled unethical before it even started. It was obvious that if you have blocked pipes, it must be best to unblock them. When it was published in 2007, it caused howls of pain from interventional cardiologists, interrupted only by the glugging sounds of words being eaten. And yet here are the latest data: “During an extended-follow-up of up to 15 years, we did not find a difference in survival between an initial strategy of percutaneous coronary intervention plus medical therapy and medical therapy alone in patients with stable ischemic heart disease.” Odd then that triallists continue to recruit patients with stable angina or ischaemia without symptoms into trials of PCI, long after COURAGE was published. There was one in the NEJM just two weeks ago.

JAMA 10 Nov 2015 Vol 314
Leg it further with a stent
1936 For some reason, stents in blocked leg pipes have never caught on in the same way as stents in blocked heart pipes. A Dutch trial compared supervised exercise alone for intermittent claudication with exercise plus selective stenting in 212 patients who got pain at around 100m. The revascularized group showed greater improvement in walking distances and health related quality of life scores at one year.

ACS is defined by troponin, so you need troponin to define ACS
1955 The key text that set up evidence based medicine in the early 1990s was the second edition of Clinical Epidemiology by David Sackett et al, and the key journal series was The Rational Clinical Examination, which started off with an introduction by Sackett a couple of years later. Looking back, it’s interesting that the main focus of these now canonical writings was diagnosis, not evidence synthesis. In particular, the book took its readers through several chapters about diagnosing myocardial infarction with the help of an inadequate biomarker, creatine kinase, which was fairly sensitive but not at all specific. Then, around the turn of the millennium, several studies appeared showing that the dying cardiac myocytes released highly specific cellular degradation markers called troponins. I can remember asking my review followers (probably about 40 at the time) why this test wasn’t available in all British hospitals. Within a year, it was. This is a great example of diagnostic technology that is so good and so important that it was adopted with lightning speed, everywhere. But every time a brilliant new diagnostic test appears, it changes the epidemiology of the condition it defines. In the old days, we didn’t talk about “acute coronary syndromes” but about myocardial infarction and unstable angina. With the universal adoption of troponin measurement, we began to talk about ST-elevation MI and non-ST-elevation MI: the acute coronary syndromes (ACS). At the same time, the standard of treatment became percutaneous coronary intervention rather than thrombolysis. I’m leading you through all this seemingly ancient history because the latest Rational Clinical Examination paper seems to go in a complete circle on this. “Among patients with suspected ACS presenting to emergency departments, the initial history, physical examination, and electrocardiogram alone did not confirm or exclude the diagnosis of ACS. Instead, the HEART or TIMI risk scores, which incorporate the first cardiac troponin, provided more diagnostic information.” Well yes, they would.

If you mess with mineralocorticoid blockers, test often
1973 More history from the dusty shelves of cardiology. Back in the 1980s, we learned that a new class of drugs could improve outcomes in heart failure caused by cardiac damage. Captopril, the first of these angiotensin converting enzyme inhibitors, had a marked first dose effect which could make people fall over from postural hypotension, especially if they were already fluid depleted. So ACE inhibitors were always initiated at very low doses and with hospital monitoring of blood pressure. I thumbed through the literature recently and found a case report from the late 1980s in which several patients over the age of 80 were given a tiny dose of captopril and half of them fainted and/or showed an acute deterioration in renal function. But as time went on, we blithely ignored these dangers and gave everyone with systolic heart failure an ACE inhibitor or angiotensin receptor blocker to take at home: first at night, and then in the morning, at ever increasing doses. These drugs have become one of the commonest causes of admission with acute kidney injury and hypotension, but on we go. Here is a research letter showing from Medicare and Medicaid billing data, renal function testing is seldom carried out adequately when initiating mineralocorticoid receptor antagonist therapy in patients with heart failure.

JAMA Intern Med Nov 2015 Vol 175
The perils of pharmacological mission creep
OL Medicine is full of mission creep. Drugs that were tested in trials for one purpose end up being used much more widely for others. Gabapentin is an example: it was developed to control epilepsy, but these days you are risk of being prescribed it for anything from mood swings to a painful ingrowing toenail. In America this is known as “off-label” use, and here is a study showing that your chances of experiencing an adverse effect are more than 40% higher if you are given a drug for an off-label indication.

