Richard Lehman reviews the latest research in the top medical journals
NEJM 28 Jun 2018
Joseph Haydn came close to being castrated as an adolescent. He was an outstanding chorister in St Stephen’s Cathedral, Vienna and they wanted him to continue as a soprano for the rest of his life. But he resisted the customary procedure, entered puberty, and was thrown out of the choir when the Empress complained that his singing had become more like crowing. Nowadays the term “castration-resistant” is confined to prostate cancers that fail to respond to androgen suppression. Enzalutamide is different from direct forms of chemical castration because like bicalutamide it binds to the androgen receptors of the tumour itself. Both drugs have been shown to slow progression in metastatic prostate cancer and this trial finds another niche for enzalutamide in the treatment of localized prostate cancer that has not yet metastasized, but is leading to rapid increases in PSA. One of the end-points is time to metastasis compared with placebo: here there is no quibbling because the median metastasis-free survival was 36.6 months in the enzalutamide group versus 14.7 months in the placebo group. But what about survival? It’s too soon to say, but “at the first interim analysis of overall survival, 103 patients (11%) receiving enzalutamide and 62 (13%) receiving placebo had died.” In other words, no difference. Yet in the conclusion the pharma-funded authors say that “enzalutamide treatment led to a clinically meaningful and significant 71% lower risk of metastasis or death than placebo.” Journals today make me long for the arrival of a replacement method of presenting data fairly and contestably in the public domain. I believe they were better at this in 1665 than they are now.
LABA/steroid inhalers: don’t mention the meta-word
Reports of interventional trials conducted by manufacturers feature prominently in the NEJM, but the world’s loftiest medical journal has maintained a taboo on meta-analyses. This one sneaks in by calling itself “Combined Analysis of Asthma Safety Trials of Long-Acting β2-Agonists” and it’s true that it is not a standard meta-analysis because it covers just four related trials. These were mandated by the FDA after four companies marketing LABAs for asthma were licensed to sell their products on condition that they perform prospective, randomized, controlled trials comparing the safety of combination therapy with a LABA plus an inhaled glucocorticoid with that of an inhaled glucocorticoid alone in adolescents (12 to 17 years of age) and adults. The conclusion: “Combination therapy with a LABA plus an inhaled glucocorticoid did not result in a significantly higher risk of serious asthma-related events than treatment with an inhaled glucocorticoid alone, but resulted in significantly fewer asthma exacerbations.” So all is well. But is it? In fact this is an ideal teaching case for the limitations of meta-analysis. The sample size of 36,010 was big enough to demonstrate a clinically meaningless, but statistically significant reduction in exacerbations of 1.9%. On the other hand it was too small to detect a significant imbalance in harms. The only two people to die of asthma in these trials were in the LABA/steroid group. So I’d suggest the conclusion should read: “Combination therapy with a LABA plus an inhaled glucocorticoid did not result in a clinically significant reduction in exacerbations compared with an inhaled glucocorticoid alone. Two asthma-related deaths occurred in the combined inhaler group, suggesting that further safety surveillance is required before clinical recommendation of this treatment.”
Getting wise with PPIs
Another meta-analysis sneaks into this week’s New England Journal by way of a masterly review of prophylaxis against upper gastrointestinal bleeding in hospitalized patients by Deborah Cook and Gordon Guyatt. It comes as a key reference providing evidence from 31 randomised trials that the use of acid suppression in hospitals is far more widespread than it should be. This article demonstrates that it probably does nothing to reduce mortality while being an important risk factor for hospital acquired infection—especially pneumonia and Clostridium difficile. I get the impression that omeprazole 20mg daily must come as a standard printed entry on all hospital and nursing home drug charts. Strike it out. Choose wisely, using this review to inform your discussion with patients.
Ann Intern Med 26 Jun 2018
Three choices for kidney cancer
One of the changes I’ve noticed in twenty years of reviewing the medical literature is a change of marketing language about cancer treatments. “Radical” and “aggressive” have given way to “targeted” and “personalized”, while “middle-of-the-road” doesn’t get much mention. For small kidney cancers without metastasis (T1a), radical nephrectomy has slowly given way to nephron-sparing nephrectomy as the middle-of-the-road standard of care, with similar high rates of cure. But now a new technique has come on the scene, in the form of percutaneous ablation. This is done by radiologists aiming needles at the tumour and ablating it with heat or cold. This paper compares outcomes from the two types of surgery and thermal ablation. Alas, this is purely observational and only in those aged 66 and over. “For well-selected older adults with T1a RCC, PA may result in oncologic outcomes similar to those of RN, but with less long-term renal insufficiency and markedly fewer periprocedural complications.” For me, this reads too much like “we don’t really know, but we hope this data helps a bit.”
