290    Bless you! Sneezing can be dangerous, if you are one of the 25% of people who go around with apatent foramen ovale, which is a little flap between the two atria. Then if you sneeze, or go diving, or fly a fighter jet, your right atrial pressure will exceed your left atrial pressure, and venous blood will pass into your arterial circulation, just as it did when you were in the womb. Venous blood can carry little clots, or fat globules, or other gunk, which the lungs can filter easily and usually harmlessly: but not so the brain. Various studies found that people with severe migraine, strokes and multi-infarct dementia had more PFOs than the general population, so some cardiac surgeons in the USA have taken to closing them whenever they discover them during cardiothoracic surgery. This study painstakingly examined outcomes in 13,092 patients who had transoesophageal echocardiography in Cleveland, Ohio between 1995 and 2006. Intraoperative PFO closure was performed in 2277 of them. But there was no benefit and there is a definite trend to harm, ironically in a higher rate of postoperative stroke. Bless you.  http://jama.ama-assn.org/cgi/content/abstract/302/3/290

298    A year ago, I promised a patient with advanced ovarian cancer that I would do a literature search on the relationship between hormone replacement therapy and ovarian cancer. I found that the studies up to that time were inconclusive, though there was a trend towards higher risk, possibly around 30%. I ran this past the top HRT expert in Oxford but she was not able to quantify it either. I think we now can, thanks to this whole-population study from Denmark: it is indeed around 30% for all types of HRT and all types of ovarian cancer. The confidence intervals are very tight, as the study covered 7.3 million women years. Sadly the patient in question died a few months later, but I think we owe it to her to put this information in our HRT advice sheets.  http://jama.ama-assn.org/cgi/content/abstract/302/3/298

NEJM  16 July 2009  Vol 361

235    What is the best way to harvest veins for coronary bypass grafting? Some Oxford purists would scoff at this question and insist that grafts should be arterial whenever possible. Other European cardiac surgeons have taken to removing the saphenous vein through an endoscope. This leaves the patient with a neater looking leg than open vein harvesting and the leg also heals faster. But the veins themselves block more frequently and at three years there is a definite mortality disadvantage to endoscopic harvesting, according to this retrospective, non-randomised study.http://content.nejm.org/cgi/content/abstract/361/3/235

245    If you carry the APOE ε4 genotype, your cognitive function will decline from before the age of 60 and you are highly likely to get Alzheimer’s disease (see the following study, p. 255). So if you have a parent with Alzheimer’s, do you really want to know your APOE status? This study attempts to ascertain the psychological effect of offering APOE genotyping to adults in this situation, and shows that in the short term there is little harm (and of course relief among those without the ε4 allele). The only adverse outcome was in people who were anxious to start with, who became more anxious if they had the bad gene. But I bet that includes more and more people as they approach 60.  http://content.nejm.org/cgi/content/abstract/361/3/245
http://content.nejm.org/cgi/content/abstract/361/3/255

264    A few weeks ago I had occasion to touch on the joys of colonoscopy, a procedure much dreaded by most of its victims, though the bowel preparation is generally the worst bit. “They don’t like it up ‘em, sir, they don’t like it at all,” as Corporal Jones once said, albeit of a somewhat sharper instrument. But what if a miracle of modern technology allowed visualisation of the colon merely by swallowing an electroniccapsule? They don’t mind it down ‘em, sir. Except that is requires even more bowel prep and misses more lesions compared with direct colonoscopy, according to this European study which did a direct comparison in 328 patients.  http://content.nejm.org/cgi/content/abstract/361/3/264

279    Bless you! You’d better go home right away – your sneeze could be the first sign of H1N1 influenza. If you do indeed get it, you will have time to do justice to this excellent historical perspective piece: if you are struggling amidst the stress and tedium of dealing with the outbreak, you’ll probably only have time for the shorter and even better account on p.225 (both attached in free full text). Pieces like these put a seal on the NEJM’s status as the best of the medical journals: none of the others has come anywhere near it in mapping, analysing and explaining this pandemic which has blown up over the last three months. It is essentially the same virus that caused the 1918 pandemic, one that mutated to cause “shift pandemics” in 1957 (H2N2) and 1968  (H3N2) with a few reassortments in between and now a reversion to a more benign version of the original. You won’t of course understand influenza epidemiology by reading these pieces – nobody does that – but you will reach a happy state of increased enlightenment. Bless you. http://content.nejm.org/cgi/content/full/361/3/279
http://content.nejm.org/cgi/content/full/361/3/225

