Buijs, C., Mom, C.H., Willemse, P.H., Marike Boezen, H., Maurer, J.M., Wymenga, A.N., … Mourits, M.J. (2009). Venlafaxine versus clonidine for the treatment of hot flashes in breast cancer patients: A double-blind, randomized cross-over study. Breast Cancer Research and Treatment, 115, 573–580.

DOI Link

Study Purpose

Comparison of venlafaxine versus clonidine for the treatment of hot flashes with regard to side effects, efficacy, QOL, and sexual functioning in patients with breast cancer.

Intervention Characteristics/Basic Study Process

Patients randomly assigned to receive venlafaxine for eight weeks, followed by a wash-out period of two weeks, then eight weeks of clonidine or vice versa

Sample Characteristics

N = 60 (30 assigned to venlafaxine and 30 to clonidine). Patients with a primary or metastatic breast cancer age 60 years or younger, allowed antitumor hormonal treatment if started one month prior and continued taking throughout treatment period.

Study Design

Double-blind, randomized, cross-over study.

Measurement Instruments/Methods

Assessment took place before the start of each drug, then at 2, 8, 12, 18 weeks after treatment began. Six questionnaires were used to compare the drugs' effects on adverse events, efficacy, QOL, and sexual functioning: daily diary on hot flashes, hot flash–related daily interface questionnaire, Medical Outcomes Study Short Form (SF-36), sexual activity questionnaire, and Zung Self-Rating Depression Scale.

Results

Forty patients completed all treatments, 12 patients only one treatment, 8 patients neither. Dropout rates during venlafaxine were 15 out of 59, versus clonidine, 5 out of 53. Withdrawal rateswere not affected by sequence of treatment. Efficacy: After eight weeks, no difference was seen between the two drugs in reduction of hot flash scores: median 49% for venlafaxine and 55% for clonidine. The drug that the patient received first caused the greatest reduction in hot flash score.

Limitations

Statistics: based on the testing used to calculate the number of patients needed to detect differences, the sample size was too small to detect difference, which may be the reason no statistical difference was  found between the interventions.