Prospective Randomized Open Blinded End-point (PROBE) Study. A novel design for intervention trials

A novel design for intervention studies is presented, the so called PROBE study (Prospective Randomized Open, Blinded End-point). This design is compared to the classical double-blind design. Among the advantages of the PROBE design are lower cost and greater similarity to standard clinical practice, which should make the results more easily applicable in routine medical care. Since end-points are evaluated by a blinded end-point committee it is obvious that there should be no difference between the two types of trials in this regard.

Hypertensive Retinal Vascular Changes: Relationship to Left Ventricular Hypertrophy and Arteriolar Changes Before and After Treatment

Eye-ground-photos were taken in twenty-eight previously untreated men with mild to moderate essential hypertension. The same eye was evaluated before and after 26 weeks of double-blind treatment with Enalapril or Hydrochlorothiazide. The vascular changes were assessed by using a more elaborate and refined grading than the Keith-Wagener-Barker scale. All photos were examined by the same observer without knowledge of blood pressure, type of treatment or the order in which the photos had been taken. There were significant positive correlations between the vascular alterations in the retina in the untreated state and left ventricular wall thickness (echocardiography), minimal vascular resistance in the calf (plethysmography) and blood pressure respectively. Treatment with Enalapril decreased the reflection of the retinal arterial wall significantly and reduced the narrowing of arteries and arterio-venous crossing phenomena non-significantly. Hydrochlorothiazide did not affect any of the retinal vascular changes. It can be concluded that this relatively simple technique of evaluating eye-ground-photos with a new grading scale, when used in non-malignant hypertension, gives a useful assessment of the degree of hypertensive target organ damage in the retina as well as in other important target organs, i.e. the heart and vascular beds. In addition, Enalapril positively affects hypertensive retinopathy in contrast to Hydrochlorothiazide, reflecting what happens to structural cardiovascular changes in the rest of the body.

Cost of treatment of schizophrenia in six European countries

BACKGROUND AND AIMS

As part of an RCT in six European sites, the direct mental health care cost for 422 patients with schizophrenia was analysed according to how total and medication costs differed across sites and which variables were likely to predict total or service-specific costs.

METHOD

Service use was recorded continuously during a 12-month follow-up. Prescribed psychotropic medication was recorded at baseline and 12 months later. Service use data were transformed into EURO, log-transformed and analysed using linear regression models.

RESULTS

Although samples were homogeneous, large inter-site cost differences were found (annual means ranging from 2958 euro in Spain up to 36978 euro in Switzerland). Psychopharmacologic costs were much more constant across sites than costs for other services. Total costs were associated more with region or socio-demographic characteristics than with disorder related parameters.

CONCLUSIONS

The findings confirm remarkable differences in direct costs of patients with schizophrenia across Europe. However, the relative stability of medication costs suggests a need to analyse mechanisms that influence service-specific costs for schizophrenia.

Circulatory effects of stress in essential hypertension

Hemodynamics and plasma catecholamines were studied during mental arithmetics and cold pressor test in thirteen patients with essential hypertension. Blood pressure was significantly increased by both forms of "stress". Cold pressor test caused a significant increase in total peripheral resistance, while mental arithmetics increased heart rate and cardiac output significantly. Thus, different forms of "stress" can increase blood pressure by totally different hemodynamic mechanisms. Acute cardioselective and non-selective beta-receptor blockade did not affect the hemodynamic reaction pattern during cold pressor test. The increase in heart rate and cardiac output during mental arithmetics was blocked by non-selective but not with cardioselective beta-blockade. Both beta-blockers were, however, inefficient in preventing the blood pressure elevation induced by mental arithmetics. Plasma noradrenalin was significantly increased during mental arithmetics after both cardioselective and non-selective beta-blockade compared to placebo. During cold pressor test there was a significant increase in noradrenalin only after non-selective beta-blockade. Adrenalin in plasma was significantly increased only during mental arithmetics after cardioselective beta-blockade. The pathogenesis of essential hypertension is complex and still not fully understood. Both hereditary (1) and environmental factors such as salt intake (2) and stress (3,4,5) are probably of importance as risk factors for essential hypertension. Animal studies have shown that different forms of stress can produce permanent hypertension due to structural changes in the resistance vessels (6). Against this background we decided to investigate acute hemodynamic effects of two well-defined types of "stress", e.g. mental arithmetics and cold pressor test in patients with essential hypertension and to study if the hemodynamic reactions could be modified by cardioselective and non-selective beta-receptor blockade. Catecholamines in plasma were studied before, during and after "stress".

