Richard Lehman’s journal review – 1 August 2011

Richard Lehman
JAMA 27 July 2011 Vol 306
The management of early invasive breast cancer has evolved gradually though improved understanding of its natural history together with improved deployment of chemotherapy and radiotherapy. We also now have better ways of detecting micrometastases in lymph nodes and bone marrow and this American study reports on their prognostic significance at a median of 6.3 years. In the context of modern treatment, the presence of micrometastases in sentinel lymph nodes has no effect on disease-free survival. The presence of bone marrow micrometastases is uncommon, and perhaps a little more ominous: but the confidence intervals are so large that it would be unwise to read too much into this. Best to wait for even longer follow-up.

David Simel’s Rational Clinical Examination has been the best series in any of the medical journals I have reviewed over the past 13 years, but recent choices of topic have strained the limits of the format somewhat. This piece which asks “Does This Patient Have Medical Decision-Making Capacity?” is an invaluable addition to the series, but it also illustrates the problems of straying into territory where there the gold standard is simply “expert opinion”. This is exactly what evidence-based medicine is supposed to deliver us from, and given such a comparator I don’t think it means much to apply the concepts “sensitivity” and “specificity” and then subject these to statistical analysis. Predictive values are always population-dependent; here they also vary from day to day and from situation to situation. We are dealing with human cognition and choice, not subarachnoid haemorrhage or pyelonephritis. But never mind: we should all be grateful for the authors for doing the best possible job, and everyone who shares decision-making with elderly patients should have the open-access Toronto instrument ACE (Aid to Capacity Evaluation) immediately to hand, while relegating the MMSE to occasional other uses in patients with gross dementia.

NEJM 28 July 2011 Vol 365
“Hats off, gentlemen, a genius!” wrote Robert Schumann when reviewing a newly published piano piece by the then unknown Polish composer, Frédéric Chopin, in 1831. Every now and again – maybe once a year – you’ll read a medical paper which gives you the same feeling. The technical aspects of this study are very simple; the numbers are not huge; but the intervention works, and people who would otherwise die will henceforth remain alive. The intervention here is desensitization to HLA in patients awaiting renal transplantation who have become sensitized to HLA through previous transplantation, blood transfusion or pregnancy. The wait for a suitable donor for such patients can be endless, so that the survival advantage for the desensitization group who are able to receive immediate transplantation gets larger as time goes on. Just as Chopin’s Variations on Là ci darem la mano stood out amongst the trashy repetitive virtuoso pieces of the early 1830s, so this study stands out among the usual garbage of statistics-squeezing and clinical irrelevance that fills most of our medical journals. “Hut ab, ihr Herren, ein Genie!”

And now back to the familiar world of market-driven drug trials and their quaint little ways. The only interventions which reduce progression to end-stage renal disease in type 2 diabetes to a clinically significant degree are blood-pressure lowering drugs. In this phase 2 trial, funded by Ream Pharmaceuticals, bardoxolone methyl was given to type 2 diabetic patients with an estimated glomerular filtration rate between 20-45. In the first few weeks, patients given the drug at various doses showed an increase in eGFR of about 10 units which was sustained for the duration of the trial. Also, compared with placebo, they had twice the rate of GI side-effects and muscle cramps, and about 80% more incident hypertension. “The improvement persisted at 52 weeks, suggesting that bardoxolone methyl may have promise for the treatment of CKD.” says the Conclusion. Lord help us.

Lancet 30 July 2011 Vol 378
One of the most depressing aspects of generalist medicine is being constant reminded of the penalties of old age, particularly dementia. I am not at all surprised at the result of this landmark British trial of two antidepressants, sertraline and mirtazapine versus placebo in patients with moderate dementia and symptoms of depression, 85% of them still living in the community. These drugs made no difference at all to those with depressive features such as lack of interest or enjoyment in food or company, or a wish to die. Such feelings, alas, may come with the territory for many old people who are aware of declining cognition.

