We must take issue with the author of the trial of butterbur versus
cetirizine for seasonal allergic rhinitis (1) on three main issues.
That the trial has failed to show that one treatment is better than
the other does not mean that it has shown them to be equally good. A
power of 80% means a 20% chance of failing to show a difference when one
is actually present. A negative result from such a small sample size
should be treated with great caution. The effect size of 0.5 in the power
analysis is given without any explanation or justification.
A negative result, however, was always likely, given the
inappropriate choice of outcome measures. The physician-rated improvement
scores do not reflect the patients’ own perceptions of whether they are
better or not. The only patient-centred outcome used is overall quality
of life, as rated on SF36. This is a blunt instrument, and very large
numbers of subjects would be required to show a difference between any two
groups, because reported quality of life varies so much between
individuals, even those with the same severity of disease. Why is there
no comparison of symptoms? 10cm visual analogues for sneeze, itching,
blockage, rhinorrhoea and hyposmia would be reasonable, as would some sort
of objective measure (peak nasal inspiratory flow rate, for example). A
disease-specific quality of life measure such as the Rhinoconjunctivitis
Quality of Life Questionnaire (2) would also have been a good addition.
As it is, we have no evidence that butterbur improves the symptoms of
rhinitis at all.
The other major concern is that the paper fails to mention the
hepatotoxicity, teratogenicity and carcinogenicity of the actve alkaloid
ingredients of butterbur . This will do nothing to correct the widely-
held fallacy that herbal remedies (unlicensed, poorly tested and poorly
standardised) are somehow safer than pharmaceuticals, just because they
are perceived to be “natural”.
Butterbur has potential dangers, and this industry-sponsored trial
shows no evidence that it is effective at all, let alone as effective as
cetirizine. It cannot be recommended for use on the basis of this
evidence.
References:
1. Schapowal A. Randomised controlled trial of butterbur and
cetirizine for treating seasonal allergic rhinitis BMJ 2002; 324: 144-146
2. Juniper EF. Rhinitis management: the patients’ perspective Clin Exp
All 1998; 28(supp 6): 34-38
Rapid Response:
A knife through butterbur
We must take issue with the author of the trial of butterbur versus
cetirizine for seasonal allergic rhinitis (1) on three main issues.
That the trial has failed to show that one treatment is better than
the other does not mean that it has shown them to be equally good. A
power of 80% means a 20% chance of failing to show a difference when one
is actually present. A negative result from such a small sample size
should be treated with great caution. The effect size of 0.5 in the power
analysis is given without any explanation or justification.
A negative result, however, was always likely, given the
inappropriate choice of outcome measures. The physician-rated improvement
scores do not reflect the patients’ own perceptions of whether they are
better or not. The only patient-centred outcome used is overall quality
of life, as rated on SF36. This is a blunt instrument, and very large
numbers of subjects would be required to show a difference between any two
groups, because reported quality of life varies so much between
individuals, even those with the same severity of disease. Why is there
no comparison of symptoms? 10cm visual analogues for sneeze, itching,
blockage, rhinorrhoea and hyposmia would be reasonable, as would some sort
of objective measure (peak nasal inspiratory flow rate, for example). A
disease-specific quality of life measure such as the Rhinoconjunctivitis
Quality of Life Questionnaire (2) would also have been a good addition.
As it is, we have no evidence that butterbur improves the symptoms of
rhinitis at all.
The other major concern is that the paper fails to mention the
hepatotoxicity, teratogenicity and carcinogenicity of the actve alkaloid
ingredients of butterbur . This will do nothing to correct the widely-
held fallacy that herbal remedies (unlicensed, poorly tested and poorly
standardised) are somehow safer than pharmaceuticals, just because they
are perceived to be “natural”.
Butterbur has potential dangers, and this industry-sponsored trial
shows no evidence that it is effective at all, let alone as effective as
cetirizine. It cannot be recommended for use on the basis of this
evidence.
References:
1. Schapowal A. Randomised controlled trial of butterbur and
cetirizine for treating seasonal allergic rhinitis BMJ 2002; 324: 144-146
2. Juniper EF. Rhinitis management: the patients’ perspective Clin Exp
All 1998; 28(supp 6): 34-38
Competing interests: No competing interests