Richard Lehman’s journal review—11 August 2014

richard_lehmanNEJM 7 August 2014 Vol 371
497  A new gene for breast cancer susceptibility? The PALB2 gene locus has been known about for several years, but this study puts it firmly on the map by intensively investigating 362 members of 154 affected families. The risk for female PALB2 mutation carriers, as compared with the general population, is 35% by the age of 70—about the same as for BRCA2 mutation carriers. The editorial explains why this is so: both genes work in concert to repair double strand breaks in DNA. This is a very fundamental process, and you would have thought that any impairment to it would lead to a whole range of cancer risks, but in the case of PALB2, the risk seems to be mainly of breast cancer (in men as well as women) and Fanconi’s anaemia. Therapeutic efforts for carriers of BRCA and PALB2 mutation carriers are focussed on inhibition of PARP, causing cells that contain broken double stranded DNA to die rather than turn cancerous.

519  Variant Creutzfeldt–Jakob disease (CJD) should be epidemic in Britain by now, according to the predictions of Richard Lacey, a professor of clinical microbiology from Leeds who wrote a book about it in 1994. Instead, it is exceedingly rare, with one or two new cases detected in the UK each year.

The situation seems different in Italy, where investigators managed to collect 31 living patients with the disease. At least I think they were alive: the narrative is exceptionally hard to follow. Autopsies jostle with biopsies. Using the latest ultrasensitive, multiwell plate based fluorescence assay, involving PrPCJD-seeded polymerization of recombinant PrP into amyloid fibrils, they detected abnormal prion protein in 30 of these patients using nasal brushings. In 43 people without CJD, nasal brushings were all negative. So: we may have a useful non-invasive test for CJD. And people with CJD are probably sneezing out prions all the time, without infecting anybody.

530  And they are peeing out prions too. A different study used a similar multiplication assay to detect misfolded prion protein (PrPSc) in the urine of 13 of 14 urine samples obtained from patients with variant Creutzfeldt–Jakob disease, and in none of the 224 urine samples obtained from patients with other neurologic diseases and from healthy controls. Yet for all the prions swimming around us, the Lacey epidemic fails to materialize. This must be depressing for those in the field, stuck with so little clinical material; and yet they remain optimistic. “Evidence suggests that variant Creutzfeldt–Jakob disease prions circulate in body fluids from people in whom the disease is silently incubating . . . A recent retrospective study of archived surgically resected appendixes in the United Kingdom estimated that the prevalence of asymptomatic variant Creutzfeldt–Jakob disease infection in the U.K. population was approximately 1 case per 2000 persons; this suggests that approximately 30 000 people in the United Kingdom might be carriers of potentially infectious variant Creutzfeldt–Jakob disease prions.” Hmm. In that case, the incidence figures suggest that prions only successfully “incubate” in about one carrier in several thousand. I’m not entirely sure we really know what we are looking at or for here.

540  The United States is deeply interested in UK models of pay for performance. Ah, see how our once exemplary primary care system is falling apart, and learn. The Quality and Outcomes Framework is not the only cause, just a major part of a more general blight. Just beware of the games you force people to play. Quality comes from commitment and continuity and is a cyclical, iterative process involving everybody—especially patients. It is the opposite of plonking in some targets and expecting immediate results. Here’s a scholarly account of the Long Term Effect of Hospital Pay for Performance on Mortality in England: “Short term relative reductions in mortality for conditions linked to financial incentives in hospitals participating in a pay for performance program in England were not maintained.” How, indeed, could they be?

573  “Young Blood” is an interesting Clinical Implications of Basic Research article. The author, Alessandro Laviano, says that “several recent studies showing the therapeutic effect of the infusion of blood obtained from young mice into old mice fuel the sanguine view that new experimental approaches to the treatment of age related diseases will emerge.” A most witty use of the word sanguine, my young friend. Perhaps you would like to visit my castle in Transylvania?

JAMA 6 August 2014 Vol 312
483  “Productive, high quality, reproducible, shareable, and translatable.” That is what John Ioannidis and Muin Khoury would like all research to be: the initials make PQRST. They plead for better rankings of research using these principles. Productivity metrics need to change from simple citation figures. Quality assessment is in the eye of the individual beholder, they concede. But shareability is easier to measure: how many papers by a given author are accompanied by shareable data, materials, or protocols? Translation refers to the likelihood of the work having further scientific or clinical value. Stem the tide of academic futility using PQRST.

492  There is a widespread delusion that primary medical services should act as guardians of human behaviour: detecting aberration, and correcting it by gentle exhortation or by more direct means. I have been at it for nearly 40 years, and I can count few successes in this department. Smoking cessation, perhaps, occasionally. Opioid substitution, if that counts. Alcohol reduction, almost never. But it has become an article of faith that we must seek out such behaviours. This week’s JAMA looks at two trials of a brief intervention for “problem drug use.” This one in Washington State recruited 868 individuals with self-confessed illegal or inappropriate drug use after sending a screening questionnaire to 39 062 people. The intervention had no discernible effect at all.

502  It was the same in the ASPIRE trial, which you can read in full for free. This one did the screening in primary care, as well as the intervention. “These results do not support widespread implementation of illicit drug use and prescription drug misuse screening and brief intervention.” If only public health experts and policy wonks would do these futile things for themselves and learn from their mistakes. That might free primary care professionals to do what they are good at, which is looking after acutely sick people and forming long term supportive relationships with those who are living with long term conditions.

