Richard Lehman’s journal review—14 March 2016

richard_lehmanNEJM 10 Mar 2016 Vol 374
Treating malaria in pregnancy
913 Here’s a tonic for those of us who lie abed with thoughts about the stupidity of the world and the pointlessness of medical research. The PREGACT trial was supported by the European and Developing Countries Clinical Trials Partnership and it tells doctors in Africa exactly what they need to know: which is the best treatment out of four commonly used combinations to treat malaria in pregnancy? There’s no point in my listing the alternatives, because anyone who treats malaria can just look up the paper. The bottom line is that currently the best choice in Africa seems to be dihydroartemisinin–piperaquine. The manufacturers of the various drugs donated them to the project, and without commercial support the trialists used simple, robust randomisation methods and good ascertainment. It can be done like this. It should be done like this.

Preventing malaria in pregnancy
926 And here is another great example of doing it like this. This time the sponsor was the Eunice Kennedy Shriver National Institute of Child Health and Human Development, and the question was to determine which prophylactic anti-malarial might best replace sulfadoxine–pyrimethamine during pregnancy. The trouble with prophylaxis is, of course, that you create selection pressure in favour of the most resistant parasites, even if you only use it intermittently. So in countries like Uganda, sulfadoxine–pyrimethamine is inevitably losing some of its effectiveness. The alternatives tested here were a three dose regimen of dihydroartemisinin–piperaquine, and a monthly regimen of dihydroartemisinin–piperaquine. The latter proved best at preventing malaria in pregnant Ugandan women.

“Incompatible” kidneys do well
940 I don’t pretend to understand the immunology of renal transplantation and I would be rather sceptical of anyone who claimed to. But here are some figures that tell a story. They demonstrate the survival benefit for 1025 recipients of kidney transplants from HLA-incompatible live donors who were matched with controls who (a) remained on the waiting list or received a transplant from a deceased donor (waiting list or transplant control group) and (b) controls who remained on the waiting list but did not receive a transplant (waiting list only control group). That’s confusing enough, so let’s just jump to the eight year survival percentages: 76.5% vs. 62.9% and 43.9% per group. If you understand stuff like flow-cytometric cross-matching versus cytotoxic cross-matching and the pharmacology of solid-organ transplant immunosuppression you can look forward to a lot of enjoyment from this article. For most of us, the take home message is: if my kidneys pack up, just give me a kidney, any kidney.

Peanuts & the babes of London
OL Peanuts are unnecessary. They have been a human food of last resort for about 7000 years in South America, and for about 80 years in North America. Any food that contains peanuts would be better off without them. I hope I have made my position on the peanut question clear. However, it is not universally shared, and total peanut avoidance has become difficult in the 21st century. Hence the need to protect future generations from peanut related diseases, mostly in the form of allergy. You can do this by feeding peanuts to your infants in the first year of life, as in the LEAP trial. That way they are 80% less likely to show allergy to peanuts when they are 5. The LEAP-on trial went on a further 12 months and sheds further light on this question, but I am losing the will to continue, unlike the 32 000+ people who have made this the second most sought after article on the NEJM website. The primary outcome in the follow-up trial was the percentage of participants with peanut allergy after 12 months of peanut avoidance. It made no difference. Personally, I would recommend a minimum avoidance period of 100 years.

JAMA 8 Mar 2016 Vol 315
Vitamin D: this week’s negative trial
1005 Vitamin D is important for all sorts of things throughout the human lifetime, probably beginning at the first week of gestation. That’s what makes it so difficult to investigate. I’ve taken a sort of brotherly interest in vitamin D, as it’s an area of interest to Harold Hin, whose companionship as a work colleague for 18 years was a constant delight. After reviewing dozens of negative short term trials, I’ve come to the conclusion that almost all of them address the wrong questions and intervene too late and over too short a period. I suspect it will take a full lifetime of observational research to begin to understand how vitamin D levels affect the function of every organ—the brain, the kidneys, the heart, and the lung—as well as musculoskeletal health. This trial examines the effect of taking extra vitamin D for two years on knee pain and knee cartilage volume in patients with symptomatic knee osteoarthritis and low vitamin D levels. It has no effect at all.