Last week we read about the failure of a “vaccine” against GAD to halt the progression of type 1 diabetes. It seems to me that the problem lies in the fact that by the time children and adolescents present with the condition, the great majority of their beta-cells are already destroyed. Unless somebody finds a beta-cell specific regeneration factor, all you can hope to do is to keep the remaining horse in the stable while the rest have bolted. So I have grave doubts as to whether the use of specific immune suppressants, such as abatacept, will ever lead to useful treatments for type 1 DM. Abatacept inhibits the generation of autoaggressive T-cells and is already used in rheumatoid arthritis and psoriasis: here it is shown to inhibit the destruction of beta-cells as assessed by stimulated C-peptide production. But the effect wore off after 6 months: clinically, this does not seem like a promising approach, unless we can find ways to spot the disease process at a much earlier stage.

BMJ 30 July 2011 Vol 343
Here is the seventh meta-analysis to show that intensive glucose reduction in type 2 diabetes is generally pointless and can be harmful. Three years after ACCORD and ADVANCE, I think it is time we now moved into a post-Copernican view of T2DM: the sun does not revolve around blood sugar levels, and the popes of diabetology who have declared otherwise should withdraw their bull. Or should that be bulls? Something like that, anyway. We now know some useful things about older white people with T2DM but our ignorance about the rest is immense. Is there anyone out there who might want to set a new research agenda around Patient-Centred Care for Type 2 Diabetes? Beginning with an open-access text summing up current knowledge from the perspective of individual patients – say a Bangladeshi woman of 39 with five children, or a Chinese man of 58 who has just had an MI, or an otherwise healthy Italian of 73: does one size fit all? If a project like this interests you, I’d love to hear from you at

Diverticular disease has been labelled a “disease of Western civilization” and it remains fashionable to attribute it to a lack of dietary fibre. The nearest thing to dietary fibre I ever saw as a child was the statutory radish which accompanied formal afternoon teas with limp lettuce, slices of tinned meat and processed cheese accompanied by white sliced bread: all vegetables were cooked to a pulp and fruit was mostly tinned peaches on special occasions. Since the introduction of edible food to the British Isles, rates of admission for diverticulitis have soared. I do not believe the fibre hypothesis, but see if you do after reading this paper from the EPIC-Oxford study. As usual in dietary studies, the authors discuss all the limitations of their data but then go on to declare that “these findings lend support to public health recommendations that encourage the consumption of foods high in fibre such as wholemeal breads, wholegrain (unrefined) cereals, fruits, and vegetables.” OK: eat grainy bread if you like it. Or avoid sitting while defecating, if you prefer that particular theory of diverticular disease causation.

Do you remember SnNout? This is the rule that a test with a high Sensitivity rules a condition out if it is Negative. It belongs to one of the earlier chapters of Dave Sackett et al’s Clinical Epidemiology in its classic second edition. Very high sensitivity is an essential quality for all tests that are used for the detection of immediately life-threatening conditions. This excellent Canadian prospective cohort study finds that this is the case for modern computed tomography performed within six hours of headache for suspected subarachnoid haemorrhage. So if you work in any setting where such patients first present, you must have a high index of suspicion, and insist on an immediate CT. If you dither, or take no for an answer, you will miss one: the track record of British GPs in this respect is nothing to be proud of.

Arch Intern Med 25 July 2011 Vol 171
Rumour has it that our dear government would like to introduce 28-day readmission rates as a quality marker for hospitals. Here over the pond they prefer to count to 30 days, and how the data will be used is anybody’s guess: the longer I spend in a unit which scrutinizes the American health system, the more I feel that it can’t be long before the Mad Hatter appears with a chorus of singing oysters. This study investigated the real-life efficacy of a the Care Transitions Intervention, which is a package of better physician communication and patient education, shown to reduce 30-day readmission by 30% in the setting of a randomized controlled trial. Applied to Medicare patients in 6 Rhode Island hospitals, it had about half that effect size. The title of the accompanying editorial says it all: Interventions to Decrease Hospital Readmission Rates: Who Saves? Who Pays? Good questions to ponder in any market-based health system. Thank goodness we don’t have any of that nonsense in the NHS.

