Machine Learning for the Prediction of Surgical Morbidity in Placenta Accreta Spectrum

Itamar D. Futterman; Olivia Sher; Chaskin Saroff; Alexa Cohen; Georgios Doulaveris; Pe'er Dar; Myah M. Griffin; Meghana Limaye; Thomas Owens; Lois Brustman; Henri Rosenberg; Rebecca Jessel; Scott Chudnoff; Shoshana Haberman

doi:10.1055/a-2405-3459

American Journal of Perinatology, Table of Contents

Am J Perinatol 2025; 42(03): 281-292
DOI: 10.1055/a-2405-3459

SMFM Fellowship Series Article

Machine Learning for the Prediction of Surgical Morbidity in Placenta Accreta Spectrum

Authors

Itamar D. Futterman

¹Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, Maimonides Medical Center, Brooklyn, New York

²Division of Complex Obstetrical Surgery, Department of Obstetrics and Gynecology, Maimonides Medical Center, Brooklyn, New York
Olivia Sher

¹Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, Maimonides Medical Center, Brooklyn, New York
Chaskin Saroff

³BioticAI Inc., San Francisco, California
Alexa Cohen

⁴Division of Fetal Medicine and Ultrasound, Obstetrics, Gynecology and Women's Health, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, New York
Georgios Doulaveris

⁴Division of Fetal Medicine and Ultrasound, Obstetrics, Gynecology and Women's Health, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, New York
Pe'er Dar

⁴Division of Fetal Medicine and Ultrasound, Obstetrics, Gynecology and Women's Health, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, New York
Myah M. Griffin

⁵Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, NYU Langone Medical Center, New York, New York
Meghana Limaye

⁵Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, NYU Langone Medical Center, New York, New York
Thomas Owens

⁶Division of Maternal Fetal Medicine, Mount Sinai West, Icahn School of Medicine at Mount Sinai, New York, New York
Lois Brustman

⁶Division of Maternal Fetal Medicine, Mount Sinai West, Icahn School of Medicine at Mount Sinai, New York, New York
Henri Rosenberg

⁷Division of Maternal Fetal Medicine, Department of Obstetrics, Gynecology and Reproductive Science, Icahn School of Medicine at Mount Sinai, New York, New York
Rebecca Jessel

⁷Division of Maternal Fetal Medicine, Department of Obstetrics, Gynecology and Reproductive Science, Icahn School of Medicine at Mount Sinai, New York, New York
Scott Chudnoff

¹Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, Maimonides Medical Center, Brooklyn, New York
Shoshana Haberman

¹Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, Maimonides Medical Center, Brooklyn, New York

Abstract

Objective We sought to create a machine learning (ML) model to identify variables that would aid in the prediction of surgical morbidity in cases of placenta accreta spectrum (PAS).

Study Design A multicenter analysis including all cases of PAS identified by pathology specimen confirmation, across five tertiary care perinatal centers in New York City from 2013 to 2022. We developed models to predict operative morbidity using 213 variables including demographics, obstetrical information, and limited prenatal imaging findings detailing placental location. Our primary outcome was prediction of a surgical morbidity composite defined as including any of the following: blood loss (>1,500 mL), transfusion, intensive care unit admission, vasopressor use, mechanical ventilation/intubation, and organ injury. A nested, stratified, cross-validation approach was used to tune model hyperparameters and estimate generalizability. Gradient boosted tree classifier models incorporated preprocessing steps of standard scaling for numerical variables and one-hot encoding for categorical variables. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), positive and negative predictive values (PPV, NPV), and F1 score. Variable importance ranking was also determined.

Results Among 401 PAS cases, 326 (81%) underwent hysterectomy. Of the 401 cases of PAS, 309 (77%) had at least one event defined as surgical morbidity. Our predictive model had an AUC of 0.79 (95% confidence interval: 0.69, 0.89), PPV 0.79, NPV 0.76, and F1 score of 0.88. The variables most predictive of surgical morbidity were completion of a hysterectomy, prepregnancy body mass index (BMI), absence of a second trimester ultrasound, socioeconomic status zip code, BMI at delivery, number of prenatal visits, and delivery time of day.

Conclusion By identifying social and obstetrical characteristics that increase patients' risk, ML models are useful in predicting PAS-related surgical morbidity. Utilizing ML could serve as a foundation for risk and complexity stratification in cases of PAS to optimize surgical planning.

Key Points

ML models are useful models are useful in predicting PAS-related surgical morbidity.
Optimal management for PAS remains unclear.
Utilizing ML can serve as a foundation for risk and complexity stratification in cases of PAS.

