Richard Lehman’s journal review—14 September 2015

richard_lehmanNEJM 10 Sep 2015 Vol 373
997 Are any readers looking for a nice short term research project in evidence based medicine? Here you have it. On 1 September, the MATRIX triallists reported the results of a trial which randomised 7213 participants with an acute coronary syndrome to receive either bivalirudin or unfractionated heparin prior to percutaneous coronary intervention. Bivalirudin costs over £300 per vial while unfractionated heparin costs £2-£5. The two groups had the same outcome, both in terms of cardiovascular events and adverse events. So what someone now needs to do is map the de-adoption of bivalirudin. How soon will the NHS, for example, start saving £300+ per patient undergoing emergency PCI? How will this play out in the US within Medicare and the large insurers? How many cardiologists will still be using bivalirudin for ACS in a year’s time? Those who do want to carry on will find some comfort from the editorial on this trial, though I didn’t. The author declares 12 possible conflicts of interest, including a grant from the makers of bivalirudin.

1021 Ciclosporin for heart attacks? Well no, actually. It didn’t work, but the more remarkable thing is that anyone thought it might. Apparently animal experiments showed that it reduced myocardial loss and reperfusion injury following induced coronary artery occlusion. But “in patients with anterior STEMI who had been referred for primary PCI, intravenous cyclosporine (Eng. ciclosporin) did not result in better clinical outcomes than those with placebo and did not prevent adverse left ventricular remodeling (Eng. remodelling) at 1 year.” This was a 970 participant trial, so there the matter will probably rest.

1040 For several years now we have been teetering on the verge of a breakthrough in curing multiple myeloma. That’s if you can teeter on the verge of a breakthrough: I’m having trouble with this metaphor. Anyway, this paper shows that the teeterers may have broken through the verge. Or the thing that is just past the verge. Whatever. They think they may have cured multiple myeloma. Their subject was a woman who was diagnosed with multiple myeloma at the age of 43 and had already had marrow ablation followed by autologous stem cell transplantation followed by lenalidomide, bortezomib, carfilzomib, pomalidomide, vorinostat, clarithromycin, and elotuzumab. Two years on, with 95% of her bone marrow infiltrated by myeloma, she underwent another round of ablation using melphelan. But this time it was followed by an experimental treatment using her own T cells which had been processed to express a CD3-zeta/CD137–based anti-CD19 chimeric antigen receptor from a lentiviral vector. CTL019 cells for short. The result was as dramatic as the first case studies of penicillin for widespread sepsis. “Autologous transplantation followed by treatment with CTL019 cells led to a complete response with no evidence of progression and no measurable serum or urine monoclonal protein at the most recent evaluation, 12 months after treatment. This response was achieved despite the absence of CD19 expression in 99.95% of the patient’s neoplastic plasma cells.”

JAMA 8 Sep 2015 Vol 314
1021 “Objective: To estimate the recent prevalence and update US trends in total diabetes, diagnosed diabetes, and undiagnosed diabetes using National Health and Nutrition Examination Survey (NHANES) data.” This looks interesting: the sampling stretches all the way from 1988 to 2012. Now there are three ways to measure “diabetes,” and they differ in their distribution and reproducibility: fasting glucose, HbA1c, and glucose tolerance testing. By any or all of these criteria, the prevalence of diabetes in the USA has risen and stands at about 12-14%. “Diabetes is a major cause of morbidity and mortality in the United States, costing an estimated $245 billion in 2012 due to increased use of health resources and lost productivity,” is the opening sentence of this paper. Now when I read most of the literature about diabetes I feel as if I’m trapped in revival meeting of a religion that I don’t belong to. Why does having abnormal glucose tests cost the USA $245bn? Does everybody beyond these threshold values pose an inevitable burden on the US economy? Wouldn’t it be better to do fewer of these tests and so reduce costs and spare a lot of people from needless anxiety and futile treatment? I turned in vain to the editorial on this study, with the alluring title “Prevalence of Diabetes in the United States: A Glimmer of Hope?”

It begins: “Obesity is a major risk factor for type 2 diabetes” and goes on to point out that between 1980 and 2000 the prevalence of obesity in the USA doubled, but since then seems to have reached a plateau with 35% of US adults aged 20 years or older estimated to be obese. And the prevalence of diabetes seems to have actually decreased a bit since 2008, despite a higher level of detection. The American Medical Association, the author notes with approval, has classed obesity as a disease. Well, anything for a crazy society to change its habits I suppose. Maybe it would help if the AMA classed gun ownership as a disease. I feel disoriented in a strange world of category errors and reversed logic. Somebody please help me. Just what is this “type 2 diabetes”?

JAMA Intern Med Sep 2015
OL We discovered this week that a distinguishing feature of human life at its earliest stage—as early as tool-making and probably much earlier than language—was ceremonial attention to those who had died. What were the thoughts and feelings of the Homo nadeli people as they buried their dead? Could they even be said to be people with thoughts and feelings? I think so—and in that case, there were human people walking the earth more than two million years ago. Honouring the dead remains an extremely powerful instinct, connected with whatever meaning we give to life itself, and sometimes even overpowering the instinct to survive. And how we remember the dying of those we love stays with us and marks us forever. If we wind forward 2-3M years we can uncover another study from a different sort of cave: the intensive care unit. Modern people die in these caves every day. They are surrounded by the ceremonies of science and technology, but the ceremony of innocence is often drowned. “Among a cohort of surrogate decision makers with a relatively high degree of religiosity, discussion of religious or spiritual considerations occurred in fewer than 20% of goals-of-care conferences in intensive care units, and health care professionals rarely explored the patient’s or family’s religious or spiritual ideas.”

