Richard Lehman’s journal review—19 August 2013

Richard LehmanNEJM  15 Aug 2013  Vol 369
603    A publicly funded trial in the USA, simply called the Prostate Cancer Prevention Trial, examined the effect of finasteride on prostate cancer incidence and mortality in men without known prostate cancer at the time of enrolment, and found that finasteride significantly reduced the risk of prostate cancer, but was associated with an increased risk of high-grade disease. The cancers that finasteride “prevented” were those detected by regular PSA testing followed by transrectal biopsy where necessary. This paper describes the results of follow-up for up to 18 years, and finds that there was no significant between-group difference in the rates of overall survival or survival after the diagnosis of prostate cancer. This inspires the best short piece I have ever read about prostate cancer detection and prevention, in the form of an editorial by Michael LeFevre. It is beautifully clear and well-argued and ends: “Men who are aware of and understand the benefits, risks, and uncertainties associated with the use of finasteride for prevention may make a rational decision to take the drug to reduce the harm of screening. Of course, another way to reduce the harm of screening is to choose not to be screened.”

630   The NEJM chooses to publish a phase 2b trial funded by Boehringer Ingelheim of a new drug regimen to treat genotype 1 hepatitis C. Like everybody who is not a specialist, I have no idea about the pros and cons of different hepatitis C treatment regimens, but the beginning of the abstract boldly informs us that “Interferon-free regimens would be a major advance in the treatment of patients with chronic hepatitis C virus (HCV) infection.” And the conclusion of the abstract duly informs us that the new Boehringer Ingelheim regimen is just such a major advance, producing a sustained virologic response 12 weeks after the completion of therapy in 52% to 69% of subjects, according to dose. As usual, “The first draft of the manuscript was prepared by the first author with support from a medical writer employed by the sponsor.” And as yet we have no idea what the cost of treatment with faldaprevir in combination with deleobuvir plus ribavirin might be: perhaps if their drug fulfils its early promise, Boehringer Ingelheim might yet surprise us and make it affordable to the 150-200 million people around the world who have hepatitis C.

JAMA  Intern Med  12/26 Aug 2013  Vol 173
1397   Last week I told you about a newly discovered gut pathogen, Bradyrhizobium enterica, which Boston microbiologists had detected using whole-genome sequencing. That happened in Boston, just across the river from Cambridge, Massachusetts; but over here in Cambridge, England, they have the same technology, and in this paper British microbiologists by the river Cam describe how it can be used to do whole-genome bacterial sequencing in real time in a laboratory run by the NHS. “The speed, accuracy, and depth of information provided by WGS platforms to confirm or refute outbreaks in hospitals and the community, and to accurately define transmission of multidrug-resistant and other organisms, represents an important advance.” Too right: the friendly microbiologists who live harmlessly in the lining of your local hospital have suddenly acquired the most powerful new tool since the days of Pasteur and Koch.

1407   The journals are currently filling up with articles showing that “biological” treatments for rheumatoid arthritis may not be the huge leap forward they first appeared to be. The standard surrogate measure in RA trials is radiological progression, but this may not translate into discernible patient benefit. In a Swedish trial, patients who did not achieve low disease activity after 3 to 4 months of methotrexate therapy were randomized to receive additional biological treatment with infliximab or conventional combination treatment with sulfasalazine plus hydroxychloroquine. There was no difference in sick leave between the two groups, even though the infliximab group showed less progression on their X-rays.

1416   The use of “biological” to describe drugs targeted at particular inflammatory mediators is pretty silly. And some biologicals are more biological than others: monoclonal antibodies are of course large molecules designed to act just like natural antibodies and attack specific receptors. And being large molecules, they themselves can stimulate somatic antibodies which attack them in turn. A systematic review finds that this is an insufficiently studied problem, best understood in treatments for rheumatoid arthritis where the “presence of antibodies against anti–tumor necrosis factor monoclonal antibodies confers a risk of discontinuation of treatment in rheumatoid arthritis.” There is also a risk of development of hypersensitivity reactions in all immune-mediated inflammatory diseases.

1429   The WHIMSY Study Group—wouldn’t you love to join a bunch who call themselves that? They looked at whether women given conjugated equine oestrogens for menopausal symptoms get less dementia. They don’t.

