Benzo-pyrones for reducing and controlling lymphoedema of the limbs

Lymphoedema is the accumulation of excess fluid in the body caused by obstruction of the lymphatic drainage mechanisms. It can be caused by a number of factors, including congenital predisposition, parasitic infection or surgery. Lymphoedema is chronic and progressive and affects a significant proportion of the population. The standard treatment regimes include compression hosiery, skin care and exercise. The use of drugs in treatment, particularly benzo-pyrones, has gained favour over the last ten years. Benzo-pyrones, originally developed for use in vascular medicine, are prescribed to reduce vascular permeability and thus the amount of fluid forming in the subcutaneous tissues. Advocates for this treatment method believe that, as a result of reducing filtration, the drugs have some beneficial effect on pain and discomfort in the swollen areas. Proponents also claim that these drugs increase macrophage activity, encouraging the lysis of protein, which in turn reduces the formation of fibrotic tissue in the lymphoedematous limb.

To assess the effectiveness of benzo-pyrones compared to placebo or to different benzo-pyrones in reducing limb volume, pain and discomfort in lymphoedematous limbs. To assess the effect of benzo-pyrones on the quality of affected tissues and on the patient's quality of life and, finally, to establish the incidence of adverse effects

We searched the Cochrane Breast Cancer Group register (September 2003), the Cochrane Central Register of Controlled Trials (The Cochrane Library, Issue 4,2003), MEDLINE, EMBASE, CINAHL, UnCover, PASCAL, SIGLE, reference lists produced by The British Lymphology Society, the National Research Register (NRR) and The International Society of Lymphology congress proceedings.

Types of studies considered for review were randomised controlled trials testing Paroven, coumarin, Venastat, Cyclo 3 Fort or Daflon versus placebo (with both groups having or not having standard physical treatment

Eligibility for inclusion was confirmed by two blinded reviewers who screened the papers independently using a checklist of criteria relating to the randomisation and blinding of the trial. Both reviewers extracted data from the eligible studies using a data extraction form.

Overall, 15 trials were included that evaluated the role of benzo-pyrones. Three trials of oxerutin were found. Each tested the drug over 6 months using the same dose of drug against placebo. Two were crossover trials and one a parallel group trial with a total number of 127 participants and data available for only 81 of them. There were insufficient data provided in any of the trials to calculate the per cent reduction or increase in baseline excess limb volume. Standard deviations or confidence intervals and the numbers in the groups at the different stages of the trial were missing for all the data in two of the reports and for much of the data in the third, making any attempt at meta-analysis impossible. One trial testing Cyclo 3 Fort (approved name) over 3 months was found and involved 57 patients but provided insufficient data to allow a proper analysis of its findings. A single trial of Daflon (approved name) was found, lasting 6 months and involving 104 participants; once again there was insufficient information provided in the report to reach a conclusion about the effectiveness of the drug. Three trials of coumarin combined with troxerutin were found and tested two different doses of the drug against each other with no placebo, however, numbers of participants in the trial groups and baseline data were not provided. Eight trials of coumarin were identified. Two of the reports were confirmed as reporting the same trial and a further trial potentially also referred to the same trial but this was unconfirmed. A further two papers appeared to refer to the same trial but this was not confirmed. Three trials involved the same researcher. Five studies were conducted in India or China and they added anti-filarial dia or China and they added anti-filarial drugs to the interventions tested. The numbers of participants withdrawn and the numbers included in the analyses in all these trials were not extractable; the reporting of outcome measures in most of the trials was not clear. Loprinzi's 1999 trial in the USA reported the conduct of the trial and its findings with more detail, however, its conclusions were very much at odds with the findings of the other trials, finding that no difference was observed between those on the active preparation (coumarin) and placebo in any of the outcomes under investigation. This trial also reported a case of hepato-toxicity in a patient receiving the active preparation.

Meta-analysis was not performed due to the poor quality of the trials. It is not possible to draw conclusions about the effectiveness of Benzopyrones in reducing limb volume, pain, or discomfort in lymphoedematous limbs from these trials.