OL There was once a Clostridium which was difficult to culture, so they gave it the Latin name difficile, neuter form of the adjective difficilis. Nothing to do with French, be it noted, and pronounced quite differently. When it became clear that C difficil-e is a common cause of persistent colitis in vulnerable people, especially following antibiotics, the clever boffins devised an immunoassay to detect the toxin produced by pathogenic strains. But other boffins, thinking themselves even cleverer, then devised a polymerase chain reaction (PCR) test to detect the bacterium itself. This important study shows that was a step too far.”Among hospitalized adults with suspected CDI, virtually all CDI-related complications and deaths occurred in patients with positive toxin immunoassay test results. Patients with a positive molecular test result and a negative toxin immunoassay test result had outcomes that were comparable to patients without C difficile by either method. Exclusive reliance on molecular tests for CDI diagnosis without tests for toxins or host response is likely to result in overdiagnosis, overtreatment, and increased healthcare costs.” Neat. Choosing the best diagnostic test can be difficilis and getting it wrong can have dire consequences. We may even be talking stool transplantation.

Lancet 12 Sep 2015 Vol 386
1041 Last December I threw in the stethoscope and thus avoided another season of triaging babies and toddlers with bronchiolitis. It would have been my 37th. The smaller the baby, the greater the worry, and each year I used to admit a number to hospital. Yet in 2013 there were just 46 deaths in the UK from all respiratory causes combined in the first 12 months of life. Bronchiolitis is fortunately a more benign condition than it often looks and sounds. So if you were doing a trial like the one here, comparing oxygen saturation targets in infants with bronchiolitis admitted to eight British paediatric hospitals, what would your outcome measure be? Clearly not mortality. Return to a respiratory rate within two standard deviations of the age-adjusted normal? Reduction of points on some validated score for respiratory distress? Actually this trial chose time to resolution of cough as its primary outcome. I am frankly baffled. Using this outcome measure, the authors say that management of infants with bronchiolitis to an oxygen saturation target of 90% or higher is as safe and clinically effective as one of 94% or higher. I leave you to judge: the paper is open access.

1066 Another open access article tells us about a trial of bronchoscopic lung volume reduction with endobronchial valves for patients with heterogeneous emphysema and intact interlobar fissures. It’s called the BeLieVeR-HIFi study, which will give joy to collectors of UnBeLieVablY-CluNky acronyms. For the rest of us, it doesn’t really amount to anything. Even in carefully selected atypical COPD patients, it sounds like this is still a hazardous and experimental procedure.

1097 In this lung centred issue of The Lancet, we get a lengthy review of community-acquired pneumonia. A lot of it is about antibiotics. But it must have been written too late to include the most startling fact about community-acquired pneumonia: we don’t know what causes the majority of it in the developed world. I refer you back to the American survey in the NEJM of 30 July which used the most advanced diagnostic techniques to detect pathogens in 2259 patients admitted with pneumonia and chest X-ray changes. In 62% they could not detect a virus, a bacterium, nor a fungus.

BMJ 12 Sep 2015 Vol 351
Reading through all the main journals, and also a lot of the lesser ones, and also all the new Cochrane reviews, and also a lot of papers editors send me for review, I get pretty fed up with clinical trials. Like most people I wish there was some easier, quicker, cheaper, less futile way to get patient-important evidence. But here’s a great teaching article which contrasts the observational data about digoxin for heart failure with the evidence from clinical trials. Observationally, patients who are given digoxin tend to die sooner. But in clinical trials—admittedly mostly before the era of full “standard treatment” for HF—they did not die sooner and tended to end up in hospital a bit less often. The latter effect is so weak (RR 0.92), and hospitalisation is such a dubious end-point, that I still think digoxin is a drug which can safely be left in museums of medical history. But this study does illustrate all the insurmountable weaknesses of observational evidence, and it inspires a delightful editorial by Graham Cole and Darrel Francis, “Trials are best, Ignore the Rest,” a must for every collection of great pieces on the principles of evidence based medicine.

Pulmonary embolism happens all the time, and most of it doesn’t matter. But when it produces symptoms, it can matter a great deal. D-dimer will tell you if there is clotted blood somewhere in the system. Clinical appraisal will tell you if someone is ill. So how do you combine the two and appraise the outcome? I won’t even try to describe the study which these diligent Dutch investigators undertook. They conclude that there are five different PE prediction models that work equally well in primary care and that going for the Wells rules produces the lowest “failure rate.” This is defined as the “proportion of pulmonary embolism cases in group of patients with low probability.” But don’t spend too long on puzzling about what that means. The editorial nicely undermines this whole exercise by puffing a different score altogether—the PERC rule—and by pointing out that “Judgment or gestalt has recently been shown to trump both the Wells criteria and revised Geneva score in a head to head study. Our years of training and experience have provided us with the skills and intuition to manage these cases. We should be comforted by this concept and trust our clinical acumen.”

Fungus of the Week: Suillus aeruginascens

I was wandering by a stand of larch when I spotted some boleti with off-white caps and dark brown-grey gills. I have been hunting fungi for decades but I couldn’t remember finding anything quite like these before. In fact these guys are not particularly rare, and in previous generations they were known as Boletus viscidus. The books class these and many other lesser boletes as “edible but not worthwhile.” My hungry Polish ancestors would probably have put them into a soup or mixed them with ground meat and sour cabbage as a filling for pierogi. I’m afraid that after enjoying the experience of finding them and identifying them, I just put them in the compost bucket.