NEJM 13 Oct 2016 Vol 375
Outcomes and choices
This week’s print NEJM contains mostly papers I’ve already commented on—notably, Gilbert Welch’s important study of mammography and breast cancer outcomes and the landmark British trial comparing surgery, radiotherapy, and watchful waiting for PSA detected prostate cancer (ProtecT).
However, I hadn’t commented on the patient reported outcomes paper that went with this study. This is immensely important because the two papers together provide all the information that men with localised prostate cancer detected by PSA need to make a decision about treatment. If they opt for surgery, they run the highest risk for impotence and urinary incontinence, and this is unlikely to improve over time. If they opt for radical radiotherapy, they run similar risks but the effects tend to improve after six months. They are also likely to have bowel effects, which may resolve to a varying extent. Interestingly, no significant differences were observed among the groups in measures of anxiety, depression, or general health related or cancer related quality of life. In reality, I suspect that few men will now opt for anything but watchful waiting, but at least they should have these choices put clearly before them.
At this point, I would normally go to the “Online First” section and take my pick of the papers awaiting print. But just look at this list:
Now I guess that some NEJM readers might be interested in one or more of these topics, but this is multiplicatio ad absurdum. Even oncologists must baulk at this overload of trials, most of which use “progression-free survival” as their primary outcome. They are all free to download in full copy. There is no conceivable reason for anyone to order a reprint. I just mention this. If you want to read a good critique of their science and clinical importance, Bishal Gyawali’s monthly blog is a must. I will move on, sadly. The NEJM is meant to be the world’s leading generalist journal, not the Annals of Expensive Oncology.
JAMA 11 Oct 2016 Vol 316
What shall we do with the broken ankle?
I’ve just been doing a session with surgical evidence based medicine Masters course students, and it was a pleasure to tell them that surgical trials are now generally better than trials of medicines because they are mostly clinician led and address good, answerable, binary questions. Here is an example: A team of orthopaedic surgeons decided to test the hypothesis that for unstable ankle fractures in people over 60, close contact casting might be a good alternative to open fixation surgery. The latter involves a lot of dodgy metalwork that has to be removed later and has a 10% infection rate. They ran a randomised trial involving 620 patients. Blinding was impossible, except for the final functional assessment, and it took 10 years from the original pilot study to completion. Malunion was commoner in the casting group, although their infection rate was very low (1%). Functional results at six months were similar. Casting took less time, and there were no differences in quality of life, pain, mobility, or patient satisfaction. Patient, take your pick: here is the evidence to inform your choice.
Changing diet of supplements in the US
The percentage of the American population who took “dietary supplements” between 1999 and 2012 didn’t change, according to the National Health and Nutrition Examination Survey. It stands at 52%, and supports a massive industry there. Imagine if all that money was spent on something useful, like burying their electricity cables in the ground or building fast efficient railways. The only chink of rationality is that people are slightly moving away from multivitamin/mineral supplements to fish oils (possibly of slight benefit to a few) and vitamin D (almost certainly beneficial to some).
JAMA Internal Med Oct 2016
Hospitals were made for C diff
Clostridium difficile spores persist for months in hospital rooms where previous patients have had the infection. Now, it seems, patients are at higher risk even if the previous occupant had received antibiotics. But the absolute risk is just 0.57%: among 100 615 pairs of patients who sequentially occupied a hospital bed, there were 576 cases. Read on if you like, but after adjustment for other factors, the added risk of 1.22 (adjusted hazard ratio) only just held on to significance.
I’m just back from the US where the fall colours in Vermont were spectacular. People who follow them southwards through New England are called “leaf peepers,” but during my sojourn further south in Connecticut I was mainly among trial peepers. Trial peeping is very much a minority sport, practised by those who like lifting the stones of medicine to see what creepy crawlies may emerge from beneath. Joe Ross, my generous host at Yale, is a leading practitioner, but the latest study comes from Harvard and examines what happens to trials of “failed” drugs. Among 640 novel therapeutic agents, 54% failed in clinical development, 36% were approved in the US, and 10% were approved only outside the US. For the 344 failed agents, pivotal trials were only published in 40% of cases. In fact, of 74 trials of agents that failed for commercial reasons, only six were published. Does this matter, seeing that the drugs never appeared on the market? Certainly, because these trials contain material, especially about safety, which could be vital to protect participants in future trials of similar agents.
Lancet 15 Oct 2016 Vol 388
Mitigating doctor burnout
For certain medical specialties, burnout comes with the territory. And it is probably increasing, to the point where over half of doctors in the US fulfil the Maslach criteria for burnout, as measured by emotional exhaustion and depersonalisation. The demands of the job deplete the economy of the soul and it goes into permanent deficit. I could feel this happening to me midway through my 35 years as a general practitioner, at about the same time as I became “senior partner” in our fast expanding practice. For my partners, I proposed six months sabbatical leave every five years, and we agreed generous annual leave arrangements; the rest of the time we were unusually open to seeing all the patients who wanted to be seen on the day. For myself, I took up a research interest and started to write these reviews every week. Here is a review of the literature on physician burnout and its cures.
