BackgroundFinding appropriate analgesics is important for a mother's recovery after a caesarean section. The aim of this study was to compare the analgesic effect of spinal morphine and fentanyl for women undergoing a caesarean section. MethodsIn this randomised, unmasked, parallel-group, controlled trial, eligible participants were women undergoing caesarean section with spinal anaesthetic at the Al-Helal Al-Emirati Hospital, Rafah, Gaza Strip, occupied Palestinian territory. Using a manually generated allocation sequence that was independently managed by medical staff, we randomly assigned women (1:1) to receive either 0·2 mg preservative-free morphine or 20 μg fentanyl, spinally, combined with 2 mL of 0·5% hyperbaric bupivacaine for spinal anaesthesia. Pain was recorded by visual analogue scales (VAS; range 0–10, where 0 means no pain and 10 means pain is intolerable) at 1 h, 6 h, 12 h, 18 h, and 24 h after the caesarean section. Time and amount of other analgesics needed were recorded (primary outcome). Secondary outcomes were nausea-vomiting scores (NVS 0–3, where 0 means no nausea or vomiting and 3 means severe nausea or vomiting that is unresponsive to antiemetics), and pruritus scores (PS 0–2, where 0 means no pruritus and 2 means severe pruritus). Data were analysed with SPSS. We compared outcomes using Mann-Whitney U-test, student's t-test, OR, or χ2 test. The study was approved by the ministry of health and the Helsinki Committee, Gaza Strip. Women provided verbal informed consent before participation. FindingsWe enrolled 121 women (mean age 28·43 years [SD 5·62]). 59 women were randomly assigned to the morphine group, and 62 women were randomly assigned to the fentanyl group. Pain was reduced with morphine compared with fentanyl at 1 h (VAS score: 1·81 for morphine vs 2·57 for fentanyl; p<0·039), at 6 h (1·84 vs 5·08; p<0·0001), at 12 h (1·45 vs 3·30; p<0·0001), at 18 h (1·62 vs 3·18; p<0·0001), and at 24 h (1·61 vs 2·68; p=0.0001). The time to needing other analgesic drug was longer in the morphine group than in the fentanyl group (9·03 h vs 2·46 h; p<0·0001). 20 (34%) women in the morphine group did not need analgesics, whereas all (100%) women with fentanyl needed additional analgesics (odds ratio [OR] 2·59, 95% CI 2·03–3·31). Mild nausea and drowsiness were seen more in the morphine group than in the fentanyl group at 6 h and 12 h, respectively (p=0·010). Nine (15%) women in the morphine group had pruritus and needed treatment compared with one (2%) woman in the fentanyl group (OR 10·98, 95% CI 1·35–89·62). InterpretationSpinal morphine was more effective than spinal fentanyl for caesarean analgesia. Morphine was associated with more side-effects than fentayl, but these were treatable and did not pose any danger to women. FundingNone.