Burr M, L., King S, Davies H, E., et al. The effects of discontinuing long-term diuretic therapy in the elderly. Age & Ageing 1977;6(1):38-45
PubMed link: https://www.ncbi.nlm.nih.gov/pubmed/402800
| Methods | Study design: Randomized double-blind placebo-controlled study with a parallel-design
Number of groups: Two groups |
| Participants | Number of participants: 106 randomized, 89 completed
· Intervention deprescribing group: 54 randomized, 41 completed · Control group: 52 randomized, 48 completed Age: 80.5 years Sex: 93 female, 13 male Participants with dementia: No Inclusion criteria: · Receiving diuretics for more than one month · Assessed as discontinuation of the treatment might not be unsafe Exclusion criteria: · Congestive cardiac failure during the previous three months · History of left ventricular failure · Hypertension controlled in hospital by diuretic therapy · Diuretics prescribed for a nephritic syndrome or glaucoma Country: United States of America Setting: Hospital – long-stay wards of six hospitals |
| Interventions | Medicine: Diuretics and potassium supplementation withdrawal
· Frusemide · Frusemide and spironolactone · Frusemide and Moduretic · Moduretic · Navidrex K · Clopamil · Chlorthalidone All the patients taking clopamide and chlorthalidone, and all except two of those on frusemide alone, were also receiving potassium supplements Withdrawal schedule: Not described Comparator: Placebo compared to continued medicine |
| Outcomes | Blood pressure and pulse
Distribution of plasma potassium levels Distribution of plasma urea levels Changes in ankle edema |
| Dates | Dates: Not described
Follow-up duration: 12 weeks |
| Funding sources | The authors state that they thank “the following pharmaceutical manufacturers for supplies of active and placebo tablets: CIBA Laboratories, Geigy Pharmaceuticals Ltd, Hoechst Pharmaceuticals Ltd, Merck Sharp & Dohme Ltd, Sandoz Products Ltd, and Searle Laboratories.” |
| Notes |
Risk of bias table
| Bias | Authors’ judgment | Support for judgment |
| Random sequence generation (selection bias) | Unclear risk | Randomization method was not described.
“Other patients were randomly allocated into two groups.” |
| Allocation concealment (selection bias) | Unclear risk | Randomization method was not described. |
| Blinding of participants and personnel (performance bias) | Low risk | “One group continued to receive the same treatment as before while the other group had their diuretic drugs replaced by placebo tablets of similar appearance. Potassium supplements were similarly replaced by placebo tablets for patients allocated to the placebo group. The active and placebo tablets were supplied individually in special containers so that the medical and nursing staff were unaware of the group to which each patient was assigned. Patients witheach type of diuretic were randomized separately so as to produce balanced groups.
“Before the code was broken at the end of the 12-week period the ward sisters were asked their opinions aboutthe identity of the tablets each patient had received…” “…shows that they were unable to distinguish between patients on the active and placebo tablets.” |
| Blinding of outcome assessment (detection bias) | Low risk | As above. |
| Incomplete outcome data (attrition bias) | High risk | The analysis was as-treated, even though the reason for drop-outs related to the intervention in many cases. |
| Selective reporting (reporting bias) | Unclear risk | The intended outcomes were not clearly stated. The outcomes are listed below:
Cardiac failure Blood pressure General condition was not reported,although results presented on whether ward nurses could identify differences that would rely on some judgment of general condition. |
| Other bias | High risk | There appears to be no conflict of interest. This study is quite old and predates the introduction of many first-linedrugs e.g. ACE-inhibitors. This does not introduce bias but does reduce the relevance of the findings to 2015 medical practice. |
Leave a Reply