Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index; 1 of the 14 sources its claims cite were checked for support.

Individualized Prediction of Recurrence Following Uterine-Preserving Pelvic Organ Prolapse Repair Using Internally Validated Machine Learning Models

Open access
Malul, Shenhav, Chill, Henry H., Yosef, Rami, Miron-Shatz, Talya, Shveiky, David · 2026 · Journal of Clinical Medicine

Abstract

Background/Objectives: Approximately 11–20% of women affected by pelvic organ prolapse (POP) will undergo surgical intervention during their lifetime and nearly one-third of these women will require an additional procedure due to recurrence or failure of the initial repair. This study aimed to identify risk factors for failure and to develop models to predict subjective, anatomical, and composite failure after primary uterine-preserving POP surgery. Methods: We performed a retrospective cohort study of women undergoing primary uterine-preserving POP repair at a tertiary academic medical center between 2010 and 2024. Failure outcomes were defined as: subjective failure (patient-reported prolapse symptoms), anatomical failure (prolapse beyond the hymen during exam), and composite failure (subjective and/or anatomical failure and/or reoperation for prolapse). Risk factors were evaluated using univariable and multivariable logistic regression with odds ratios and 95% confidence intervals. Prediction models were trained in Python and internally validated using bootstrap optimism correction (500 resamples). Discrimination, overall accuracy, and calibration were evaluated. Results: Among 277 women, subjective failure occurred in 36 (13.0%), anatomical failure in 24 (8.7%), and composite failure in 46 (16.6%). In multivariable risk-factor models, larger genital hiatus was associated with higher odds of failure across endpoints and posterior compartment descent was independently associated with anatomical and composite failure. After bootstrap optimism correction, discrimination remained high for the best-performing models. Conclusions: In this single-center cohort, independent risk factors for failure after primary prolapse surgery were identified and prediction models showed strong internal discrimination. However, calibration limitations necessitate recalibration and external validation before clinical implementation.

Overview

CoChat Summary

## TL;DR

This retrospective single-center study of 277 women undergoing primary uterine-preserving POP repair found that larger genital hiatus and posterior compartment descent were the most consistent predictors of recurrence/failure across subjective, anatomical, and composite endpoints. Internally validated machine learning models showed strong discrimination but suboptimal calibration, so the authors present them as preliminary rather than ready for clinical use.

## Key findings

  • Among 277 women (mean age 60.4 years; mean BMI 26.0 kg/m²), failure rates were 13.4% for subjective failure (36/269), 9% for anatomical failure (24/268), and 17% for composite failure (46/270).
  • In multivariable logistic regression, larger genital hiatus (GH) was independently associated with all three outcomes:
  • subjective failure: adjusted OR 1.94 per 1 cm increase (95% CI 1.02–3.66; p = 0.041)
  • anatomical failure: adjusted OR 3.55 per 1 cm increase (95% CI 1.75–7.58; p = 0.001)
  • composite failure: adjusted OR 1.79 per 1 cm increase (95% CI 1.02–3.12; p = 0.040)
  • Posterior compartment descent was also predictive:
  • anatomical failure: Bp adjusted OR 1.53 per 1 cm increase (95% CI 1.14–2.12; p = 0.007)
  • composite failure: Bp adjusted OR 1.31 per 1 cm increase (95% CI 1.03–1.70; p = 0.031)
  • Other notable adjusted associations included hyperthyroidism with anatomical failure (adjusted OR 23.24; 95% CI 1.33–623.79; p = 0.027), while the machine learning models had “strong optimism-corrected discrimination” but calibration intercepts and slopes that “deviated notably from ideal values.”

## Methods

  • Retrospective cohort study at a tertiary university-affiliated teaching hospital; included women undergoing primary uterine-preserving apical prolapse repair from 2010–2024, excluding those with prior pelvic surgery for prolapse/incontinence.
  • Data came from electronic medical records and operative reports, including demographics, obstetric/surgical history, lifestyle factors, menopausal status, POP-Q measurements, procedure type, and concomitant procedures.
  • Outcomes were binary subjective, anatomical, and composite failure; associations were assessed with logistic regression, and prediction models were trained in Python 3.10 using penalized logistic regression, random forest, and LightGBM with 500-bootstrap internal validation, evaluating AUC, PR-AUC, Brier score, and calibration.

## Limitations

  • The paper states the study is limited by its single-center design, modest event counts (especially anatomical failure), absence of external validation, suboptimal calibration, and analysis of outcomes as binary variables without accounting for variable follow-up duration.
  • The authors also note that including statistically significant baseline comparisons as candidate predictors may introduce outcome-driven variable selection and inflate apparent model performance.
  • Evident from the text: the anatomical-failure model had only 24 events and hyperthyroidism’s very wide CI (1.33–623.79), indicating unstable estimates; the authors themselves caution that several findings should be considered preliminary and require prospective external validation and recalibration.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/jcm15166366

Cites

12 of 12 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

1 Valid · 0 Not valid · 7 Not checked · 74 Own results and interpretations

39 more claims

Create a free account to see all 82 claims and to verify the 13 references still open.

See all claims — free account
  • Probably not a substantial claim. 20
  • It may state two separate claims as one. 27
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Research works printed but not linked: 2

  1. Maher, 2016, Surgery for women with apical vaginal prolapse, Cochrane Database Syst. Rev., 10, CD012376
  2. Friedman, 2018, Risk factors for prolapse recurrence: Systematic review and meta-analysis, Ultrasound Obstet. Gynecol., 29, 13

Cited by

Still being checked.

Metadata

Authors
Malul, S., Chill, H. H., Yosef, R., Miron-Shatz, T., & Shveiky, D.
Affiliations
  • Ben-Gurion University of the Negev (IL)
  • Hebrew University of Jerusalem (IL)
  • Faculty of Business Administration, Ono Academic College, Kiryat Ono 5545001, Israel
Corresponding author
Chill, Henry H.
Venue
Journal of Clinical Medicine
Publisher
MDPI AG
Published
2026-08-18
Volume
15
Issue
16
Pages
6366
ISSN
2077-0383 (ISSN-L)
OpenAlex
W7203671130
Cited by
0
References
12
FWCI
0.00
Type
journal article
Language
en
Open access
Locations
  • Journal of Clinical Medicine (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
Topics
  • Health Sciences › Medicine › Urology › Pelvic floor disorders treatments
  • Health Sciences › Medicine › Obstetrics and Gynecology › Uterine Myomas and Treatments
  • Health Sciences › Medicine › Surgery › Hernia repair and management
Keywords
pelvic organ prolapse, uterine-preserving surgery, prolapse recurrence, surgical failure, risk factors, prediction model, genital hiatus, reoperation, machine learning

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.