Richard Lehman’s journal review—23 September 2013

Richard LehmanNEJM 19 Sep 2013 Vol 369
1106 I’m starting with the second paper about colonic cancer screening in this week’s New England Journal of Medicine, because it takes us to one of the places where it first began: the state of Minnesota. Up there, just under Canada and just west of the Great Lakes, 46,551 citizens were enrolled from 1976 to 1978 to take part in annual or biennial colon cancer screening using guaiac testing for faecal occult blood. Crude though it is by any standard, this form of screening did produce a reduction in colon cancer mortality which is still detectable to this day. However, there was no difference in all-cause mortality between the screened and unscreened population.

1095 At the same time as the Minnesotans began sending in their little brown cards to the researchers at their state university, the large (female only) Nurses’ Health Study was getting under way in the whole of the USA, and the (male only) Health Professionals Follow-up Study followed ten years later. These ladies and gentlemen were free to do as they pleased for the duration of the studies, and some chose to undergo colon cancer screening by flexible sigmoidoscopy or colonoscopy. Others had these procedures because they were symptomatic in some way. Altogether, there were 1815 colon cancers diagnosed over a 22 year period, and tissue samples of them exist to allow advanced genotyping. This immensely painstaking observational study not only produces figures for the possible protective effect of lower bowel endoscopy, but also describes the genomic features of the fast-growing cancers which slipped through the net. In line with previous observational studies, it shows impressive reductions in the hazard ratio for colon cancer in those who underwent sigmoidoscopy (HR 0.57) and colonoscopy (HR 0.44).

1115 Interventional cardiologists are gun dogs. A cardiologist who spots a stenosis wants to rush off and do something, like a springer spaniel hearing the rustle of a pheasant in the covert. Following the COURAGE study, we are trying to retrain our cardiology springers to ignore stable angina on optimal medical treatment: “No, Doc! Sit! Stay!” we tell them, as they quiver and make whimpering noises. But the PRAMI study will let them off the leash to stent at their hearts’ content when faced with acute myocardial infarction. Patients with ST elevation MI in five large British hospitals were randomised to receive immediate percutaneous intervention for the “culprit lesion” alone, or for as many significant stenoses as the cardiologist felt worth stenting. The trial was stopped early when it was clear that there was a big benefit from the latter approach: “Hazard ratios …were 0.34 (95% CI, 0.11 to 1.08) for death from cardiac causes, 0.32 (95% CI, 0.13 to 0.75) for nonfatal myocardial infarction, and 0.35 (95% CI, 0.18 to 0.69) for refractory angina” during a mean follow-up of 23 months. Woof!

1134 Atul Gawande has written wonderfully about the factors which determine outcomes in surgery, and is one of the co-authors of this survey of readmission rates following surgery in 3004 hospitals across America, using Medicare data about six procedures. The authors then compare readmission rates with surgical volume and surgical mortality. There is some correlation, but it is quite marginal: “hospitals in the highest quartile for surgical volume had a significantly lower composite readmission rate than hospitals in the lowest quartile (12.7% vs. 16.8%, P<0.001), and hospitals with the lowest surgical mortality rates had a significantly lower readmission rate than hospitals with the highest mortality rates (13.3% vs. 14.2%, P<0.001).”

JAMA 18 Sep 2013 Vol 310
1135 The AQUARIUS trial is just the latest in which Novartis fails to find a market for aliskiren, a renin blocker which was first licensed six years ago. They hoped to use it in people with “pre-hypertension” and coronary disease, and used a surrogate outcome: coronary atheroma measured by intravascular ultrasound. I can’t do better than to quote a tweet from Ken Covinsky: Rx of pre-htn (not a disease) wth aliskiren doesnt reduce atheroma volume (not an outcome) That was helpful. Not.

1145 At the moment my bedtime reading is Peter Gøtzsche’s damning indictment of the drug industry, Deadly Medicines and Organised Crime, which has just been published by Radcliffe. It’s a must-read. He describes all the ways that companies running trials manipulate the results or hide them altogether. But this week I am going to praise three industry run trials for their integrity, because they all show the opposite of what the manufacturer was hoping for. The first is the AQUARIUS trial described above,  run by Steve Nissen, with company sponsorship, but without company control of the data; and the third is the HERA trial in The Lancet. In the middle lies this trial of otamixaban. It sounds like a treatment for otitis externa, but otamixaban is actually the latest direct factor X inhibitor to reach a crowded market. The TAO trial was designed by its manufacturer to “compare the clinical efficacy and safety of otamixaban, a novel intravenous direct factor Xa inhibitor, with that of unfractionated heparin plus downstream eptifibatide in patients with NSTE-ACS undergoing a planned early invasive strategy.” But sadly for Sanofi, “Otamixaban did not reduce the rate of ischemic events relative to unfractionated heparin plus eptifibatide but did increase bleeding.” So onwards to another trial for some other indication—though it’s a fair bet it won’t involve a direct comparison with another drug in the same class.