Circulatory effects of noise

Thirteen patients with mild essential hypertension, mean age 44 years (range 21-59), were studied during "stress" before and after postsynaptic alpha-adrenoceptor blockade and combined postsynaptic alpha- and non-selective beta-adrenoceptor blockade. Loud broad band noise (100 dBA for 10 min) was used as the stress stimulus. Exposure to noise caused a significant increase in systolic (7%, p less than 0.05), diastolic (9%, p less than 0.01) and mean arterial pressure (6%, p less than 0.01). The blood pressure elevation was caused by an increase in total peripheral resistance (12%, p less than 0.05). There was no significant change in heart rate, stroke volume or cardiac output. The blood pressure response during noise stimulation was not affected by postsynaptic alpha-adrenoceptor blockade (prazosin, 2 mg orally). The hemodynamic reaction pattern, however, was totally reversed. Thus, the cardiac output increased significantly (9%, p less than 0.05), while the total peripheral resistance tended to decrease. Combined postsynaptic alpha- and non-selective beta-adrenoceptor blockade (labetalol, 200 mg orally) inhibited the increase in systolic blood pressure caused by noise, while the diastolic and mean arterial pressures still increased significantly (5%, p less than 0.01). Labetalol effectively blocked the stress-induced increase in total peripheral resistance and there was no significant increase in cardiac output after combined alpha- and beta-adrenoceptor blockade. Exposure to noise caused a significant increase in circulating noradrenaline (20%, p less than 0.05). Plasma adrenaline and plasma renin activity were not affected by noise stimulation. These results suggest that blood pressure elevation is essential during "stress" but that the hemodynamic pattern causing blood pressure elevation may vary and may be affected by pharmacological blockade of various parts of the sympathetic nervous system.

Captopril or atenolol in essential hypertension

Twenty-five patients with essential hypertension were randomly assigned to treatment with either captopril or atenolol. There were 15 males and 10 females and their mean age was 53 years, range 32-66 years. Systolic and diastolic blood pressures were significantly reduced by atenolol 50-100 mg once daily. Captopril 25-50 mg 3 times daily caused a significant decrease in supine diastolic but not in systolic blood pressure. After the addition of hydrochlorothiazide (25-50 mg) to those who did not become normotensive (supine diastolic blood pressure less than 95 mm Hg) on captopril or atenolol alone, blood pressure was further reduced. Captopril combined with hydrochlorothiazide significantly reduced both supine and standing blood pressure by 31/17 mm Hg (p less than 0.01) and 33/18 mm Hg (p less than 0.001) respectively. Atenolol combined with hydrochlorothiazide caused a significant decrease of both recumbent and standing blood pressure by 21/10 mm Hg (p less than 0.01) and 23/13 mm Hg (p less than 0.05 systolic, p less than 0.001 diastolic). The reduction of systolic blood pressure was significantly better with the captopril/hydrochlorothiazide combination, while there was no difference between the groups as regards diastolic blood pressure reduction. Thus, hydrochlorothiazide potentiates the blood pressure lowering effect of captopril more than of atenolol. This could be due to a synergistic interaction between captopril and hydrochlorothiazide.

Mode of action of beta-adrenoceptor blocking agents in hypertension. A comparison between metoprolol and pindolol with special reference to peripheral vascular effects

beta-Adrenoceptor blocking drugs are generally recognized as being effective in the treatment of hypertension. The mechanisms whereby these drugs reduce blood pressure are, however, not fully understood. In a double-blind, randomized study either metoprolol, 100--300 mg/day, or pindolol, 5--15 mg/day, was given for 6 months and the effects on blood pressure, heart rate and vascular resistance in the calf and forearm were investigated. Measurements were made at rest, during and after physical exercise, and during postischaemic hyperaemia. Both drugs reduced blood pressure to the same extent both at rest and during and after exercise. Metoprolol reduced heart rate to a greater extent than pindolol at rest and after exercise, whereas no difference was seen during physical exercise. Pindolol reduced the vascular resistance in the calf at rest by 14% (p less than 0.05), whereas metoprolol tended to increase vascular resistance, the difference in effect being highly significant (p less than 0.005). During and after leg exercise, there was no difference in forearm vascular resistance between the two drugs. It may be concluded that pindolol reduced resting blood pressure partly through peripheral vasodilatation. This was probably an effect of beta 2-adrenoceptor stimulation linked to the pronounced intrinsic sympathomimetic activity (ISA) of pindolol. Metoprolol on the other hand, acted mainly through cardiac mechanisms, as suggested by its pronounced reduction of heart rate.

