Richard Lehman’s journal review—2 April 2013

Richard Lehman JAMA  27 Mar 2013  Vol 309
1241    Have you ever heard of someone who’s recently had a heart attack going off and having 40 infusions of disodium ethylenediaminetetraacetic acid, ascorbate, B vitamins, electrolytes, procaine and heparin? No, I hadn’t either, until I went to give a talk to some cardiac rehabilitation patients about 12 years ago.

I said my little piece about all the lovely medical progress being made, and then took questions. They were all about chelation therapy, and I was reduced to desperately scanning the large room for means of escape while saying that I looked forward to reading the results of large randomized trials of this improbable mumbo-jumbo – though I might not have used those precise words. And here, at last, it is! A large, well-conducted trial (TACT) of chelation versus intravenous placebo in 1708 survivors of myocardial infarction. The trend to benefit is consistent and adds up to a statistically significant reduction in adverse cardiovascular events, driven largely by a reduction in the perceived need for revascularization. But this cannot be true, argues Steve Nissen in a commentary. They must have cheated. Harlan Krumholz has fun taking the opposite view in a Forbes Weekly piece: “This study has opened my mind to the possibility that there may be something more to this therapy than I originally thought. And given what I thought about it before, I can hardly believe I am writing that.”

1251   What happens when you give a low-dose antibiotic to a lot of people for a long time? They may get fewer infections, but they are likely to breed resistant organisms: biology behaves as expected in two trials of long-term low-dose macrolide antibiotics. So what is the overall benefit likely to be from adopting this strategy in real life for people with non-cystic-fibrosis bronchiectasis? It’s hard to say from the trials published here. BAT was a small Dutch trial lasting a year and using azithromycin 250mg daily: BLESS was a larger Australian trial using erythromycin ethylsuccinate 400mg bd. Surprisingly, azithromycin seemed to cause a lot more tummy ache than EES. Other than that, there was a definite reduction in exacerbations, especially in the BAT trial. It’s impossible to say if this would be maintained if macrolide-resistant airways pathogens became a lot more common. Bless my bat, I don’t think we’re all that much further on.

1278  For much of the twentieth century, immunotherapy of one kind or another enjoyed a wide vogue, and when I first went into general practice much nursing time around this time of year was given over to injecting dilute grass solutions to prevent hay fever. This they did at the cost of a few deaths per year in the UK from anaphylaxis. The safer method of using sublingual administration has failed to catch on, at least in the UK. In this systematic review, the authors struggle with a wildly heterogeneous literature of mostly industry-sponsored RCTs using different dilutions, preparations, schedules and end-points. They could not do a meta-analysis, and all they manage to do is pick up some signals of benefit and none of serious harm.

NEJM  28 Mar 2013  Vol 368
1179   Did I imagine it, or did I really see a tweet from a cardiac surgeon describing off-pump coronary bypass grafting as “like making love while standing up in a canoe: it can be done, but why would you want to?” Well, there’s the showing off of technical skill – no, wait, I am getting mixed up here – cardiac surgeons are not like that; it must be to achieve better neurocognitive outcomes for patients. Forget the canoe: let’s look at what happens when these surgeons stand on the terra firma of their operating rooms and perform CABG, using a cardiac bypass pump or not according to randomization schedule. The first large trial in this week’s NEJM randomized 4752 patients, and conclusion is “At 1 year after CABG, there was no significant difference between off-pump and on-pump CABG with respect to the primary composite outcome, the rate of repeat coronary revascularization, quality of life, or neurocognitive function.”

1189   The second trial looked at outcomes in 2539 Germans aged 75 or over, where cognitive effects might be expected to show up more clearly. But oddly enough, they were not looked for. The primary end-points that were measured showed “no significant difference between on-pump and off-pump CABG with regard to the composite outcome of death, stroke, myocardial infarction, repeat revascularization, or new renal-replacement therapy within 30 days and within 12 months after surgery.” So that’s it for off-pump CABG. It’s just showing off. Oops darling, splash!

1199   It was only a matter of time before we developed the ability to monitor metastatic cancer by direct measurement of circulating tumour DNA, and this study from Cambridge UK shows that we can now do this for breast cancer. It is a highly labour-intensive process, which used targeted or whole-genome sequencing to identify somatic genomic alterations and designed personalized assays to quantify circulating tumour DNA in serially collected plasma specimens. This is a breakthrough, but like most cancer breakthroughs it points the way to more arduous and expensive work rather than a ready cure.

1210   Anaemia is a very common accompaniment to heart failure, and each decrement of haemoglobin is associated with worse survival, so it would seem logical to try and do something about it. But this randomized trial shows that using darpoetin alfa to achieve a haemoglobin of 13 in patients with systolic HF does not reduce death or hospitalization for HF. It does not even improve symptoms.

