Richard Lehman’s journal review – 18 June 2012

Richard LehmanJAMA  13 June 2012  Vol 307
2383    It is surprising how long and valuable a period people can survive for with malignant pleural effusion. I worked beside such a person for 18 months while she adjusted her hectic schedule to lie down occasionally and drain her indwelling pleural catheter. To read more about her last days, you need to look at Personal View in this week’s BMJ. Another person still dearer to me lived for a similar period of time after having talc pleurodesis. These people were exceptional in every way, however: the average life expectancy following the development of pleural effusion due to secondary cancer is 4 months. This useful British randomized trial (TIME2) compared results from the two methods and found that there is no clear superiority for either. This is what we call “preference-sensitive care”: the patient should be given the opportunity to decide, though no patient with advanced cancer should ever have to carry the burden of decision if they prefer to hand it to others.

2400    We know surprisingly little about the long-term effects of low-dose ionizing radiation. This became a matter of intense debate during and after the era of above-ground nuclear weapons testing, and much of what we know about the subject is based on extrapolation from the effects of the weapons detonated over Hiroshima and Nagasaki. It is very hard to quantify the cumulative effects of exposure to diagnostic X-rays and to work out the benefit/harm ratio from this exposure: all that this study and others can tell us is that such exposure in advanced health systems is steadily increasing, mainly due to the use of CT scanning. In large integrated health systems in the USA, use of this modality more than doubled between 1996 and 2010.

2418   The largest contribution to increased CT radiation exposure in the future is likely to be from lung cancer screening in smokers, which was shown to have a mortality benefit in the National Lung Screening Trial (NLST). In the NLST, the number needed to screen to prevent 1 lung cancer death was 320 persons undergoing 3 annual low dose CTs. The pick-up rate for lung nodules across the studies was about 20%, leading to further investigation (often involving more radiation) in this substantial group, to detect the 10% of nodules that are malignant. Some people who undergo CT lung cancer screening will undoubtedly suffer from radiation-induced cancer many years later, but it is impossible to know how many, as this paper and the editorial make clear.

NEJM  14 June 2012  Vol 366
2247   This week is diabetes-heavy in the main journals, and the NEJM leads with a paper about the newest and most worrying kind of diabetes: type 2 starting between the ages of 10 and 17. The TODAY trial recruited 699 American children with this condition, all of them with age-adjusted BMI over the 85th percentile. All were treated initially with metformin, and after six months they were randomized to continue with metformin alone, to have rosiglitazone added, or to continue with metformin plus a lifestyle intervention programme. About half of these children achieved durable control (HbA1c less than 8 ) with metformin alone. Lifestyle intervention did not achieve much extra. Rosiglitazone did, but that’s not a finding that is likely to have much relevance for the future. The authors conclude sombrely that “these results suggest that a majority of youth with type 2 diabetes may require combination treatment or insulin therapy within a few years after diagnosis.” Sadly, you can see them waddling outside fast food outlets on every American street corner.

2257    Oral contraception began as a massive natural experiment in the 1960s, and it is one we are still learning from. Where better to go than Denmark, which has a superb National Registry of Patients and Register of Medicinal Products Statistics: combine the two and you can get data relating to 1,626,158 women, with 14,251,063 person-years of observation. Øjvind Lidegaard and a team of four crunched the numbers to give us the most detailed picture yet of the risk of myocardial infarction and thrombotic stroke attributable to each type of oral contraception. The type of progestin seems to make little difference, but the dose of oestrogen probably does. Pills with the standard dose of 30 mcg of ethinyl oestradiol may roughly double the risk of MI and thrombotic stroke in this population. This equates to about 1-2 extra events per 10,000 women annually. The title of the editorial sums it up: Hormonal Contraceptives and Arterial Thrombosis — Not Risk-free but Safe Enough.

2294   “Since poliomyelitis has nearly been eliminated, the Guillain–Barré syndrome is currently the most frequent cause of acute flaccid paralysis worldwide and constitutes one of the serious emergencies in neurology. A common misconception is that the Guillain–Barré syndrome has a good prognosis—but up to 20% of patients remain severely disabled and approximately 5% die, despite immunotherapy.” Be aware. This is a never-miss-this diagnosis: check the tendon reflexes in anyone who complains of feeling weak following a flu-like or diarrhoeal illness. Campylobacter may be the commonest precipitant. Here is an excellent update that every generalist should read: pride yourself on getting the diagnosis within a minute, every time: which may only be once or twice in any working GP’s lifetime.

