From January 2015 to August 2018, 405 women were treated for pelvic tumors at the University Hospital Aachen. A total of 95 patients died during this period. 87 patients could not be reached by telephone. One patient had a neobladder. The phone numbers of 83 patients were incorrect, and 18 patients refused to participate in the study. Therefore, a total of 284 patients were excluded and 121 patients included.
The ages ranged from 29 to 90 years, the average age was 62.3 years. The BMI was between 16 and 63.2, the mean was 27.3.
Eighty-eight patients (72.7%) had at least one child and seventeen of them (14%) had at least one cesarean section.
The distribution of tumor types among the 121 women with malignancy was as in Table 1.
Overall, 59 patients (48.8%) reported no UI, 16 women (13.2%) had UI prior to their oncology therapy and reported an exacerbation due to treatment, and 46 (38%) patients had a de novo UI.
The prevalence of each category of ICIQ scores among patients is presented in Table 2.
In this study, 116 patients (95.9%) underwent gynecological surgery: 52 patients (43%) had the surgery performed laparoscopically, 57 patients (47.1%) had laparotomies; 31 patients (25.6%) had vaginal or vulvar surgery; 84 women (69.4%) underwent hysterectomy; 22 patients (18.2%) had radical hysterectomies; 56 patients (46.3%) had pelvic and para-aortic lymphadenectomy; and CTx, RTx, and ABT were performed in 51 patients (42.2%), 31 patients (25.6%), and 11 patients (9.1%), respectively.
The recurrence rate was also recorded: 22 patients (78.6%) had one recurrence, 3 patients (2.5%) had two recurrences, 1 patient (0.8%) had three recurrences, 1 patient (0.8%) had four recurrences and 1 patient (0.8%) had seven recurrences.
Regarding the preoperative information, 57 women (47.1%) indicated that they were well informed about the potential risk of preoperative UI as one of the possible complications of their oncological therapy.
In addition, 22 patients (18.2%) were uncertain whether they had been informed of this risk and 42 patients (34.7%) confirmed that they had not been informed preoperatively.
Of the patients with UI, 15 (24.2%) reported that UI did not affect their lives.
This study shows that 67 women (55.4%) have a very good QoL. Furthermore, we show that 25 (40.3%) patients with UI have a QoL score ≥ 5.
Table 3 shows the severity of the impact UI had on patients’ lives.
The distribution of reported QoL scores among patients is shown in Table 4.
Fifty percent of our collective had an ICIQ score between 3 and 21, with a median of 3. Bivariate analysis showed that the median ICIQ score in the cervical cancer group, at 6, was higher than this value in other groups of Types of tumors (Fig. 2a). In addition, bivariate analysis between ICIQ score and surgical therapy showed that the women who had at least one pelvic or pelvic floor surgery had lower median ICIQ scores than those who had no surgery (1.5 vs. 9 ). However, 25% of patients in the subgroup with surgery had an ICIQ score between 9 and 21, and the sample size in the subgroup without surgery was small (five women) (Fig. 2b).
In addition, the bivariate analysis showed that the patients who underwent vulvar or vaginal surgery had higher median ICIQ scores than those who did not have such surgical access (3 vs. 0.5), and that they even higher had median ICIQ scores than those who underwent laparotomy or laparoscopy (3 vs. 0 and 3 vs. 1.5, respectively) (Fig. 3a,b,c).
The intensity of UI has also been observed in women who underwent radical hysterectomy or simple hysterectomy, or even in women who underwent neither. The study showed higher median ICIQ scores in patients who underwent radical hysterectomy than in patients who had undergone simple hysterectomy and those who had neither (median ICIQ scores of 7, 0, and 3, respectively) ( Fig. 4).
The ICIQ score was also evaluated in patients who received RTx, CTx or ABT. In all of these groups, the median ICIQ scores (4, 3, and 3, respectively) were higher than those who had not undergone these therapies (median ICIQ score 0) (Fig. 5a,b,c).
A subgroup analysis of the ICIQ scores according to tumor and therapy type resulted in the median ICIQ scores shown in Table 5.
This study found no significant association between UI intensity and tumor types. Furthermore, no significant correlation was found between ICIQ score and age, birth and cesarean section.
According to the best model, multivariate analysis, including all variables as possible predictors, revealed that BMI (p= 0.05), radical hysterectomy (p= 0.02), vulvar or vaginal surgery (p= 0.05) and presence of UI before treatment (p= 0.004) have a significant impact on the presence of UI (ICIQ score > 0). They reduced the likelihood of having an ICIQ score of 0. This statistical model also showed a slight association between malignancy recurrences (p= 0.07) and an ICIQ score of 0. The higher the rate of malignancy recurrence, the lower the likelihood of having an ICIQ score of 0 (Table 6).
The best model also assessed which factors played a role in increasing the ICIQ score in cases with UI (ICIQ score > 0) and showed that ABT (p= 0.001), laparoscopic surgery (p= 0.01) and laparotomy (p= 0.03) have a negative impact on increasing ICIQ score in cases with UI (ICIQ score > 0). However, simple hysterectomy (p= p= p= 0.02) and CTx (p= 0.03) have positive effects on increasing the ICIQ score and thus worsening UI in cases with UI (ICIQ > 0) (Table 7).
According to the best model, multivariate analysis showed no association between tumor types and hysterectomy with quality of life, but showed that the following factors reduce the likelihood of a very good quality of life (QoL score of 7) in cases with pelvic malignancy: high BMI (p= 0.09), high rate of malignancy recurrence (p= 0.04), vulva or vaginal surgery (p= 0.03), pelvic and para-aortic lymphadenectomy (p= 0.18), the presence of UI before treatment (p= 0.01) and the time between initial diagnosis and collection of questionnaire responses (p= 0.18). However, both factors, surgery and no cesarean section, increased the likelihood of having a very good QoL (Table 8).
The quality of life of patients with pre-existing UI before diagnosis and treatment of their gynecologic cancer improved after therapy but was not statistically significant (p= 0.16). This study also showed that a high BMI (p= 0.17), vulvar or vaginal surgery (p= 0.002) and pelvic and para-aortic lymphadenectomy (p = 0.14) worsened quality of life in those with impaired quality of life (Table 9).
Spearman’s correlation analysis revealed a strong correlation between ICIQ and QoL scores, showing that patients with low ICIQ scores tended to have higher QoL scores and vice versa (correlation coefficient of −0.80).