1229   Salt in health and disease: an interesting topic, but try as I might I couldn’t really find much interest in this review. If salt has any effect on human health it is a very small one.

Lancet  30 Mar 2013  Vol 381
1099    When I bathe – in a bath, or the sea, or a river – I get into the water. Anything else is washing, or showering, or dabbing. So I think it is wrong to entitle this paper “Daily chlorhexidine bathing to reduce bacteraemia in critically ill children: a multicentre, cluster-randomised, crossover trial.”  The fact is that the children in the SCRUB study conducted in American pediatric intensive units did not get dunked in chlorhexidine: nor did they get scrubbed: they were washed using cloths. Perhaps I am being pedantic, as the result of growing up in a religious sect which placed great emphasis on baptism by total immersion. The kids who got wiped with chlorhexidine every day had fewer episodes of bacteraemia, and that’s what counts.

1107   Patients who are taking oral anticoagulants and who need to have percutaneous coronary intervention are, it seems, often loaded up with both aspirin and clopidogrel prior to the procedure, and not surprisingly they have a high incidence of bleeding. This trial from Belgium and the Netherlands tried the effect of omitting the aspirin and just using clopidogrel in a randomised comparison with standard triple therapy. “Use of clopiogrel (sic) without aspirin was associated with a significant reduction in bleeding complications and no increase in the rate of thrombotic events.”

1125  For a variety of reasons, I have absolutely no sense of nationality except European. But of course I have no idea of the name of my European MP, and not the slightest idea about health law and policy in the European Union; and although I wish I could promise you that I will spend the rest of this cold grey day reading the three articles on these matters in this week’s Lancet by the admirable Martin McKee, I suspect that somehow I may not get round to it. Or perhaps I will, just to spring a surprise test on Muir Gray when I next meet him, since he has tweeted to his followers that they are essential reading. So Muir, tell me briefly about the difference between EU directives and regulations, and give examples of how these have been embodied into the law of member states, with special reference to Estonia.  Here are the links:

BMJ  30 Mar 2013  Vol 346
I love the consultation, and I would have loved my career as a GP had there not been so infinitely many pressures and distractions from the core business of understanding ill people and helping them. I was good at timekeeping, but conscious that I was cutting out a great deal of what should have been done had there been more than ten minutes to do it. Shared decision-making, for example: or behaviour change counselling, the subject of this primary care study from Cardiff. They went to a lot of trouble training doctors and nurses in the techniques of focussed motivational interviewing: and proved that it made no difference to patients’ reported behaviour at three months or in biometric or biochemical measures. It takes rather longer to change the habits of a lifetime.

I listened to the great Peter Gøtzsche in Oxford last week and was convinced that screening mammography is not just futile but actually harmful. The argument is complex and only part of it relates to overdiagnosis, but here are some figures from Denmark to inform the quantitative debate: the probable overdiagnosis rate in screening participants is 1-5%. This seems very small until you work out the consequences. I won’t try to do so here: but bear in mind always that mammography screening has no effect on all-cause mortality. For every “life saved” by early diagnosis of breast cancer, another is lost.

Hip replacement surgery is perhaps the most successful elective operation ever, making life tolerable and even normal for millions of older people the world over. But the coming of metal-on-metal prostheses converted an almost uniformly successful procedure into one with a 50% failure rate. And we know surprisingly little about the economics of the various types of THR. Here’s a big modelling study based on large NHS databases from England. “Cemented prostheses were the least costly type for total hip replacement, but for most patient groups hybrid prostheses were the most cost effective. Cementless prostheses did not provide sufficient improvement in health outcomes to justify their additional costs.”

JAMA  Intern Med  25 Mar 2013  Vol 173
407   Performing tests for patient “reassurance” is a fool’s game, but we all get sucked into it. In the US, where you get paid for doing more, and sued for doing too little, it is an intractable problem. A couple of authors strive nobly to go through the randomized trials that have been conducted in the field, and conclude that “diagnostic tests for symptoms with a low risk of serious illness do little to reassure patients, decrease their anxiety, or resolve their symptoms, although the tests may reduce further primary care visits. Further research is needed to maximize reassurance from medically necessary tests and to develop safe strategies for managing patients without testing when an abnormal result is unlikely.”

418   Here’s a bold attempt to look at medical records to determine what causes serious diagnostic errors in two US primary care settings. Commonly missed conditions included pneumonia, acute renal failure, heart failure, cancer and urinary tract infection: and the great majority of failures were attributed process breakdowns in the initial patient-practitioner encounter. Listen properly, examine carefully and order the right tests: do that perfectly, and you’ll never make a mistake.

Plant of the Week:

You must be joking.