Following implementation of our exclusion criteria a total of 107 women were included in the study, 73 women underwent PME (PME group) as opposed to 34 women who were treated by TME (TME group). A comparison of basic and pre-operative characteristics is presented in table 1. Mean age was 60.7 (SD=9.5) and 61.2 (SD=12.6) in the PME and TME groups, respectively (p=0.813). Women in the TME group had higher preoperative ASA score compared to the PME group. Twenty-five women in the TME group underwent radiation therapy prior to surgery as opposed to none in the PME group (p<0.001). Other parameters assessed such as BMI, smoking status and other comorbidities did not differ between the groups.
Intra-operative and post-operative data as well as questionnaire scores are presented in table 2. Mean length of surgery and complication rate were similar between groups. Women in the PME group had larger tumor size (4.0± 1.9 vs. 2.6 ± 1.2 cm, p<0.001) and were discharged earlier from the hospital (mean hospital stay 7.6± 2.8 vs. 9.2± 3.2 days, p=0.013) compared to women in the TME group. No difference was found with respect to need of adjuvant therapy between groups.
Urinary dysfunction following surgery as assessed using the UDI-6 questionnaire did not differ between groups (11.6± 19.5 vs. 10.2± 16.9, p=0.989, for PME and TME groups, respectively). Similar findings were recorded upon comparison of the impact of urinary dysfunction on quality of life (QOL) as assessed using the USIQ questionnaire (table 2).
Subgroup analyses were performed comparing TME and PME groups focusing on women with any urinary dysfunction (UDI-6>0) as well as women with more severe urinary dysfunction (UDI-6>25). Proportion of women with any urinary dysfunction did not differ between groups (28.8% vs. 35.3%, p=0.496, for PME and TME groups, respectively). Similar findings were noted with respect to more severe cases of urinary dysfunction (15.1% vs. 17.6%, p=0.734, for PME and TME groups, respectively).
We further analyzed the effect of time elapsed from surgery on urinary dysfunction. Women who underwent surgery within the previous year had similar UDI-6 scores compared to women with a time interval since surgery of over one year (14.5±21.3 vs. 9.8± 17.3, p=0.23, for <1 year and >1 year, respectively).
Univariate and multivariate analyses were performed, in the attempt of detecting risk factors associated with any urinary dysfunction within our cohort. Results of the stepwise logistic regression model can be found in table 3. The logistic regression model was implemented for women with any level of urinary dysfunction (UDI-6>0), women with more severe urinary dysfunction (UDI-6>25) and for women with any impact of urinary dysfunction on quality of life (USIQ>0). Following multivariate analysis women with longer hospitalization (OR=1.38, CI 1.1-1.72, p=0.005) had increased risk of having some level of urinary dysfunction. Regarding impact of urinary dysfunction on quality of life, increased hospital stay was associated with a detrimental effect on quality of life (OR=1.36, CI 1.0-1.84, p=0.05). Higher level of anastomosis (OR=0.90, CI 0.83-0.98, p=0.013) and absence of comorbidities decreased the risk of urinary dysfunction (OR=0.19, CI 0.05-0.81, p=0.025).