Concussed kids
1014 A big Canadian study sought to find out what clinical features might predict persistent postconcussion symptoms (PPCS) in kids who have had a bang on the head. Surprisingly, the number of children presenting to US emergency departments with concussion has doubled over the past decade. Most of them have no symptoms at two weeks but a third go on to get PPCS, which can result in school absenteeism, impaired academic performance, depressed mood, loss of social activities, and lower quality of life. It also probably makes them less fit to bear arms. The investigators did their job thoroughly, collecting 2006 concussed kids from nine emergency departments to design their prediction tool (derivation cohort) and 1057 to test it (validation cohort). But they ended up disappointed with their own efforts. It showed “modest discrimination to stratify PPCS risk at 28 days. Before this score is adopted in clinical practice, further research is needed for external validation, assessment of accuracy in an office setting, and determination of clinical utility.”

Glucose control in older diabetics
1034 The greatest pleasure I get out of medicine these days is seeing brilliant youngsters doing stuff that I have long yearned to see done. After I wrote a BMJ editorial on glycaemic control in stable type 2 diabetes with Harlan Krumholz in 2009, I made a trip to Yale with John Yudkin and met a young endocrinologist and scholar called Kasia Lipska. And now, eight years after the key trials appeared, Kasia has published several articles, which should finally overturn the current model of overtreatment for type 2 diabetes in older patients. “High quality evidence about glycemic treatment in older adults is lacking. Optimal decisions need to be made collaboratively with patients, incorporating the likelihood of benefits and harms and patient preferences about treatment and treatment burden. For the majority of older adults, an HbA1c target between 7.5% and 9% will maximize benefits and minimize harms.”

Spyware for diabetics
1051 There are now 271 diabetes apps that allow you to make personal measurements all day long. Why? Well, one reason is that the app developers sense a lucrative market in collecting and selling personal data. This brief study shows that the great majority of these apps embed tracking cookies, which collect personal measurements and share them with third parties. Got diabetes? Buy an app and become a commodity.

JAMA Intern Med Mar 2016
App-roximate blood pressure
OL If you’d still like to go ahead and buy an app to keep track of your blood pressure, beware. A study of the most popular one shows hair-raising levels of inaccuracy when compared with validated readings. I don’t know what market regulation of these things is like in the UK, but I would stay away from them until you know the answer.

Reducing readmission
OL To derive the HOSPITAL readmission score, data were crunched from 117 065 adults consecutively discharged from medical wards in the US, Canada, Israel, and Switzerland. The investigators feel pleased with the result: “The HOSPITAL score identified patients at high risk of 30-day potentially avoidable readmission with moderately high discrimination and excellent calibration when applied to a large international multicenter cohort of medical patients.” But the area under the ROC curve (illustrated in Fig 2—the study is open access), otherwise known as the c-statistic, is 0.72. Put crudely, that’s 44% better than tossing a coin. And despite the “international” range of this study, I struggle to know whether it has any meaning for NHS medical patients.

Avoidable readmission
OL Did you notice the words “potentially avoidable” in the paper above? A second paper examines this concept in relation to patients readmitted within 30 days of discharge to 12 US academic medical centres between 1 April 2012, and 31 March 2013. The paper has enough figures in it to qualify for publication in The BMJ, but at its heart lie qualitative judgments based on surveys of patients and physicians, reviews of documentation, and two-physician case review to determine preventability of and factors contributing to readmission. In other words, people talking to people, reading stuff, and pairing opinions to reach subjective judgments and ball-park estimates. This is meant as a compliment, not a criticism. We need more of this kind of thing, instead of meaninglessly precise measurements of epiphenomena, such as readmission, which ignore context and personal meaning to patients and clinicians. And the conclusion makes obvious sense: “Approximately one quarter of readmissions are potentially preventable when assessed using multiple perspectives. High priority areas for improvement efforts include improved communication among health care teams and between health care professionals and patients, greater attention to patients’ readiness for discharge, enhanced disease monitoring, and better support for patient self-management.”
Note that this does not apply to 75% of readmissions. There are grave limitations to better disease monitoring and self-management for the majority of sick older patients. The junior doctors’ strike is not about money but about despair. How can we even begin to tackle the real issues of care within the NHS while the same inadequate workforce is being spread across more hours of work; social service support is being slashed; and primary care is falling apart through imposed priorities, which privilege the well at the expense of the needy? What future do our young professionals have under the lash of bullies who couldn’t care less about the real needs of complex sick individuals, people who will never make their friends rich by popping into treatment centres for day-case procedures? It reminds me of Moses and the Pharaoh who wouldn’t give the slave- children of Israel enough straw to bind mud to make bricks. “Let my people go!” It is time to reshape the political oversight of the NHS. At the current rate, there will be nothing left for our medical patients to get readmitted to.

