Jones, E., Koyama, T., Ho, R.H., Kuttesch, J., Shankar, S., Whitlock, J.A., . . . Frangoul, H. (2007). Safety and efficacy of a continuous infusion, patient-controlled antiemetic pump for children receiving emetogenic chemotherapy. Pediatric Blood and Cancer, 48, 330–332.

To evaluate the safety and efficacy of antiemetic medication administration via patient-controlled infusion pump

Patients who failed a standard antiemetic regimen (including a scheduled 5HT3 and one or more additional breakthrough medication of a different class) were placed on a patient-controlled pump containing diphenhydramine 4 mg/ml, lorazepam 0.16 mg/ml, and dexamethasone 0.27 mg/ml mixed in 30 ml of normal saline. The pump ran with a basal rate of 0.1–0.2 ml per hour with a demand dose of 1–2 ml every 30 minutes. The infusion pump was used along with a scheduled 5HT3 agent.

• N = 30  
• MEDIAN AGE = 14.5 years (range = 4–18 years)
• MALES: 47%, FEMALES: 53%
• KEY DISEASE CHARACTERISTICS: Twenty-five patients with solid tumors (sarcoma and neuroblastoma) and five with acute leukemia
• OTHER KEY SAMPLE CHARACTERISTICS: 72% received highly emetogenic chemotherapy and 28% received moderately emetogenic chemotherapy. Dexamethasone was not used in 30% of the cycles for various reasons.

• SITE: Single site    
• SETTING TYPE: Not specified    
• LOCATION: Nashville, TN

• PHASE OF CARE: Active antitumor treatment
• APPLICATIONS: Pediatrics

This was a quasi-experimental trial with no comparison group. Data were from a retrospective chart review of all patients who used an antiemetic pump to control chemotherapy-induced nausea and vomiting (CINV) between March 1999 and April 2004.

Complete response (CR) was defined as no emesis and no rescue medications. If emesis occurred but no rescue medications were used, it was considered a partial response (PR). Failure was defined as the use of rescue medications. Total patient days in CR, PR, and failure were counted for cycles with and without the antiemetic pump. The total number of rescue medications used and the total number of hospital days also were compared.

Patients with an antiemetic infusion pump had more days in CR than without (45 versus 21 days). A larger number of rescue medications were required to control breakthrough nausea and emesis for the conventional prophylaxis versus the antiemetic pump (108 versus 24). Patients with an antiemetic pump had significantly shorter hospital stays.

Using an antiemetic continuous-infusion pump with demand dosing may reduce the total daily dose of each antiemetic medication used individually. Adverse events occurred in 4% of courses, and all side effects resolved with a decreased rate of infusion. No patients required the discontinuation of an infusion pump.

• Small sample (< 100)
• Baseline sample/group differences of import 
• Risk of bias (no control group)
• Risk of bias (no blinding)
• Risk of bias (no random assignment) 
• Unintended interventions or applicable interventions not described that would influence results
• Measurement validity/reliability questionable
• Other limitations/explanation: Retrospective study data; potential for placebo effect of patient-controlled pump

Preliminary data showed that using an antiemetic infusion pump with basal and demand dosing may be an effective way to manage CINV in pediatric patients who fail standard antiemetic therapy. Additional research is needed.