JAMA 8 Sep 2010 Vol 304
1073 “Adenocarcinoma of the pancreas is arguably the most challenging of human malignancies”, as the commentary on this study of adjuvant chemotherapy points out. And in fact the ESPAC-3 study doesn’t get us any further, except to rule out any benefit from adding gemcitabine to a regimen of fluouracil plus folinic acid. Still, the incremental acquisition of knowledge by large randomised trials needs to go on: there are now twice as many 5-year survivors of pancreatic cancer than there were before adjuvant chemotherapy, and gemcitabine is probably the best single agent overall. The editorial on p.1124 provides a useful summary of where we stand and where we’re going.

1091 Marshfield, Wisconsin: could this be another little American town with a family consisting of Virgil Stimson, with wife Madge and son Mart, precocious daughter Lina etc. And when the Stimsons get a cough, or a chill, or a sneeze, they troop to jovial Dr Belongia to get a nasopharyngeal swab and a blood test so that he can tell whether they have seasonal (H3N2) flu or new pandemic (H1N1) flu and publish his findings in JAMA. These show that in a community of 50,000 people, H1N1 flu symptoms and outcomes were just the same as for seasonal flu in the year that preceded the pandemic. The new strain hit a younger age group but was no more severe than the seasonal kind. About 90% of Marshfield participated in the study, and the paper is nicely written by Dr Belongia; though some regulars in Mo’s Bar have questioned whether he really belongs here.

1099 If memory serves me right (an unlikely event), there are about 90 different pneumococci, of which seven were identified as the most pathogenic and targeted in a vaccine which is now given to every child in several countries, including the Netherlands (sometimes also known as Marshfield). And despite the high standards of domestic hygiene for which the Dutch are legendary, the pneumococci are winning: the serotype 19A is becoming prevalent in vaccinated subjects and this tends to show high levels of antibiotic resistance.Time to develop a better polyvalent vaccine; and also, as the Dutch are always keen to remind us, to use fewer antibiotics.

NEJM 9 Sep 2010 Vol 363
1005 The sputum smear test for tuberculosis has been around for 125 years and is highly unreliable, especially in people with HIV; culture is slow and also unreliable; molecular testing is expensive and has its own problems. But it is definitely the way forward if we are ever to tackle the growing epidemic of TB in the poorest countries where HIV is endemic and drug resistance is an increasing problem. Xpert MTB/RIF sounds like the best tool so far: an answer within 2 hours, and rifampicin resistance detected at the same time. This kit, and the expensive third-line drugs which need to go with it, could halt the spread of drug-resistant TB in the poor world; all we need is one of the 960 billionaires who haven’t yet taken the charity pledge to do so. See also p.1050: MDR Tuberculosis: critical steps for prevention and control.

1016 Apparently one adult in twenty over the age of 40 has Fuchs’s corneal dystrophy and yet I knew Fuchs’s all about it until I read this paper. It can’t be that important, and indeed it doesn’t usually do very much unless people with the condition undergo refractive or cataract surgery. Then it can cause corneal oedema and visual damage and require corneal transplantation. “It is observed in 38% of the first-degree relatives of the probands”, according to this paper which is hailed as a major breakthrough in the genomics of a common condition (see editorial on p.1072). “Alleles in the transcription factor 4 gene (TCF4) encoding a member of the E-protein family (E2-2) were associated with typical FCD (p=2.3×10^-26).” Reader, I’m afraid you’ll have to get used to this terminology: this tells us useful stuff about the mechanism of Fuchs’s dystrophy (mediated by E2-2 protein) and the kind of statistical power you need when you’re looking at the human genome.

1038 If you enjoy taking a SWIPE at gullible surgeons, then here is a trial to please you, though the first SWIPE was delivered by JAMA three weeks ago. SWIPE-1 and SWIPE-2 were phase-3 trials designed to prove the value of a gentamicin-collagen biodegradable sponge for the American market. Since 1985, Innocoll has sold over 2 million of these sponges in 54 countries, but sales are due to plummet since both the US trials indicate that the sponges actually increase wound infections when placed in the fascia beneath the incision – following thoracic surgery in the first trial, and following colorectal surgery in SWIPE-2.

Lancet 11 Sep 2010 Vol 376
875 Trials of new interventions for heart failure have tended to show diminishing returns in an era when everyone with systolic dysfunction is already on at least an ACE inhibitor (or ARB) and a β-blocker if they can tolerate it. This trial recruited over 6500 patients with an ejection fraction under 35% and who either couldn’t tolerate a β-blocker or else still had a pulse rate over 70 despite taking one. Ivradabine slows the heart rate by a direct action on the sinus node, and in this pretty huge trial it showed a benefit in symptoms and survival which just reached significance for this atypical group.

BMJ 11 Sep 2010 Vol 341
513 This editorial about rosiglitazone started as a personal commentary and finished as a piece written to a tight deadline with two eminent co-authors both on holiday. All the while I kept thinking, “This is so blindingly obvious, why has nobody said it before?” The point of treating diabetes is not to lower sugar but to prevent vascular harm, so why on earth are we accepting drugs which cause the very thing they are meant to prevent? Why don’t licensing bodies demand evidence of vascular benefit before marketing drugs for diabetes, instead of dithering about the exact level of vascular harm ten years after a drug has been licensed? I had never intended to sound off about diabetes in my old age, but somebody needs to ask these questions; and quite a few more.

