Low-dose intrathecal opioid has been used for early postoperative pain co1ntrol. This study was designed to assess effect intrathecal morphine on postoperative pain control for total knee arthroplasty (TKA) under combined spinal-epidural analgesia (CSE) in elderly patients.

Fifty four patients over 60 years, undergoing TKA were randomly allocated to three groups. M₀ group for control group did not received intrathecal morphine, M₅₀ and M₁₀₀ group received intrathecal morphine 50 µg and 100 µg respectively. The pain scores (verbal numeric rating scale, VNRS) at rest and coughing, analgesic consumption, patient satisfaction and side effects such as nausea, vomiting, pruritus, headache, dizziness, sedation, respiratory depression, and urinary retention were recorded immediately before and at 1, 3, 6, 12, 24, 48 hour after the initiation of patient-controlled epidural analgesia (PCEA).

VNRS were low at each time, and were not exceeding 2 in all groups. M₅₀ and M₁₀₀ group revealed significantly less analgesic consumption compared to M₀ group (P < 0.05). PCEA first injection time after PCEA connection was shortest in M₀ group compared to M₅₀ and M₁₀₀ group. The incidence of pruritus increased in M₅₀ and M₁₀₀ group with dose-dependence, but no significant differences were noticed in other side effects.

Intrathecal morphine use showed no significant analgesic effect except pruritus compared to control group. Further studies are required into the effective intrathecal morphine without side effects in elderly patients for TKA.