Defining and explaining overdiagnosis
If you can’t define a word in one sentence, don’t use the word. That’s why we need to rethink using the term overdiagnosis except in very specific contexts. Here is a nice definition in the context of cancer screening: “the detection of a (histologically confirmed) cancer through screening that would not otherwise have been diagnosed in a person’s lifetime had screening not been done.” But as this thoughtful article makes clear this is only the start of the problem: how can you quantify overdiagnosis or explain it to people? Excellent attempts have been made, but they have needed whole books to succeed. To paraphrase something I said about “heart failure” long ago, when a term confuses both doctors and patients, it is time to think of a better term. Or to rethink the concept entirely. If it isn’t helping the shared understanding of medicine, it isn’t doing its job.
The Lancet 29 Jun 2018
Diabetic drug development
So if we should drop terms that confuse both doctors and patients, how about “type 2 diabetes”? Archie Cochrane called for its abolition 40 years ago. Why define a risk cluster by an arbitrary threshold of a single biochemical component? The answer is simple and depressing: because it provides such a simple common narrative for doctors and patients and a massive market for sugar-lowering drugs. Of late the marketing ploys have become ever more sophisticated, with drugs that promise dual hormonal actions and maybe weight loss too. Here’s a new example, which The Lancet seems to think is sufficiently exciting to publish a phase 2a trial of. It doesn’t even have a name yet. “Interpretation: MEDI0382 has the potential to deliver clinically meaningful reductions in blood glucose and bodyweight in obese or overweight individuals with type 2 diabetes. Funding: MedImmune.” I may be wrong, but I suspect this drug belongs to a large and heterogeneous class of drugs called reprintins. Reprintins bind to the financial receptors of leading journals, opening their cell membranes to the inward transfer of money in exchange for the outward transfer of paper.
The non-metrics of decent care
Next time you get a letter saying “Your care and attention meant so much to us,” how about sending back a bill detailing the exact amount and asking them to itemize any services that you may have missed. No, honestly, please don’t. I was trying to be ironic, but that is a dangerous game in the NHS these days. (Do you realize that we have just had an NHS coproweek? Seriously? You can go to the hashtag #coproweek and find all about it; or there’s #coprocommitment for serious coprophiles). I’m trying to make the old point—which can never be repeated enough—there are values in medicine which cannot be captured by simple metrics, let alone by money. The point is made more elegantly by Chris Dowrick in a commentary on this trial of a complex intervention for people living with comorbidity in primary care. It was built on the best current ideas about self-efficacy and personal support, but the authors conclude that “To our knowledge, this trial is the largest investigation of the international consensus about optimal management of multimorbidity. The 3D intervention did not improve patients’ quality of life.” Chris Salisbury and his team deserve congratulations. Patients felt better supported and listened to, and the trial did massive good by demonstrating what we can and can’t conclude about interventions for people living with complex illnesses using the current metrics.
The BMJ 29 Jun 2018
Readers who have stayed with me through these reviews will be aware that “hypertension” is also on my bin-list of terms which confuse both doctors and patients. There is so much circularity in most papers about high blood pressure that if you don’t get vertigo, you can’t be reading them properly. The obvious solution is not to read papers about high blood pressure. Ask for some risk figures instead, with direct trial evidence about particular risk-reducing strategies. Stuff that actually applies to the person with the sphyg round their arm. Again, if it takes more than a couple of sentences to explain, don’t bother. Here is a trial of a new strategy for detecting “hypertension” using criteria based on cardiovascular risk. So far so good: “this new triaging approach accurately classified hypertensive status for most, with half the utilisation of ABPM compared with usual care.” But what does “hypertensive status” mean in terms of a shared strategy? This is not just a purist question, asking for long-term trials on this newly defined population with “hypertensive status”; it’s also a plea for nuance in discussing continuous variables with healthy people and helping them make personal choices between multiple drugs which over a hundred people need to take to prevent one event.
Plant of the Week: Dregea sinensis
We used to visit a mournful Cockney nurseryman near Oxford, who sold some wonderful plants very cheaply. On one visit he came up to me and said “Ere, smell this”.
It was a climber with small umbels of sweetly scented flowers which I recognised as Wattakaka sinensis, so I replied “What a cracker!”
“What did you just say? Yer, it’s a Wattakaka. It’s called a wattakaka” he said, slightly puzzled. He was not a man of puns.
Since those days the Chinese wattakaka has changed its name to Dregea, but not its scent, which is still sweet and intense and old-fashioned, redolent of hot summers in the 1950s. But in fact it was never a common English garden plant, because it usually dies after a winter or two. We’ve been through four so far.
Our latest one is tucked behind a bench in a sheltered spot close to jasmine and a very fragrant rose. Despite this competition it still makes its mark on the nostril. This is good: olfactory overload is what you look for when sitting on a sunny bench in high summer. Time to put crushed borage in the lemonade.