Containing the use of diagnostic tests.

since it delays conduction through the normal tissues only; in such a case amiodarone would be appropriate. In atrial fibrillation associated with hyperthyroidism digitalis may be ineffective and a f blocker preferred. And with little evidence that digitalis is of value in preventing attacks of paroxysmal atrial fibrillation quinidine would be first choice. Secondly, in ventricular arrhythmias the choice is from a larger range of drugs of broadly similar efficacy. In acute ventricular tachyarrhythmias lignocaine is still widely preferred as first choice, with intravenous amiodarone being given if lignocaine fails. Refractory arrhythmias may require other drug treatment: it would make sense, in the absence of other guidance, to substitute a drug of class Ia (for example, procainamide) and then one of class Ic (for example, flecainide). For chronic ventricular arrhythmias a drug of class Ia would be first choice. Amiodarone, while highly effective' and fashionable, has long term adverse effects which are frequent and may be serious,9 and its future use seems likely to be only short term, except in selected cases where the benefit is thought to outweigh the long term risks. Of other established drugs, such as mexiletine, none has any clear cut advantages, and it is too soon to say whether tocainide and the class Ic drugs such as flecainide will prove to be any better.'0 The precise choice of drug in cases such as these may be further influenced by contraindications (anticipated adverse effects or interactions)-for example, disopyramide and quinidine (because of their anticholinergic actions) in patients with glaucoma or prostatism; drugs of class Ia in patients with cardiac failure or heart block (which they will tend to exacerbate); and combinations of drugs which prolong the Q-Tc interval (drugs of class Ia and amiodarone, because of the risk of ventricular tachyarrhythmias, particularly torsade de pointes"). Further guidance may be given by the technique of inducing a patient's arrhythmia (for example, by rapid pacing or by delivering precisely timed stimuli during the cardiac cycle) and then observing its response to selected drugs. The responsiveness to class I drugs in these controlled circumstances has been claimed to be a good predictor of the subsequent responsiveness during long term treatment for both ventricular and supraventricular tachyarrhythmias,'2 and in some series this method has predicted effective treatment in over 70% of patients,'3 '4 though in the case of amiodarone its value is controversial.' There are, however, disadvantages: the technique requires special skill and is not universally available; only a few drugs can be studied at a time; the induced arrhythmia may not prove to be the same as that requiring treatment; and the induction of arrhythmias may be hazardous (in one series just over half of patients required DC cardioversion to terminate the induced arrhythmia, though none died'3). Finally, the successful long term suppression of arrhythmias will depend on the achievement of similar plasma concentrations to those effective during the short term study. Because of these problems this technique is likely to be used only in selected patients in skilled centres. Despite large advances in our understanding of the pathogenesis of cardiac arrhythmias 16 and of the pharmacological effects of antiarrhythmic drugs3 1' drug treatment of cardiac arrhythmias remains, if not entirely unsatisfactory, far short of ideal. In most studies a single drug has been given, often to patients thought (though not always proved) to be resistant to other treatment, and has been shown to be effective in a variable proportion of cases. In fewer instances controlled comparisons have been made of two drugs, and even then it has been difficult to determine any features (relating to the patient or his disease) which would lead to a rational preference of one drug to another. Until such information is available we shall continue to depend mostly on empiricism and the avoidance of adverse effects and interactions.

[1]  I. Wickings Putting it together. The patient, the purse, and the practice. , 1977, Lancet.

[2]  S. Gehlbach,et al.  Effects of an educational feedback strategy on physician utilization of thyroid function panels. , 1979, The Journal of family practice.

[3]  E. V. Vaughan Williams A classification of antiarrhythmic actions reassessed after a decade of new drugs. , 1984, Journal of clinical pharmacology.

[4]  J. Fisher,et al.  Cardiac pacing and pacemakers II. Serial electrophysiologic-pharmacologic testing for control of recurrent tachyarrhythmias. , 1977, American heart journal.

[5]  L. Horowitz,et al.  Recurrent Sustained Ventricular Tachycardia 3. Role of the Electrophysiologic Study in Selection of Antiarrhythmic Regimens , 1978, Circulation.

[6]  C G Fraser,et al.  Effect of feedback to clinical staff of information on clinical biochemistry requesting patterns. , 1981, Clinical chemistry.

[7]  L. A. Thibodeau,et al.  A trial of two strategies to modify the test-ordering behavior of medical residents. , 1980, The New England journal of medicine.

[8]  J A Stilwell,et al.  Evaluation of Laboratory Tests in Hospitals , 1980, Annals of clinical biochemistry.

[9]  R. Winkle,et al.  Electrode-Catheter Arrhythmia Induction in the Selection and Assessment of Antiarrhythmic Drug Therapy for Recurrent Ventricular Tachycardia , 1978, Circulation.

[10]  J. Hampton,et al.  Relative contributions of history-taking, physical examination, and laboratory investigation to diagnosis and management of medical outpatients. , 1975, British medical journal.

[11]  M. B. Rosenbaum,et al.  Ten years of experience with amiodarone. , 1983, American heart journal.

[12]  B. Littenberg,et al.  Does Cost Information Availability Reduce Physician Test Usage?: A Randomized Clinical Trial With Unexpected Findings , 1982, Medical care.

[13]  P R Fleming,et al.  Work-loads in chemical pathology: too many tests? , 1981, Health trends.

[14]  G. Sandler Do emergency tests help in the management of acute medical admissions? , 1984, British medical journal.

[15]  G Sandler,et al.  Costs of unnecessary tests. , 1979, British medical journal.

[16]  G D Lundberg,et al.  Laboratory request forms (menus) that guide and teach. , 1983, JAMA.

[17]  D. Krikler,et al.  Torsade De Pointes, an atypical ventricular tachycardia. , 1976, British heart journal.

[18]  J. Aronson Cardiac glycosides and drugs used in dysrhythmias , 1984 .