Hypothesis / aims of study
Anti-incontinence (AI) surgery at the time of pelvic organ prolapse repair has been the subject of several large clinical trials, including the Colpopexy and Urinary Reduction Efforts (CARE) trial. Primary outcomes following an AI procedure frequently include subjective and objective measures of stress and urgency urinary incontinence. While effective at treating both storage and voiding symptoms, the impact of an AI procedure on changes in post-void residual over time has been not been fully reported in the literature. To improve patient counseling in women considering an AI procedure at time of prolapse repair, we created a Kaplan-Meier cumulative incidence model to better characterize the transient and long term changes in post-void residual following surgery.
Study design, materials and methods
After obtaining IRB approval, we identified sequential patient records, which contained pre-operative urodynamic (UDS) procedure codes followed by anterior and/or apical prolapse repair at our institution over a 6 year period. Inclusion criteria included those women with at least a POP-Q stage 2 cystocele and at least 7 days of follow-up after surgery. Cases were stratified by those who underwent sling placement versus those who did not. Women who underwent sling removal were excluded. A two-reviewer case-by-case chart review was performed for: demographics; UDS parameters; UDS tracings; pre-operative POP-Q stage; date of surgery; operation; date and volume of all post void residuals (PVR) after surgery. This identified 143 women with verified pre-operative UDS followed by prolapse repair with or without sling placement. Our primary outcome of interest was time to first instance of PVR greater than the pre-operative PVR prior to surgery. Our secondary outcome of interest was longest follow-up PVR greater than pre-operative PVR. Data were analyzed in SAS (Cary, North Carolina, USA) using chi-square test, t-test, and Kaplan-Meier cumulative incidence methods. Data are presented as mean ± standard deviation. A p-value <0.05 was defined as significant.
One hundred and forty three women with POP-Q stage 2 or greater prolapse (mean age 62.3 ± 11.5 years) underwent anterior and/or apical prolapse repair and had preoperative UDS tracing, surgical and follow-up data available for analysis. Pre-operative UDS revealed a mean: Pdet@Qmax 20.4 ± 13.6 cmH2O, Qmax 17.6 ± 10 mL/s, capacity 524.9 ± 216.4 mL, PVR 115.1 ± 180.4 mL. Women underwent anterior only (n=66), apical only (n=21) or anterior/apical (n=56) prolapse repair for overall POP-Q stage 2 (n=57), 3 (n=79) or 4 (n=7) pelvic organ prolapse. From this group of sequential cases who underwent prolapse repair at our institution, a majority underwent sling placement (n=83; Pdet@Qmax 18.1 ± 10.7 cmH2O, Qmax 18.9 ± 9.7 mL/s, capacity 518.8 ± 213.4 mL, PVR 84.4 ± 162.6 mL) at the time of surgery, versus the remaining women who did not undergo sling placement [n=60; Pdet@Qmax 23.6 ± 16.4 cmH2O (p = 0.027), Qmax 15.8 ± 10.2 mL/s (p = 0.067), capacity 533.3 ± 222.0 mL (p = 0.694), PVR 157.5 ± 196.2 mL (p = 0.016)]. Following surgery, at least one PVR was recorded in all women with and without sling placement (3.0 ± 1.8 versus 2.3 ± 1.2, p = 0.010). Women were followed with a PVR measurement for a mean 587.7 ± 730.4 days after surgery.
To look at the association between time and first instance of elevated residual greater than pre-operative PVR, we applied Kaplan-Meier cumulative incidence methods to model the short term (event = first instance of follow-up PVR greater than pre-operative PVR) and long term (event = longest follow-up PVR greater than pre-operative PVR) effect of sling placement in women undergoing surgical prolapse repair. Using this approach, sling placement was associated (HR 1.53; CI 1.01-2.30; p = 0.0475) with a greater proportion of women with a first instance of follow-up PVR worse than their pre-operative PVR at 30 to 90 days of follow-up (Figure 1A-B). While the first instance of elevated PVR was most pronounced during the first year after surgery, on longest follow-up to 2,000 days, there was no significant difference (HR 1.26; CI 0.78-2.02, p = 0.3494) in the proportion of patients with longest follow-up PVR worse than prior to surgery (Figure 2A-B).
Interpretation of results
Sling placement was associated with a transiently worse PVR after surgery, with no significant difference in the proportion of women with versus without sling placement who had a long term follow-up PVR worse than their pre-operative PVR after pelvic organ prolapse repair.