JAMA  17 July 2013  Vol 310
270   There are three papers in this week’s JAMA which would make good teaching material for a course on critical reading. The first is a randomised, double-blind trial of an intervention for in-hospital cardiac arrest, carried out in three large Greek hospitals. The “placebo” was epinephrine (adrenaline) in saline, and the “active” intervention was combined vasopressin-epinephrine and methylprednisolone: “Vasopressin and methylprednisolone were prepared in study center pharmacies in identical, preloaded, 5-mL syringes and placed along with epinephrine ampules in boxes bearing patient codes. At the time of patient enrollment, a box was opened and study drugs were injected intravenously according to protocol. Drug injection was followed by 10 mL of normal saline.” I hope the crash team understood what they were doing because I’m not quite clear from this description. And given the small differences in survival rates, a few mistakes could make a large difference in the odds ratios for return of spontaneous circulation and survival to discharge. So if you like doing maths in the summer heat, you could go on and work out the actual potential differences; but I have already spent too long on a study which at best needs replicating before it changes practice.

280   Cardiovascular disease is undergoing a very steep decline in most developed countries, but as we are told every day, this might go into reverse as the current generation of obese young individuals reaches late adulthood. I like studies which challenge orthodoxy, but here we have an observational report which just provides weak surrogate support for the “obesity time-bomb” hypothesis. People who joined the CARDIA study between the ages of 18 and 30 were not obese, but 40% of them became so over the following 25 years. They each had a coronary calcium score done at 15 years and at 20 or 25 years. You can examine the statistics for yourself: there is a slight added risk of coronary calcium score progression according to years of obesity. What does this tell us? That some useful knowledge might emerge from this cohort if we wait for real events to happen.

289    I was quite a fan of the UK General Practice Research Database, which recently morphed into the Clinical Practice Research Datalink, but then I read the recent survey of myocardial infarction in the UK which seemed to show that the CPRD fails to record about 50% of MI. So now I am hesitant to believe everything I read from this source; and that’s not the fault of the hardworking Israeli authors of this article, who clearly think they have been using a wonderful resource. For prescribing that is true: very little escapes recording on GP computers. In this case the drugs they searched for were highly specific for prostate cancer: GnRH agonists (leuprolide, goserelin, triptorelin), and oral antiandrogens (cyproterone acetate, flutamide, bicalutamide, nilutamide). They looked at men who had been prescribed these drugs for newly diagnosed localized prostate cancer, to see if testosterone suppression carries a risk of acute renal injury, as judged by a first admission for renal failure. It’s here that the data become dodgy. The CPRD seems to show an increase in renal failure admissions, driven mainly by dual androgen suppression. But men receiving such drastic treatment for localised disease can’t be typical. I can’t take home a clear message from this study.

NEJM  18 July 2013  Vol 369
213    Two protons plus two neutrons make a helium nucleus. The “alpha rays” emitted by the decay of radioactive elements aren’t really rays at all but helium nuclei thrown out over short distances. So what you need for cancer therapy is an alpha-emitting isotope which will place itself within the cancer and bombard it with these helium nuclei, which don’t travel far enough to harm surrounding tissue. It should be an isotope with a short half life: radium-233 will do, but so would samarium-153 if you could get hold of it. The radium has a half-life of 11.4 days as it emerges from the cyclotron, and it’s easy to transport because ordinary containers stop alpha-particles. Put radium-233 into men with advanced metastatic prostate cancer and it makes for the bones and irradiates the metastases. Very neat, hugely expensive, largely free of adverse effects. Buys three months of life.

224   Twenty-three years ago I found myself sailing on a yacht in San Francisco harbour with an expatriate Brit who was making a fortune out of designing insulin pumps. I asked him how close we were to getting one with automatic glucose sensing and feedback, and he fell silent. That is my anecdote: I felt you should know. As for self-regulating insulin pumps, this despatch from the front line seems to indicate that progress has been slow. The trial shows that a new kind of pump will successfully switch off at night if the sugar level drops too low. This, believe it or not, reduces nocturnal hypoglycaemia.

245   The National Lung Screening Trial was a randomized trial that compared the efficacy of low-dose CT for lung-cancer screening with that of chest radiography in 53,454 smokers between the ages of 55 and 74 years with a minimum of 30 pack-years of smoking and no more than 15 years since quitting. When its results appeared two years ago, even hardened anti-screeners like me had to admit that the mortality benefit was impressive; but even hardened pro-screeners had to admit that the benefit was dependent on the baseline risk of lung cancer. The roll-out of systematic screening programmes needed to depend on further analysis of the NSLT data: and here it is. 88% of the benefit from screening—lung cancer deaths prevented—came from the 60% at highest risk. Screening the lowest quintile of risk had a 1% lung cancer mortality benefit.

Lancet  20 July 2013  Vol 382
209    Every year 800,000 children die from diarrhoea. The rate at which this falls towards zero will be a good measure of the progress of civilization. Most of these deaths occur in sub-Saharan Africa and South Asia, and this paper reports on the main causative agents. The top five are rotavirus, Shigella, enterotoxigenic Escherichia coli producing heat-stable toxin (ST-ETEC), Cryptosporidium and typical enteropathogenic E coli. These were identified by careful fieldwork carried out by the Global Enteric Multicenter Study (GEMS), who certainly deserve top slot in the Lancet. It’s interesting to see that two causes of diarrhoea we seldom worry about in well-nourished British children—rotavirus and Cryptosporidium—can be major killers in the poor world. We need a better world as well as better interventions for diarrhoea.

