Background: Control of acute postoperative pain remains a serious problem. Postoperative pain is associated with an increase in thrombotic or respiratory complications. In the other hand the association between surgery, acute postoperative pain and ongoing chronic pain is well defined.

Target: To evaluate the incidence of severe pain after surgery for degenerative lumbar pain, with two analgesic techniques; intravenous analgesia (i.v.) (group 1) and patient controlled analgesia (PCA) (group 2).

Study design: Retrospective study with dates obtains prospectively.

Patient sample: We studied 206 patients operated between january 04 and june 05. Group 1 (intravenous) 80 patients and 126 in group 2 (PCA).

Outcomes mesasures: Percentage of patients with severe pain, number of rescues and complications.

Materials and methods: The type of postoperative analgesia administrated was decided by the anaesthesiologist. To measure pain intensity the VAS was assessed every 6 hours and worst score was used, excluding recovery room. Type of rescue analgesia was the same in all patients and these was recommended in VAS > 3. We evaluate sex, age and comorbidity by ASA. We have defining analgesic ‘failure’ by the overall incidence of pain intensity in two categories: the percentage of patients who experienced moderate-severe pain (VAS > 3) and severe pain (VAS> or = 7). As the pain scores were not normally distributed we not used mean and SD of VAS. The number of rescues and complications were also evaluated.

Results: There was not differences in median age (group 1 50’85 sd 15’4; group 2 52’44 sd 15’4 p=0’47), ASA (group 1 1’89 sd 0’75; group 2 1’90 sd 0’57 p=0’88) or diagnosis between two groups. There were differences in percentages of sexes, group 1 with 40% of women and 62’69% in group 2 (p=0’013). There were not differences in incidence of patients with moderate-severe pain (group 1 15/80; group 2 30/126. p=0’392) neither in incidence of patients with severe pain (group 1 33/80; group 2 51/126. p=0’912). There were not differences in number of rescues (p=0’912) neither in number of complications between groups. Global incidence of VAS > 3 were 40’8 but the incidence of rescues were 25’2%

Conclusions: Our findings suggest that both techniques have similar effectiveness, although the global incidence of severe pain was not optimal (21’8%). It is important to remark the different between incidence of patients with VAS > 3 and number of rescues administrated.