Richard Lehman reviews the latest research in the top medical journals

NEJM  24 May 2018

Dupilumab for steroid-dependent asthma

Forty years ago, I believed that asthma might be sorted by the time the new century arrived. Sadly not: we have a few more mechanistic insights, but then as now treatment depends largely on corticosteroids and ß-adrenergic drugs. For a couple of decades people have been developing monoclonal antibodies to various bits of the cytokine pathways which play a part in most inflammatory conditions, including asthma. It’s then a matter of trying them out and finding the ones that work best; a job which is strangely left largely to their manufacturers. Sanofi and Regeneron Pharmaceuticals have developed dupilumab, a fully human anti–interleukin-4 receptor α monoclonal antibody that blocks both interleukin-4 and interleukin-13 signalling. In one trial they got to compare it with placebo injection in 210 people with asthma treated with oral glucocorticoids.

In another trial they randomised 1902 patients 12 years of age or older with uncontrolled asthma in a 2:2:1:1 ratio to receive add-on subcutaneous dupilumab at a dose of 200 or 300 mg every 2 weeks or matched-volume placebos for 52 weeks.

Dupilumab showed a helpful effect in both groups. But, as the trenchant accompanying editorial points out, “in the past few years mepolizumab, reslizumab, benralizumab, and dupilumab—(all) directed largely against type 2 inflammation, broadly writ, will have been introduced for the treatment of asthma… how will we decide which of these treatments should be used now? Without more data, these decisions are likely to be driven by the marketing arms of the various entities that produce these drugs. We envision that there will be a cottage industry based on analysis and reanalysis of the current data, each driven by the desire to show greater effectiveness, better safety, or more convenience to the patient, for one product versus another. These approaches will inevitably be flawed. Instead, clinical researchers should design and mount pragmatic head-to-head trials of these four new asthma-treatment biologics.”

Wow, yes indeed. And who is the author of these inflammatory words about inflammatory pathways? No less than the editor of the NEJM, the journal that sells bulk reprints to the marketing arms of the various entities that produce these drugs.

Procalcitonin fails to guide

Forty years ago I also thought that someone was bound to come up with a test that could distinguish between viral and bacterial infections. In fact they already had without knowing it. Procalcitonin was discovered in the 1970s as a precursor of calcitonin, but was only recognised as possible discriminant between viral infection and bacterial infection about 20 years later. Some European trials in primary care suggest that antibiotics can safely be withheld from patients with signs of lower respiratory infection but normal levels of calcitonin. This has led to suggestions that procalcitonin, despite its rather poor predictive characteristics, should be widely used to reduce antibiotic prescribing. But a US hospital-based study specifying a strict protocol for a procalcitonin-based prescribing regimen in people with strictly defined lower respiratory infection showed minimal effect on antibiotic use. Or patient outcomes. This is an idea whose time came a while ago, but has never quite arrived.

Antiphospholipid syndrome

To continue in old-man-looking-back mode, I became aware of antiphospholipid syndrome in the 1990s, about ten years after Hughes first described it. It was one of the conditions which led me to think of the Easily Missed series for The BMJ. In fact the fulminant kind is not easily missed, and remains mysterious. “ Patients with catastrophic antiphospholipid syndrome present with thrombosis involving multiple organs.” as this latest review points out. It was a case like this that woke me up to its existence. But then I became aware of what this article calls “obstetrical antiphospholipid syndrome”, characterized by fetal loss after the 10th week of gestation, recurrent early miscarriages, intrauterine growth restriction, or severe preeclampsia. I started looking for that too, but gradually gave up sending of bloods for antiphospholipid tests. They rarely lead anywhere because unfortunately there is little useful treatment, other than warfarin for those who have had venous events.

JAMA 22/29 May 2018

Annual surveillance after resected colorectal cancer

Two studies, one from Sweden, Denmark and Uruguay and the other from the USA, look at the effect of actively detecting recurrence of colorectal cancer after surgery, either intensively or by routine follow-up. In the first study, which was a randomised trial, there was astonishingly little difference in time to recurrence, whereas in the observational study the more closely followed-up patients had the return of cancer detected a little earlier. But the great thing was that it made no difference to survival rates in either study: five-year survival in the first was 86.4% and in the second was 73.7%. These seem wonderfully good figures to me.

Closing the L atrial appendage during cardiac surgery

Just as surgeons used to remove the vermiform appendix during laparotomy for other reasons (“while we’re in here”), so cardiac surgeons have taken to closing off the left atrial appendix while they are doing other things inside the heart. And this appears to be a good thing. A retrospective study based on Medicare data from the US finds used propensity testing to match large cohorts of patients who did or did not have L atrial appendage occlusion (LAAO) while having cardiac surgery for a different primary reason. Those having occlusion constituted only 5.9% of the total, so whatever the matching attempted, I don’t think any watertight comparisons can be made. But for what it’s worth, those who had LAAO went on to have fewer strokes and live longer after adjustment for the confounders the authors considered.

JAMA Intern Med May 2018

Aldosterone blockade for all after MI?

Of late, I’ve taken to whingeing about the uselessness of most meta-analyses as a guide to individual treatment. But here is one which does something else: it examines an important hypothesis and finds prima facie evidence that it may be true. People who were given aldosterone antagonists after ST-elevation myocardial infarction in ten trials showed a reduced mortality rate, even though their left systolic ejection fraction was normal. The absolute mortality reduction in these 4147 patients was from 3.9% to 2.4%. So there is a now a convincing case for a very large randomised trial of aldosterone antagonists following MI of any kind, and it would be very good to know which groups are most likely to benefit. But alas, this is not an individual patient meta-analysis, so we don’t know.