Lancet  18 July 2009  Vol 374

210    Golly! Golimumab works where previous TNF-α blocking agents have failed in rheumatoid arthritis. Although this study, like so many in The Lancet, comes into the pend category (Promotes Expensive New Drug), it does seem to represent an important advance in a difficult group of patients. There was a highish drop-out rate (59 out of 461), but over the 24 weeks of the double-blinded study the rate of adverse effects was higher in the placebo than in the active groups. Golilumab achieved worthwhile remission in more than a third of these patients who had already had a variety of major treatments, in all cases including failed or discontinued TNF-α blockade. A fairly good head-to-head trial, then, but as the editorial (p.178) points out, we need longer-term safety data.  http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)60506-7/abstract

222    We badly need an effective treatment for idiopathic pulmonary fibrosis, and this multinational trial was meant to INSPIRE hope that interferon gamma-1b would improve survival. But it didn’t: patients receiving the interferon had a lot of flu-like symptoms and died at the same rate as the placebo group. http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)60551-1/abstract

229    I was initially deeply sceptical about the place of circumcision as a population strategy against HIV transmission. Then two randomised studies showed a useful, if far from complete, protective effect against HIV infection in men. This study, on the other hand, shows a complete lack of protection tofemale partners of HIV-infected men; in fact transmission is increased in the weeks following circumcision, due to incomplete healing. So circumcision cuts both ways, so to speak. Tell it not in Gath, publish it not in the streets of Askelon, lest the uncircumcised rejoice.  http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)60998-3/abstract

239    It’s 33 years since I was let loose with a cystoscope on patients with superficial bladder cancer, on the basis that it probably didn’t matter too much whether I spotted and zapped their new lesions correctly. Most bladder cancer is relatively benign, unless it has spread through the lamina propria, when it can start to require all manner of nasty treatments, ranging from intravesical cytotoxics and BCG to total cystourethrectomy. This American review is very comprehensive and includes a follow-up schedule somewhat more rigorous than is standard in the UK, including whole urinary-tract imaging every 12-24 months because of the 5% lifetime risk of an associated upper tract tumour. http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)60491-8/abstract

BMJ   18 July 2009  Vol 339

144    For the last couple of years, there’s been much debate about the relative merits of theFramingham versus the QRISK multifactor cardiovascular risk prediction scores in the UK population. This independent validation exercise shows that QRISK explains more of the variation at the cost of slight underprediction, whereas Framingham considerably overpredicts risk in UK men and slightly overpredicts it in UK women. In a theoretical world where every doctor and every patient did what they were told, this might matter. We are supposed to base our prescribing of things like statins on a risk equation of some sort; but in fact doctors divide into bulls and bears, and patients likewise. Some would like anything that reduces risk; others hate pills on principle. If we go by either of these scores, we will still miss the opportunity of preventing most cardiovascular disease. They are just very rough guides to be used in shared decision making with patients. http://www.bmj.com/cgi/content/abstract/339/jul07_2/b2584

152    I remember when proton pump inhibitors first appeared and were terribly expensive. I had a young patient with severe symptoms who told me he couldn’t possibly manage without them, and I was so appalled at the potential cost that I sent him to be considered for open fundoplication – also a new thing then, and costly, but I did the maths and persuaded the surgeon that this would be much less than the cost of a lifetime on omeprazole. Fortunately he did well. Now cheap omeprazole is Britain’s most popular drug and fundoplication is done through a laparoscope. So how do the two compare in 2009? In terms of cost, this study argues that operative treatment is still more effective than a lifetime of PPIs. But reading the head-on trials, I’d be pretty reluctant to undergo it because of the high incidence of medium-term effects like bloating and discomfort after eating. http://www.bmj.com/cgi/content/abstract/339/jul14_2/b2576

