POP-Q Phenotypes of Pelvic Organ Prolapse

Objectives: A limitation of the Pelvic Organ Prolapse Quantification (POP-Q) is the stages do not correlate with symptoms and do not adequately describe important prolapse subtypes. We sought to develop clinically meaningful prolapse “phenotypes” (subtypes) utilizing POP-Q measurements. Primary aim was to define the phenotypes and their frequency in women undergoing prolapse surgery. Secondary aims: compare demographics, medical characteristics, and symptoms between phenotypes.

Methods: Patients with POP-Q exams and Pelvic Floor Distress Inventory (PFDI-20) scores undergoing prolapse surgery between 2017 and 2023 were included. Patients were categorized into one of eight phenotypes based on POP-Q measurements (Table 1). Two principles were used for defining mutually exclusive phenotypes: (1) prolapse exists when the anterior or posterior vaginal wall descend to the hymen or the apex descends to half total vaginal length, and (2) prolapse may exist in anterior, posterior and/or apical compartments. Demographics, medical characteristics, and PFDI-20 responses were compared between phenotypes, using AC (anterior-predominant and apical) phenotype as the reference group. Linear or logistic regression models were used for comparisons while controlling for POP-Q severity (mild: stage I or II, severe: stage III or IV).

Results: The AC phenotype was most common (231/501 patients, 46.1%). Only 15 (3%) patients were in the least common AP phenotype (anterior and posterior) (Table 2).

Compared with AC phenotype (mean (SD) age 64 (11) years), the no prolapse, P (isolated posterior), C (isolated apical) and PC (posterior-predominant and apical) phenotypes were younger, and the A (isolated anterior) phenotype was older; and P, PC, and APC (anterior and posterior and apical) phenotypes had greater body mass index. Differences were seen in frequency of prior hysterectomy and prolapse surgery (Table 2).

PFDI-20 responses differed between phenotypes. The strongest differences were in the P phenotype with 5 Colorectal-Anal Distress Inventory (CRADI-8) questions having higher scores than the AC phenotype. Additionally, Pelvic Organ Prolapse Distress Inventory (POPDI-6) #4 (splinting with bowel movements), Urinary Distress Inventory (UDI-6) #17 (stress incontinence) and UDI-6 #18 (small drops of incontinence) were all higher in the P phenotype. Similarly, the PC phenotype had higher scores for splinting and rectal prolapse (CRADI-8 #14). Conversely, the C phenotype total PFDI-20 scores were lower (93 (50-109), p=0.01) than the AC phenotype. This effect was driven primarily by the POPDI-6 and UDI-6. Only the APC phenotype had no significant differences in any PFDI-20 question compared with the AC phenotype.

Table 1. Phenotype definitions.

Table 2. Results. Shaded cell indicates significant result. Arrows indicate direction and strength (one = p< 0.05, two = p< 0.01) of difference compared to the AC phenotype (reference group).

Conclusions: We present a novel method for characterizing prolapse. These phenotypes may allow for improved understanding, communication, and individualized counseling about prolapse and prolapse treatment in the future.