3.2. Impact of routine vaginal packing after vaginal hysterectomy
Vaginal packing may have little or no effect on vaginal bleeding. In a study involving 77 women, postoperative blood loss was greater in the group receiving vaginal packing (median, 16 [interquartile range, 13.4] g) compared with the control group (5 [8.2] g; p < 0.01) (15). In another study involving 173 participants, the rate of vaginal bleeding was similar among groups (one outcome without packing, relative risk [RR] = 0.34; 95%CI 0.14–8.16) (16).
Vaginal packing probably does not affect postoperative pain, with one study reporting no differences in postoperative pain visual analogue scale (VAS) scores in either the early postoperative period or before discharge. Differences in pain scores among groups did not reach the minimal relevant difference used in the sample size calculation (14 mm): in the immediate postoperative period, a median of 41.6 (25.7) in the packing group versus 46.3 (26.2) in the control group; and at discharge, a median 35.0 (36.0) in the packing group versus 40.0 (39.0) in the control group. More women in the control group needed ketorolac analgesia (median 15 mg [SD 45 mg] in women with packing versus 45 mg [SD 75 mg] in women without [p < 0.001]) (15). Another study did not show differences in McGill Pain Questionnaire scores the morning after the surgical procedure (median 11 [rank IQ, 3–18] in the packing group versus 10 [rank IQ, 4–16] in the control group) (16). Finally, a third trial did not show differences in VAS scores 24 h after the surgical procedure (median 1.0 [1.9] in the packing group versus 0.81 [1.5] in the control group) (13).
Vaginal packing likely does not increase the risk of haematoma. Combined analysis of the three trials describing haematoma as the outcome revealed no statistically significant differences among the studied groups. However, haematomas were more frequent in the group without packing (3.3% versus 6.7%; RR 0.53, 95%CI 0.22–1.31). These results should be considered imprecise given that approximately 450 adverse events would have been necessary to find a significant reduction.
A study involving 173 participants showed a trend towards a greater risk of vaginal haematoma in the no packing group, although differences were not significant (4/86 versus 9/87; RR 0.44, 95%CI 0.14–1.37); 2 of these cases corresponded to an infected vaginal cuff haematoma. This study also showed similar proportions of women suffering from urinary infection in each group (9.3% versus 10.3%; RR 0.88, 95%CI 0.35–2.17) (16). In another study involving 43 women, a non-statistically significant greater proportion of women without packing also developed a haematoma (9.1% versus 19.1%; RR 5.0, 95%CI 0.25–98.27). This study further revealed a greater proportion of women with urinary tract infection in the packing group (not significant, 9.1% versus 0.0%; RR 0.23, 95%CI 0.03–1.96) (14). Finally, a study of 144 women also showed a non-significantly greater proportion of women with haematoma in the group without packing (0.0% versus 4.28%; RR 0.13, 95%CI 0.01–2.57) (13).