Richard Lehman’s journal review—22 October 2012

Richard LehmanJAMA  17 Oct 2012  Vol 308
1545   “Between 1988 and 2010, favorable trends in lipid levels have occurred among adults in the United States.” That may seem pretty amazing, but there is a lot we don’t understand about these things. Remember that cardiovascular disease is also falling steeply, even as the population gets more obese. The authors of the NHaNES analysis speculate that the cause of the fall in total cholesterol and LDL-cholesterol may be linked with a fall in sugar consumption and smoking. Greater use of lipid-lowering drugs has contributed very little to the overall decline.

NEJM  18 Nov 2012  Vol 367
1487   Some clinical issues are sufficiently important to warrant several randomized controlled trials. One such is whether patients with leukaemia or myelodysplasia who need to have “bone marrow” transplantation benefit more from bone marrow or peripheral blood stem cells. Like previous studies, this one shows identical survival, with a small trade-off of harms and benefits. There were remarkable few differences but there was a trend to better graft survival with peripheral blood cells, but a possibly higher risk of graft-versus-host disease.

1497   If you look after patients with Alzheimer’s disease—many of them in nursing homes—you may find that following an episode of agitation or psychosis, someone (yourself perhaps, or a psychiatrist) has put them on risperidone and forgotten to take them off it. Is this bad practice? Possibly: these “atypical” antipsychotics have a reputation for numerous harms. But this study of discontinuation of risperidone shows that the majority of such patients relapse after being taken off it.

1508   Linezolid for most of us is a small-print drug we have barely heard of, but it has good antimycobacterial properties and is increasingly used as a treatment for extensively drug resistant tuberculosis. Hopefully this will soon be a thing of the past as new agents of greater promise appear. This small trial from South Korea shows that linezolid usually works, but it is a nasty toxic drug which causes peripheral neuropathy in most patients and myelosuppression and optic neuritis in a significant number.

1519   Speaking of nasty toxic drugs, do you remember ustekinumab? Me neither: I can’t keep up with these monoclonal antibodies. Jeff Aronson will patiently take you through the logic of their nomenclature, but I hope he agrees that it was a bad idea to lumber a whole class of diverse drugs with such silly names. I get in a tizz with mabs, and when I see yet another I am tempted to mutter something like bugrofumab. Anyway, ustekinumab is targeted against the inflammatory modulators interleukin -12 and -23; in this trial it was compared with placebo for adults with Crohn’s disease which had failed to respond to blockers of tumour necrosis factor. At six weeks, it produced some response in a significant number of patients. What happens when this drug is continued for longer in such patients is not known, but from other trials, we may expect an increase in infections and cancers. It would be interesting to know if the NEJM is selling reprints of this paper to Jannsen, who ran this trial: it would clearly be unethical to promote the use of this drug on the basis of this short term study, given high possibility of harm from the drug in the medium term. Surely the NEJM would not hide behind the argument that it has no responsibility for the way reprints of its articles are used?

1562   The most popular paper on the NEJM website for the last week has deservedly been this one on chocolate consumption, cognitive function, and nobel laureates. It’s great that this solemn journal could show its sense of humour, but not so great that it’s behind a paywall. It’s like an exceptionally good Christmas BMJ paper, and shows an incontestable correlation between per capita chocolate consumption and the award of Nobel prizes, with Switzerland at the top. The author Franz Messerli denies any conflict of interest, though his name sounds suspiciously Swiss and he admits to eating a lot of Lindt chocolate. I think Lindt should award him a lifetime’s supply, to accompany his upcoming IgNobel Prize.

Lancet  20 Oct 2012  Vol 380
1387   Here is a truly obscure phase 2 trial of a drug which doesn’t yet have a name in a group of patients hitherto largely ignored by the cardiology community, especially in the UK—people with heart failure and preserved systolic ejection fraction. The drug is classed as an angiotensin receptor neprilysin inhibitor (ARNI)—I won’t try and explain but you can think of it as an angiotensin receptor blocker with bells on. The comparator was an ordinary ARB (valsartan), and the end point was N-terminal proBNP at 12 weeks. So far, so bad: what is this trial doing in The Lancet, and why am I bothering to tell you about it? I don’t know the Lancet’s reason, but mine is that I want to go on about heart failure and BNP yet again. This hormone is an excellent marker for ventricular strain but it is very variable physiologically and not a reliable way to monitor treatment in individuals. So of course this trial is worthless in itself and needs to be followed by one that has hard end-points. BNP is raised equally in all patients with heart failure, whether or not they have reduced ejection fraction, and is an excellent predictor of death. It is high time that we used highly elevated BNP despite optimal treatment as an indicator that patients should receive palliative and supportive care. As for these new ARNI drugs and their place in the great scheme of things, you can learn more from the editorial by John Cleland and Andrew Clark, which usefully challenges the primacy of systolic ejection fraction as a measurement in heart failure.

