King's College London

Research portal

Prevalence and main outcomes of placenta accreta spectrum: a systematic review and metaanalysis

Research output: Contribution to journalReview article

Eric Jauniaux, Catey Bunce, Lene Grønbeck, Jens Langhoff-Roos

Original languageEnglish
Pages (from-to)208-218
Number of pages11
JournalAmerican Journal of Obstetrics and Gynecology
Issue number3
Early online date1 Feb 2019
Publication statusPublished - 1 Sep 2019

King's Authors


OBJECTIVE: The objective of this study was to evaluate the prevalence of placenta accreta spectrum in general population studies and the main maternal outcomes at delivery.

STUDY DESIGN: Data sources: We searched PubMed, Google Scholar, and MEDLINE between 1982 and 2018.

STUDY ELIGIBILITY CRITERIA: Articles providing data on the number of cases of placenta accreta spectrum per pregnancies, births or deliveries in a defined population.

STUDY APPRAISAL AND SYNTHESIS METHODS: Study characteristics were evaluated by two independent reviewers using a predesigned protocol. Primary outcomes were the prevalence of placenta accreta spectrum and clinical diagnostic at birth and pathologic criteria used to confirm the diagnosis. Secondary outcomes included cases requiring transfusion, incidence of peripartum hysterectomy and maternal mortality rates. Heterogeneity between studies was analysed with the Cochran's Q-test and the I2 statistics.

RESULTS: Of the 98 full-text studies identified, 29 articles met the defined criteria including 22 retrospective and 7 prospective studies comprising 7,001 cases of placenta accreta spectrum out 5,719,992 births. Prevalence rates ranged between 0.01 and 0.1% with an overall pooled prevalence of 0.17% (95% CI 0.14-0.19). Only 10 studies provided with detailed histopathologic data. The pool prevalence for the adherent versus the invasive grades was 0.5 (95% CI 0.3-0.36) and 0.3 (95% CI 0.2-0.4) per 1000 births, respectively. The pooled incidence for peripartum hysterectomy was 52.2% (95% CI 38.3-66.4; I2= 99.8%) and 46.9% (95 % CI 34-59.9, I2= 98.8%) for haemorrhage requiring transfusion. The pooled estimate of maternal death was 0.05% (95% CI 0.06-0.69, I2=73%). We found large amounts of heterogeneity between studies for all parameters and further quantifying was limited because of methodological inconsistencies between studies with regards to clinical criteria used for the diagnosis of the condition at birth and the histopathologic confirmation of the diagnosis and differential diagnosis between adherent and invasive accreta placentation.

CONCLUSIONS: This meta-analysis indicates wide variation between studies for the prevalence rate of placenta accreta spectrum and for the different grades of accreta placentation, highlighting the need for consistency in definitions used to describe placenta accreta spectrum at birth and in reporting on this increasing common obstetric complication.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454