1 Measurements of the binding of 12‐alpha‐[3H]‐digoxin to the membranes of intact erythrocytes, erythrocyte 86rubidium uptake and intraerythrocytic sodium concentrations have been made in the red cells of various groups of patients‐those who have not received digoxin, those during the early phases of treatment, those during chronic (greater than 2 months) treatment, and those toxic. 2 The values of those measurements in the patients in the early phases of therapy and in the toxic patients differed significantly from those of the untreated patients. 3 However, the values in the chronically treated patients were not different from those of the untreated patients. 4 The results suggest that the biochemical pharmacological effects of digoxin which occur during the early phases of therapy do not persist in the long‐term. 5 The possible clinical significance of these observations is discussed.

1 Measurements of the binding of 12‐alpha‐[3H]‐digoxin to the membranes of intact erythrocytes, erythrocytic 86rubidium uptake and intraerythrocytic sodium concentrations have been made in the red cells of patients receiving digoxin in the short‐term for atrial fibrillation or cardiac failure in regular rhythm. 2 During the first few days of treatment [3H]‐digoxin binding and 86rubidium uptake fall and intraerythrocytic sodium concentrations rise. 3 Subsequently parallel fluctuations occur in [3H]‐digoxin binding and 86rubidium uptake but not in intraerythrocytic sodium concentrations and the significance of the fluctuations is discussed. 4 The values of all three measurements correlate significantly with the response of the heart in sinus rhythm as measured by QS2I. 5 Plasma digoxin concentrations do not correlate with QS2I.

1. Previous electrophysiological studies have suggested the presence of KCa and Kv channels in human platelets. However, the pharmacology of these channels has not been defined. 2. We have studied potassium channels in human platelets by measuring the efflux of 86Rb+ (a marker for K+) from 86Rb+-loaded cells, and have defined their responses to stimulation by the platelet agonist thrombin and the calcium ionophore ionomycin. 3. Thrombin (0.1–0.6 i.u./ml) stimulated an increase in 86Rb+ efflux from the platelets in a concentration-dependent manner. This efflux was significantly inhibited by apamin (100 nmol/l), charybdotoxin (300 nmol/l) and α-dendrotoxin (100–200 nmol/l), blockers of SKCa channels, KCh channels and Kv channels respectively. Iberiotoxin (300 nmol/l), a specific inhibitor of BKCa channels, had no effect on the thrombin-stimulated 86Rb+ efflux. Although glibenclamide, an inhibitor of KATP channels, inhibited the thrombin-stimulated efflux, it did so only in a high concentration (20 μmol/l). 4. Ionomycin (1–5 μmol/l) stimulated an increase in 86Rb+ efflux from the platelets in a concentration-dependent manner. This efflux was significantly inhibited by apamin (100 nmol/l) and charybdotoxin (300 nmol/l). However, iberiotoxin (300 nmol/l) had no effect on the ionomycin-stimulated 86Rb+ efflux. 5. These findings suggest that 86Rb+ efflux from platelets stimulated by thrombin and ionomycin occurs via two types of KCa channel: SKCa and KCh channels. Thrombin also stimulated efflux via Kv channels.

We have investigated the interaction of azapropazone with phenytoin in five healthy volunteers. From steady‐state plasma phenytoin concentrations of about 17 mumol/l there was at least a two‐fold increase following the introduction of azapropazone. The main mechanism of the interaction was a decrease in phenytoin clearance, attributable to competitive inhibition by azapropazone of phenytoin p‐hydroxylation. Protein‐binding of phenytoin in the plasma (as assessed by salivary phenytoin concentrations) was significantly reduced from 92 to 90% by azapropazone and similar changes occurred in in vitro studies of [3H]‐ phenytoin protein binding.