Lancet 21 Sep 2013 Vol 383
1021 Trastuzumab, you will remember, is marketed as Herceptin and used as adjuvant treatment for HER2-positive breast cancer. It can be a powerful heart toxin. In HERA, an open-label Roche-funded trial, 5102 patients with early HER-2 cancer were randomised to standard chemo alone, or combined with one or two years of Herceptin. “We recorded 367 events of disease-free survival in 1552 patients in the 1 year group and 367 events in 1553 patients in the 2 year group.” You don’t need to be a statistician to see that amounts to absolutely no difference: but if you look at serious cardiac damage, you find 67 more patients in the two-year group. And it is hard to put a price on that extra year of Herceptin, but I would be surprised if it is less than $70K.

BMJ 21 Sep 2013 Vol 347
It never ceases to amaze me that more than half the young women of the developed world have been taking hormones for the last 40 years and the only major harm has been a few deep vein thromboses. And for most of that time there has been one clear leader in the market, which still comes out favourably in every meta-analysis: ethinyoestradiol 30mg plus levonorgestrel 150mg, meaning Microgynon or Ovranette. This new Dutch meta-analysis confirms that this is still the 30mg pill that carries the lowest risk of VTE.

A massive study of suicides in 54 countries shows that they have risen in males since the economic crash of 2008. But considering the severity of the crisis and the size of the populations studied, the excess of only 5000 suicides above the number predicted seems incredibly small. Hope springs eternal in the human breast, even when bankers try their best to destroy it.

John Ioannidis is one of the most valuable critics of medical orthodoxy alive, and anything from his team becomes required reading from the moment it appears. A title like “The effect of excluding treatments from network meta-analyses: a report” is hardly calculated to lure the eye of a jobbing clinician, but lured you should be. The reason is given with characteristic lucidity in the conclusion: “Excluding treatments in network meta-analyses sometimes can have important effects on their results and can diminish the usefulness of the research to clinicians if important comparisons are missing.” You cannot know the best treatment unless you know about all the treatments, including doing nothing.

Ann Intern Med 17 Sep 2013 Vol 159
390 I am baffled at the way health systems allow themselves to be ripped off by pharmaceutical companies, on the basis that there will be a net influx of money to the economy because these companies sell so many of their rip-off products abroad. I’d like to read a clear analysis of this from someone reliable like John Appleby or Andrew Dilnot. These musings arise from reading a study of varenicline, a drug which blocks the effect of nicotine on the brain receptors which cause people to crave for a smoke. In the UK, the basic price for a month’s treatment is £56: not huge, you might say, but think of the size of the target market: every person in the world who is trying to stop smoking. Varenicline (Champix/Chantix) is just a modification of cytisine, a laburnum extract which has been used for more than 50 years in eastern Europe and is extremely cheap: but cytisine isn’t even available in the UK. The main worries about varenicline have been (a) possible cardiovascular harm and (b) a tendency to worsen depression and even cause suicide. A trial here tries to address point (b): it took 38 centres in 8 countries to recruit 525 adult smokers with stably treated current or past major depression and no recent cardiovascular events. “Pfizer funded the study and was involved in study design and in collection, analysis, and interpretation of data with the authors.” The conclusion of the paper reads: “Varenicline increased smoking cessation in smokers with stably treated current or past depression without exacerbating depression or anxiety.” I haven’t space to go into all the possible sources of bias in this study. Try and spot a few yourselves (the full paper is free). Here’s a little starter from the Abstract:” Some data were missing, and power to detect differences between groups was low in rare events. Smokers with untreated depression, with co-occurring psychiatric conditions, or receiving mood stabilizers and antipsychotics were not included.”

Plant of the Week: Cyclamen hederifolium

There are two woodland plants which will succeed almost anywhere and should be grown by everybody. One is the wood anemone, to give joy in spring, and the other is this cyclamen, to make the heart leap in autumn.

Cyclamens grow from woody corms which can become immense. They also have a happy habit of spreading from the seeds which follow the flowers, in seed-pods borne on strange curly twists of stem. We have stands of cyclamen growing all over the place where we can’t remember planting them. That could mean one of two things, but I prefer to believe it is because they have planted themselves.

These Neapolitan cyclamens are the earlier flowering ones, bearing flowers from soft pink to almost pure white. Some of them are scented. The leaves are attractively mottled. In fact there is nothing this plant can be faulted for. It doesn’t even seem to deplete the soil when planted next to other choice subjects.

We recently planted the new blue flowered geranium hybrid “Rozanne” in a shady part of the garden, unaware that two nearby clumps of cyclamen were about to produce abundant white flowers flushed with pink. The effect is magical.