Lancet 14 Nov 2015 Vol 386
Take these pills and hope that nothing happens
OL Now it’s time to think again about high blood pressure. It’s the commonest cause for giving drugs to healthy people, and most of them will never benefit. So what are the most pressing issues in blood pressure research? I’d say they are (a) the identification of those who do not need to take drugs and (b) the best way to share decisions with individuals about treatment based on accurate personal estimates of benefit and harm. At present we are a long way from being able to address these issues adequately, but at least they are beginning to appear as dim shapes beyond the fog of population studies. There must be a tipping point in lowering BP at which the harms outweigh the benefits. A meta analysis like the one here tries to establish this for whole populations, but we can only start practising person centred medicine when we start taking into account the tipping point for each individual. The SPRINT trial is quite good because it gives us enough information to do this for a limited group of people over a short time, but we need to be able to do it for everybody over a lifetime. And SPRINT also vindicates the conclusion of this meta analysis: “In high risk patients, there are additional benefits from more intensive blood pressure lowering, including for those with systolic blood pressure below 140 mmHg. The net absolute benefits of intensive blood pressure lowering in high risk individuals are large.” Personally, I shy from using the words “patients” and “benefits” in this context. The benefits of lowering BP lie in the things that never happen, and so can never be known to these individuals, however high their risk may be. And of course their risk can be lowered by other things besides BP lowering drugs.

BMJ 14 Nov 2015 Vol 351
Behave, or you will lose your eyes and your legs
Diabetes is the commonest cause of blindness and limb amputation in developed countries. The prevalence of diabetes is increasing in developed countries. Blindness and limb amputation are rare complications of diabetes. I am giving you this cluster of statements—all equally true—to continue the theme of population versus individual benefit. To lower the population risk of blindness and amputation, we need to manage diabetes better, though blood sugar control is only a small part of that, and there’s a lot that we don’t understand. To lower an individual’s risk of blindness or amputation, we expect that person to take several medications which reduce a tiny risk to one that is a bit tinier. As with control of blood pressure, we’ve traditionally been very cavalier in the way we have conveyed these risks to individuals. We’ve measured the BP and said “I think you should take these pills because we don’t want you to have a stroke,” without ever specifying the actual risk of stroke. And we’ve all used similar catastrophizing language to improve “compliance” in people newly diagnosed with diabetes, with the result that the healthtalk website used to feature a grown man crying as he recalled that being told this made him feel worse than if he had cancer. So two cheers for a new instrument that enables better prediction of the people with diabetes at most risk of blindness and limb ischaemia. Let’s hope that, properly and cautiously used, it helps a few individuals without inducing crippling anxiety in the great majority who will never suffer these outcomes.

The computer will console you now
“Guérir quelquefois, soulager souvent, consoler toujours.” To cure sometimes, to relieve often, and to comfort always: these are the three tasks of medicine, in a saying attributed to nearly every famous doctor since Hippocrates. We are told by no less an authority than Jeremy Hunt that computers will soon replace the doctor in working out how to cure people. No doubt they will also be quite good at selling symptom relief products. And, finally we must ensure that they can take over the business of comforting too. It is just a question of British industry coming up with right cognitive behavioural programme. Alas, we are not quite there yet. “Supported computerised CBT does not substantially improve depression outcomes compared with usual GP care alone. In this study, neither a commercially available nor free to use computerised CBT intervention was superior to usual GP care.” Sadly, we are still some way short from the happy day when the NHS can run with no doctors at all, so that we may all be able to lead fulfilled lives within the financial sector.

Plant of the Week: Skimmia “Kew Green”

The further North you go, the fiercer become the seasonal myths of death and rebirth. Here trees lose their leaves, animals hibernate, the ground is covered in ice and snow, the sun nearly or actually disappears. It is a time of monsters and trolls. The wind howls and the sea rages. The old and the weak perish, the traveller departs and never returns. The hurdy gurdy man stands on the ice with naked feet.

In fact November this year has worn a benign aspect in England. There is still the odd rose and hollyhock flower in the garden, and apples still hang on the tree. But you know these will soon be gone. There is just one plant that shouts resurgam—I will rise again. And that is the lovely skimmia called Kew Green, which I have praised countless times before in these reviews, because it grows a few feet from where I write them.

Pretty in habit, its lanceolate evergreen leaves shine in a dome beneath the pallid sun. And all among them are spikes of flower bud which seem about to open. You know that they will carry an amazing scent that will fill the garden for weeks, and that bees and insects of every kind will crowd them in due season. Not before four barren months have gone by. But it is good to dream on these things.