Background: On a global scale, cases of placenta accreta spectrum (PAS) are often only identified at caesarean section (CS), because they were missed at antenatal care screening. Routine operating teams, not trained in management of PAS, are then faced with difficult surgical situations and have to take decisions that may define the clinical outcomes. Although there are general recommendations on the intraoperative management of PAS, there are no studies that describe the clinical reality of unexpected PAS cases in resource poor settings. The present study describes the maternal outcomes of previously undiagnosed PAS managed in resource poor settings in Colombia and Indonesia.Methods: Retrospective case series of women with histologically confirmed PAS treated in two PAS centers after referral from remote resource poor hospitals.Clinical outcomes were analyzed according to the initial type of management: (1) no caesarean; (2) placenta left in situ after caesarean; (3) partial removal of the placenta after caesarean; and (4) post caesarean hysterectomy. Additionally, we evaluated the use of telemedicine by comparing the outcomes of women in hospitals which used the support of the PAS center during the initial surgery.Findings: Twenty-nine women initially managed in Colombia (n=2) and Indonesia (n=27) were included. The lowest volume of blood loss and the lowest frequency of complications were in women who had their caesarean section deferred (n= 5; 17·2%) and in those who had a delayed placental delivery (n= 5; 20·7%). There were five maternal deaths (14%), all in the group without the help of telemedicine. Four women died due to irreversible shock from uncontrollable bleeding.Interpretations: Previously undiagnosed PAS in a resource poor hospital had a high risk for maternal mortality. Open/close abdominal surgery and leaving the placenta in-situ seem the best choices for unexpected PAS management in resource poor settings. Telemedicine with a PAS center may improve the prognosis.Funding: The Authors declare no funding resource for this submission.Declaration of Interest: The Authors declare that they have no conflicts of interest.Ethical Approval: The study is registered under No.1711-2019 and 016-2022, respectively, and approved by the Institutional Review Board of the SAGH and FVL