Walma EP, Hoes AW, van Dooren C, et al. Withdrawal of long term diuretic medication in elderly patients: a double blind randomised trial. British Medical Journal 1997;315:464-68
PubMed link: https://www.ncbi.nlm.nih.gov/pubmed/9284668
Full text article: https://www.bmj.com/content/315/7106/464.long
| Methods | Study design: Randomized double-blind placebo-controlled study with a parallel-design
Number of groups: Two groups |
| Participants | Number of participants: 202 randomized
· Active deprescribing group: 102 randomized · Control group: 100 randomized Age: 76 ± 1 year Sex: 151 female, 51 male Participants with dementia: No Inclusion criteria · Aged 65 or more · Receiving diuretics for at least six months · No overt heart failure · No hypertension, where hypertensionwas defined asmean of three blood pressure values (two measured at successive home visits and one obtained from the medical file) > 180/100 mm Hg Exclusion criteria: · History of acute heart failure · Symptoms of heart failure during the previous three months · Manifest heart failure · Use of frusemide at dosages over 80 mg daily · Hypertension: mean of three blood pressure values (two measured at successive home visits and one obtained from the medical file) > 180/100 mm Hg · Hypercalciuria · Nephritic syndrome · Glaucoma · Use of fixed combinations of diuretics with beta-blockers or ACE-inhibitors · Combination therapy of alpha-blockers, diuretics, and vasodilators for hypertension · Use of a diuretic for which no placebo was available · Noncompliance during the run-in phase · Regular general practitioners refused to cooperate Country: The Netherlands Setting: Community |
| Interventions | Medicine: Diuretics – frusemide 32%, thiazides (alone or in combination with triamterene) 66%, or triamterene alone in 2%)
Withdrawal schedule: Patients with baseline frusemide dose: · 40 mg daily halved the dose for one week · 80 mg daily halved the doe for two weeks Comparator: Continued therapy compared to placebo |
| Outcomes | Successful deprescribing
Changes in systolic and diastolic blood pressures |
| Dates | Dates: Not described
Duration: Six months |
| Funding sources | Dutch Organization for Scientific Research (NWO, Research Grant nr 92001173). |
| Notes | The authors declare: “Conflict of interest: None.” |
Risk of bias table
| Bias | Authors’ judgment | Support for judgment |
| Random sequence generation (selection bias) | Low risk | After stratification by age (65 – 79 and > 80 years) and type of diuretic, each patient was randomly assigned to placebo (the withdrawal group) or continuation of diuretic therapy (the control group).
Blocks of four sets of study medication each consisted of two placebo and two genuine packages, which were consecutively assigned to enrolled patients. |
| Allocation concealment (selection bias) | Low risk | Randomization lists and numbered sets of study medication were generated by the trial pharmacist of the Academic Hospital, who also produced sealed envelopes with decoding information for emergencies. |
| Blinding of participants and personnel (performance bias) | Low risk | “The patients correctly guessed whether they were allocated to the withdrawal or the control group in 62% of cases while their doctors guessed correctly in 65%. The data analysist (EPW) was blinded until the allocation codes had been entered in the SPSS spreadsheet.”
“The similarity of genuine and placebo tablets ensured the impossibility of recognizing them by color, form, or taste.” |
| Blinding of outcome assessment (detection bias) | Low risk | As above |
| Incomplete outcome data (attrition bias) | Low risk | 100% follow-up |
| Selective reporting (reporting bias) | Low risk | All stated outcomes in the paper are reported. |
| Other bias | Low risk | Independent funding |
Leave a Reply