161    I remember being astonished over ten years ago to learn that the maternal circulation is teeming with fetal cells, but it’s even more spooky to think that it is also awash with cell-free fetal DNA and RNA. At the moment this can only be used to detect rhesus incompatibility, but in the next five years it might be the basis of a very reliable early pregnancy test for Down’s syndrome and other aneuploidies, plus single gene disorders and fetal sex determination. http://www.bmj.com/cgi/content/extract/339/jul06_2/b2451

166    Lithium is a good drug, but it is not friendly to the kidney. This piece emphasises the wide areas of uncertainty which exist in the assessment and management of lithium-related renal impairment. Beware the “creeping creatinine” in these patients and remember to check their eGFR as well as their TSH at least annually. http://www.bmj.com/cgi/content/extract/339/jul03_1/b2452

175    I don’t normally comment on Personal Views, but Ann McPherson’s courageous piece about facing death is required reading for everybody, and I am glad that Fiona Godlee made it the basis of her editorial and put the words “Assisted Dying – The Fight Goes On” on the front cover of this week’s BMJ. Ann’s use of the word “dignity” has already produced a typical response from a palliative care doctor who implies that the dignity she wants is a middle-class privilege to be claimed at the expense of the “vulnerable majority”. But the majority are already completely vulnerable to the whims of the “double effect” principle, formulated to protect the tender consciences of the palliative care theocracy. As a paediatrician points out in the Rapid Responses, this allows doctors to decide when they can give large doses of drugs such as opioids without any scrutiny, provided they claim that their primary intention is to alleviate symptoms rather than hasten death. It is high time we had a safer, kinder, and more accountable system for assisting death, decided by patients rather than doctors.  http://www.bmj.com/cgi/content/extract/339/jul15_1/b2827

Arch Intern Med  13 July 2009  Vol 169

1188    The attraction of coronary artery calcification screening is a complete mystery to me – big bills and big X-ray doses in order to provide big anxiety. But it seems that everyone is free to tout it, and when I examined a few websites, I found no mention of any hazard from the radiation dose, which can vary tenfold. This paper attempts to translate this into cancer risk. That’s pretty difficult as we have little to go on except extrapolation from the effects of much higher doses of radiation, but the figures are about 42 cancers per 100,000 men and 62 cancers per 100,000 women. http://archinte.ama-assn.org/cgi/content/abstract/169/13/1188

1195   One reason we take antihypertensive drugs is to prevent cognitive decline, and it’s been proposed that angiotensin-converting enzyme inhibitors might be particularly good at this. Results from the Cardiovascular Health Study fail to show this as a class effect but they do show weak evidence of protection from centrally active ACE inhibitors such as lisinopril and ramipril, which are probably the ones you most prescribe.  http://archinte.ama-assn.org/cgi/content/abstract/169/13/1195

1203    It’s from Leiden, so this paper has to be about thrombosis. Or its opposite, which isanticoagulation. Here is a trawl through four years of results from their warfarin clinic, and the results may not be what you expect. In my part of the world, the target INR for atrial fibrillation is 2.5, but these data support a target of 3.0 to 3.4. Conversely, we try to keep our patients with mechanical heart valves around 3.5, but this study suggests 2.5 to 2.9; and for myocardial infarction, we never anticoagulate at all, unless there is a mural thrombus, but in Leiden they seem to anticoagulate lots of patients and find it best to keep their INR between 3.5 and 3.9.  http://archinte.ama-assn.org/cgi/content/abstract/169/13/1203

1210    We lost a loved one to a warfarin bleed two years ago, and the memory is still raw. Here a Swedish study suggests that measuring thrombomodulin may help to identify the patients on anticoagulants at most risk from bleeding complications. But perhaps by the time we know how to use this biochemical marker in clinical practice, warfarin will have been superseded.  http://archinte.ama-assn.org/cgi/content/abstract/169/13/1210

John Donne for the Week:

As sickness is the greatest misery, so the greatest misery of sickness is solitude.

