Richard Lehman’s journal review – 26 September 2011

Richard LehmanJAMA  21 Sep 2011  Vol 306
1205   I don’t know why spammers have me down as so interested in imitation Rolex watches and erectile function: neither is particularly true. But I guess that if I had a diagnosis of localised prostate cancer, I might want to know the chances of each treatment option causing me to become impotent, and this is what this study does. We now know a lot about this patient reported outcome measure through careful follow-up of patients with well categorised characteristics from 9 academic centres in the USA, after they received various kinds of prostatectomy, external beam radiotherapy or brachytherapy between 2003 and 2006. The accompanying editorial rightly hails this as a good example of patient-centred outcomes research, and also uses the paper to illustrate some of the methodological problems of this kind of research. The big omission in this study, for example, was a group managed by watchful waiting. To measure the effect of an intervention on an outcome, you also have to measure the untreated progress of the disease: a principle first set forth by James Jurin, Secretary of the Royal Society of London, in 1722. When he called in witnesses from Yorkshire and New England to present their statistics on the mortality of smallpox with and without prior variolation, before himself and the president, Sir Isaac Newton, medical outcomes research was born.

1215   By the banks of the Arno, where Dante first caught sight of Beatrice, there now walk men and women with distracted frowns, oblivious to the beauty around them. These are the Florentine researchers who have produced the latest paper to show that high residual platelet activity after clopidogrel predicts increased risk of further cardiovascular events. Not by much, though, and there is not the slightest evidence that one can do anything about it, though they tried increasing the clopidogrel dose in these patients. I long ago gave up trying to follow the detail of this debate: it is a Dark Wood in which the true way is easily lost. If you are determined to press on, however, an editorial by Dominick J Angiolillo provides a very good summary of the evidence and the issues around clopidogrel and platelet function testing (which is a futile exercise). Let Dominick be your Virgil as you explore these Infernal regions.

NEJM  22 Sep 2011  Vol 365
1079   Idiopathic pulmonary fibrosis is a horrible condition: sooner or later it is bound to kill you by depriving you of oxygen, but you cannot tell when. My last patient with IPF took to his bed for two years before he died. So full marks to Boehringer Ingelheim for trying to develop a tyrosine kinase inhibitor  (BIBF 1120) which might improve the outlook of this condition. Unfortunately it doesn’t, so far as we can tell from this trial, which seems to have been mostly about dose-finding. There was a tendency to improvement in some indices of lung function at the optimal dose, but this never reached statistical significance and patients felt worse than on placebo. Boehringer Ingelheim “provided writing and editorial support” through an agency for the writing of this paper, and the conclusion of the abstract reads:
“In patients with idiopathic pulmonary fibrosis, BIBF 1120 at a dose of 150 mg twice daily, as compared with placebo, was associated with a trend toward a reduction in the decline in lung function, with fewer acute exacerbations and preserved quality of life.”
I hope the FDA and EMA take a good hard look at this drug before anyone starts trying to market it as a gleam of hope for patients with this nasty disease.

1088   Somebody needs to write a book about the future of medicine along the lines of Galileo’s Dialogue concerning the two chief world systems of 1632. In one system, the world of medicine is centred on the laboratory, which produces new drugs and genomic tests, and the patient revolves in distant orbit around this blazing source of golden light. This conception is variously known as genomic medicine, translational medicine, or even, by an Orwellian twist, personalised medicine. It generates vast revenues for industry and ploughs some of these back into academic centres: it produces studies like this one of lebrikizumab, a monoclonal antibody which only works (and then not much) in adults with asthma who have a particular genomic signature. It is a world view celebrated in the accompanying editorial, “Moving closer to personalized medicine,” and no doubt also by the various employees of Roche and Genentech like the one who “reports that Roche/Genentech is developing a periostin assay as a potential companion diagnostic for lebrikizumab.” The contrasting world system holds that medicine revolves around the patient: it is variously called user-driven healthcare or patient-centred medicine or outcomes-based medicine. It starts and finishes with the experiences of individual patients as participants in their own care, seeking all the evidence that can contribute to shared decision-making aimed at achieving the patient’s own preferred outcomes. The Vatican of the medical research establishment smiles indulgently at such fancies: yes, my child, such things await us in Heaven if we are obedient in this fallen world; come, do not dabble in heresy, or we will have to show you the instruments of the Inquisition: you may find it hard to get funded, to get published, to get promotion; perhaps we may try a little vexatious litigation if you become troublesome.

Lancet  24 Sep 2011  Vol 378
The world is teeming with medical researchers whose future depends on getting published in a high impact journal: a stupid, unjust, archaic system, which cannot end soon enough. This week, a throng of petitioners go empty-handed as The Lancet strides through them, like Cardinal Wolsey holding an orange studded with cloves to his nose: such commoners can wait, this week we will turn our gaze on the great world and those high themes on which it pleases us best to pontificate.
So, if you want to learn even more about Japan and its health services, longevity etc than you did last week, here is your chance. Yet another survey tells us how far we are still short of reaching Millennium Goals 4 and 5 on maternal and child mortality.

