Emergency cesarean hysterectomy for placenta accreta spectrum: Estimation of maternal morbidity and mortality

Placenta accreta spectrum refers to morbidly adherent placenta to the uterus.1 The incidence of placenta accreta spectrum is gradually increasing.2 Patients with suspected placenta accreta spectrum frequently undergo planned hysterectomy at cesarean delivery.1 Cesarean hysterectomy for placenta accreta spectrum is associated with significant maternal morbidity and mortality.3 Equally important, its impact on maternal morbidity when performed in an emergency setting is under active investigation.4-6 Given mixed findings from prior investigations,4-6 the objective of this study was to assess the estimated maternal morbidity and mortality for patients undergoing emergency cesarean hysterectomy for placenta accreta spectrum. This retrospective study queried the American College of Surgeon's National Surgical Quality Improvement Program.7 The study population included 164 patients who underwent cesarean hysterectomy for a diagnosis of placenta accreta spectrum from 2016 to 2022. The World Health Organization's International Classification of Diseases, Tenth Revision, Clinical Modification codes were used to identify placenta accreta spectrum per prior investigations (O43.21, O43.22, and O43.23 for accreta, increta, and percreta subtypes, respectively).2, 3 The exposure was the surgical setting classified as: (i) non-emergency, elective; (ii) non-emergency, non-elective; and (iii) emergency. The first step grouping followed the study variable “EMERGNCY” to assign emergency versus non-emergency surgery. Among the non-emergency surgery, the second step grouping followed the study variable “ELECTSURG” to assign elective or non-elective status that represents the preplanned or unplanned admission. The main outcome measures were the estimated probability of maternal morbidity and mortality, identified by the study variables “MORBPROB” and “MORTPROB”. The modeled probability estimates were developed by the program for all cases based on a logistic regression analysis using the patient's preoperative characteristics as the independent or predictive variables. Simple descriptive analysis was performed across the three exposure groups. A multivariable generalized linear model was created to adjust the exposure–outcome association by patient age and placenta accreta spectrum type. Statistical interpretation was based on a two-tailed hypothesis, and a Pvalue of less than 0.05 was considered statistically significant. SPSS Statistics (version 29.0; IBM, Armonk, NY, USA) was used for all analyses. The University of Southern California Institutional Review Board deems this study exempt due to the use of a publicly available, deidentified, secondary database (HS-17-01019; patient informed consent, not required). The most common surgical setting was non-emergency, elective (n = 87, 53.0%), followed by non-emergency, non-elective (n = 53, 32.3%) and emergency (n = 24, 14.6%; Table 1). Patients who underwent emergency cesarean hysterectomy had lower body mass index (calculated as weight in kilograms divided by the square of height in meters) than those in the non-emergency groups (28.0 vs 32.6–34.6, P = 0.007). The non-emergency, elective surgery group had the lowest rate of increta/percreta compared with the other groups (37.9% vs 54.2%–58.5%, P = 0.035). The rate of blood transfusion and/or urinary tract repair was higher in the emergency group than in the non-emergency groups (75.0% vs 48.3%–49.1%, P = 0.057). Estimated morbidity for emergency cesarean hysterectomy for placenta accreta spectrum was higher than for non-emergency procedures (11.8% for emergency, 9.2% for non-emergency, non-elective, and 8.5% for non-emergent, elective procedures, P < 0.001; Table 1). In an adjusting model, both emergency surgery (adjusted-rate ratio increase per increment unit 3.9, 95% confidence interval [CI] 2.6–5.4, Padjusted < 0.01) and non-emergent, non-elective surgery (adjusted-rate ratio increase per increment unit 1.2, 95% CI 0.1–2.2, Padjusted = 0.028). Placenta accreta subtype was not associated with morbidity in this multivariable model (Padjusted = 0.167 for increta vs accreta; Padjusted = 0.804 for percreta vs accreta). Results were similar for estimated mortality probability (12.5 per 10 000 for emergency, 6.8 per 10 000 for non-emergent, non-elective, and 3.8 per 10 000 for non-emergent, elective, P < 0.001; Table 1). After controlling for patient age and placenta accreta spectrum subtype, emergency cesarean hysterectomy was associated with increased morbidity compared with nonemergent, elective cesarean hysterectomy (adjusted-rate ratio increase per increment unit 15.1, 95% CI 9.7–20.6, Padjusted <0.001) but not nonemergent, non-elective cesarean hysterectomy (Padjusted = 0.063). Placenta accreta subtype was not associated with mortality in this multivariable model (Padjusted = 0.714 for increta vs accreta; Padjusted = 0.112 for percreta vs accreta). Lastly, when assessed for placenta accreta subtypes, placenta accreta had higher rates of non-emergent, elective surgery compared with placenta increta/percreta (62.1% vs 42.9%), whereas placenta percreta had higher rates of emergency surgery compared with placenta accreta/increta (22.4% vs 7.1%–12.6%) (P = 0.035). These data suggest that patients with placenta accreta spectrum frequently undergo emergency cesarean hysterectomy, with nearly one in seven surgeries being performed as emergency procedures. Given the possibility of higher morbidity and mortality associated with emergency cesarean hysterectomy for placenta accreta spectrum, so that this factor may be more important than the disease severity, any strategy to reduce emergency or unplanned surgery is recommended.8 Limitations of the study included lack of information on number of prior cesarean deliveries, timing of placenta accreta spectrum diagnosis (antenatal vs undiagnosed), indication of emergency surgery (fetal or maternal), modest sample size, and decision for hysterectomy (preplanned vs unplanned) or other adjunctive intervention (e.g. ureteral stent placement). Conceptualization: Ko.M. Data curation: Ka.M. Formal analysis: Ko.M. Funding acquisition: Ko.M. Investigation: all authors. Methodology: Ko.M. Project administration: Ka.M. Resources: Ko.M., Ka.M., S.M. Software: Ko.M., Ka.M. Supervision: J.G.O., J.D.W. Validation: Ko.M. Visualization: Ko.M. Writing—original draft: Ko.M. Writing—review & editing: all authors. Ensign Endowment for Gynecologic Cancer Research (Ko.M.). The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. All authors were unrelated to the work: consultant, research grant, Merck, royalties, UpToDate, honorarium, American College of Obstetrics and Gynecology (J.D.W.). The manuscript's corresponding authors (Ko.M./Ka.M.) affirm that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained. The National Surgical Quality Improvement Program is sponsored by the American College of Surgeons, and the program is the source of the deidentified data used; race/ethnicity was grouped by the program; and the program has not verified and is not responsible for the statistical validity of the data analysis or the conclusions derived by the study team. The data on which this study is based are publicly available upon request at The American College of Surgeons National Surgical Quality Improvement Program (https://www.facs.org/quality-programs/data-and-registries/acs-nsqip/).