Lancet  16 Jun 2012  Vol 379
The Lancet does a special issue on type 2 diabetes, boldly facing both ways at once. There is a typical editorial on which drug as add-on to metformin? written by a Danish diabetologist who declares “I have been a consultant or adviser to Novartis Pharma, Novo Nordisk, Merck Sharp and Dohme, Sanofi-Aventis, AstraZeneca, Johnson and Johnson, Roche, Mannkind, Boehringer-Ingelheim, Zeeland, Lilly, and Intarcia Therapeutics, and have received fees for speaking from Novo Nordisk, Merck Sharp and Dohme, Johnson and Johnson, Roche, Schering-Plough, Sanofi-Aventis, Novartis Pharma, Lilly, Bristol-Myers Squibb, and AstraZeneca.” For reasons we can only speculate about, he comes down heavily in favour of exenatide twice-daily (made by Lilly), based on weak observational evidence. Exactly the same subject is covered by another editorialist, this time from the Netherlands, and she concludes that insulin glargine (made by Sanofi) may be the answer. “I serve on advisory boards for Abbott, Eli Lilly, Merck Sharp & Dohme (MSD), Novo Nordisk, and Poxel Pharma; am a consultant for Sanofi; and a speaker for Eli Lilly, MSD, and Novo Nordisk. Through me the VU University Medical Center receives research grants from Amylin/Eli Lilly, MSD, Novo Nordisk, and Sanofi.” And then there is a wonderfully scathing letter from three friends of mine, lambasting Richard Horton for publishing misleading conclusions to inconclusive NovoNordisk studies of expensive new insulins, and asking what are editors for? What indeed.

2243    The diabetes industry is so huge and all-pervasive that we who dare to criticize it are often on the look-out for following cars and men in raincoats with bulging pockets. And it can expect to get huger yet, now that pre-diabetes is opening up vast new markets. The Diabetes Prevention Program Outcomes Study is not pharma-funded, but seems honestly aimed at discovering at what point we can identify a group with borderline glucose levels who are likely to progress to beta-cell failure and “real” type 2 diabetes. In fact there is a straight line risk graph so this distinction is arbitrary. Every year about 10% of those defined as “pre-diabetic” because of borderline fasting and post-prandial sugar levels will move up the line and get labelled diabetic. This study, unsurprisingly, finds another group who, for one reason or another, fail to move up the line but achieve normal glucose levels: and believe it or not, these people have the best outcomes. That’s it: I don’t see that this tells us anything new, or anything about the best interventions to use in this heterogeneous group defined by a single biochemical marker.

2252   Type 2 diabetes is a high-risk state and remains a leading cause of blindness, cardiovascular disease, amputations, and kidney failure, we are told all the time. Then we are told that it is our fault (or sometimes the patient’s) for not trying harder. The things we are urged to lower are HbA1c, LDL-cholesterol, and blood pressure. In fact this is misleading: BP is far more important than HbA1c, and the LDL-C level is irrelevant since every person with type 2 diabetes should be on a high potency statin without reference to lipid levels. Dozens of trials have examined the effect of quality improvement strategies on surrogate end-points in T2DM, and their effect is minimal. In fact studies of the UK Quality and Outcomes Framework show a clear levelling off of effort once the system was introduced.

2262    And now let us descend into the abyss from which most of our knowledge about diabetes treatment emerges. This malodourous declivity is filled with rank vegetation planted by the pharmaceutical industry, supported by a fungal network of academics with financial interests. In this open label trial (EASIE), glargine insulin (Lantus) was compared with sitagliptine (Januvia) as second-line treatment in T2DM inadequately controlled with metformin. “The funding source participated in initial discussions about trial design, participated in the respective study steering committees, and undertook the data analysis and preparation of study reports. Company representatives are named as authors and contributed to development of the report as described in the authors’ contributions.” Well I never: these authors conclude that Lantus made by Sanofi controls sugar better than Januvia made by Merck.

2270   Next: a trial of exenatide twice-daily compared with a sulfonylurea, glimepiride. The outcome measure? Gosh, it’s glycaemic control, once again. The sponsors? Gosh, they’re the manufacturers of exenatide. And the design and reporting? “The sponsor took part in study design, data collection, data analysis, data interpretation, and writing of the report.” And the conclusion? “These findings provide evidence for the benefits of exenatide versus glimepiride for control of glycaemic deterioration in patients with type-2 diabetes inadequately controlled by metformin alone.” Gosh.