535  I winced a bit when I saw the title. Does This Man With Lower Urinary Tract Symptoms Have Bladder Outlet Obstruction? The Rational Clinical Examination: A Systematic Review. When I was a urology SHO at St Thomas’ in 1976, the only physical examination I did was palpation for the bladder followed by a rectal examination. I think this was rational, although it did not tell me whether the man needed a prostate operation. That depended on his symptoms, which were likely to fluctuate anyway. But in those days we suffered from the delusion that if we didn’t operate on big prostates, their owners would inevitably suffer acute retention or renal damage. We persuaded innumerable night pee-ers to have unnecessary transurethral prostatectomies, and ignored the fact that many of them were unhappy with the results in terms of sexual function, retrograde ejaculation, and the rest. Jack Wennberg eventually sorted all this out at Dartmouth in the 1990s. Most men who are given the facts don’t want surgery. And whether they have “bladder outlet obstruction” really doesn’t matter very much. If you need to know, the rational answer is of course to do an ultrasound scan, not a physical examination.

Lancet 9 August 2014 Vol 384
OL  Awaiting appearance in the printed Lancet is a re-analysis of data from the ACCORD trial of intensive glucose lowering in high risk type 2 diabetes, showing that those who attained the target reduction of blood glucose showed a reduction in cardiovascular events. Uh? Wasn’t the ACCORD trial stopped early because of an excess of cardiovascular deaths in the intensive treatment group? Why yes, but this is not enough to deter the authors of this paper from contesting the evidence. Those deaths, they argue, occurred in those whose blood sugars did not respond to the additional treatment. People who attained the desired level of HbA1c actually had fewer myocardial infarctions (by a just significant 15%), they observe, adding in observational data from after the study had been curtailed. So intensive treatment, they argue, works for those whose blood sugar falls. Yes, I can accept that may be true, but that doesn’t alter the main message of ACCORD, which is that for each one who benefits there is another who is harmed. We know that, observationally, blood sugar is a continuous risk factor for myocardial infarction: what we still don’t know is how best to reduce this risk in individuals. This analysis certainly does not reverse the message of ACCORD, but suggests a hypothesis that needs a new trial in clinical practice.

OL  General practice is the art of doing medicine in 10 minute chunks. A general practitioner’s job is to remain able, amiable, and available. I always thought that telephone triage, especially when done by non-doctors, was a waste of time and potentially dangerous. The latest and biggest trial confirms that placing this barrier to patients wanting to see a doctor the same day actually increases their use of services. And working, as I do, mainly out of hours, I can tell you that it decreases their respect for their GP providers, increases their anxiety and frustration, and leads to inappropriate use of emergency services. General practice may be facing catastrophic overload, but telephone triage only worsens the situation.

OL  I have followed the saga of thrombolysis for stroke from its beginnings in the early 1990s, and I remain deeply ambivalent. Doctors typically want less treatment for themselves than they inflict on their patients. If I am severely hemiplegic, or have difficulty swallowing, or impairment of consciousness, do not give me thrombolysis. Do not give me fluids or food. Do not give me antibiotics. Take it that I want to die, and help me to die reasonably quickly and reasonably well. I know there is a small chance that I might recover and live some kind of life, but I would rather not take it. The thrombolysers, however, ignore all this—or rather downplay it. There is an “average absolute increase in disability free survival of about 10% for patients treated within 3.0 h and about 5% for patients treated after 3.0 h, up to 4.5 h.” These figures come from a meta-analysis of individual patient data from 6756 patients in nine randomised trials. “The proportional benefits were similar for patients aged older than 80 years compared with younger patients, and for patients with minor or severe strokes compared with other patients.” But I don’t really understand how you can apply these figures to individuals with their own personal, evolving strokes, let alone share the decision making in real time.

The BMJ 9 August 2014 Vol 349
“Albert Calmette, a French physician and bacteriologist, and his assistant and later colleague, Camille Guérin, a veterinarian, were working at the Institut Pasteur de Lille (Lille, France) in 1908. Their work included subculturing virulent strains of the tubercle bacillus and testing different culture media. They noted a glycerin-bile-potato mixture grew bacilli that seemed less virulent, and changed the course of their research to see if repeated subculturing would produce a strain that was attenuated enough to be considered for use as a vaccine . . . The BCG vaccine was first used in humans in 1921.” I love Wikipedia. I even went to a session of the Wikimania conference last Friday. It was great. But I need to tell you about this latest meta-analysis of BCG for protecting kids. It works. It stops kids getting TB infection and it stops TB infection progressing. Calmette and Guérin, we salute you! Ninety three years on, we have still not improved on the product of your patient efforts.

The Nottingham department of primary care is famous for producing risk scoring instruments based on UK data. General practitioners in the UK are famous for not using them, and for leaving general practice in droves because they can no longer cope with the number of tasks they have been set. Here is a lovely new QBleed score from Julia Hippisley-Cox and Carol Coupland to predict the risk of upper GI bleeds and intracranial haemorrhages in people taking anticoagulants. But how do we actually get GPs to share decision making with patients using this score? Only by embedding it in one keystroke decision aids that pop up with each prescription. And perhaps, alas, not even then, if the surgery is running 45 minutes late.

The Clinical Review this week is about the management of spasticity in adults. It deserves to be read widely, but it will not be. It is too long. It is behind a paywall. Where would a review like this do most good? Why, in Wikipedia of course. Then it would come up top if you did a Google search on “spasticity.” Patients, carers, and health professionals would all read from the same reliable source. But it can’t happen because there is no system for making authoritative, up to date clinical reviews freely available to those who most need them. Somebody needs to do something about this. James Heilman and I want to hear from you. Emails: and

Plant of the Week: Ceratostigma willmottianum

Almost by accident, we have created a section of front garden dominated by lovely blues in August. This little shrub has abundant little flowers of the sharpest, brightest, cleanest blue imaginable. It consists of a bundle of sticks, which come to life at the end of the growing season, and it gives pleasure for several weeks.

As its flowers go over, its leaves turn a startling red. Once they drop off in November, cut the whole thing to the ground.