We recently observed a patient in whom we suspected an interaction of azapropazone with phenytoin resulting in phenytoin toxicity. This and a report of a similar case' led us to investigate the effect of azapropazone on steady-state plasma phenytoin concentrations in five healthy volunteers. Subjects and results INITIAL CASEA 60-year-old man who had had grand mal epilepsy for three years was receiving maintenance treatment of phenytoin 300 mg daily. He was given azapropazone 600 mg twice daily for arthralgia and two weeks later developed increasing confusion, nausea, diplopia, and vertigo. Examination showed nystagmus on lateral gaze. Plasma phenytoin concentration was 148 Hmol/l (37 isg/ml). Phenytoin and azapropazone were stopped, and his condition returned to normal within a week. Phenytoin 300 mg daily was restarted without recurrence of toxicity. Two months later his plasma phenytoin concentration was 32 ,umolll (8 tLg/ml). rose to 27 JLmol/l (6&7 itg/ml) and then fell to 10 jLmol/l (2-5 jLg/ml). CommentAzapropazone added during steady-state administration of phenytoin doubled plasma phenytoin concentrations. The initial fall in phenytoin concentrations when azapropazone was added, followed by a gradual rise, resembled the results of Neuvonen et al,3 who observed similar changes in epileptic patients starting phenylbutazone, a pyrazolidine derivative related to azapropazone. Phenytoin is 90% and azapropazone 95% plasma protein bound, and probably azapropazone displaces phenytoin from protein-binding sites, leading to an increase in the free fraction of phenytoin in the plasma and an increase in the rate of clearance of total phenytoin with a decrease in plasma total phenytoin concentration. The reverse would happen on withdrawal of azapropazone (see figure).The subsequent rise in plasma phenytoin concentrations was probably due to decreased clearance of phenytoin. Metabolism of phenytoin is the main mechanism of clearance of the drug, and there is good evidence that some drugs-for example, chloramphenicol and isoniazid-inhibit metabolism of phenytoin.4 Azapropazone decreases the rate of clearance, and therefore presumably inhibits the metabolism, of tolbutamide.5 An effect on absorption of phenytoin is unlikely since phenytoin (Epanutin) is almost completely absorbed. We cannot rule out altered tissue distribution occurring during this interaction, but that alone would not account for the changes in steady-state plasma phenytoin concentrations during azapropazone treatment.We used doses of phenytoin that produced initial plasma phenytoin concentrations well below the therapeutic range (40-80 ,umol/l; 10-20 pg/ml). The concentrations doubled during azapropazone treatment, and even greater changes might be expected in patients starting with concentrations within the therapeutic range because of the non-linearity of phenytoin pharmacokinetics. This is therefore a potentially dangerous interaction. We advise avoiding azapropazone in patients treated with phenytoin.The Committee

1 m-Chlorophenylpiperazine (m-CPP), a S-HTic-receptor agonist, induces migrainelike headaches when taken orally by migraine sufferers.

1. Na+/K+/2Cl− co-transport mediates a bidirectional symport of Na+, K+ and Cl−. The important properties of the co-transport system are its requirement for Na+, K+ and Cl− and its inhibition by loop diuretics such as bumetanide. This co-transporter has been described in a number of animal and human tissues. However, its presence in human platelets, although inferred, has not been demonstrated directly. 2. We have studied the efflux of 86Rb+ (a marker for K+) from Rb+-loaded platelets, and have defined their response to stimulation by high concentrations of external K+. 3. KCl (30–120 mmol/l) stimulated a concentration-dependent increase in 86Rb+ efflux from the platelets. This efflux was completely inhibited by bumetanide (10 μmol/l) but was insensitive to ouabain and R(+)-[(dihydroindenyl)oxy]alkanoic acid. It also required Cl− in the external medium, but did not depend on the presence of extracellular Na+. 4. These observations suggest that 86Rb+ efflux from platelets stimulated by external K+ occurs via Na+/K+/2Cl− co-transport acting in a K+/K+ (K+/Rb+) exchange mode. 5. Non-stimulated efflux of 86Rb+ from the platelets (i.e. in the presence of 5 mmol/l K+) had the characteristics of Na+/K+/2Cl− co-transport acting in normal mode.