NEJM  16 Feb 2017  Vol 376
Periviable infant outcomes
“Periviable” is a new word to me. I think I shall start using it for pieces of cheese that have been slightly forgotten in the back of the fridge. For neonatologists, it has a more serious meaning: it refers to infants born on the borderline of viability at 22 to 24 weeks of gestation. In most developed countries, survival rates have improved for these extremely premature babies, but the current study looks at their associated neurodevelopmental outcomes too, using data on 4274 infants from 11 centres that participated in the National Institute of Child Health and Human Development Neonatal Research Network. Between 2000 and 2011, survival increased from 30% to 36%, but the percentage of infants who survived with neurodevelopmental impairment did not change significantly. Even when cared for in a subset of America’s leading academic centres, 43% of surviving periviable babies will show neurological damage in childhood.

STAMPEDE towards bariatric surgery for diabetes
STAMPEDE stands for Surgical Treatment and Medications Potentially Eradicate Diabetes Efficiently. Yes. It does. Here are the five year results for 134 of the original 150 stampeders. They all had type 2 diabetes with a body mass index of 27 to 43, and were randomly assigned to receive intensive medical therapy alone or intensive medical therapy plus Roux-en-Y gastric bypass or sleeve gastrectomy. In other words, they ranged in weight from normal to obese, but not super obese. The primary endpoint was a glycated haemoglobin of six or less, which may be traditional but is fairly meaningless in itself. The trial was not powered to detect meaningful events in the five year period. Gastric bypass achieved the biggest reductions in weight and fasting glucose, but sleeve gastrectomy was nearly the same and produced a marked reduction in blood pressure. I hope these data are available for individual participant meta-analysis, as evidence grows that bariatric surgery is the best treatment for high risk type 2 diabetes and we badly need to know who might benefit most. In the end, as this article concludes: “The potential benefits of bariatric surgery on clinical endpoints, such as myocardial infarction, stroke, renal failure, blindness, and death, as suggested in non-randomized trials, can be adequately assessed only through larger, multicenter trials.” Acronyms, anyone? Don’t all stampede at once.

JAMA  14 Feb 2017  Vol 317
Keeping a cool head during chemo
SCALP is the acronym for our next trial, which tested the effectiveness of a scalp cooling device to prevent alopecia in women undergoing chemotherapy for breast cancer. Good. This is an interim analysis, but it shows that keeping a cool head during chemo with anthracyclines or taxanes reduced the need for a wig or head wrap from 100% to 63%. A very similar trial looked at women with breast cancer receiving non-anthracycline chemotherapy. In both trials, scalp cooling to a chilly 3°C began 30 minutes before chemo administration and continued for 90-120 minutes afterwards. The results were the same as in SCALP: a very useful 50% diminution in hair loss.

Sublingual grass
Whenever I read an article about sublingual grass pollen for seasonal allergy, I’m reminded of William Blake’s amazing print of Nebuchadnezzar eating grass. Readers versed in the Holy Scriptures will remember that the Assyrian king continued in this state until “seven times” passed over him, by which time his hairs were grown like eagles’ feathers and his nails like birds’ claws (Daniel 4;33). Adults who use two years of sublingual grass therapy for moderate to severe allergic rhinitis are unlikely to suffer the same fate. In fact, nothing is likely to happen to them at all. In the GRASS randomised controlled trial, the active group showed no significant difference from the placebo group in nasal response to allergen challenge at three years’ follow-up. Maybe you need to do the full Nebuchadnezzar.

Waist-to-hip ratio
I’m old enough to remember when the body mass index was a new thing. You worked it out on a pocket calculator when you had weighed and measured your patient. Before that, you just looked at their middle. This is, in fact, a better indicator of risk, as people have been pointing out for decades. Best of all, you can do both, as in this study of nearly 120 000 individuals on the UK Biobank database. And then you can see how much of the risk appears to be genetic, as you have their full genome as well. The conclusion of this study is that “A genetic predisposition to higher waist-to-hip ratio adjusted for body mass index was associated with increased risk of type 2 diabetes and coronary heart disease. These results provide evidence supportive of a causal association between abdominal adiposity and these outcomes.” Indeed, though in clinical practice I find jeanomics to be a better guide. The key question is “Have you had to buy a new pair of jeans lately and what is their waist size?”

JAMA Intern Med  Feb 2017
Subgroups and “precision medicine”
In my retirement I’ve come to mix with the kind of people who greet you in the corridor saying, “Hey, have you read the latest Ioannidis?” I don’t mind: they are among the best sort of people, and it makes a pleasant change from talking about Trump. The latest paper from John I’s team at Stanford should really be credited to its first author, Josh Wallach, and it poses the question “How often are subgroup claims reported in the abstracts of randomized clinical trials supported by a statistically significant interaction test result and corroborated by subsequent randomized clinical trials and meta-analyses?” This is not an anorak question, but something all clinicians should worry about. Somebody should invent a term for the urban myths that spread so rapidly through medicine despite the best attempts of the evidence based medicine community. Hint that some new thing may perhaps be better for a certain group of patients and we all start trying it on them. This should never be the case. The article concludes that “Attempts to corroborate statistically significant subgroup differences are rare; when done, the initially observed subgroup differences are not reproduced.” The study was based on 64 randomised controlled trials, which made a total of 117 subgroup claims in their abstracts.

Drugging the ageing CNS
One way to monitor prescribing trends in the US is through the National Ambulatory Medical Care Survey, an open access survey of “office based physicians.” This study used the Beers criteria to measure changes in polypharmacy with three or more drugs affecting the central nervous system in people age 65 and over. For these older Americans, such polypharmacy more than doubled between 2004 and 2014, even though there was no change in the consultation rate for anxiety, depression, or insomnia. It would be useful to compare trends in different countries: this seems almost incredible, and could even be responsible for the lowered age of death in some parts of the US.