From Devotions Upon Emergent Occasions and Severall Steps in My Sickness, 1624, quoted in the week’sLancet, p.194.

JAMA  22-29 July 2009  Vol 385

385    This highly complex multinational study recruited 750 people with mild cognitive impairment, 529 with a diagnosis of Alzheimer’s disease, and 304 controls and tookcerebrospinal fluid from all of them in order to identify clinically useful CSF biomarkers for incipient Alzheimer’s disease. The markers were, for your interest, Aβ42, T-tau and P-tau.That’s CSF β-amyloid_1-42, total tau protein, and tau phosphorylated at position 181 threonine. You can find the sensitivity, specificity, positive and negative likelihood ratios on the link provided. Here’s a simpler test: if you can follow this then you certainly won’t get Alzheimer’s disease any time soon.  http://jama.ama-assn.org/current.dtl

394    What we call heart failure is the end stage of many possible processes which singly or severally cause circulatory impairment sufficient to produce a familiar cluster of symptoms and signs. The only subtypes of heart failure we have any clear idea of how to treat are those caused by myocardial infarction and cardiomyopathy. Most of the rest is caused by longstanding hypertension and age-related stiffening of the heart and capacitance vessels. We all assume that if you smoke and you are fat your lifetime risk of heart failure increases. Other factors thought relevant in this study were exercise, 5 or more alcoholic drinks per week (for protection), breakfast cereals and fruit and vegetable consumption. In fact obesity was the biggest predictor in this study, though paradoxically the obese survive heart failure best once they’ve got it. The rest of these factors show small effects which are additive.  http://jama.ama-assn.org/cgi/content/abstract/302/4/394

412   At the end of this review of lipoprotein (a) concentration and the risk of coronary heart disease, stroke and nonvascular mortality there is a list of all the members of the Emerging Risk Factors Collaboration, who allegedly joined forces in writing this piece. Since they number hundreds I guess each one must have contributed about ten words. These eminent researchers conclude that there is a continuous, independent and modest association between Lp(a) concentration and the risk of CHD and stroke. So it might be weakly causative for vascular disease, or weakly associated with it. But do we really need any new Risk Factors to Emerge? I’d suggest they take Milton’s advice to Melancholy:

In Stygian Cave forlorn

    ‘Mongst horrid shapes, and shrieks, and sights unholy,

Find out som uncouth cell,

     Were brooding darknes spreads his jealous wings,

And the night-Raven sings,

     There under Ebon shades, and low-brow’d Rocks,

As ragged as thy Locks,  

     In dark Cimmerian desert ever dwell.

From L’Allegro, POEMS of Mr. John Milton, 1645
http://jama.ama-assn.org/cgi/content/abstract/302/4/412

NEJM   23 July 2009  Vol 361

335    Several generations of British boys were inspired to become doctors by reading A.J. Cronin’s 1937 novel The Citadel.  I say “boys” because it’s a tale of male bonding and heroic actions like promoting amateur surgery and blowing up water mains, little calculated to appeal to girls of a scientific bent. The idealistic young doctor-hero, Andrew Manson, discovers that impoverished Welsh households supplied by a certain water company are being stricken with “the enteric”, i.e. typhoid. Fed up with cycling from one dying patient to another, he decides to take direct action. More than seventy years later, typhoid still kills the poor of many developing countries, like those in Kolkata, India where this cluster-randomised trial of Vi typhoid vaccine was carried out amongst slum-dwellers. The paper even contains a disease map eerily similar to John Snow’s famous cholera map of Broad Street in 1854. The remedy here was not to remove the pump handle or blow up the pipe but to vaccinate all children under 5 with a single dose of Vi polysaccharide. Not only does this demonstrate an 80% reduction of typhoid in the vaccinated, but also a 44% reduction of typhoid in the unvaccinated, through herd protection. There is another effective typhoid vaccine which is given in three oral doses on alternate days: the problem is not that there aren’t effective vaccines but that they don’t reach the people who need them. Oh, and clean water would also help.  http://content.nejm.org/cgi/content/abstract/361/4/335