Here is another illustration of the same phenomenon with patients readmitted to hospital with heart failure. The intervention was “an advanced practice nurse–led transitional care program for patients with heart failure”. We have been here before, starting with Lynda Blue in Glasgow circa 1998. Like all supportive interventions for heart failure, this started with modest success in small trials and became less successful with time. And why? Because heart failure patients inevitably get worse from time to time, and call for an ambulance when they are feeling about to drown, knowing that outside office hours that is surest way to get help. In hospital they get intravenous diuretics and oxygen, they are monitored and nursed, and discharged as soon as possible on much the same treatment as they were on before. Unless we develop miraculous new interventions for failing hearts or
better models of 24-hour care in the community, this pattern will continue.

Plant of the Week by Eric Larson of Marsh Botanical Gardens

June 23, 2010
Dublin Bay Rose
Rosa x ‘Dublin Bay’

I have had more questions about this plant than a zookeeper during elephant mating season. It is sited perfectly to provide an entrance “wow” as you come off the street on Mansfield, admittedly not our best first impression. If you can look past the dump area that grounds maintenance uses for organic waste, a thirty yard dumpster, a five yard dumpster, another container for storage of tools, the sand and salt mixing area with metal and canvas canopy, you will raise your eyes to the completely charming and very assuming Dublin Bay Rose.

This wonderful rose has been blooming for six weeks now, which is very unusual for a climbing rose. I don’t spray it, I don’t prune it, I don’t pamper it, I hardly even tie it to the fence (although I will this week, I promise). It rewards my benign neglect with nothing short of extraordinary blooms of the deepest red.

Did I say that is also fragrant? When you cut a bouquet of these roses, you won’t be sneaking up on anybody. They extend their presence into the room like a queen sending her entourage ahead of her. The darkly scented aroma is a favorite in our household, because it takes a queen to know one. That queen being my lovely wife, of course.

Dublin Bay Rose, like all roses, prefers full sun, well-drained soil and good fertility. Like most climbing roses, it will grow pretty rampantly up and to the side. Unlike many roses of any kind, Dublin Bay seems to enjoy a life unfettered by the debilitating effects of fungal and bacterial problems. As I mentioned earlier, if there is a truly carefree rose, it is Dublin Bay.

The rose is a member of the eponymous Rose family, Rosaceae, which includes a wide array of plants from apples, cherries and similar fruits to Raspberries and other brambles, from Hawthorns to the lovely little Potentilla. There are about a hundred genera within its ranks, and almost three thousand species. Most species in the genus Rosa are native to Asia, with a few from Europe, even smaller number from North America and a very few from northwest Africa. There are numerous cultivars and hybrids of rose adding to the genetic pool, creating a botanical nomenclaturists nightmare or job security, depending on your viewpoint.

For instance, if you google Dublin Bay Rose, you’ll get alternate names, including “Improved Blaze” and “Don Juan.” I prefer Dublin Bay for some reason, don’t ask me why. The naming of plants, especially specific cultivars, is complicated. If you “discover” a “sport” or a variant genotype of a plant, you can name it. But that same sport may have already been discovered and named by someone else, setting up a potential Jerry Springer-type conflict. Speaking of naming rights, James Thurber wrote a wonderful piece about a woman whose husband studied insects. She complained that his colleague, a botanist, named wonderful flowering plants after HIS wife, while her name gets immortalized on some creepy crawly thing. Ah the foibles of our species.

Come see the Dublin Bay rose soon, or plan to visit in September, when its lovely blooms will reappear in that special time for all roses, June and September (and for some on into October). The flush of bloom at those times seems to be a phenological indicator of Homo sapiens’ need to ritualize our relationships. Nice to have roses for the wedding.


“Plant of the Week” and “Liquid Sunshine” are sent out almost-weekly, sometimes weakly. They do not reflect the corporate or organizational views of Yale University, Marsh Botanical Gardens or any other rational entity. For complaints, contributions and more information, please contact: Eric Larson at 203-432-6320 or

RL’s Note for British Readers: Eric is right: this is a really great rose. It is so intensely red that photographs of it look unreal. It is reliably repeat flowering – continuous from June to October or even November in our English garden. Eric is also absolutely right about its freedom from disease, and its general imperturbability: we grow it on limy clay in cool cloudy Oxfordshire, he grows it on acid sand in baking New England, and it flourishes on both. But I think it must need the New England climate to produce scent; back in Old England we can only just pick up a hint of apples on warm evenings.