Siegel 1992.
| Methods | Randomized, double‐blind, placebo‐controlled trial. Wash‐out period = 1 month | |
| Participants | DBP 95 to 105 mmHg. Mean age range: 58.1 to 62.2 years. All male patients with abnormal resting ECG (e.g. arrhythmias) | |
| Interventions | HCTZ 50 mg/d + potassium 40 mmol/d (N = 32), HCTZ 50 mg/d + potassium 40 mmol/d and magnesium 400 mg/d (N = 35), HCTZ 50 mg/d + triamterene 100 mg/d (N = 32), HCTZ 50 mg/d (N = 66), chlorthalidone 50 mg/d (N = 34) or placebo (N = 33) Trial duration = 8 weeks | |
| Outcomes | 24‐hour Holter monitoring (for arrhythmias). Serum and intracellular potassium and magnesium levels. Data on glucose and insulin levels available from duplicate publication | |
| Notes | A sample size calculation was not provided. All of the patients were males with resting electrocardiographic abnormalities (stratified by the presence or absence of left ventricular hypertrophy). It is not known how many patients were randomized to each treatment group; a range of 42 to 48 per group was given. The study authors stated that there were no statistically significant differences (P value = NS) in the baseline patient demographics and characteristics across all treatment groups. No data on DBP or SBP. WDAEs, mortalities, SAEs and total AEs were not reported Additional publication: Siegel et al. Hypertension 1994;23(part 1): 688‐94; measurement of glucose and insulin levels only |
|
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | "Participants were then assigned to a study medication using a randomised block design, stratified by the presence or absence of left ventricular hypertrophy on ECG." (line 15 from top of p.1084, middle column). No further information was given |
| Allocation concealment (selection bias) | Unclear risk | Not stated by study authors |
| Blinding (performance bias and detection bias) All outcomes | Low risk | "Treatment assignment was blinded from participants, clinicians and laboratory staff by having another member of the staff dispense and count the blindly labelled medications, which were identically packaged." (line 19 from top of p.1084, middle column) Comment: no further information was given |
| Incomplete outcome data (attrition bias) All outcomes | Unclear risk | The primary efficacy analysis was based on an intention‐to‐treat (ITT) technique Exclusions: It is not known if there were any patients who were excluded from the study during the placebo run‐in period prior to randomization Attrition: 21/233 (9%) of patients withdrew from the study (data were not available for each treatment group). Reasons for withdrawing were not specified WDAEs: not given |
| Selective reporting (reporting bias) | High risk | BP was not measured. Except for serum potassium, glucose and insulin levels, biochemical data were not reported on. Mortalities, SAEs and total AEs were not reported |
| Industry sponsorship | Low risk | Supported by grant HL‐36821 and by National Heart, Lung and Blood Institute Preventive Cardiology Academic Award HL‐02081 |