Richard Lehman’s journal review—24 September 2012

Richard LehmanJAMA  19 Sep 2012  Vol 308
1122    This week’s JAMA is devoted to obesity. It’s a bit like global warming: we can see it happening around us, we can foresee terrible consequences, we pretend to ourselves we’re doing something about it, but we know that in the end only drastic solutions will work, and these are quite likely to have unforeseen consequences. The drastic cure for obesity is bariatric surgery, but scaling this up to meet demand is something no health system has dared do to date. In fact the consequences of bariatric surgery are now well known, and for most individuals they are overwhelmingly beneficial, as this 6 year follow-up study of gastric bypass surgery shows. A cohort of mainly female subjects with BMI>35 who had Roux-en-Y gastric bypass (RYGB) surgery is compared with two control groups: a group who sought surgery but did not get it, and a control group of weight-matched individuals who did not seek surgery. The differences are startling, though hard to pick out of the abstract. Two thirds of the RYGB cohort who were already diabetic ceased to be so, and the incidence of diabetes in the rest was reduced ninefold. Systolic BP fell by more than 5mmHg in the surgical group but rose in the controls, and a similar pattern was seen with cholesterol. As for weight itself, that dropped by a fifth and stayed down

1132   Longer-term information comes from a Swedish study of health care utilization in the 20 years following bariatric surgery – this time mostly vertical-banded gastroplasty, an effective method but one which has fallen out of fashion. This study doesn’t look at clinical outcomes but at health care costs, and finds that for the first 6 years, these patients had higher utilization and spending, but from then on the pattern was reversed.

1142   What bariatric surgery doesn’t do, however, is reduce sleep apnoea. The OSA sufferers in this trial lost nearly 30kg following surgery as opposed to 5.1kg if they went on a conventional diet. And yet they still needed the same amount of continuous positive pressure ventilation. It’s not rational.

NEJM  20 Sep 2012  Vol 367
1082   In the next few weeks, we can look forward to dignified murmurs of offended innocence from the pharmaceutical industry in response to Ben Goldacre’s brilliant and damning new book, Bad Pharma. But it is difficult to maintain dignity when caught with your pants down, as GlaxoSmithKline has been over the mis-selling of Paxil and Wellbutrin, and the hiding of safety data relating to Avandia. However, all that’s needed is to pay up $3BN and your pants can quickly be raised and adjusted. Nobody need be called to account, and business can go on as usual: after all, $3bn is about half the profit that these three drugs produced every year while they were being mis-marketed. Here’s a swingeing perspective piece titled Punishing Health Care Fraud—Is the GSK Settlement Sufficient? The answer of course is no. There is nothing to stop fraud continuing as a lucrative business plan, and nothing to bring those responsible to court. And is GSK just a bad apple? No way—there are 25 others who are under special measures, including most of the big names in US pharma.

1087   As chemicals go, you can’t get much simpler than dimethyl fumarate which has long been used as a mould suppressant in leather furniture. It is one of the most potent sensitizers known, and gave rise to the poison chair incident when over a hundred people in the UK—including a patient of mine—developed severe eczema from sitting on a newly purchased leather sofa made in China. This editorial drily follows the story of dimethyl fumarate from this to the two trials in multiple sclerosis which appear in this week’s NEJM. The CONFIRM study run by Biogen shows that dimethyl fumarate (here called BG-12) at either of two oral dosage regimens is no better at preventing disability progression than placebo or daily injections of glatiramer acetate in relapsing-remitting MS. However, it does reduce the number of relapses and particularly the appearance of new MRI lesions. So if you were an MS patient without a personal MRI facility, you would notice nothing, as you would be unaware of the statistical reduction in relapses but very aware of progressive disability. You might well be aware of adverse events such as flushing, diarrhoea and abdominal pain, as they each affect a third or more of recipients.

1098   Biogen clearly believe that there are profits to be made from selling this simple molecule as a treatment for MS, because they also ran a second trial called DEFINE, this time with a placebo control only. Here they did find a reduction in time to disability progression, equivalent to that shown by existing agents. Mixed messages then, but probably sufficient for the licensing agencies to let this substance loose on the crowded market of relapsing MS sufferers, as yet another difficult choice amongst a list of largely ineffectual agents with various toxicities. By rights, at least this should be the cheapest: but that, of course, remains to be seen. Whatever the market can bear…

1119   Take a look at this study of how US medical interns rate hypothetical studies of new drugs: it’s free to everybody courtesy of Jeff Drazen, who also produces an editorial that you can read in full. Lampytinib, bondaglutaraz, provasinab: what fun to make up drug names like that, and trial designs that varied from a proper large RCT to a footling open-label study with no safety data. Not surprisingly, the interns had little difficulty in grading the quality of these mock-ups. But now comes the interesting bit: they were much less inclined to believe the results if the studies were marked as industry funded. This earns them a lofty reproof from headmaster Drazen: “Patients who put themselves at risk to provide these data earn our respect for their participation; we owe them the courtesy of believing the data produced from their efforts and acting on the findings so as to benefit other patients.” Amen to that—but that’s exactly why we mistrust data that have been gathered and interpreted by people with billions of dollars riding on the result—and who have a track record of deceit. If Drazen wants to show true courtesy to the trial participants, he needs to insist that for every trial published in the NEJM there is a full database of raw, de-identified patient-specific data, with the trial protocol and any variations from it, and all other meta-data required to interpret the study, open to all bona-fide investigators. And let’s not forget that the NEJM itself has a vested interest in putting a positive spin on industry-funded trials, as it gets a substantial (undisclosed) proportion of its income from the sale of reprints to pharma companies. There have been some terrific responses to this editorial, especially from Harlan Krumholz and Joe Ross.