2300   And what is the elephant in this room? Why, he is none other than Barry the Surgeon. Bariatric surgery cures type 2 diabetes. So why is almost nobody getting it? Perhaps because it brings in huge immediate costs for health systems and no profits for Novo Nordisk, Merck Sharp and Dohme, Roche, Schering-Plough, Sanofi-Aventis, Novartis Pharma, Lilly, Bristol-Myers Squibb, and AstraZeneca, to name but a few. Here is a good review of the topic which is careful not to overstate its evidence base – although the evidence base is actually far better than for any other intervention in T2DM, and the effect size is an order of magnitude greater. Surgery should be offered as first-line treatment for most obese people with type 2 diabetes.

BMJ  16 June 2012  Vol 344
“The variation in outcome after surgery and its association with the volume of patients treated by an institution or individual surgeon has been extensively examined over the past 50 years.” It’s actually gone on for much longer than that: it was first proposed in the 1730s and had reached a degree of sophistication by the time of Lister and Nightingale in the 1860s. A study of biliary surgery from the Mayo Clinic in 1900 reported a mortality exceeding 50%, whereas for cholecystectomy in this Scottish survey of 1998-2007 it was typically 0.5%. High-volume centres did best on a range of outcome measures, but not so much so as to make much difference to low-risk patients coming to elective surgery.

I don’t know why I haven’t grown to hate medical journals by now, but the fact is that I still keep a fondness for them, especially on the rare occasions that I penetrate a library and handle paper copy. In even the most obscure journals there is usually something to catch the eye: some chance insight, some forlorn attempt to capture the intangible, or just the immense pathos of interminable futile effort.

Full many a gem of purest ray serene
The dark unfathom’d caves of ocean bear:
Full many a flower is born to blush unseen,
And waste its sweetness on the desert air.
(from Thomas Gray: Elegy Written in a Country Churchyard,1751)

To gather these gems and to bring in these flowers from the desert is the task of the journal scanner, of which great tribe I have the honour to be a humble representative. Paul Glasziou often encouraged me in this task in days gone by, and here he and his colleagues look at what it would really require to keep up to date in the major subspecialties of medicine. If COPD is your interest, a mere 10 journals may suffice; whereas if you are a general neurologist, you should really be looking at 896. Rationalizing and harmonizing the flow of medical knowledge are major tasks for the twenty-first century. The whole of the present system is unfit for purpose.

Arch Intern Med  11 June 2012  Vol 172
837    Richard Doll famously said that death in old age is inevitable, but death before old age is not inevitable. Amongst the avoidable factors he was referring to, smoking remains the most obvious, and after sixty years of preaching on the subject, he must be sighing in his grave to read yet another study proving that this is still so, regardless of where you set the threshold for “old age.” This systematic review shows that the mortality benefit of giving up can be seen even beyond the age of 80.

854   Less is More is the most essential series in any journal: not perhaps for its originality, not for its academic content, but simply because it carries the most important messages for all developed health systems. The drive to do useless things is nowhere near as strong in the NHS as in the pay-by-item American systems, but it happens all the time for all that. Exercise echocardiography in asymptomatic patients following coronary revascularization at any time following the procedure makes no difference to outcomes. It is a waste of resources and should not be done.

873   The same can be said for a lot of stress testing following admission for chest pain in US hospitals. In one tertiary hospital, the authors found that “most patients who are admitted with low-risk chest pain undergo stress testing, regardless of pretest probability, but abnormal test results are uncommon and rarely acted on.” The need to factor in pre-test probability is a lesson that never seems to get learnt sufficiently.

878    Many trials have established beyond doubt that acupuncture is a moderately effective treatment for a wide range of conditions, working by suggestion quite independently of where you put the needles. This Japanese trial of “real” versus “sham” acupuncture seems to buck the trend, finding a positive effect in chronic obstructive pulmonary disease only if the acupuncture was conducted using traditional methods. The “real” acupuncture patients gained weight and had improved exercise capacity. There are a number of explanations and confounders, such as drug treatment and awareness of allocation. Don’t let this one study put you off setting up as a sham acupuncturist: this is a potentially lucrative and even useful form of employment, requiring only a set of needles and a plausible manner.

Plant of the Week: Rosa “Albéric Barbier

My fingernails have been clean of soil for a year now, living in the US with little connection to gardens besides memories of the small and crowded plots that surround our house. Nobody has gardens like that in New England. Space is seldom at a premium, so great areas are given up to monocultures of ground cover plants or for the display of large and beautiful trees. We will miss those when we get back this week: but what are we most looking forward to back in England?

I think it must be the prospect of climbing roses of every kind. They scramble over all our walls and fences in a messy, un-American way, by which exuberance becomes beauty. Old Alberic is a barbarian of this sort: he just gets on with putting out lots of untidy fragrant flowers of yellow turning white, against excellent dark glossy foliage. Somehow he manages to turn the whole thing into a class act. We look forward to greeting him, knowing that he never disappoints.