The Lancet  18 Feb 2017  Vol 389
In praise of cynicism
Diogenes, the original cynic, had fun turning up at Plato’s lectures naked and munching bread to disturb the solemn tone of those occasions. He famously walked around Athens carrying a lamp in daytime, saying he was trying to find an honest man. Cynicism, in this good sense, is to do with the puncturing of pomposity: it demonstrates a sense of the impossibility of idealism in a corrupt world, but without any personal claim to superiority. It is nice to see this celebrated in a Lancet essay with the title “Cynicism as a strategic virtue.” Read it if you can gain access. We all need a bit of cynicism as a coping mechanism, and it looks as if we will need ever more in coming years.

Redoing joint replacement
Woo, here is culture change in action: “Our study used novel methodology to investigate and offer new insight into the importance of young age and risk of revision after total hip or knee replacement. Our evidence challenges the increasing trend for more total hip replacements and total knee replacements to be done in the younger patient group, and these data should be offered to patients as part of the shared decision making process.” Offering data to patients as part of a shared decision making process? In orthopaedics? Political correctness gone mad. In my day you told the patient what operation they needed, how many months it would take to get done on the NHS, how many days it would be to get done privately, plus the number of my private secretary. Next! This new study was based on the UK Clinical Practice Research Datalink and found that the lifetime risk of requiring revision surgery in patients who had total hip replacement or total knee replacement over the age of 70 years was about 5% with no difference between sexes. But for those who had surgery younger than 70 years, the lifetime risk of revision increased for younger patients—up to 35% for men in their early 50s—with large differences seen between male and female patients (15% lower for women in the same age group). Those are certainly figures that patients should know about.

Combined faecal transplant for UC
A distinguished follower of these reviews recently told me that I talked too much about bottoms. I could only apologise and explain that I must go where I am led. For example, the faecal microbiome is a frequent subject in journal articles, which are difficult to comment on without involving bottiness of some kind. Here is an extreme example: “We randomly allocated patients with active ulcerative colitis (Mayo score 4–10) in a 1:1 ratio, using a pre-established randomisation list, to either faecal microbiota transplantation or placebo colonoscopic infusion, followed by enemas 5 days per week for 8 weeks. Patients, treating clinicians, and other study staff were unaware of the assigned treatment. Faecal microbiota transplantation enemas were each derived from between three and seven unrelated donors.” The primary outcome was steroid-free clinical remission with endoscopic remission or response. It was achieved in 29% of the patients who received their 40 enemas of mixed faeces and by 8% of those assigned to placebo. Active UC is a nasty condition that often leads to total colectomy, so these results are not to be pooh-poohed at, if you will pardon the expression.

The BMJ  18 Feb 2017  Vol 356
Sneezing at vitamin D
Individual participant data (IPD) meta-analysis is a good thing. In theory, it should mitigate the “small trial effect” by providing clearer outcome data for individuals. In small trials, vitamin D supplementation reduces the number of upper respiratory tract infections. In large trials, it doesn’t. Pool them all, and you’re left with a small but statistically significant benefit, NNT=33. It’s a great exercise in IPD meta-analysis and it’s up to you whether you believe its clinical significance. Personally, I’m happy to take cheap over the counter vitamin D whenever I remember to, for this and its other putative benefits. In the lottery of life, this is a 1p coin that you may or may not be bothered to pick up. That said, The BMJ has hedged its bets by printing an editorial, which comes down against taking vitamin D unless you are at risk of osteomalacia. Given how little this matters, I am sure it will be the subject of prolonged and heated debate.

A SurePath to cytological sensitivity
It’s amazing to think that nearly 90 years have passed since Georgios Nikolaou Papanikolaou first described an intravaginal cytological test to detect cancer of the uterine cervix, and yet we’re still working out how best to do it. Here a Dutch team compares the cumulative incidence of cervical cancer diagnosed within 72 months after a normal screening sample between conventional cytology and liquid based cytology tests SurePath and ThinPrep. The absolute differences are tiny: they are expressed as numbers per 100 000 samples, and they come out slightly better for SurePath (44.6) than for ThinPrep (66.8) or conventional cytology (58.5).

Pneumonia followed by heart failure
All patients aged more than 17 years with community acquired pneumonia admitted during 2000-02 to any of the six hospitals or seven emergency departments covering a million people in Canada were enrolled in a clinical registry. That’s quite an interesting cohort, and it excluded people who had cystic fibrosis or were immunocompromised. Their mean age was 55 and two thirds were treated as outpatients. Over a follow-up period of nearly 10 years, post-pneumonia patients under 65 were twice as likely to develop heart failure as matched controls (4.8% v 2.2%). Those over 65 naturally had a much higher absolute incidence of heart failure, although the relative difference was less (24.8% v 18.9%).

Plant of the Week: Eranthis hyemalis
Winter aconites are members of the buttercup family with bright yellow flowers, which come out in late January and are gone by the end of February, at which point the celandines take over. They are designed to take advantage of the light in deciduous woodlands before the trees break leaf. By mid-spring, their low rosettes of leaf will have done their work and put enough food into their rhizomes for them to form a bigger plant the next season. You can see them in abundance in many English woods on limy soil, often flowering together with snowdrops.

That gives you a clue about where to plant them in the garden. Dig them in where the sun shines now, but where they will be overshadowed by trees and shrubs later in the season. You may wish to put in a few snowdrops too. And do not eat these so called aconites: they are full of cardiac glycosides and even bufadienolide toxins, otherwise best known from toads.