Lancet 12 Mar 2016 Vol 387
NSTEMI in the very old
1057 If my present levels of fatalism persist, I shan’t very much care if I live or die after the age of 80. On the other hand, I wouldn’t like to inconvenience my loved ones by having a disabling stroke or repeated heart attacks. So this Norwegian trial gives me some useful information to use if I happen to get a non ST elevation myocardial infarction at or beyond the age of 80. It makes no difference to my likelihood of death over a year and a half whether I have immediate invasive treatment or just medical treatment. But if I go for the latter, there’s a much greater chance of having to go back for an emergency procedure, and it gives me a 48% greater risk of another MI and a possibly higher risk of stroke. So I’d probably opt for PCI.

The BMJ 12 Mar 2016 Vol 352
Height matters
It’s odd what gets into medical journals. It’s undoubtedly interesting to know that “height and BMI play an important partial role in determining several aspects of a person’s socioeconomic status, especially women’s BMI for income and deprivation and men’s height for education, income, and job class.” And it’s certainly legitimate for medical journals to try and be interesting. But to say that “These findings have important social and health implications” seems to me to be stretching it a bit, if only in the sense that stretching people to make them taller is difficult.

Non-arrhythmic floxacins
If you pool prescribing and outcome data from two countries, you can be pretty certain of reaching the highest possible level of significance, even for a negative outcome. Over many years in the whole of Denmark and Sweden, people taking fluoroquinolone antibiotics did not show any excess serious cardiac dysrrhythmia compared to people prescribed phenoxymethylpenicillin. These dogs deserve a good name.

Hypoglycaemia and quality
American medical journal titles are often plain wrong, and unfortunately The BMJ is showing signs of the same ailment. “Hypoglycemia as an indicator of good diabetes care” is an excellent analysis piece, which should be read by everyone who looks after type 2 diabetes. Kasia Lipska on its authorship, once again. But its title really should read “Hypoglycemia as an indicator of bad diabetes care.”

Travelling the US by rail this week in March: a warning from Dickens

Last Sunday evening I left the Falls of Niagara for this and two intervening places. As there was a great thaw, and the melted snow was swelling all the rivers, the whole
country for three hundred miles was flooded. On the Tuesday afternoon (I had read on the Monday) the train gave in, as under circumstances utterly hopeless, and stopped at a place called Utica; the greater part of which was under water, while the high and dry part could produce nothing particular to eat. Here, some of the wretched passengers passed the night in the train, while others stormed the hotel. I was fortunate enough to get a bed-room, and garnished it with an enormous jug of gin-punch; over which I and the manager played a doubledummy rubber. At six in the morning we were knocked up: “to come aboard and try it.” At half-past six we were knocked up again with the tidings “that it was of no use coming aboard or trying it.” At eight all the bells in the town were set agoing, to summon us to “come aboard ” instantly. And so we started, through the water, at four or five miles an hour, seeing nothing but drowned farms, barns adrift like Noah’s arks, deserted villages, broken bridges, and all manner of ruin. I was to read at Albany that night, and all the tickets were sold. A very active superintendent of works assured me that if I could be “got along,” he was the man to get me along: and that if I couldn’t be got along, I might conclude that it couldn’t possibly be fixed. He then turned on a hundred men in seven-league boots, who went ahead of the train, each armed with a long pole and pushing the blocks of ice away. Following this cavalcade, we got to land at last, and arrived in time for me to read the Carol and Trial triumphantly. My people (I had five of the staff with me) turned to at their work with a will, and did a day’s labour in a couple of hours. If we had not come in as we did, I should have lost £350, and Albany would have gone distracted. You may conceive what the flood was, when I hint at the two most notable incidents of our journey :—1, We took the passengers out of two trains, who had been in the water, immovable all night and all the previous day. 2, We released a large quantity of sheep and cattle from trucks that had been in the water I don’t know how long, but so long that the creatures in them had begun to eat each other, and presented a most horrible spectacle.

Charles Dickens, letter to John Forster, mid-March 1868.