1166   The only research paper in this issue is a descriptive case-series of 12 patients with severe haemolytic-uraemic syndrome and neurological symptoms due to Shiga-toxin producing E coli in the outbreak which killed 50 North European adults last year. They were treated with immunoadsorption and with the complement blocking antibody eculizumab. Ten were left with no neurological damage.

BMJ  24 Sep 2011  Vol 343
It is heartening to see the return of more useful clinical research to the BMJ, like this Belgian trial to determine the effect of manual lymph drainage in addition to written guidance and exercise therapy on lymphoedema following breast cancer with unilateral axillary lymph node dissection. Thank goodness for a nicely designed study on an important topic, reported clearly and with due modesty, showing that manual drainage makes very little difference to this outcome.

I like the next paper too, even though it is written largely by statisticians. They examine the hourly effects of ambient air pollution on myocardial infarction, describing their efforts as a “time stratified case crossover study linking clinical data from the Myocardial Ischaemia National Audit Project (MINAP) with PM10, ozone, CO, NO2, and SO2 data from the UK National Air Quality Archive.” Excellent: this is what statisticians and epidemiologists are for – to supply interesting talking points for clinicians and policy-makers. The conclusion here (which I would not presume to interrogate) is that air pollution brings forward acute coronary events which were likely to happen anyway.

Again, it is very useful to be reminded that “The total burden of premature deaths from natural causes in people with schizophrenia or bipolar disorder is substantial.” Also that “there is a need for better understanding of the reasons for the persistent and increasing gap in mortality between discharged psychiatric patients and the general population, and for continued action to target risk factors for both natural and unnatural causes of death in people with serious mental illness.” Could it be something to do with the increasingly poisonous medication we are giving them? Drugs like olanzapine, quetiapine, and risperidone call to mind Oliver Wendell Holmes’ saying that if all of them “were thrown into the bottom of the sea, it would be so much the better for mankind, and all the worse for the fishes.”

I am not quite so enamoured of this next paper, also written largely by statisticians, in which they draw on copious data from different sources to construct a Bayesian model for the fluctuations of severity in the 2009 pandemic of influenza A/H1N1 in England. The conclusion here seems counter to clinical experience: the model suggests a mild pandemic “characterised by case and infection severity ratios increasing between waves.” Maybe we all just got used to the presence of H1N1 flu in our midst, and were oblivious to these changes. And I still haven’t come across a clear summary statement on whether the arrival of H1N1 increased or decreased total influenza mortality.

An excellent editorial by Andrew Farmer and Robin Fox, a GP colleague from Bicester, calls for a better recognition in general practice of the atypical kinds of diabetes, and the importance of this is starkly illustrated in this study which shows that “the proportion of deaths caused by acute complications of diabetes has increased in patients with late onset type 1 diabetes,” in contrast to diminishing mortality in type I diabetic patients who developed the disease in childhood. I had such a patient who in her 70s twice developed acute ketoacidosis due to infection, because although she was on insulin, we mistakenly tended to regard her as a type 2 diabetic. But such patients have no reserve beta-cell function whatever, and spiral out of control with dangerous rapidity.

Ann Intern Med  20 Sep 2011  Vol 155
345    The American obsession with diagnosing streptococcal throat infection continues in this study which adds real-time biosurveillance to the Centor score. Apparently knowing the local prevalence of group A strep improves diagnostic accuracy by altering the pre-test probability of various features of the score. But it’s not diagnostic accuracy we’re looking for: what we need is a predictor of useful response to antibiotic treatment, which is not quite the same thing.

361    For as long as I have been writing these reviews, data have been accumulating about the time-dependent benefits of percutaneous intervention for myocardial infarction.  If the adage that time equals myocardium is true, then delay should predict increased heart failure as well as increased mortality. And so it proves from these data from Western Denmark, between 1999 and 2010.

389   But if system delays are still common in Denmark, a flat peninsula that you can cross in a couple of hours, are they not even more inevitable in most other countries? And given that fact, should we not be thinking in terms of an integrated dual reperfusion strategy to best serve all patients with STEMI? That’s certainly the view of this discussion piece, which deals mainly with the situation in North America. Here is an area where detailed modelling and analysis are still needed to guide policy.

Plant of the Week: Ficus carica

I fear that this is a dismal season for British fig growers: your little trees of “Brown Turkey,” carefully grown with a restricted root-run by a sunny wall, will now be bearing many a green bulge that is due to drop off when the first frosts arrive. I think that in twenty years of attempted fig-growing in the UK, I have produced the same number of barely edible fruits. We Britons persevere because of this plant’s Mediterranean and indeed biblical connections, and when we curse our barren fig trees we can plead that we are simply following the example of Our Lord.

It is much different here in the USA, I am happy to report. The food shops now abound in little plastic baskets containing a dozen or more ripe black locally grown figs, which I find hard to resist. And British fig lovers are of course perfectly able to buy nice ripe imported figs in local markets for several months of the year. I never tire of them, though my wife is less keen: but even she acknowledges the supreme merit of sliced ripe figs with Parma ham, or, best of all, interleaved with smoked duck and drizzled with a few drops of finest balsamic vinegar.