223    The beginning of the nineteenth century brought with it a vogue for post-mortem examinations as the key to medical knowledge, and although this gave rise to much macabre humour, it was accepted by the public to a surprisingly wide extent. We live in more squeamish times now, especially where children are concerned, as the Liverpool body parts scandal demonstrated. Parental consent for autopsy is declining, as is the number of autopsies performed on foetuses, infants and children, despite a large increase in the number of requests. One way to overcome this may be the “minimally invasive autopsy” carried out by MRI scanning with limited tissue sampling as required. This comparative study shows that it has a very high degree of correlation with full open autopsy.

234   Surgeons continue in a state of rest or constant motion, unless acted upon by an external force. For a common operation, like caesarean section, they will use exactly the same technique every time, from the time when they were first taught to the time when they finally abandon the scalpel for the golf club. The CORONIS trial dared to find out whether this matters in caesarean section: “CORONIS was a pragmatic international 2×2×2×2×2 non-regular fractional, factorial, unmasked, randomised controlled trial that examined five elements of the caesarean section technique in intervention pairs. CORONIS was undertaken at 19 sites in Argentina, Chile, Ghana, India, Kenya, Pakistan, and Sudan. Each site was assigned to three of the five intervention pairs: blunt versus sharp abdominal entry; exteriorisation of the uterus for repair versus intra-abdominal repair; single-layer versus double-layer closure of the uterus; closure versus non-closure of the peritoneum (pelvic and parietal); and chromic catgut versus polyglactin-910 for uterine repair.” This is pretty awesome, and it involved nearly 16,000 women. They all had roughly similar outcomes.

BMJ  20 July 2013
Four Chinese students and their professor examine the evidence that fish oils and n-3 polyunsaturated fatty acids (n-3 PUFA) prevent breast cancer. Their wonderfully intricate meta-analysis suggests a dose-responsive effect which averages out at about 14% between the highest and lowest categories of n-3 PUFA intake; but you have to take this in capsules, because total fish intake by itself seems to count for little. This is the opposite of the case with cardiovascular protection, where you have to eat the fish and not the capsules to get the benefit. As a man, I shall just eat fish.

By the time you have reached late middle age, you are very likely to have had a parent admitted to hospital before death, and you almost certainly carry with you some awful memory of what happened to them. Some of this may be inevitable: most care may have been good; but the bad bits remain in your mind. That is the moral of this study of patients with cognitive impairment or delirium admitted to a large hospital and randomized to usual general ward care or a specialist medical and mental health unit. The usual “objective” measures showed no difference, but the authors conclude “Specialist care for people with delirium and dementia improved the experience of patients and satisfaction of carers, but there were no convincing benefits in health status or service use. Patients’ experience and carers’ satisfaction might be more appropriate measures of success for frail older people approaching the end of life.” And this may not require specialist units, but better training in the recognition of delirium and kindness towards confused patients. Bring it on soon, please.

A useful Therapeutics review looks at non-steroidal anti-inflammatory drugs. Each article of this kind gives slightly different odds for GI bleeds and cardiovascular harm from specific drugs, but this one is very well referenced and is bang up to date, so I believe it. There is no excuse for prescribing long-term diclofenac in the elderly. There is no excuse ever for prescribing any NSAID at any dose to anyone with heart failure. That little bit of naproxen or ibuprofen can tip them into renal failure, if it doesn’t send them to hospital with pulmonary oedema: just don’t do it. I wish these messages came out even more strongly in this article. And I also wish they were clearer about what risk—if any—ibuprofen carries for people with a diagnosis of asthma.

Fish of the Week: Psetta maxima

I have been told that turbot has the highest concentration of n-3 polyunsaturated fatty acids of any fish, though I have been unable to confirm this. Eat it because it is delicious, and also because it can be farmed sustainably. Though I have to say that I also buy big wild turbot when I can, because larger fish have the best texture and flavour.

Cooked turbot can have a slightly earthy taste, which is more pronounced in small sized fish cooked with their skin. Here is a way of “cooking” turbot that gets round that problem.

Get as big and fresh a turbot as you can and invite as many people as necessary to consume it. Four hours before they arrive, fillet the turbot as demonstrated on you-tube, then carefully peel off the skin by laying the fillets skin-down and slipping a sharp knife under the flesh and directing it forwards onto the skin.

As cook’s treat, you can now heat oil and chilli nearly to smoking point, drag the skin through seasoned flour, throw it in the oil and eat it immediately, sprinkled with lemon juice and accompanied by a soothing glass of sauvignon blanc.

Returning fortified to the raw turbot fillets, cut them into chunks approximately 2cm across. Put these into a bowl and mix with chopped red onion, finely diced red chilli, generous sea salt, and finely chopped chervil with a small amount of lovage (too much will make it bitter). For about 800G of turbot, add the juice of four limes and one lemon, and stir. Add more if necessary to cover the fish.

Put this in the refrigerator and serve when your guests are ready. If the fish is very firm, it may need five hours in the marinade: rarely longer. The Peruvian accompaniment to ceviche is boiled or grilled sweet corn. Their Cusquena beer is excellent with every kind of ceviche, as is any crisp white wine.

The chervil/lovage mixture is the gold standard as far as I am concerned, and I like the onion marinated with the fish: but you can use other herbs, especially coriander (cilantro) and/or parsley and/or fennel, either in the mixture of sprinkled at the time of serving. You may wish to be sparing with onion in the marinated mixture, and then add more thin rings of extra red onion as a final garnish. It is up to you. Use halibut, or farmed sea bream, or wild sea bass: just eat ceviche as often as you can.