Richard Lehman’s journal reviews—6 November 2017

Richard Lehman reviews the latest research in the top medical journals

richard_lehmanNEJM2 Nov 2017  Vol 377

Albuminuria in type 1 diabetic adolescents

Here’s a trial which seeks to find out if giving an angiotensin converting enzyme inhibitor together with a statin might avert deterioration of renal function in adolescents with type 1 diabetes and high albumin excretion. The primary end-point chosen was the albumin-to-creatinine ratio calculated from three early-morning urine samples obtained every 6 months over 2 to 4 years. Does this make sense? I can’t say, because nothing in nephrology really makes sense to me. If I wanted to track a person’s renal function, I would measure their cystatin C, not an inaccurate number like the eGFR or the amount of protein in their urine. Or if I really wanted to know what happened to a group of them in real terms, I would just have to wait a long time. Based on this study, there might be no difference between those given a statin and ACE inhibitor early on, and those left alone. At least, there was no difference in albuminuria.

Astronaut brain

I get the impression that medical journals want to be loved. Or if not loved, at least read; or if not read, at least reported on. Every now and again they publish papers that they know will hit the headlines of the lay press, like this one about the brains of people who have spent time being weightless in space. “Narrowing of the central sulcus, upward shift of the brain, and narrowing of CSF spaces at the vertex occurred frequently and predominantly in astronauts after long-duration flights.” My doctorly advice: don’t go there.

JAMA Nov 2017  Vol 318

Sad about sertraline

Chronic disease is often accompanied by low mood, and low mood is often labelled depression. Here is a trial of antidepressant medication in people who were classed as having chronic kidney disease stage 3-5. Their entry level of eGFR was below 45, meaning many were asymptomatic from their kidney dysfunction. But given sertraline at doses up to 200mg over 3 months, quite a few became symptomatic. Sertraline was chosen because it is cleared by the liver, not the kidneys, and it caused acute liver dysfunction in one patient out of 97 patients. As for the depression, the effect of sertraline was modest and exactly the same as placebo. And we are not told how many patients then found it difficult to come off sertraline. My advice again: don’t go there.

Ann Intern Med  31 Oct 2017  Vol 167

Deaths from SLE

Between 1968 and 2013 just over 100 million US citizens died, and just over 50,000 of those had “systemic lupus erythematosus” on their death certificates. So SLE is a relatively rare cause of death if these figures are to believed. However, while all-cause mortality showed a steady downward trend over this period, there was a rise in SLE-related death between 1975 and 1999, since when it has followed the overall decline in mortality. It would be interesting to get figures from other developed countries: they shouldn’t be hard to come by. Was there something that doctors were doing over those 24 years that was actually hastening death in SLE? High dose corticosteroids?

Type 2 diabetes: no hidden iceberg in US

With an estimated 850,000 cases of undiagnosed cases of Type 2 diabetes in the UK, around one person in every 74 has an undiagnosed case of the condition. This means that most people will have a friend or family member who has the condition but does not know it.” Diabetes UK made this claim in 2012, though I can’t find it on their website today. The idea of a hidden iceberg of diabetes persists, however. But an iceberg is 90% below water, whereas according to a new population study in the USA, only 10% of type 2 diabetes is hidden in the ocean, even though American primary care is much more hit-and-miss than ours. “When a confirmatory definition is used, undiagnosed diabetes is a relatively small fraction of the total diabetes population; most US adults with diabetes (about 90%) have received a diagnosis of the condition.”

The Lancet 4 Nov 2017  Vol 390

Surrogate-led treatment of Crohn’s

Expensive monitoring to drive expensive treatment: that’s what I first thought when I skim-read the abstract of this article entitled “Effect of tight control management on Crohn’s disease (CALM): a multicentre, randomised, controlled phase 3 trial.” And to some extent that is true: the expensive treatment used here was adalimumab, made by AbbVie, who sponsored the trial, and it costs £350 per 40mg shot. The monitoring, however, was comparatively cheap: faecal calprotectin comes at about £25 per test, while plasma CRP costs very little. “Tight control” consisted of treatment guided by these surrogate markers of disease activity, while clinical management consisted of treatment guided largely by symptoms. The primary endpoint was mucosal healing (CDEIS <4) with absence of deep ulcers 48 weeks after randomisation. This was achieved more frequently in the tight control group, who also ended up on higher doses of adalimumab and azathioprine. “Future studies should assess the effects of such a strategy on long-term outcomes such as bowel damage, surgeries, hospital admissions, and disability.” Indeed, and just to remind you: “AbbVie funded the study, contributed to the study design, participated in the collection, analysis, and interpretation of the data, and in preparation and approval of this report.”

