JAMA  5 June 2013  Vol 309
2223   It’s nice to see some meaty stuff in JAMA this week: I was beginning to grow despondent. It’s true that we are expected to take an interest in the Association Between the MUC5B Promoter Polymorphism and Survival in Patients With Idiopathic Pulmonary Fibrosis after this article, but first let’s rejoice in a good trial (REDUCE) with a nice clear clinical message: “In patients presenting to the emergency department with acute exacerbations of chronic obstructive pulmonary disease, 5 day treatment with systemic glucocorticoids was noninferior to 14 day treatment with regard to reexacerbation within 6 months of follow-up but significantly reduced glucocorticoid exposure. These findings support the use of a 5 day glucocorticoid treatment in acute exacerbations of COPD.” I know that some of my older and more sensitive readers will flinch at the use of “noninferior” and “reexacerbation”, but hey, Shakespeare invented some pretty silly words, and at least with these we get the meaning. When treating exacerbations of COPD, give the prednisolone 40mg (or prednisone over the Atlantic) for 5 days not 12.

2240   Last week, Edwin Gale’s brilliant Lancet essay reminded us that type 2 diabetes is a category error. It can be treated with glucose-lowering agents and “cured” by bariatric surgery, but in the end only one thing matters: do these treatments improve life for patients? Measurements like fasting blood sugar, HbA1c, body mass index, and blood pressure are after all just surrogates. Age also comes into the equation: if, in retirement, my enjoyment of food causes me to get fatter and develop “type 2 diabetes,” I really won’t care very much. But if I were a Bangladeshi woman aged 40 with a BMI of 38 and diabetic neuropathy, expected to stay at home and look after five children and a couple of elderly relatives, I might take a quite different view. It is long term quality of life which counts. Quantity of life is also a factor. And for bariatric surgery, as for every other diabetes treatment, we know too little about these to make a clear judgement. The results of this trial are nonetheless very useful, because the comparator was the most intensive lifestyle modification possible, and the population from the US and Taiwan had a mean age of 49 and BMIs ranging from 30 to 40. The patients randomized to Roux-en-Y gastric bypass lost 26% of their body weight and ended up requiring much less medication for glucose, blood pressure, or lipids. But there were some unpleasant surgical complications, including one case of brain and limb damage. This was a smallish trial (n=120) with a huge number of exclusions: further individual patient follow-up data, hopefully available to all, will go into the great evidence pool which will eventually give us some clear answers about how to treat diabetes.

2250   Here is the same thing, expressed in the words of systematic reviewers who went through as many articles as they could in this fast-developing field (32 surgical studies, 11 systematic reviews on nonsurgical treatments, and 11 large nonsurgical studies published after those reviews). “Current evidence suggests that, when compared with nonsurgical treatments, bariatric surgical procedures in patients with a BMI of 30 to 35 and diabetes are associated with greater short-term weight loss and better intermediate glucose outcomes. Evidence is insufficient to reach conclusions about the appropriate use of bariatric surgery in this population until more data are available about long-term outcomes and complications of surgery.” In fact only three medium sized randomised trials really met the original criteria for this systematic review, and individual patient data were only available for one of them. The science of medicine really cannot advance while so much data that should inform decision making remains beyond reach.

2274   And there is a very good editorial on the subject as well. It ponders this very question—how can you do a systematic review that is both rigorous and clinically useful? “Too limited inclusion criteria result in no guidance for clinicians caring for certain patients and criteria that are too broad lead to recommendations that represent expert opinion. Expert opinion cannot be viewed as evidence or serve as the basis for practice guidelines.” Amen. Is anyone at NICE listening?

NEJM  6 June 2013  Vol 368
2192   Oops, I decided to skip over the JAMA paper on the importance of promoter polymorphism (rs35705950) in MUC5B and prognosis in interstitial pulmonary fibrosis, only to find MUC5B cropping up in the NEJM as well. And there is so little for the generalist in this week’s issue that I might as well try and tell you about MUC5B and its more general importance in all interstitial lung diseases. Here is a study that describes the correlation between this polymorphism and the appearance of CT lung changes in the general population—who, as it happens, live in Framingham, Mass. It turns out that if you carry the offending MUC5B polymorphism, you’re 6 times more likely to have definite pulmonary fibrosis and about three times more likely to have vaguer interstitial lung changes, sufficient to cause some shortness of breath. The older you are, the more likely these are to be apparent, with or without a history of smoking. But the JAMA study interestingly shows that if you have the misfortune to have interstitial pulmonary fibrosis, and carry this MUC5B polymorphism, you actually live longer than if you don’t. A fighting editorial in the NEJM uses these discoveries to prove that genomics is not boring: “It is fashionable to decry genetic studies as dull and stereotyped, but to do them as effectively as the investigations of idiopathic pulmonary fibrosis reported in the journal is demanding and difficult. These results are an exemplar of the robust foundations that genetic studies can provide to the understanding of complex diseases.”