368    The pay for performance experiment in British primary care has attracted a lot of international attention, particularly in the USA – hence the appearance of this Special Article in the most exclusive of its medical journals. The investigators from Manchester and Cambridge (original UK versions) find that the introduction of the Quality and Outcomes Framework saw a scramble to reach targets, which could be equated with an improvement in care, provided the targets were appropriate; followed by a levelling off of effort once the target had been reached. Pavlov’s dogs stopped salivating once they were fed. Prior to QOF there had been a steady linear improvement in outcome measures and greater continuity of care, and we can conclude that without QOF we would have been happier doctors with better treated patients and much lower administrative costs. The professional motivation of doctors is such that, left to themselves, most of them will try to improve their whole range of clinical care according to best evidence, whereas if you dangle bits of meat in front of them they will turn into Pavlovian dogs. It is time that QOF was abolished. http://content.nejm.org/cgi/content/abstract/361/4/368

379    A good practical review of ectopic pregnancy reminds me of the days when obstetrics and gynaecology were practised without ultrasound scanning; and also of the bizarre fact that in my area it is still unobtainable on a Saturday or a Sunday. Both Jewish and Christian teaching enjoin the saving of life on the Sabbath, but the British Weekend must never be disturbed. However, it’s just as well that five-days-a-week ultrasound did come along between 1970 and 1996, because that period saw a six-fold increase in the incidence of ectopic pregnancy. At the same time, due to greater vigilance and better imaging, the mortality rate has fallen steeply. Remain vigilant. It is a condition in the “Never Miss This” category, seven days a week. http://content.nejm.org/cgi/content/extract/361/4/379

Lancet  25 July 2009  Vol 374

301    The PApilloma TRIal against Cancer In young Adults yields the acronym PATRICIA: faintly ridiculous, but perhaps tolerable if it described the study accurately. However, this is not a trial of some magic papilloma which when grown on young people stops them getting cancer, but of an anti-papillomavirus vaccine which when given to young women stops them getting HPV-16/18 infection. This will, it is hoped, prevent most of them getting cancer of the cervix. The vaccine, from GlaxoSmithKline Biologicals who ran the trial, showed high efficacy, and according to the authors, many of whom own stock in GSK, it promises much for the reduction of HPV infection by universal mass immunisation. But why stop at vaccinating the world’s entire young female population? The accompanying editorial, written without conflicts of interest, suggests that we should vaccinate all male teenagers before their “sexual debut” as well, and we might get success on the scale of smallpox eradication. http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)61248-4/abstract

324    The patients I have seen with multiple myeloma over the years have generally been diagnosed at the stage when most of their bones look moth- eaten and they have all died within five years, mostly after many episodes of severe pain. Plasma-cell malignancy remains nasty, but this seminar on the subject is determinedly upbeat. One of the stars of the show isthalidomide, a drug developed and marketed as a hypnotic in the 1950s without any evidence for either its efficacy or its safety, as related by Druin Burch in Taking The Medicine (2009). This drug is uncommonly good at producing fetal malformations and neuropathy but it is also very good at stopping myeloma in its tracks and is coming to be used as a first-line treatment, though the less toxic lenalidomide may be a better option, especially in combination withbortezomib. Perhaps we can look forward to a time soon when myeloma will no longer mean four years of pain and debility followed by death. http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)60221-X/abstract

340    Considering it’s the code of life, Nature seems astonishingly cavalier about how may copies of the genome it leaves lying around in us. Lots of advanced plants like apple trees have two or three or even four copies of their chromosomes in every cell, and although humans remain resolutely diploid, we can carry many duplicate copies of individual genes. Up to four is the usual European maximum, but up to ten can be seen in some Africans. I tell you all this stuff in the hope that it might arouse your interest in this paper on genomic copy number variation, human health, and disease. If you’re like me, you’ll skim through it looking for (a) the bits you can understand and (b) the bits that seem to explain some clinical phenomenon, like the wide range of CYP2D6 drug metabolism, meaning that some mothers break down codeine so fast that they can poison their breastfeeding infants. The rest is meat and drink solely for the gene-gnomically inclined. http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)60249-X/abstract

