PREDICTORS OF INCONTINENCE SEVERITY IN A COHORT OF STRESS INCONTINENT WOMEN  UNDERGOING SURGICAL TREATMENT

H.E. Richter, K.L. Burgio, A. Markland, V. Mallett, P.A. Moalli, L. Brubaker, S.D. Menefee, H.W. Johnson, Jr., M. Boreham, K.J. Dandreo, and A.M. Stoddard for the Urinary Incontinence Treatment Network (NIDDK)

University of Alabama, Birmingham, AL

OBJECTIVE:  The aim of this study was to identify predictors of urinary incontinence severity at baseline in a cohort of women undergoing surgery for stress incontinence.

MATERIALS ANDMETHODS:  Baseline data were obtained from 655 women with pure or predominant stress incontinence participating in a randomized clinical trial comparing the Burch retropubic urethropexy to autologous rectus fascial sling. The baseline evaluation included a 3-day bladder diary, medical, obstetric, and surgical history, physical examination, and the PISQ-12 sexual function questionnaire. Severity of incontinence was defined by the number of incontinence episodes recorded in the bladder diary. The relationship between incontinence severity and a number of variables was explored using bivariate and multivariable linear regression modeling when appropriate. The primary dependent variable was log of mean number of accidents per day. Independent variables included demographic (age, ethnicity, socioeconomic status), medical history  (presence of diabetes, any  fecal incontinence, current smoking, urinary tract infections), obstetric history  (number of pregnancies, vaginal deliveries, weight of largest baby), gynecological history (menopausal status, current hormone therapy, hysterectomy), previous treatment for incontinence, sexual dysfunction score, body mass index, Q-tip displacement and Pelvic Organ Prolapse Quantification stage (POP-Q).

RESULTS: Mean age at enrollment was 52 years (sd 10.3) with a range of 28 to 81 years; mean number of accidents per day was 3.2 (sd 2.9), range 0-26, median .3. Bivariate analysis revealed no association of severity with parity, number of vaginal deliveries or weight of the largest baby. In a multivariable model, increased number of incontinence episodes per day was significantly associated with lower prolapse stage (p<.0001), greater body mass index (p=.0003), current smoking status (p=.02), and lower Q-tip displacement (p=.04). Severity was not significantly related to age, socioeconomic status, or previous hysterectomy.

CONCLUSION: Incontinence severity in a surgical population was independently associated with two modifiable factors, obesity and tobacco use, as well as pelvic support.

Key Words: incontinence severity, risk factors, surgery

Disclosure - Speakers Bureau: H. Richter, Pfizer, Novartis; V. Mallett, Pfizer, Ortho-McNeil, Berlex; M. Boreham, Ortho-McNeil, Pfizer, Watson; S.D. Menefee, Boston Scientific; Consultant: K.L. Burgio, Lilly, Ortho-McNeil, Novartis, Yamanouchi, Merck; L. Brubaker, Pfizer, Lilly; Advisory Board: V. Mallett, Lilly; L. Brubaker, Pfizer, Q-Med; Grant/Research: L. Brubaker, Pfizer, Q-Med, Life-Tech.