The authors found 2617 articles and looked in depth at 230: they ended up reviewing 15 randomised controlled trials and 37 cohort studies (NB: do not attempt a systematic review as a cure for burnout). The results are frankly rather banal: a reduction in working hours helps; you can also try mindfulness or stress management. Well yes, but life goes on for a long time; it is not a rehearsal; and in medicine your job is most of your life, and deeply affects the lives of others. Among the titles of books that I will never write is The Economy of the Soul, proposing a model of supported autonomy as the cure for burnout. But such a model cannot work in a society that has turned its back on cooperation and kindness.
LMW heparin and the placenta
As I said, systematic reviews are not a cure for burnout. Individual participant data (IPD) meta-analysis is bound to deplete the economy of anyone’s soul, and I’m just glad that there are people willing to do it. Here is a great example, which looks at the effect of low molecular weight heparin on placenta mediated pregnancy complications, which include pre-eclampsia, late pregnancy loss, placental abruption, and birth of a small for gestational age neonate. Long and hard did they labour through many a weary day and night. And the result: “Low-molecular-weight heparin does not seem to reduce the risk of recurrent placenta-mediated pregnancy complications in at-risk women. However, some decreases in event rates might have been too small for the power of our study to explore.” These guys are heroes of science. There is more rejoicing in heaven over one definitive, negative IPD meta-analysis than over badly designed cancer trials finding “breakthroughs” that turn out to be all hype and no substance.
Probing pile procedures
Hurray for another clinically important, publicly funded British trial comparing two common surgical options. Traditional excision surgery for piles works, but leaves people with a sore botty for weeks; stapled haemorrhoidopexy is less painful in the short term. But over 24 months, the excision results are better for pain and quality of life. Moral: if you have piles, try to avoid surgeons. If you have to see one, ask for the traditional procedure. Provided he/she is good at it.
The BMJ 15 Oct 2016 Vol 355
A couple of research articles in this week’s The BMJ examine the “off-target” effects of childhood vaccination. Observational evidence, especially from resource poor settings, has suggested that some vaccines confer protection beyond their specific diseases: the usual examples are measles vaccine and Bacillus Calmette-Guérin (BCG). A systematic review from Bristol finds some support for this, importantly from randomised trials as well as observational studies. But it goes on to say that “Receipt of diphtheria-tetanus-pertussis vaccine (almost always with oral polio vaccine) was associated with a possible increase in all cause mortality on average (relative risk 1.38, 0.92 to 2.08) from 10 studies at high risk of bias.” This raises the “Cochrane dilemma” I mentioned last week: when the evidence is frankly crappy, should you accord it any airtime at all? Especially when the result is alarming as well as unsubstantiated. The BMJ’s answer is to publish a parallel systematic review that puts it more clearly: “The quality of the evidence, however, does not provide confidence in the nature, magnitude, or timing of non-specific immunological effects after vaccination with BCG, diphtheria, pertussis, tetanus, or measles containing vaccines nor the clinical importance of the findings.” This is also the message of the accompanying editorial.
Antenatal corticosteroids revisited
The Cochrane Collaboration began after Iain Chalmers had performed an exhaustive review of the evidence base of obstetrics and neonatology in the 1980s. A key finding was that oral corticosteroids given to mothers before premature delivery reduced respiratory distress syndrome in their babies. This is confirmed for later term pregnancies by a new systematic review and meta-analysis of the effectiveness of antenatal corticosteroids given at ≥34 weeks’ gestation.
But this is not the end of the story. Iain has recently written a wonderful blog with the title “Should the Cochrane logo be accompanied by a health warning?” He points out that last year the Lancet published a new study that estimated that for every 1000 women exposed to a steroid use scale up strategy, there was an excess of 3.5 neonatal deaths and increased maternal infection, and secondary analysis suggests that this increased mortality was attributable to differential exposure to prenatal steroids. These figures come from low to middle income countries. In the application of evidence, context is all. This applies (as here) to populations; it applies to individuals; it applies to diagnostic categories; it applies to everything in medicine. Forget this at your peril, and especially your patient’s peril.
Plant of the Week: Geranium wallichianum “Buxton’s Variety”
Known commonly as “Buxton’s Blue,” this is an absolute must have for any garden. It is not named for the chilly upland town in the Peak District, but for EC Buxton, who gardened in the scarcely less chilly settlement of Betwys-y-Coed in North Wales and produced this marvel around 1900. As usual, Graham Stuart Thomas says it perfectly: “It is a pearl beyond price, producing a non-stop display from the end of June onwards of lovely Spode-blue flowers with large white centres and dark stamens over a luxuriantly leafed plant. It makes a sound clump and enjoys any fertile drained soil in sun or shade.”
Our plants are still producing flowers, which make lovely splashes of blue among the trailing red foliage of a Virginia creeper. They will carry on for some weeks yet.