ORBITA beams message of COURAGE

In 1959, the NEJM published a tiny sham-controlled trial of internal mammary artery ligation for the relief of angina pectoris. The idea was that by tying off the internal mammary arteries (which fill during systole) you could increase blood flow through the entirely unconnected coronary arteries (which fill during diastole). Many patients felt an improvement in their symptoms, and the procedure was starting to become popular after being puffed in the Reader’s Digest. But it was stopped dead in its tracks by this study of 17 patients, who were randomised to true ligation or skin incision only. It was clear that the operation had a transient placebo effect and nothing more. Nobody claimed that the trial was too small, because the burden of proof had shifted to those advocating the procedure, and they didn’t even have a plausible narrative to offer.

Contrast this with the widely held belief that angina is caused by narrowed coronary arteries. This is an entirely plausible narrative: in fact it is probably true. It therefore follows that by widening the coronary arteries using stents, you will provide better symptomatic relief than you can possibly provide using drugs. Even the COURAGE trial, which proved that medical treatment was as good as stenting for preventing myocardial infarction and death, left open this possibility for a few patients. But now comes the British ORBITA study, which moved beyond COURAGE by using an active (sham) comparator to assess the effect of stenting in people with stable angina who showed severe stenosis on their angiograms. The result has sent shock waves across the interventional cardiology community: “In patients with medically treated angina and severe coronary stenosis, PCI did not increase exercise time by more than the effect of a placebo procedure.” But should we be surprised? For the second week running I direct your attention to Ending Medical Reversal by Vinay Prasad and Adam Cifu (2015), p.29 “Could stenting for people with angina from stable coronary artery disease be like ligation of the internal mammary artery? Could the small, relatively short-lived effect simply be a placebo effect?..It should be tested. If stenting were found to be no better than sham stenting then…(this) would be the greatest example of reversal, at least in terms of cost, in the last 20 years”. Yes.    

The BMJ 4 Nov 2017  Vol 359

Towering Tessa

In 2004, Tessa Richards was given a diagnosis of widespread invasive adrenal cancer. You can read her typically clear-headed, but deeply moving account of the experience in an article she wrote nine years later. Being Tessa, in the interim she just got on with what she thought most worthwhile: fiercely advocating for patients as an editor at the BMJ. Now she updates us on her journey, but only for the sake of the lessons it teaches about the current shortcomings of cancer care, and in particular how even the best-informed and empowered patients can feel terrified and excluded from decisions which profoundly affect the remaining portion of their lives. No quotation can do justice to this piece: everybody needs to read it in full, and come back to it each time they have anything to do with the care of a person with cancer.  

Can I have that hip, please?

By comparison with the treatment of advanced cancer, sharing decisions about a hip prosthesis or procedure should be a piece of cake. Ah, but which piece of cake? Because once you have eaten hip cake, it stays inside you for a long, long time. Figure three in this systematic review and meta-analysis of implant combinations in hip replacement says it all. Perhaps you would like to try the Belgian chocolate acetabulum over here, madam, which goes very well with the filou pastry and Bramley apple femoral compartment on the tier below? Certainly you can have the salted caramel gateau sir, and may I suggest the sour cherry flan with chef’s custard to go inside the femur? Oh orthopaedic waiter, I can’t make up my mind: you decide.   

Patient-centred diagnosis

For several years now, The BMJ has supported conferences called “Preventing Overdiagnosis”, and the Overdiagnosis group of UK GPs is the liveliest forum of medical debate that I know of. So what is the opposite of Overdiagnosis? It would be rather worrying if it were Underdiagnosis, but I think the right term is in the title of this long overdue and extremely welcome Analysis piece. It is Patient-Centred Diagnosis. Every label has potential consequences for the people labelled. Every test can lead to further tests. This excellent piece should be the first of a whole series.

  • kidmugsy

    Thank you for pointing to the Prasad and Cifu book. For years I have asked what the evidence is that high BP causes deaths. Their second anecdote suggests that perhaps it doesn’t. For if two drugs reduce BP to the same extent but only one extends life, that does suggest that BP isn’t the killer. Other explanations are possible of course; all sorts of things can be explained away by the imaginative. Has any more light been shed on this in the last dozen years?

  • kidmugsy

    And thank you for your wonderful quotation “With an estimated 850,000 cases of undiagnosed cases of Type 2 diabetes in the UK, around one person in every 74 has an undiagnosed case of the condition.” It got the chortle it deserved.

  • Craig Medred

    one would hope to God the journals want to be read. if they’re not read, they’re just a waste of everyone’s time and resources.