The Lancet 26 May 2018

How deadly is diabetes?

“Diabetes has been associated with an approximate 75% increase in mortality rate in adults, and the average 60 year old person diagnosed with diabetes loses 5 years of his or her life to the disease” says the opening sentence of this paper. Where do those figures come from and what do they mean? I have absolutely no idea. And it doesn’t matter anyway because the whole point of this study is to show that anything that has previously been said about diabetes and mortality is wrong. Using data from the US National Health Interview Survey Linked Mortality files from 1985 to 2015, the authors find that all-cause death rates declined by 20% every 10 years among US adults with diabetes (from 23·1 to 15·2 per 1000 person-years), while death from vascular causes decreased 32% every 10 years (from 11·0 to 5·2 per 1000 person-years), deaths from cancers decreased 16% every 10 years (from 4·4 to 3·0 per 1000 person-years), and the rate of non-vascular, non-cancer deaths declined by 8% every 10 years (from 7·7 to 7·1). That drop in CV mortality is spectacular, and beyond what has happened in the population at large. I won’t say anything more about this one epidemiological study relating to “diabetes”: maybe when I have freed up a bit of time by dropping these reviews I’ll be able to write at greater length about this delusive label.

Renal denervation revisited

Radiofrequency denervation of the renal artery usually produces a large drop in blood pressure, but the place of this procedure is contentious. Medtronic bought up rights to the technique several years ago following successful unblinded trials, but then paused to conduct its own trial, which did not meet its prespecified end-point. Now it’s doing another trial, wisely allowing drug treatment to continue while at the same time monitoring drug adherence.  

The six month results are promising, but perhaps the most telling aspect is that medication adherence was about 60% and varied for individual patients throughout the study. “Resistant hypertension” is a category which resists simple definitions.

A second company, ReCor Medical, manufactures an ultrasound-based renal denervation device and used a much simpler approach to recruit for its trial.

803 patients were screened for eligibility and 146 were randomised to undergo renal denervation (n=74) or a sham procedure (n=72) across 21 centres in the USA and 18 in Europe. This is baffling: the entry criteria are so common that you could collect enough participants from a single large practice (patients with combined systolic–diastolic hypertension aged 18–75 years were eligible if they had ambulatory blood pressure greater than or equal to 135/85 mm Hg and less than 170/105 mm Hg after a 4-week discontinuation of up to two antihypertensive medications and had suitable renal artery anatomy). Conclusion: “Compared with a sham procedure, endovascular ultrasound renal denervation reduced ambulatory blood pressure at 2 months in patients with combined systolic–diastolic hypertension in the absence of medications.” Two months follow-up for a lifelong irreversible intervention ? Oh please.

The BMJ 26 May 2018

BNP: so promising, so disappointing

I gave ten of my best years to BNP. I read the earliest papers in 1993 with great excitement: here, resounding through the bloodstream, was the very voice of the myocardium. All you needed to do was sample some blood and you could hear the stretched ventricles shouting in distress or the happy heart murmuring contentedly. Surely here was the simple key to detecting heart failure in primary care and to monitoring treatment. A quarter of a century later, people are still pursuing this attractive phantom. Here’s a summary of their efforts in the form of a meta-analysis of the diagnostic accuracy of point-of-care natriuretic peptide tests in patients with chronic heart failure, with a focus on the ambulatory care setting.

It concludes that “large scale trials in primary care are needed to assess the role of point-of-care natriuretic peptide testing and clarify appropriate thresholds to improve care of patients with suspected or chronic heart failure.” On the basis of my own experience, I couldn’t agree less. Heart failure is a complex, comorbid condition in which BNP bounces all over the place from hour to hour. And BNP will go up with anything else that strains either ventricle, such as RV strain due to lung infection. It is mostly just noise.

Bad eczema is a cardiovascular risk

“Patients with severe atopic eczema had a 20% increase in the risk of stroke, 40% to 50% increase in the risk of myocardial infarction, unstable angina, atrial fibrillation, and cardiovascular death, and 70% increase in the risk of heart failure.” This unwelcome news comes from a population based matched cohort study using three UK databases. The confidence intervals are rather large, but I think these data substantially increase the  cardiovascular risk score of everyone with severe eczema. Like everyone else, they should have access to risk-reducing treatment if they wish.

Plant of the Week: Meconopsis cambrica

The Welsh poppy is a thuggish weed, or a lovely gift from the gods that every garden should treasure. After a quarter of a century, my wife has persuaded me mostly towards the latter point of view. Our garden is dotted with papery flames of yellow and they look wonderful, especially where blue hardy geraniums are beginning to appear.

If we had a garden with a semi-wild patch, this combination—Welsh poppies with meadow geraniums—would be a great idea, because they could both flower together without hindrance over several months. In a small and ordered garden they both need a bit of discipline, but that is easy to provide. If they arrive in the wrong place, just pull them up. With permission, of course.

Could you write a weekly journal review?

Richard Lehman will be retiring from writing his weekly reviews in the Summer 2018. We are looking for someone to replace Richard and write a weekly journal review. We are interested in candidates with excellent knowledge of research and critical appraisal, and an understanding of how research impacts on clinical practice.

We are particularly interested in candidates who have clinical experience and an excellent standard of English. You must be motivated, creative, and able to write the weekly reviews to a tight deadline every week.

This freelance position offers a salary per column.

Please apply with a CV and covering letter and an example of how you would write a research review.

Please send applications to: Juliet Dobson, Digital content editor, The BMJ—jdobson@bmj.com