LeBlanc, J.K., Al-Haddad, M., McHenry, L., Sherman, S., Juan, M., McGreevy, K., . . . DeWitt, J. (2011). A prospective, randomized study of EUS-guided celiac plexus neurolysis for pancreatic cancer: One injection or two? Gastrointestinal Endoscopy, 74(6), 1300–1307.

To compare the pain relief achieved by one injection of alcohol versus two injections of alcohol during endoscopic ultrasound–guided (EUS) celiac plexus neurolysis (CPN); to compare the safety associated with the one- and two-injection techniques; to determine the extent to which the number of injections affects the onset and duration of pain associated with pancreatic cancer

Each CPN procedure involved injecting 20 ml 0.75% bupivacaine and 10 ml 98% alcohol into either one or two sites at the celiac trunk. The total sample was composed of 50 patients. Investigators conducted follow-up interviews by telephone to evaluate onset and duration of pain relief and complications. Interviews occurred at 24 hours after the procedure and then weekly.

• The sample was composed of 50 patients.
• Mean patient age was 63 years.
• Of all patients, 48% were male and 52% were female.
• All patients had pancreatic cancer.

• Single site
• Outpatient
• Indiana University Medical Center, Indianapolis, Indiana
   

• Phases of care: multiple
• Clinical applications: end of life, palliative care, late effects, survivorship
   

Prospective single-blinded randomized parallel-group study

• Pain relief: Chi-square–measured difference in proportion of pain relief, subjective pain relief, and reduction in the use of pain medication.    
• Onset of pain relief: Time from date of injection until 30% reduction of the pain reported at baseline. Investigators used Kaplan-Meier estimate of median time. Log rank test measured time until pain relief between groups. Mann-Whitney test compared time until pain relief in responders.
• Duration of pain relief: Time from date of injection to the date the pain returned to within 30% of the baseline value. Investigators used Kaplan-Meier estimate of median duration of pain relief. Log rank test compared duration of pain relief between groups and between those who achieved pain relief.
• Predictors of pain relief: Two-sample t tests were used to compare baseline pain score, days until onset of pain relief, and total reduction in pain relief between groups.
• Survival: Survival was calculated from date of injection to date of death.
   

Fifty patients were enrolled and randomized. Pain relief was observed in 37 patients (74%): 20 (69%) in the one-injection group and 17 (81%) in the two-injection group. Median onset of pain relief in both groups was one day. Median duration of pain relief in the one-injection and two-injection groups was 11 weeks and 14 weeks, respectively. Complete pain relief was achieved in four patients (8%), two in each group. No long-term complications occurred.

The two techniques were not associated with any significant difference in patients' experience of pain.

• The study had a small sample, with fewer than 100 patients.
• The study was a single-blinded test.

Authors observed no difference in the safety associated with the two techniques or the survival rate. Nurses would be justified in advocating for the single-injection technique, to minimize patients' discomfort.