Consecutive inpatients with hypokalemia requiring intravenous potassium replacement were enrolled in the study after they gave informed consent. This consent was given while they were in a combined medical and surgical ward. The study protocol was reviewed and approved by the Hamilton Health Sciences/McMaster University Research Ethics Review Board.
Patients who had been prescribed one or more intravenous mini-infusions of KCl (10 mmol in 100 mL) through the peripheral vein were randomized into 3 groups: (i)10 mmol KCl in 100 mL sterile water, (ii) 10 mmol KCL in 100 mL 0.9% sodium chloride (NaCl), and (iii) 10 mmol KCl in 100 mL 0.45% NaCl. The KCl solutions for all 3 groups were prepared by the hospital pharmacy (Hamilton Health Sciences, McMaster University Medical Centre Site, Hamilton, Ontario) in identical mini-infusion bags. The randomization code was available in the pharmacy in case of infusion- related problems.
An intravenous cannula was inserted at least 1 h before the start of the KCl infusion, and all sites were inspected before the first infusion to ensure that there was no obvious pre-existing local pain, tenderness, or phlebitis. Each 100-mL infusion of KCl (10 mmol) was administered over 1 h with an IV infusion pump, and patients received 1, 2, or 3 infusions, as ordered by their physicians. A numeric rating scale (NRS) ranging from 1 (no pain) to 10 (worst pain) was used to rate pain at the infusion site. The nurse asked patients to rate the severity of pain at the infusion site using the NRS before the start, at 15 min after the start, and at the end of each 10-mmol KCl infusion. The nurse, who was unaware of the randomization code, recorded the results on a reporting sheet designed specifically for this study. At the completion of each mini-infusion, the nurse inspected the infusion site and recorded any signs of redness or heat. The nurse also recorded the name and dose of any other drug that was being infused through the same cannula at the same time.
The osmolality and pH of the KCl solutions were determined for 3 samples of each group of solutions. Osmolality was determined by freezing-point depression with the Advanced Micro-Osmometer model 3MO Plus (Advanced Instruments, Norwood, Massachusetts). The pH was determined with the Accumet Basic Meter (Fisher Scientific, Hampton, New Hampshire).
A study sample size of 23 for each group was calculated, assuming a 25% reduction in pain and an estimated standard deviation of the difference in pain of 30% (a = 0.05, 6 = 0.2). Based on these assumptions, 20 patients were to be enrolled in each group. If there were evidence of significant differences in pain responses when half of the patients had been recruited, an interim analysis would be conducted. The aim of the interim analysis was to ascertain whether the occurrence of significantly greater infusion-site pain in one group was sufficient to require that the study be discontinued. buy levitra 20 mg
The Kruskal-Wallis statistical test was used to analyze the pain scores of the 3 groups. Pain scores were compared between 2 groups with the Mann- Whitney U-test. A p < 0.05 was used as the cut-off for statistical significance. Repeated-measures analysis of variance was used for pain assessment at different time points. Linear regression was used to explore the relationship between solution osmolality and pain scores.