Lancet  22 Sep 2012  Vol 380
This week The Lancet is mostly about surgery, and Richard Horton visits Australia and generally approves of what he finds. The Australians must be so pleased.

1059   Rates of mortality following surgery have been studied for centuries: in fact The Lancet in the 1810s built up its circulation by hounding out surgeons who killed unusual numbers of patients. Variation in surgical mortality was the subject of some astounding studies by James Simpson in the 1850s, showing that patients were far more likely to survive amputation carried out by a county doctor than by a distinguished surgeon in an Edinburgh hospital (where the mortality exceeded 40%). Sixty years later, Codman’s studies of variation in US surgical outcomes were so damning that he was rapidly ostracized by his colleagues. But this 2011 study of 7-day mortality after surgery across Europe actually shows far more outrageous variation than any of these: mortality varies more than 15-fold between European countries, even after adjustment for known confounders. If you develop a surgical problem in Poland, board a plane to Finland at once. It’s a really frightening study which calls for action at a multinational level.

BMJ  22 Sep 2012  Vol 345
Ranibizumab, bevacizumab: they’re the talk in every cab. Same mode of action: hundred-fold difference in the tab. We know that head-to-head, these vascular endothelial growth factor A inhibitors are equally effective for “wet” age-related macular oedema. This indirect comparison by systematic review concludes that the same is probably true of diabetic macular oedema. And the actions of Genentech/Roche in relation to these products should dispel any notion that pharma companies can be shamed out of squeezing as much money out of health systems as they can get away with.

“It is refreshing to read an article highlighting the risks of tighter blood pressure control in diabetic patients. But we have to remember that it is a retrospective study, so we don’t know when a patient’s glycaemic index breached the diabetes threshold and when patients were diagnosed. Furthermore, who measured patients’ blood pressures and by what method?” It’s really weird that the BMJ flagged up and printed this rapid response because it is mistaken on almost every count. This is a database study of the blood pressure as measured by UK general practitioners in the first year of diagnosis of type 2 diabetes (from 1990 to 2005), and how it relates to mortality over that period. It tells us nothing about treatment effects. The reason that the glycaemic index of the patients is not reported is presumably because nobody wanted to eat them.

Ann Intern Med  18 Sep 2012  Vol 157
398   Idiopathic pulmonary fibrosis is a long drawn out death sentence, and anything that can improve quality of life has to be a good thing. Does this apply to thalidomide? I can’t really tell from this 24 week cross-over trial. Cough-related quality of life scores definitely improved, but there was a very high incidence of adverse effects such as constipation, dizziness and malaise. I guess the way to tell in individual patients is either by a blinded n-of-one trial or simply by letting them try the stuff and decide for themselves.

439    I have spent a fair bit of space this week haranguing industry for its shortcomings as a source of evidence for improving healthcare: and for more, much more, read Ben Goldacre’s book plus David Healy’s Pharmageddon. It’s not just the biased evidence, it’s the distorted priorities and disease-mongering which arise out of our willingness to let the profit motive dominate medical research. So how can we change the culture? How should we really go about setting the research priorities for helping to treat and prevent real disease? You won’t find clear answers in this confused and bureaucratic American paper but you will find them if you look at the James Lind Alliance Framework.

446   The key is Seeing Through the Eyes of Patients, eloquently argued for in this paper about the US Patient Centered Outcomes Research Institute by Harlan Krumholz and Joe Selby. It is vitally important to bring patients into direct dialogue with health professionals and to ensure that they have real power in the process. As doctors, our only business is to be able and willing servants to patients; and the only business of medical academia is to be a servant both to patients and to health professionals. In the current situation, industry and academe sit with linked arms like heavy parasites dominating and feasting on systems of health provision in every country. Changing this to a healthy symbiosis is the most urgent task in health care. It means a complete reversal of existing power structures—otherwise known as a revolution. The obstacles to this—political and conceptual—are discussed in a companion editorial by Pignone.

Plant of the Week: Daphne x transatlantica “Beulah Cross”

The daphnes are generally known as temperamental plants, liable to sulk or die at the least provocation. I can probably count ten daphnes for every one that survives in our garden, yet until I am reported to the Daphne Protection Society I shall continue to try and grow them in every corner I can find. The tough British species will usually thrive anywhere—the deciduous D mezereum with mauve flowers in March, and the woodland evergreen D ponticum, with hidden yellow flowers in winter. But these aren’t really very worthwhile. You want a daphne that will produce fragrant flowers in long succession and that has leaves worth looking at all the year round.

Fortunately (though to the annoyance of those who like exclusivity) there are now quite a few tough hybrid daphnes that will succeed in most gardens, even ours. We owe thanks for these to American hybridists who first produced the excellent green-leaved “Jim’s Pride” by marrying D. collina with D. caucasica. Beulah Cross is a sport of Jim with lovely grey leaves variegated with white. Hebraists amongst you will know that beulah means “married land” (cf. Isaiah 62;4), but the Cross in this name is not a reference to hybridity or religion but to an American lady who happened to have the surname Cross and the first name Beulah.

Thinking this would be a small, miffy plant, I planted it too close to a big skimmia. Now it looks set to grow a metre high and would be a shapely dome had I not overcrowded it. The best thing is that it keeps producing lovely clusters of palest pink flowers all season long, and particularly just now. Do get this lovely, hardy small shrub—though it is not easy to find—and plant it somewhere raised up so that you can constantly pass by and take deep draughts of its amazing scent.