1396  Blimey. I thought that the Stent Wars were over, and that all sensible earthlings long since clambered into their starships and returned to the home planet. Not so, apparently: far out there, sirolimus and zotarolimus are busy eluting in competition with another, and their light beams are now reaching Earth. It’s an exciting draw: there is no difference in stent thrombosis: but the plaintive cry of the Medtronic-funded investigators is that “Time analysis suggests a difference in definite or probable stent thrombosis between groups is emerging over time, and a longer follow-up is therefore needed given the clinical relevance of stent thrombosis.” What, will the Stent Wars stretch out to the crack of doom?

BMJ  20 Oct 2012  Vol 345
“How to avoid precipitating an acute adrenal crisis: most importantly, heed patients’ requests for hydrocortisone.” The title of this editorial is your take-home message for the week: somebody’s life may depend on it. I was really glad to see this piece by John Wass and a colleague appear in the BMJ, two years after he first approached me with case histories of people with Addison’s who had nearly died when junior hospital staff and nurses ignored their pleas for prompt steroid treatment during acute illness. This editorial lays out the risk situations very clearly, and you may remember that Addison’s also featured early in the Easily Missed series. Above all, don’t be a prat: believe the patient.

It’s strange how obscure Greek words invade medicine at every level: if you try to stop premature labour with drugs you are said to be performing tocolysis. Is it a good idea to do this? Nobody knows. It is entirely possible that we may be doing more harm to the fetus by delaying its exit than by letting it come out, however prematurely. In my day on the obstetric wards, we used intravenous salbutamol or alcohol; nowadays the most effective tocolytics, according to this systematic review and network meta-analysis, are calcium channel blockers and prostaglandin inhibitors.

Miscarriage is a common and distressing event, and I always feel bad about the fact that we don’t locally have the means to diagnose it definitively except between 9 and 5 on weekdays—just the times when I never do any doctoring these days. This systematic review, led by a Dutch medical student, shows that a single progesterone measurement for women in early pregnancy presenting with bleeding or pain and inconclusive ultrasound assessments can rule out a viable pregnancy.

I have to say that among the many career paths in medicine that opened up before me long ago, proctology was never one that captured my fancy. I still have difficulty working out the precise course of all the possible anal fistulas that the proctologist might have to lay open; and I squirm at the butchery which must have been the rule before MRI imaging. There are some nice pictures of tracks in this excellent clinical review—diagrams rather, you may be relieved to hear. But the best ones are in a medieval treatise by John of Arderne.

Ann Intern Med  16 Oct 2012  Vol 157
542    Ever since I worked alongside two lovely young Indian doctors at Yale for a year, I have puzzled about how best we might help their huge and miraculous country to achieve a humane, comprehensive health service. The stories they told me were of a few inspirational idealist teachers stranded in a desert of paternalism, corruption, and wilful ignorance; a place where no capable doctor would consider a career in primary care, where all continuing education is provided by drug companies, and where most payments for medical care are out-of-pocket and likely to be catastrophic. This paper from India is a report of reality on the ground. Few patients can afford the cost of a new implantable cardioverter-defibrillator, so it seems criminal to throw these devices away after patients have died if there is a chance that they might still work for several years. So one cardiac centre in Mumbai decided to remove ICDs from dead patients, and if they still had at least three years’ battery life remaining, sterilize them and offer them for reuse by patients who had class 1 indications for ICD use but could not afford a new device. They worked very well for their remaining lifetime and there were no adverse effects.

571   Last year, the results of the National Lung Screening Trial were published and showed that in high-risk subjects, three annual rounds of CT screening could reduce all-cause mortality by 7% over 6 years. “When the average applies to nobody” is the neat title of this analysis which looks at the number needed to screen to avoid one lung cancer death for various of the subgroups involved. It varies from 82 to 1236 within the selection criteria of the NLS, according to age and duration of smoking. If this screening is ever introduced on a wide scale, it will be very important to ensure that each individual will be properly informed about the tiny likelihood of personal benefit and the possibility of anxiety or harm. The lessons of mammography need to be translated into an entirely different approach to “selling” screening.

Plant of the Week: Liquidambar styraciflua

This isn’t the kind of tree that most of us can afford garden space for, but we should be grateful to others who plant them for our enjoyment. My thanks here go to Messrs J Sainsbury and Co for their generous planting of liquidambars in the car park of their Banbury outlet. Between marvelling at the illogic of their traffic flow system and hurling oaths at various all-wheel-drive vehicles parked across two spaces, it is a joy to admire the low sun shining through leaves of burning yellow, red, and orange.

This is a good preparation for an imminent trip to New England. They don’t have many liquidambars there, but boy do they have some good fall colors. I will have a certain amount of work to get done, but I also intend to do a bit of leaf-peeping and fungus hunting.