1439   Over in the USA, they treat to a target of HDL-cholesterol: over here, we treat to a target of total cholesterol. Let’s call the whole thing off. Both strategies are stupid. Treat to risk, not to target. Once an at-risk person is taking a maximum tolerated dose of statin—as a personal choice—there is no earthly reason ever to measure their lipids again.

1458   JAMA Intern Med is easily my favourite US medical journal these days, and this issue is packed with goodies: here is a medical writer, Stephen Braun, talking about the way pharma companies are free to promote biased messages under the cloak of continuing medical education, and how “patient education materials continue to be created that, while factually accurate, subtly shift attitudes by including only selected facts and/or omitting ideas that would undermine the funder’s preferred paradigm.” This in turn inspires a great commentary piece by Lisa Schwartz and Steven Woloshin.

JAMA  Vol 310  14 Aug 2013
591   The main article in ordinary JAMA this week is called “The State of US Health 1990-2012” and you can read the whole thing for free. In return for health spending that is threatening to become 20% of GDP, America’s crude outcomes are improving, slightly. But as value for dollars, they are appalling: “Among 34 OECD countries between 1990 and 2010, the US rank for the age-standardized death rate changed from 18th to 27th, for the age-standardized years lost to premature death from 23rd to 28th, for the age-standardized years lived with disability rate from 5th to 6th, for life expectancy at birth from 20th to 27th, and for healthy life expectancy from 14th to 26th.” What is the pusillanimous Obama administration doing about this? The solutions are entirely political. America certainly needs better “healthcare,” e.g. fewer specialist doctors competing to overtreat patients, and cheaper medicines—but even more it needs less tobacco, less corn syrup, fewer guns, and streets that are safe to walk and cycle on. Oh, and some politicians who will ignore lobbyists and work together for the common good.

617   I am a fan of cardiac resynchronization (CRT) for heart failure: it helps many people to feel better, and the evidence seems to point to underuse, whereas for implantable cardioverter-defibrillators, the evidence points to overuse. It’s crazy that patients are expected to have both together in nearly all cases. In this large registry study, there was no attempt to distinguish between them, because dual chamber pacing (CRT) was accompanied by defibrillator placement in all 24,169 subjects. It seems to me it should have been accompanied by very careful shared decision making with each patient. But that’s how it was, and this study shows that the benefit of CRT was greatest in those with either left bundle branch block or a QRS duration of 150 ms or greater. As expected.

BMJ  17 Aug 2013  Vol 347
We British GPs may feel harassed, our earnings may be falling, but we know there will always be work for us. But job insecurity is an increasing threat for most people, particularly in the underclass that our present government seems so intent on burdening and disempowering. The poorer you are, the less secure is your employment, so the innumerable authors of this attempted systematic review of the link between perceived job insecurity and coronary heart disease are forced to conclude that the two are essentially inseparable: “The modest association between perceived job insecurity and incident coronary heart disease is partly attributable to poorer socioeconomic circumstances and less favourable risk factor profiles among people with job insecurity.”

And these days, the poorer you are, the more obese you tend to be. But was that true in Scotland sixty years ago? Or even thirty? I find myself making no sense of this record linkage cohort analysis which attempts to link maternal obesity during pregnancy with increased mortality from cardiovascular events in adult offspring in 37 709 Scots with birth records from 1950 to present day. The authors conclude that, “Maternal obesity is associated with an increased risk of premature death in adult offspring. As one in five women in the United Kingdom is obese at antenatal booking, strategies to optimise weight before pregnancy are urgently required.” Hmm. And just what would those strategies be?

