Research highlights blog – 23 September 2011

Research questions“Research highlights” is a weekly round-up of research papers appearing in the print BMJ. We start off with this week’s research questions, before providing more detail on some individual research papers and accompanying articles.

Treatments for arm lymphoedema related to breast cancer
Last year the BMJ published a single blind randomised controlled trial of early physiotherapy to prevent lymphoedema after surgery for breast cancer (BMJ 2010;340:b5396). The intervention included manual lymph drainage (gentle massage to drain lymph vessels and nodes), massage of scar tissue, shoulder exercises, and education about self management of lymphoedema; the control group was offered only the education. The trial was probably underpowered and the external validity was limited. But we published it because the research question was so important and the study provided some proof of concept for treating this debilitating and intractable condition.

Nele Devoogdt and colleagues’ trial has partly unpicked the tangle of interventions by randomising women to a treatment programme comprising education, exercise therapy, and manual lymph drainage or to the same programme without manual lymph drainage. This trial had greater statistical power and longer duration of treatment, and it found that adding manual lymph drainage had little effect on the development of arm lymphoedema during the first year after axillary lymph node dissection. The authors call for a further trial, now that there’s enough evidence to do a more meaningful sample size calculation. Meanwhile, they suggest, women should be warned that the jury is out on manual lymph drainage.

Mortality in patients discharged from psychiatric care
On psychiatric wards, doctors are well placed to observe and document an array of harmful lifestyle factors and physical health issues in patients with severe mental illness: tar stained fingers and a smell of stale smoke; complaints of coughs, pains, and weaknesses; and findings of rashes, wheezes, crackles, murmurs or hypertension. And, although some physical health issues might be addressed in an inpatient setting, what should be prioritised, and how will things progress once the patient goes home?

This paper by Uy Hoang and colleagues documents a persistent and increasing gap in mortality between inpatients discharged with a diagnosis of schizophrenia or bipolar disorder and the general population over 15 years in England. As explained in the accompanying editorial by Brian Miller, the results should be interpreted in context and with caution. Most importantly, there has been a move towards keeping patients out of psychiatric hospitals, so the inpatients of 2006 may have been more severely unwell than those in 1999. Miller believes more should be done to modify risk during inpatient admission. Perhaps it is beholden upon all primary and secondary care clinicians to try and find new ways to improve the management of such patients.

Air pollution and short term risk of myocardial infarction
Associations between common environmental pollutants and cardiovascular mortality are well established, but results for myocardial infarction specifically have been inconsistent. Most studies have analysed daily measurements of exposure to air pollution and health outcomes, but a handful that used data recorded at hourly intervals have observed effects within a few hours of exposure to some pollutants. Krishnan Bhaskaran and colleagues have now quantified these very short term effects using data from urban air quality monitoring stations in England and Wales, and from a national register for hospital admissions in the same areas.

They found that risk of myocardial infarction was transiently increased up to six hours after exposure to higher levels of traffic associated pollutants, but there was no increase in the overall risk over 72 hours. The authors conclude that air pollution may be associated with having a myocardial infarction sooner rather than the overall risk of having one. In their linked editorial, Simon Hales and Richard Edwards consider the challenges in undertaking such a study and applaud this “state of the art analysis [which] tests the epidemiological … limits of detection.”