2210   If genomics thrives on the idea of rewards to be obtained in 20-50 years’ time (or perhaps only in Heaven: it’s all the same to me), the story of HIV/AIDS is a great illustration of medical progress made in real time in response to a lethal pandemic. The NEJM brings this story to you, free of charge, in a well-written article describing how the challenge of HIV has helped to shape a new model of Global Health. Do take time to read it, and ponder the lessons for global cooperation in other areas.

Lancet  8 June 2013  Vol 381
This week’s Lancet is about China. China is vast. China is rich. China is fascinating. China is the future. Everybody wants to go to China, myself included. But as I haven’t been there yet, I shall shut up. If you want to learn all about health in China, I have nothing to say. You will have to read the Lancet.

BMJ  8 June 2013  Vol 346
Oseltamivir does something for influenza, though nobody is quite sure what, thanks to Roche’s persistent failure to reveal data from its trials. Observational evidence suggests that it reduces mortality in very severe influenza, and animal evidence suggests that high doses work better than standard doses. So a trial of standard versus high dose oseltamivir was organised in four South East Asian countries with Vietnam predominating: most of the patients were children. The high dose group did no better on any outcome measure than the standard dose group. As the editorial suggests, we simply need better drugs for influenza. And when they emerge, we need all the evidence relating to every trial they have undergone.

Statins lower plasma HDL-cholesterol markedly in everybody and raise plasma glucose a little in some. These measurements mean nothing in themselves: the only important fact is that statins lower cardiovascular risk in everybody, irrespective of its absolute starting point. The choice of whether to use a statin should be left to the patient on the basis of a calculation of individual risk reduction. “Diabetes” is just a glucose threshold: we have no idea what “diabetes” implies when it used to designate a person who has moved from having a fasting glucose of 6.7 to 7.2 because they are taking a statin. The population study from Canada in this week’s BMJ simply grades the various statins according to their likelihood of raising blood glucose, and shows that this corresponds to their degree of HMG-CoA reductase inhibition. That’s still quite an achievement, and this paper continues the marked improvement in quality of the research section in the BMJ.

But for me the best article in this week’s BMJ is Andrew Moore’s Analysis piece about pain relief: “Expect analgesic failure, pursue analgesic success.” The lady who brings back her pills and says “I don’t know what you gave me these for, doctor, they’re no use at all” is not defying some sacred corpus of analgesic knowledge, but simply telling the truth. Most painkillers either work well or hardly at all for any given individual. This is an article that every doctor who treats pain needs to read. It brings logic and science to bear on what we end up doing all the time, which is trying this and then trying that. And it also shows what is never worth trying, such as oxycodone for chronic low back pain: it has a failure rate of 100%. Most analgesics have failure rates above 50%. In fibromyalgia, 85% of sufferers get no relief from pregabalin: but that still makes it a useful drug for 15%.

Ann Intern Med  4 Jun 2013  Vol 158
Aspirin of course began life as an analgesic, but today if you stopped any doctor in the corridor and asked what the action of aspirin is, you would be told “platelet inhibitor” 19 times out of 20. Aspirin’s mode of action makes it a much weaker anticoagulant for preventing thromboembolism than low molecular weight heparin, but there seems to be one situation where the two are equivalent: in extended prophylaxis for patients who have had total hip replacement. A large Canadian trial randomized patients to receive either dalteparin or aspirin for thromboprophylaxis after ten days on dalteparin following hip arthroplasty. There was no significant difference in either venous thromboembolism or major bleeding.

Plant of the Week: Meconopsis cambrica

This is the time of year when all sorts of choice perennials begin to flower, but this week let’s hear it for the common Welsh poppy, which has already been providing patches of clear yellow cheer in the garden for some weeks. It really is a delightful plant in leaf and flower, and if it were not such a spreading weed, I think we might pay quite good money for it. In fact you can part with a few pounds if you want the fancier orange and double forms, but in the end the papery single four petals of the usual type give the most pleasure.

At the moment I am splitting some unwanted larger Welsh poppies and digging up any number of young seedlings to plant on a bank in dry shade. Although I am generally no fan of “wild gardens,” I think the poppies will look very pretty among meadow geraniums, which also favour us with plenty of seedlings to dig up and divide. You really can’t go wrong with good yellow next to good blue in the garden, wild or posh.