Keywords

placenta accreta spectrum - surgical morbidity - machine learning

Full Text

References

References
1 American College of Obstetricians and Gynecologists, Society for Maternal-Fetal Medicine. Obstetric care consensus no. 7: placenta accreta spectrum. Obstet Gynecol 2018; 132 (06) e259-e275
2 Silver RM, Branch DW. Placenta accreta spectrum. N Engl J Med 2018; 378 (16) 1529-1536
3 Jauniaux E, Bunce C, Grønbeck L, Langhoff-Roos J. Prevalence and main outcomes of placenta accreta spectrum: a systematic review and meta-analysis. Am J Obstet Gynecol 2019; 221 (03) 208-218
4 Wu S, Kocherginsky M, Hibbard JU. Abnormal placentation: twenty-year analysis. Am J Obstet Gynecol 2005; 192 (05) 1458-1461
5 Matsuzaki S, Mandelbaum RS, Sangara RN. et al. Trends, characteristics, and outcomes of placenta accreta spectrum: a national study in the United States. Am J Obstet Gynecol 2021; 225 (05) 534.e1-534.e38
6 Erfani H, Fox KA, Clark SL. et al. Maternal outcomes in unexpected placenta accreta spectrum disorders: single-center experience with a multidisciplinary team. Am J Obstet Gynecol 2019; 221 (04) 337.e1-337.e5
7 D'Arpe S, Franceschetti S, Corosu R. et al. Emergency peripartum hysterectomy in a tertiary teaching hospital: a 14-year review. Arch Gynecol Obstet 2015; 291 (04) 841-847
8 Clark SL, Belfort MA, Dildy GA, Herbst MA, Meyers JA, Hankins GD. Maternal death in the 21st century: causes, prevention, and relationship to cesarean delivery. Am J Obstet Gynecol 2008; 199 (01) 36.e1-36.e5 , discussion 91–92, e7–e11
9 Fox KA, Shamshirsaz AA, Carusi D. et al. Conservative management of morbidly adherent placenta: expert review. Am J Obstet Gynecol 2015; 213 (06) 755-760
10 Shellhaas CS, Gilbert S, Landon MB. et al; Eunice Kennedy Shriver National Institutes of Health and Human Development (NICHD) Maternal–Fetal Medicine Units Network (MFMU). The frequency and complication rates of hysterectomy accompanying cesarean delivery. Obstet Gynecol 2009; 114 (2 Pt 1): 224-229
11 Warshak CR, Ramos GA, Eskander R. et al. Effect of predelivery diagnosis in 99 consecutive cases of placenta accreta. Obstet Gynecol 2010; 115 (01) 65-69
12 Shamshirsaz AA, Fox KA, Erfani H. et al. Multidisciplinary team learning in the management of the morbidly adherent placenta: outcome improvements over time. Am J Obstet Gynecol 2017; 216 (06) 612.e1-612.e5
13 Shamshirsaz AA, Fox KA, Salmanian B. et al. Maternal morbidity in patients with morbidly adherent placenta treated with and without a standardized multidisciplinary approach. Am J Obstet Gynecol 2015; 212 (02) 218.e1-218.e9
14 Collins SL, Alemdar B, van Beekhuizen HJ. et al. Evidence-based guidelines for the management of abnormally invasive placenta: recommendations from the International Society for Abnormally Invasive Placenta. Am J Obstet Gynecol 2019; 220 (06) 511-526
15 Angstmann T, Gard G, Harrington T, Ward E, Thomson A, Giles W. Surgical management of placenta accreta: a cohort series and suggested approach. Am J Obstet Gynecol 2010; 202 (01) 38.e1-38.e9
16 Sentilhes L, Seco A, Azria E. et al; PACCRETA Study Group. Conservative management or cesarean hysterectomy for placenta accreta spectrum: the PACCRETA prospective study. Am J Obstet Gynecol 2022; 226 (06) 839.e1-839.e24
17 Han X, Guo Z, Yang X, Yang H, Ma J. Association of placenta previa with severe maternal morbidity among patients with placenta accreta spectrum disorder. JAMA Netw Open 2022; 5 (08) e2228002
18 Erfani H, Salmanian B, Fox KA. et al. Urologic morbidity associated with placenta accreta spectrum surgeries: single-center experience with a multidisciplinary team. Am J Obstet Gynecol 2022; 226 (02) 245.e1-245.e5
19 Shazly SA, Hortu I, Shih JC. et al; Middle East Obstetrics and Gynecology Graduate Education (MOGGE) foundation – Artificial intelligence (AI) Unit. Prediction of clinical outcomes in women with placenta accreta spectrum using machine learning models: an international multicenter study. J Matern Fetal Neonatal Med 2022; 35 (25) 6644-6653
20 Nieto-Calvache AJ, Palacios-Jaraquemada JM, Aryananda RA. et al. How to identify patients who require aortic vascular control in placenta accreta spectrum disorders?. Am J Obstet Gynecol MFM 2022; 4 (01) 100498
21 Mitric C, Desilets J, Balayla J, Ziegler C. Surgical management of the placenta accreta spectrum: an institutional experience. J Obstet Gynaecol Can 2019; 41 (11) 1551-1557
22 Amsalem H, Kingdom JCP, Farine D. et al. Planned caesarean hysterectomy versus “conserving” caesarean section in patients with placenta accreta. J Obstet Gynaecol Can 2011; 33 (10) 1005-1010
