George P. Lewis

The epidemiology of digoxin: A study in three boston hospitals☆

Two thousand and ninety-eight consecutive patients have been studied in context of a comprehensive surveillance program which monitored all the drugs which they received. The results with digoxin given to 441 consecutive patients have been reported. The drug was judged to have been effective in 91.7 per cent and poor in 2.8 per cent of all patients in whom a consistent evaluation was made. Adverse reactions were reported in 18.4 per cent of all patients receiving the drug, and these were confirmed in 82.8 per cent of all cases investigated. The major reaction rate was 3.7 per cent. Increased admission weight (p < 0.05), potassium depletion due to diuretics (p < 0.01), and increased admission BUN levels (p < 0.05) were significantly associated with toxicity attributed to digoxin. Within the dosage ranges employed in this study, adverse reactions were not related to mean daily dosage, age, sex, or race. The same held true for efficacy which in addition was unrelated to weight. No deaths were attributed to digoxin. On scanning all drugs given in association with digoxin, toxicity attributed to the drug was found to be significantly associated with the following: meperidine, morphine, heparin, hydrochlorothiazide (all p < 0.05), furosemide, aminophyllin (all p < 0.01), and prochlorperazine (p < 0.001). The findings with meperidine, morphine, and heparin were unexpected, and for the present, are reported without comment. It is suggested that aminophyllin might have an irritative effect on the myocardium and thus potentiate digoxin toxicity. It is emphasized that associations do not necessarily imply cause and effect relationships and that in the context of the drug surveillance program, they could be due to chance, or to the effects of hidden variables. The interaction between digoxin toxicity and the diuretics, hydrochlorothiazide and furosemide, quantitatively supports current clinical and pharmacological knowledge. Prochlorperazine was found to be associated with toxicity because of its use in the treatment of nausea and vomiting caused by digoxin. Given the methodology employed by the surveillance system, this was logical and expected.

Diurnal blood pressure in patients with mild‐to‐moderate hypertension treated with once‐daily benazepril hydrochloride

This study evaluated the blood pressure effects of administration of once daily oral benazepril hydrochloride, a new angiotensin‐converting enzyme (ACE) inhibitor, for mild‐to‐moderate hypertension. After a 2 to 4 week placebo baseline period, patients with diastolic blood pressure between 95 and 114 mm Hg, were randomized to receive either placebo or benazepril hydrochloride, 5, 10, 20, or 40 mg, once daily in double‐blind fashion for 28 days. Blood pressure was measured predose and at 1, 2, 3, 4, 6, 8, 12, 16, 20, and 24 hours after the dose during inpatient observation days at the end of the placebo baseline period, and on the first and last day of the double‐blind treatment period; and 24 hours after the dose at weekly outpatient visits. All doses of benazepril hydrochloride resulted in clinically important reductions in diastolic and systolic blood pressures that lasted between 12 and 24 hours after both the first dose, and following the last dose after 4 weeks of treatment. The findings indicate that benazepril hydrochloride may be clinically useful as once‐daily monotherapy in many patients with hypertension.

THE ROLE OF GENETIC FACTORS AND SERUM PROTEIN BINDING IN DETERMINING DRUG RESPONSE AS REVEALED BY COMPREHENSIVE DRUG SURVEILLANCE

Individual variation in response to drugs has long been recognized at the clinical level. Such variation results from a combination of numerous genetic and acquired factors of which relatively few have as yet been precisely defined. The specialty of pharmacogenetics emerged when a number of distinct examples of variation in response to specific drug challenges were explicable on a genetic basis.’ Careful metabolic and pharmacokinetic studies of many such phenomena revealed the existence of distinct enzyme or metabolic differences sufficient to explain the observed variation in drug response.2 Additional valuable information concerning the influence of hereditable factors on drug response can be obtained by approaching the problem from the opposite direction. Such an approach has already been utilized to define subpopulations showing an increased propensity to the development of certain disease states, e.g. the finding of unusual ABO blood type distributions in patients with diseases such as gastric carcinoma, duodenal ulcer, rheumatic heart disease, pernicious anemia, diabetes mellitus, and ~ t h e r s . ~ A unique opportunity allowing the study of genetic characterization and variation in response to drugs arose when a comprehensive drug surveillance program was started in Boston in 1966. Initially it involved all patients admitted into two 30 bed wards at the Lemuel Shattuck Hospital, B o ~ t o n . ~ Since this time, the program has gradually expanded so that currently 12 medical and two psychiatric wards involving eight hospitals in three countries are under surveillance: To each ward is attached a specially trained nurse (nurse monitor) who, using self-coding forms, collects all required data. In addition to obtaining precise details as to each patient’s vital statistics, past drug experience, and definitive diagnoses, she tests patients for certain genetically determined traits such as color blindness and phenylthiourea tasting, and draws blood, (which she sends to a laboratory operating within the program) for the defining of a genetic profile. Included in this profile is measurement of the total serum protein and albumin fraction.

A new method for assessing the clinical effects of oral analgesic drugs

A randomized double‐blind trial comparing aspirin and a placebo in the treatment of mild to moderate pain was introduced into a comprehensive drug surveillance program. The programs standard methods of data collection were judgments of attending physicians on the efficacy and adverse effects of drugs. The napsylate and hydrochloride salts of propoxyphene were also compared. The already established effectiveness of aspirin was demonstrated. No evidence of a difference in efficacy between the two salts of propoxyphene was found. A randomized, double‐blind, large‐scale evaluation of certain marketed and unmarketed drugs within a comprehensive drug surveillance program is feasible and can be of considerable value in their clinical assessment.

Evaluation of haptoglobin phenotype 0-0 in cirrhotic and non-cirrhotic hospital populations.

The haptoglobin phenotypes of 3,332 individuals, consisting of 2,930 caucasians and 402 negroes living in the greater Boston area, were determined. Of these, 3,222 were hospitalized medical patients fully documented regarding diagnoses. One hundred and twentyeight of the total population studied were shown to exhibit starch gel anhaptoglobinaemia (3·7%). Re-evaluation on acrylamide gel of 118 0-0 samples revealed that the majority (94%) were derived from patients exhibiting hypohaptoglobinaemia rather than anhaptoglobinaemia.