BMJ  25 July 2009  Vol 339

206    Mammographic screening is having a hard time in the medical press these days, but then it deserves it. It has been oversold on the paternalistic (why not ever, in these days of equality, maternalistic?) assumption that women need to be given clear encouragement to have it done for their own good - and also because the statistics will look better, because you need an awful lot of women to demonstrate a survival benefit that is much less than 1% in absolute terms. The most combative opponent of screening mammography is Peter Gøtzsche, head of the Nordic Cochrane Collaborative, co-author of this study of overdiagnosis of breast cancer.  His style of debate offended Nick Wald and colleagues in the BMJ’s Rapid Responses, which have become the best place to follow this debate after Fiona Godlee’s editorial and Iona Heath’s opinion piece. He’s more formal and restrained here, but concludes that one in three of the cancers diagnosed by screening is not really a cancer. http://www.bmj.com/cgi/content/abstract/339/jul09_1/b2587

214    If I discuss the preferred place of death with a terminally ill patient, I either have to be brutally frank about the limitations of the care I can promise them for 115.5 hours in the week, or else try and fill that gap myself. On evenings, nights, and throughout the British Weekend, general practice and the palliative services abandon all attempts at continuity, or sometimes even adequacy, of care. I don’t know what the answer is, and I don’t think this qualitative study of general practitioners and community nurses takes us much closer to it. Perhaps what we need are joint rotas of GPs and palliative doctors covering large areas, but anything that smacks of heroics cannot provide a lasting answer.  http://www.bmj.com/cgi/content/abstract/339/jul15_1/b2391

220    Caught in the frantic paradoxes of the current influenza A/H1NI pandemic, I don’t think there’s a single GP in Britain who would share the sunny optimism of this paper on our successes in assessing its severity. But I’m glad to learn that somebody is out there trying to do some science, while the rest of us answer the telephone to a population encouraged to demand oseltamivir if they have the least symptom of anything that could be a virus. And I commend to your disbelieving eyes the accompanying editorial (p.181) on how wonderfully well the UK is managing this outbreak of swine flu related behaviour - more reminiscent of Pangloss, who said that everything is for the best in the best of all possible worlds, than Pandemic. http://www.bmj.com/cgi/content/short/339/jul14_3/b2840

Ann Intern Med  21 July 2009  Vol 151

103    As I’ve remarked on numerous occasions, colonoscopy is not a procedure that’s very popular with patients, and I imagine the prospect of another colonoscopy list has also made many a gastroenterologist inwardly mutter “how did I end up doing this”. So a way of doing fewer of them is to be welcomed, albeit cautiously in the case of this prospective cohort study using the results of two colonoscopies to predict recurrent adenomas with high-risk characteristics. It supports the common-sense view that if you take the results of two colonoscopies into account rather than just the latest one, you can improve your identification of low-risk individuals. But it doesn’t fully validate this hypothesis. http://www.annals.org/cgi/content/abstract/151/2/103

121    Can Helicobacter pylori eradication reduce the risk of gastric cancer? Surely we don’t need a meta-analysis to give us the answer? But like so many things we take for granted in medicine, the answer is not as clear as you might suppose. The Italian meta-analysts found seven studies that met their rather broad inclusion criteria: all but one were carried out in high prevalence areas in Asia, only two assessed gastric cancer incidence, and only two were double-blinded. Sure, H pylori treatment probably does protect against stomach cancer, but we don’t appear to have any accurate NNTs for the kind of patients we see in Europe or the USA.  http://www.annals.org/cgi/content/abstract/151/2/121

Plant of the Week: Macleaya cordata “Spetchley Ruby”
This enormous perennial needs about 10m² of garden space, and even then you need to watch out because it’s invasive. But if you have that amount of room, it’s worth every inch (OK, centimetre) for its big olive green lobed foliage and its great, self-supporting stems carrying plumes of red flower. It’s hard to believe how big it will get when you first plant it, and despite its great merits I shall be trying to dig out our plume poppy when it has finished flowering in October. If you want a bit, then is the time to ask, but you have been warned.