23 Aryananda RA, Aditiawarman A, Gumilar KE. et al. Uterine conservative-resective surgery for selected placenta accreta spectrum cases: surgical-vascular control methods. Acta Obstet Gynecol Scand 2022; 101 (06) 639-648
24 El Gelany S, Mosbeh MH, Ibrahim EM. et al. Placenta accreta spectrum (PAS) disorders: incidence, risk factors and outcomes of different management strategies in a tertiary referral hospital in Minya, Egypt: a prospective study. BMC Pregnancy Childbirth 2019; 19 (01) 313
25 Mohamed Siraj SH, Tan KH, Wright AM. Conservative surgical approach towards placenta accreta spectrum disorders for uterine preservation. BMC Pregnancy Childbirth 2023; 23 (01) 28
26 Srinivasan B, Balasubramanian N, Vijayaraghavan J. et al. Study on outcomes of pregnancy in women with placenta accreta spectrum: a 10-year study in a tertiary care center. J South Asian Feder Obst Gynae 2021; 13 (03) 94-97
27 Üstünyurt E. Local uterine resection with Bakri balloon placement in placenta accreta spectrum disorders. Turk J Obstet Gynecol 2020; 17 (02) 108-114
28 Karaçor T, Bülbül M, Nacar MC. et al. The parameters affecting the success of uterus-sparing surgery in cases of placenta adhesion spectrum disorder. J Matern Fetal Neonatal Med 2021; 34 (07) 1091-1098
29 Kutuk MS, Ak M, Ozgun MT. Leaving the placenta in situ versus conservative and radical surgery in the treatment of placenta accreta spectrum disorders. Int J Gynaecol Obstet 2018; 140 (03) 338-344
30 Lional KM, Tagore S, Wright AM. Uterine conservation in placenta accrete spectrum (PAS) disorders: a retrospective case series: is expectant management beneficial in reducing maternal morbidity?. Eur J Obstet Gynecol Reprod Biol 2020; 254: 212-217
31 Nieto-Calvache AJ, Zambrano MA, Herrera NA. et al. Resective-reconstructive treatment of abnormally invasive placenta: Inter Institutional Collaboration by telemedicine (eHealth). J Matern Fetal Neonatal Med 2021; 34 (05) 765-773
32 Wang Q, Ma J, Zhang H. et al. Conservative management versus cesarean hysterectomy in patients with placenta increta or percreta. J Matern Fetal Neonatal Med 2022; 35 (10) 1944-1950
33 Fyfe EM, Thompson JM, Anderson NH, Groom KM, McCowan LM. Maternal obesity and postpartum haemorrhage after vaginal and caesarean delivery among nulliparous women at term: a retrospective cohort study. BMC Pregnancy Childbirth 2012; 12: 112
34 Dias M, Dick A, Reynolds RM, Lahti-Pulkkinen M, Denison FC. Predictors of surgical site skin infection and clinical outcome at caesarean section in the very severely obese: a retrospective cohort study. PLoS One 2019; 14 (06) e0216157
35 Ayres-de-Campos D. Obesity and the challenges of caesarean delivery: prevention and management of wound complications. Best Pract Res Clin Obstet Gynaecol 2015; 29 (03) 406-414
36 Cohen A, Lambert C, Yanik M. et al. Investigation of health inequities in maternal and neonatal outcomes of patients with placenta accreta spectrum: a multicenter study. Am J Obstet Gynecol MFM 2024; 6 (07) 101386
37 Vestal NL, Sangara RN, Mandelbaum RS. et al. Racial and ethnic disparity in characteristics and outcomes of women with placenta accreta spectrum: a comparative study. Reprod Sci 2022; 29 (07) 1988-2000
38 Jauniaux E, Fox KA, Einerson B, Hussein AM, Hecht JL, Silver RM. Perinatal assessment of complex cesarean delivery: beyond placenta accreta spectrum. Am J Obstet Gynecol 2023; 229 (02) 129-139
39 Munoz JL, Blankenship LM, Ramsey PS, McCann GA. Importance of the gynecologic oncologist in management of cesarean hysterectomy for placenta accreta spectrum (PAS). Gynecol Oncol 2022; 166 (03) 460-464
40 Touhami O, Allen L, Flores Mendoza H, Murphy MA, Hobson SR. Placenta accreta spectrum: a non-oncologic challenge for gynecologic oncologists. [published online ahead of print, April 27, 2022] Int J Gynecol Cancer 2022; ijgc-2021–003325
41 Eller AG, Bennett MA, Sharshiner M. et al. Maternal morbidity in cases of placenta accreta managed by a multidisciplinary care team compared with standard obstetric care. Obstet Gynecol 2011; 117 (2 Pt 1): 331-337
42 Hecht JL, Baergen R, Ernst LM. et al. Classification and reporting guidelines for the pathology diagnosis of placenta accreta spectrum (PAS) disorders: recommendations from an expert panel. Mod Pathol 2020; 33 (12) 2382-2396
43 Jauniaux E, Ayres-de-Campos D, Langhoff-Roos J, Fox KA, Collins S. FIGO Placenta Accreta Diagnosis and Management Expert Consensus Panel. FIGO classification for the clinical diagnosis of placenta accreta spectrum disorders. Int J Gynaecol Obstet 2019; 146 (01) 20-24

Supplementary Material

Supplementary Material (PDF)