Richard Lehman’s journal review—3 November 2014

richard_lehmanNEJM 30 October 2014 Vol 371
1685 The treatment of childhood and adolescent cancer is territory that most of us don’t trespass on, but we’ll need to go there this week just to have something to read about. JAMA is taking a Halloween break and the others seem to be going into an end of year lull. Anyway, here’s a less is more type study of how much cord blood you need for stem cell transplantation in haematological malignancies. This cord blood is precious stuff, and its use is limited by the finite number of haematopoietic progenitor cells that can be collected from a placenta. So it’s good to know that one unit of it may be better than two: “Survival rates were similar after single unit and double unit cord blood transplantation; however, a single unit cord blood transplant was associated with better platelet recovery and a lower risk of graft versus host disease.” In fact, the Kaplan-Meier chart shows that 35% of children in the single unit group and 40% of children in the double unit group died within two years: these are still very nasty diseases.

OL And now—although I have so far tried not to—we might as well start looking at Ebola virus. Mr Murdoch’s newspaper, The Times, has taken to calling it ebola without an initial capital. This, like most of The Times, is a mistake. The current epidemic is owing to an ebolavirus called Ebola, or EBOV, Zaire species. I hope I have made this clear to you now, Mr Murdoch. As for the basic facts about the infection, I have read little that makes sense in the lay press, Murdoch’s or otherwise, so I would direct your attention to an article which has just appeared on the NEJM website, giving a detailed account of the clinical features of patients admitted with Ebola to Kenema Government Hospital in Sierra Leone since May this year. Unfortunately, it is behind a paywall, and if anyone from the NEJM is reading this, I’d urge them to put this right. Don’t expect to be other than appalled, even though these people were in hospital receiving nursing care and intravenous fluids. The incubation period was estimated to be six to 12 days, and the case fatality rate was 74%. If you get Ebola over the age of 45, the case fatality rate is 94%. The symptoms are surprisingly non-specific, with fever in 89% of the patients, headache in 80%, weakness in 66%, dizziness in 60%, diarrhoea in 51%, abdominal pain in 40%, vomiting in 34%, and bleeding in only one patient out of 87. There is a mismatch between figure 2 and its legend. The figure seems to show that every patient who experienced vomiting, confusion, or oedema died, whereas the text says, “The only symptoms that were significantly associated with a fatal outcome were weakness, dizziness, and diarrhoea.” Whatever your symptoms, it’s probably best to lose no time in telling all the relevant people that you love them, others (or the same) to forgive you, and try and get hold of a few bottles of champagne to go out like Chekhov. The average time from onset of symptoms to death was 9.8 days.

JAMA Intern Med October 2014
OL Metformin is the politically correct first treatment for “type 2 diabetes,” although there is only indirect evidence that it is any better than the others. To the existing collection of observational data, we can now add: “LESS IS MORE: Initial Choice of Oral Glucose-Lowering Medication for Diabetes Mellitus: A Patient-Centered Comparative Effectiveness Study.” That sounds like a great study to set up: offer people with newly discovered hyperglycaemia a choice between various drug treatments or none, and then observe the outcomes. Unfortunately, this would have to take a decade or more for collection of long term vascular endpoints. Instead, here we have a retrospective cohort study of a large US health insurance database, and the endpoints are time to addition of a second oral agent or insulin, each component separately, hyperglycaemia, other diabetes related emergency department visits, and cardiovascular events over a prescribing period of four years. The 57.8% of patients who started off with metformin showed reduced subsequent treatment intensification, without differences in rates of hyperglycaemia or other adverse clinical events. That’s a little useful knowledge, but we need much more.

Lancet 1 November 2014 Vol 384
1577 Less is more is becoming quite a theme this week. Can you remember what thienopyridine is? Basically, it’s a grel. A platelet inhibitor that is not aspirin. In the ARTIC-Interruption trial it meant clopidogrel or prasugrel, given after coronary stenting. Now we know that when the magic 12 months have elapsed after the placement of drug eluting stents, patients are advised that they can stop their grel and carry on taking aspirin. This French trial randomised some patients to do this, and some to carry on with their clopidogrel or prasugrel for a further six to 18 months. But they have shown that this does not prevent cardiovascular events while it does increase major bleeding. So current practice seems right. A year is the holy grel.