Multiple overlapping meta-analyses of randomized trials published on the same topic. Oh dear, I am in danger of getting dweeby about this, because as a member of the YODA team I helped to organize the first deliberate parallel meta-analysis of the same datasets, consisting of 17 trials funded by Medtronic to test their products containing bone morphogenetic protein. Have I lost you yet, O Common Reader? Bear with me, because this affects the way you inform and treat patients. If all the meta-analyses strongly agree, then you have better evidence than if you depend on a single one, right? But if they disagree, you don’t know whom to believe and you start to wonder if evidence based medicine is all it is cracked up to be, right? During the YODA project, we were astonished to hear from certain EBM advocates that we were setting ourselves up to undermine EBM by showing that different teams produced different meta-analyses. The message seemed to be that there must be only one daddy, and he is always right. In this week’s BMJ, the matter is debated at length in several places. The discussion is based on this important paper from the Ioannidis team, which found there is often duplication and redundancy in meta-analyses, but rarely fundamental disagreement. Daddies usually say the same thing, so EBM is fairly safe. But the most important contribution comes from left field in a letter from Carl Heneghan, who argues “that systematic reviews based on journal publications alone are likely to be seriously flawed in terms of assessing the trade offs between the risks and benefits of the intervention. If journal publications and conferences abstracts are the only data available for analysis, then an interim solution may be to downgrade the quality of the evidence until more substantive data are made available and included in the analysis.” In other words, full patient data and accompanying meta-data are required for any dependable meta-analysis of an intervention. That is what we did in YODA, by the way, if anyone is still reading this.

Lancet  17 Aug 2013  Vol 317
571  The best thing in this week’s Lancet is the much publicised editorial entitled The NHS: free and caring or a market commodity? The last paragraph begins, “The role of Chief Executive NHS England was advertised last week with a deadline of 6 September for applications. Astonishingly, the candidate does not have to have experience in, or knowledge of, healthcare systems.” Alas, I am not astonished. The appointee will not be there to run the system, but to sell it off.

597   Ischaemic preconditioning is a phenomenon so simple and so weird that it is taking people a long time to come to terms with it. It is as simple as squeezing any part of the body so that its arterial supply is compromised for a short while. Then if you cut off the arterial supply in any other part of the body, you will cause less damage, even if the blockage is permanent. The effect of preconditioning is immediate and lasts for days. In this German trial, they tried it on patients undergoing coronary artery bypass grafting, a procedure which is often accompanied by subclinical cardiac ischaemia, as demonstrated by troponin rises. In the intervention arm, patients under anaesthesia were subjected immediately before CABG to three 10 minute cycles of arm compression to 200mg Hg followed by decompression. Not only were there lower rates of troponin elevation, but “All-cause mortality was assessed over 1•54 (SD 1•22) years and was lower with remote ischaemic preconditioning than without (ratio 0•27, 95% CI 0•08—0•98, p=0•046).” Note the wide confidence intervals, but this is nonetheless pretty amazing.

624   With you in mind as ever, dear reader, I waded through this cardiology-heavy issue of The Lancet, trying to make sense of special pleading in two industry-funded trials which sought to promote prasugrel and platelet reactivity testing, respectively. Even if I were a cardiologist, this would have been a waste of time. But the clinical review articles are much more worthwhile, and the first one addresses a question which has bugged us all for over a decade: given that timely immediate percutaneous intervention is not always possible for myocardial infarction, what might be the continuing role of immediate thrombolysis as an alternative?  The article goes through the evidence meticulously, and ends up sitting on the fence somewhat, but the message of the conclusion is worth chewing over: “The preferred reperfusion option for patients with STEMI is timely primary PCI, although the recommendation for this approach was based on comparisons with inhospital fibrinolysis alone. Delay to delivery of reperfusion, which predominantly affects primary PCI, leads to an attenuation of benefit compared with fibrinolysis. When the difference in delivery between the two strategies is more than 60 min they seem equal. Furthermore, a pharmacoinvasive strategy of prehospital fibrinolysis plus planned angiography (at 6—24 h in haemodynamically stable patients) and rescue angioplasty for failed fibrinolysis has now been shown to be equivalent (by results of the STREAM trial) or better (by results of the CAPTIM trial) to primary PCI in patients who present early.”

Plant of the Week: Geranium “Rozanne”

This hardy cranesbill was named for Rozanne Waterer, a splendid name for a keen gardener. She noticed a seedling at the bottom of her garden which grew better than all the others and bore blue flowers twice as big as those of other hardy geraniums.

It has become the geranium to have ever since it was named “Plant of the Centenary” at the hundredth Chelsea Flower Show this year. We’ve only just bought it so we can’t vouch for its keeping power, but it is said to flower from June to November, and its mauve-tinged blue flowers are certainly abundant and attractive. Nobody knows its true parentage, but judging from the veining of its leaves and the colour of its flowers, “Johnson’s Blue” must be in there somewhere. We’re hoping it will be as tough as most geraniums and divide easily to provide pools of blue all over the late summer garden.