Hemodynamic and hormonal changes induced by noise

Eighteen healthy male volunteers with normal hearing were exposed to industrial noise at different sound levels (75, 85 and 95 dB A) in a noise laboratory. Blood pressure, heart rate, stroke volume and cardiac output were recorded with noninvasive techniques. Adrenaline and noradrenaline concentration in venous plasma were analyzed before and during noise exposure. The mean resting blood pressure of the whole group was 120/70 mm Hg. During noise stimulation diastolic blood pressure increased (12.2%, p less than 0.001) as did mean arterial pressure (6.6%, p less than 0.001) and total peripheral resistance (12.7%, p less than 0.001). Stroke volume (7.3%, p less than 0.001) and cardiac output (5.0%, p less than 0.01) were both reduced at 95 dB A. Heart rate and systolic blood pressure did not change significantly. At 75 and 85 dB A there were similar but smaller changes in the hemodynamic parameters. There were no changes in adrenaline and noradrenaline in plasma during maximal noise exposure. The noise induced hemodynamic changes remained 5 minutes after the noise stimulation was stopped but had disappeared after 10 minutes of rest.

Predictors of stroke in the elderly

Hypertension, diabetes mellitus, coronary heart disease and cigarette smoking have repeatedly been identified as risk factors for stroke in young and middle-aged individuals. In order to find predicting factors for stroke in the elderly we assessed health characteristics in 55 stroke victims in the age range 65-75 years (mean 70.7 +/- 2.7) allocated to our stroke unit at Ostra University Hospital in Gothenburg. For comparison we used data from 2,009 individuals participating in the ongoing longitudinal population study "70-year-old people in Gothenburg, Sweden". Among the stroke victims we found a higher prevalence of hypertension (63.5% vs. 27.8%, p less than 0.001), diabetes mellitus (21.8% vs. 6.2%, p less than 0.001) and a history of previous myocardial infarction (12.7% vs. 4.8%, p less than 0.01), thus confirming previous findings. There was no difference with regard to smoking habits (32.7% vs. 27.5%, NS), which is at variance with findings in the young and middle-aged.

Long-term hypotensive effect of atenolol (ICI 66.082), a new beta-adrenergic blocking agent

A report is given from an on-going multicenter trial in Sweden, in which 117 hypertensive patients have been treated with a new cardioselective beta-adrenergic blocking agent, atenolol (ICI 66.082, Tenormin) for an average of six months (range 2-21). Statistically significant reductions of BP were observed, recumbent by 29/19 mmHg (p less than 0.0001) and standing by 28/18 mmHg (p less than 0.0001). Few and comparatively mild side-effects were seen.

Malignant hypertension--improving prognosis in a rare disease

A follow-up was made of 69 patients diagnosed as having malignant hypertension during 1969--76, essential in 26, secondary in 39 and unclassified in 4. A clear male dominance was seen (41 men, 28 women), particularly in the group with essential hypertension (18 men, 8 women). The mortality in this series was less than in previously published series. Thus, the 5-year survival rate was 75% in the patients with essential and 72% in those with secondary hypertension. In part this was due to haemodialysis and renal transplantation. The importance of renal function at the time of diagnosis was evident in this study. In most patients with essential hypertension and serum creatinine levels below 300 mumol/l, renal function could be maintained or improved when antihypertensive treatment was instituted, whereas progression of the renal damage was seen in those with serum creatinine levels above 300 mumol/l in spite of antihypertensive treatment with 3 or more drugs. The incidence of new cases of malignant hypertension tended to decrease during the observation period, particularly as regards essential hypertension.

Long-term effects of captopril and atenolol in essential hypertension

Fifty patients with mild or moderate essential hypertension were randomized (double-blindly) to treatment with either captopril (n = 26) or atenolol (n = 24). Their mean supine diastolic blood pressure after placebo was 100-125 mmHg. The study included an initial dose finding phase (12 weeks) during which the dosages of captopril and atenolol were increased stepwise every second week in order to obtain normotension (supine diastolic blood pressure less than 95 mmHg). Hydrochlorothiazide was added when necessary. During the second phase of the study the patients were followed on active treatment for 2 years. After the initial 12 weeks of active treatment, recumbent and standing blood pressures had fallen significantly both in the captopril group (by 31/20 and 33/19 mmHg, p less than 0.001) and in the atenolol group (by 24/18 and 30/20 mmHg, p less than 0.01 (systolic), p less than 0.001 (diastolic)). The antihypertensive effect was maintained in both groups during long-term treatment. The antihypertensive effect of both agents was potentiated to the same extent by addition of hydrochlorothiazide. Side-effects were few and mild. It can be concluded that both captopril and atenolol are safe and effective antihypertensive drugs.