1586 Para-acetylaminophenol is a bit of a mouthful. How you abbreviate it depends on where you live. In the United States, Canada, and Japan, it is called acetaminophen; everywhere else it is called paracetamol. It is the first thing we tell people to take when they have any sort of pain, including low back pain, but randomised controlled trials of paracetamol are a rarity. So good on the Ozzies, who designed this trial to see if paracetamol did anything to shorten the duration of acute back pain. I think I may have mentioned this previously when it first appeared on the website. Paracetamol, of course, does not shorten the duration of recovery from back pain. If anyone discovers something that does, please let me know. I will patent it, and become famed as a billionaire philanthropist from my humble palace in Venice.

1607 Who but the Lancet could commission an article with a title as vague and bombastic as “Culture and Health?” Here is a paragraph from its executive summary:
“We believe, therefore, that the perceived distinction between the objectivity of science and the subjectivity of culture is itself a social fact (a common perception). We attribute the absence of awareness of the cultural dimensions of scientific practice to this distinction, especially for macrocultures and large societies, which define only small-scale, microcultures as cultural. We recommend a broad view of culture that embraces not only social systems of belief as cultural, but also presumptions of objectivity that permeate views of local and global health, health care, and healthcare delivery.” Executives, take note. I’m afraid I couldn’t stay the course to find out where this hot air balloon eventually came to rest.

OL Let’s turn to a superb study by Oxford colleagues, who have also proved themselves outstandingly helpful and compassionate clinicians over the many years I have known them. In treating patients with subarachnoid haemorrhage owing to ruptured intracranial aneurysm, there has long been uncertainty about the comparative long term results of endovascular coiling or neurosurgical clipping. With this 18 year follow-up of the UK cohort of the International Subarachnoid Aneurysm Trial (ISAT), the scales tip slightly in favour of the less invasive option: “The probability of death or dependency was significantly greater in the neurosurgical group than in the endovascular group. Rebleeding was more likely after endovascular coiling than after neurosurgical clipping, but the risk was small and the probability of disability free survival was significantly greater in the endovascular group than in the neurosurgical group at 10 years.”

The BMJ 1 November 2014 Vol 349
Can you remember what a thienopyridine is? All together now: “A thienopyridine is a GREL!” Very good, children. Now in the land of grels, boys and girls, you can make a lot of money by inventing a new grel and selling it to lots of people. This is known as economic activity, and that is why we protect our pharmaceutical companies like dear little lambs. Now you remember that grels stop platelets sticking to each other, so they stop blood clots forming. That’s why they have been sold for people to use when they are having a kind of clot in the heart called non-ST elevation myocardial infarction. The trouble is that when you look at all the trials, you find that these grels don’t do any good but cause more of these people to bleed. So you can see, girls and boys, that it’s very important to have all the results of all the trials, because most of them are conducted by people wanting to sell grels. We don’t think they might want to tell fibs, but we need to be sure.

In our local prescribing guidelines, we are told to prescribe nitrofurantoin or trimethoprim as first line treatments for urinary tract infections. This is to reduce the risk of Clostridium difficile, especially in frail older patients in nursing homes. Every day, someone from a nursing home rings us up to say Mrs P or Z has become more confused and they’ve got three plusses of “nitrates” (sic) on their urine dipstick. Wearily, you write a prescription for trimethoprim, fully aware that (a) 50% of nursing home residents have dipstick positive urines and (b) the trimethoprim probably won’t do anything anyway. But the trimethoprim does do something. Five days later, Mrs P is dead. You fill out her death certificate with words like dementia and frailty. Trimethoprim at a dose of 400mg daily blocks the epithelial sodium channel (ENaC) in the distal nephron, impairing renal potassium elimination. And Mrs P, aged 87, was already taking losartan 50mg for high blood pressure that somebody had detected a few years previously. She probably died of hyperkalaemia induced ventricular fibrillation. An excellently conducted Canadian population study looks at the incidence of sudden death after prescriptions of common antibiotics. Over there, trimethoprim still comes routinely combined with sulfamethoxazole. Compared with amoxicillin in patients over 65 and taking an ACE inhibitor or ARB, co-trimoxazole is associated with a 50-80% higher risk of sudden death over the next 14 days.

Plant of the Week: Rosa “Absolutely Fabulous”

I will keep this short, because I have praised this bush rose before. The reason I’m doing so now is because I can see it from my study window, and it is still covered in healthy dark foliage and carries seven new flowers. In the warmth of this early November, they still carry a full scent of fruit salad and liquorice. I do hope someone breeds a climbing form.

You may remember me telling you that when this rose was first bred in the US, it was named for the great cookery writer Julia Childs. It acquired its British name when the grower who imported it saw it in flower, because of his spontaneous exclamation, not the television programme. I repeat this story in case you may have grown weary of Joanna Lumley on the television: you are unlikely ever to grow weary of Absolutely Fabulous in your garden.