Perception of equivalent doses of neuroleptic drugs

An anonymous questionnaire was sent to 67 senior and junior psychiatrists enquiring about their perception of equivalent antipsychotic dosages of three commonly used neuroleptic drugs. Thirty-one questionnaires were returned and revealed a wide variation in perceived potencies for the specific drugs. Increased experience in psychiatry was not associated with a decreased variation. On average clinicians saw haloperidol and flupenthixol decanoate as substantially less potent, relative to chlorpromazine, than the available literature would suggest.


R. Mullen, A.W. Coon and S. Smith
An anonymous questionnaire was sent to 67 senior and junior psychiatrists enquiring about their perception of equivalent antipsychotic dosages of three commonly used neuroleptic drugs. Thirty-one questionnaires were returned and revealed a wide variation in perceived potencies for the specific drugs. Increased experience in psychiatry was not associated with a decreased variation. On average clinicians saw haloperidol and flupenthixol decanoate as substantially less potent, relative to chlorpromazine, than the available literature would suggest.
The adverse consequences of excessive or in adequate dosage of neuroleptics are well recog nised. Despite gaining scant empirical attention, the problem of choosing the correct dose of an antipsychotic drug in a clinical setting is com mon. Appropriate dosing requires a recognition that antipsychotic drugs vary in potency.
Except for some atypical antipsychotic drugs, it is accepted from radioligand studies that effec tive antipsychotics displace ligands from dopamine receptors with a facility that correlates highly with their antipsychotic potency (Peroutka & Snyder, 1980). From this robust psychopharmacological finding, the concept of equiva lent doses of different neuroleptics arises.
Recent publicity about high rates of side effects and occasional deaths on high doses have brought the issue of dosage into focus. It has been postulated (Krakowski et al, 1993) that some patients receive high doses of neuroleptics because of ignorance about the efficacy of low doses. Our concern here is that clinicians may not be fully aware of what constitutes a high, or a low, dose of these drugs.
The study reported here sought to audit clinicians' knowledge of neuroleptic equivalence.

The study
We used a postal questionnaire survey. Sixtyseven psychiatrists working at junior and senior level at the Bethlem Royal & Maudsley Special Health Authority, with responsibilities in adult psychiatry, were sent a vignette involving a patient transferred from chlorpromazine (the usual reference drug for antipsychotic equiva lence) to other widely used neuroleptics. The vignette read: A young man with chronic schizophrenia is well stabilised on a total daily dose of 500 mg of chlorpromazine. Because of oversedation, you wish to substitute a different neuroleptic. Please indicate what dose or dose range you would consider to be of equivalent antipsychotic potency for the following drugs: Drugs specified were haloperidol and trifluoroperazlne (both as total daily dose) and flupenthixol decanoate (as an intramuscular injection every two weeks). Respondents were requested not to refer to any other source before answering. They were also asked to indicate how long they had practised in clinical psychiatry and invited to comment further if they wished. The questionnaires were returned anonymously.
Data were entered on a database using the Statistical Package for Social Sciences, and parametric and non-parametric tests used to evaluate possible relationships between sug gested doses and experience.
Comparisons were made with the suggested equivalents in British National Formulary (BNF) and the ABPI Pharmaceutical Industry Data Sheet Compen dium (Association of the British Pharmaceutical Industry, 1990).
Years in practice showed no significant corre lation (product-moment) with any dose estimate (flupenthixol decanoate, r=0.08; haloperidol, r=0.30; trifluoperazine, r=0.27), nor did one-way analysis of variance find any significant effect of experience on the range of estimates made. Neither the least experienced doctors (six months to one year, n=4) nor the most experienced (more than five years, n=7) differed significantly from the others on any dose estimates (Student's t-test

What are equivalent doses of neuroleptics, really?
On the face of it, this is a straightforward ques tion. As neuroleptic drugs (clozapine excepted) are held to be of equivalent antipsychotic effi cacy, the idea that different doses of different drugs may be equivalent presents no intellectual problem. However, the clinical utility of these drugs lies not only in their effects on narrowly defined psychotic phenomena but also in effects such as sedation, anxiolysis and immediate behavioural effects. Such effects are often the immediate goal of prescribing, for example, when a patient becomes agitated and violent. These actions may not be directly related to the effects of the drugs on central dopamine systems. Forster (1989) has suggested that these factors may account for the poor agreement about the equivalent dosage of haloperidol, noting also that the relationship between haloperidol and chlorpromazine is unlikely to be linear. Such caveats aside, there is general agreement in the literature about what doses are equivalent. The BNF 25 (1993) offers the following as equivalent total oral daily doses: chlorproma zine 100 mg=halopertdol 2-3 mg=trifluoperazine 5 mg. The authors have found this to be in accordance with other sources. However, other wise detailed formularies rarely suggest equiva lent doses for depot neuroleptics. The BNF 25 (1993) has a useful conversion table for clin icians changing a patient from one depot to another, but their guidance offered concerning replacement of oral with depot neuroleptics makes no reference to equivalence of antipsy chotic effect. Schulz et al (1989) offer advice on this, based on detailed pharmacokinetics.
The ABPI (1993-94) indicates that 8 mg of flu penthixol decanoate, two weekly, is equivalent to 100 mg daily of chlorpromazine. From the above, we can assume chlorpromazine 500 mg daily= haloperidol 10-15 mg daily=trifluoperazine 25 mg daily=flupenthixol decanoate 40 mg two weekly to be equivalent.
How do these equivalences compare with the results of our study? On average, our respon dents thought that chlorpromazine is approxi mately two to three times more potent in relation to haloperidol and flupenthixol decanoate than the literature suggests. Mean dose of trifluo perazine was in good accordance with recom mendations. For the chlorpromazine equivalents of the three drugs, one doctor gave doses all consistent with the formularies' model (15, 25 & 40 mg), none of the 31 gave consistently lower values for all three equivalents and ten (33%) gave consistently higher doses for all three drugs. If consistent miscalculation were present, it was significantly more likely to be overesti mating doses than underestimating (sign test P<0.001). Overestimates were especially com mon for doses of flupenthixol (67%) and halo peridol (73%), compared with trifluoperazine (41%). While a (weak) correlation appeared to exist between trifluoperazine and flupenthixol doses, linear regression suggested that any possible scaling was from 1:2.1, not 1:1.6 as In the equivalence suggested above.

Comment
This study, crude in design and execution, achieved a low response rate. It would be ill advised to use the results to come to firm con clusions about the psychopharmacological acumen and prejudices of the whole group of clinicians surveyed. The wide range of doses perceived as equivalent is striking. It commands attention as a disquieting finding, difficult to explain in terms of the method. For example, the high concordance across respondents shows a non-random ranking of the drugs, i.e. some knowledge of their properties. However, the quantitative relationships were not understood by most respondents, and doctors were more likely to overestimate consistently rather than underestimate consistently. In this era of con cern about prescription of high doses of neuroleptics, training in rational prescribing is of recognised importance. This study indicates that such education may currently be inadequate.
These findings may not necessarily reflect neuroleptic prescribing nationally. Further studies, perhaps comparing different centres, are needed. We agree with the conclusion of Rey et al (1989), that the concept of equivalence is clinically useful and offers the best basis for predicting antipsychotic effect.