530 I don’t usually comment on the magazine pages of the BMJ, but if you want to know what good medical investigative journalism looks like, take ten minutes out to read Deb Cohen’s piece on the Avandia story. For years I’ve been wondering why there isn’t a good weekly magazine about medicine for the general public, along the lines of New Scientist. With talent like this, the BMJ needs to go ahead and start one.

543 Perhaps what we need for type 2 diabetes is not Big Pharma but Big Farmer. This is not a reference to my esteemed colleague Andrew Farmer, but to the importance of fruit and vegetables in preventing T2DM. This meta-analysis of the benefits of market farming produce concludes that regular intake of leafy green vegetables produces an overall protective effect of about 14%, though the confidence interval teeters on the brink of 95%. I bet you get a better effect if you grow your own. Dig for Victory against Diabetes.

545 Heyup, what’s this, as the Yorkshireman of my youth would say when he spotted a slug on a diabetes-preventing prize lettuce. A month ago I wrote “Cancer of the oesophagus is getting commoner, and so is the prescription of regular bisphosphonates, which frequently cause oesophagitis: are the two by any chance linked? The reassuring message of this study of data from the UK General Practice Research Database is a firm negative: bisphosphonate intake for any length of time is not associated with a higher risk of cancer either of the oesophagus or the stomach.” Now I’ve got to write: “Cancer of the oesophagus is getting commoner, and so is the prescription of bisphosphonates, which frequently cause oesophagitis. This disturbing study from the UK General Practice Research Database suggests that anyone who has had more than 10 prescriptions for bisphosphonates, or has taken them for five years or more, has an almost doubled risk of oesophageal cancer.” This isn’t the first time that two studies based on the GPRD have reached opposite conclusions: much depends on the length of follow-up and the selection of a control group, on which criteria the editorial (p.516) awards the prize to this latest study by Val Beral and her team. The absolute risk remains small, however; and we need to ask ourselves whether there is any real reason to prescribe oral bisphosphonates for more than three years. Or indeed whether we should switch to once yearly intravenous preparations.

546 If we were serious about treating type 2 diabetes, we would be offering bariatric surgery to tens if not hundreds of thousands of people every year, providing a high proportion of them with a permanent cure. But this survey shows that from 2000 to March 2008, fewer than 7,000 Britons had bariatric surgery, although numbers have been increasing in recent years. I do hope that when my former GP colleagues get to spend their £80bn NHS budget, with the help of lunchtimes off and the backs of used envelopes, they make bariatric surgery a priority. It undoubtedly saves lives, and may even save money.

Ann Intern Med 7 Sep 2010 Vol 153
289 The Nurses’ Health Study provides us with yet more observational evidence about diets to argue about. At least the authors are modest enough to point out that very few people are ever likely again to follow the diets they devised back in 1986, which bear little relation to any popular low glycaemic diet. Some nurses were supposed to follow a low carbohydrate diet which allowed unlimited animal fat and meat; another group were given a diet where more of the protein and fat came from vegetable sources. The latter fared distinctly better over 26 years.

307 Doctors have been very slow to accept that just because somebody’s coronary arteries are narrowed, causing stable angina, it doesn’t mean that they will fare better if they have a percutaneous coronary intervention. But if doctors have been slow to take COURAGE, patients are even more reluctant. This single-centre study from Boston showed that patients were four times more likely to attribute prognostic benefit to PCI than cardiologists. I await a study of covertly recorded patient advice given British interventional cardiologists, according to whether the procedure will or will not earn them a handsome private fee.

325 Magnetic resonance angiography of the lower limb will be coming soon to somewhere near to you. Welcome it: this meta-analysis studyshows it is good at picking up clinically important stenoses and it saves on radiation.

Fungus of the Week: Entoloma sinuatum
In thirty-five years of sniffing around autumn woodland, I thought I had become familiar with all the poisonous fungi, but a couple of days ago I was nearly caught out by this handsome and pleasant-smelling large white fungus. Even the great French mycologist Quélet was deceived and suffered abdominal cramps and diarrhoea for several days. He gave it the soubriquet of “the millers’ purge”, because it smells strongly of newly milled flour when freshly picked. It is also known more flamboyantly as le grand empoisonneur de la Côte d’Or.
A large troop of them was growing by a hidden woodland ditch, looking very pretty. In Britain they are classed as rare, and I had never seen one before. They look and smell very like the St George’s mushroom, which is quite common in spring. You have to look carefully to see that the gills are a pale yellow turning pink, and less crowded, and the cap slightly fibrillose. You are actually quite safe picking white mealy-smelling fungi in spring, because this Entoloma is an autumn species, so the two fungi never appear in the same season. I took home a few in case they were of an unrecognised edible species, but realised my mistake when the book I consulted declared “Poisonous, but not deadly”. On the other hand, Wikipedia quotes an alleged death. I did not eat, and live to tell this cautionary tale, like an Eve who resisted the serpent.