Co‐Occurrence of Osteogenesis Imperfecta Type III and Chronic Abruption‐Oligohydramnios Sequence: A Case Report Suggesting a Possible Role of Type I Collagen Fragility

ABSTRACT

Pregnancy complicated by severe osteogenesis imperfecta (OI) is rare, and chronic abruption‐oligohydramnios sequence (CAOS) is an uncommon obstetric disorder characterized by persistent bleeding and oligohydramnios without membrane rupture. To our knowledge, this is the first report describing the coexistence of type III OI and CAOS. A 24‐year‐old woman with OI developed recurrent bleeding early in pregnancy. Genetic testing revealed a novel COL1A1 splice‐site variant (c.1876‐2A>C). Ultrasonography showed a circumferential subchorionic hematoma and progressive fluid loss, leading to CAOS diagnosis at 19 weeks. Despite amnioinfusion, the amniotic cavity collapsed, and pregnancy was terminated via cesarean section at 20 weeks. Placental pathology showed hemosiderin and fibrin deposition. In this case, both functional and structural factors possibly associated with type I collagen abnormalities may have contributed to CAOS development. Although causality cannot be proven, persistent bleeding in severe OI may warrant greater caution for progression to CAOS.

1. Introduction

Osteogenesis imperfecta (OI) is a hereditary connective tissue disorder caused by type I collagen abnormalities due to COL1A1 and COL1A2 mutations. Clinical features include recurrent fractures, bone deformities, short stature, scoliosis, blue sclerae, and hearing loss [1]. Type III OI is a severe form, characterized by marked skeletal deformities, short stature, and respiratory dysfunction due to thoracic narrowing [1, 2]. Type III OI is extremely rare, with an incidence of 1 in 70 000–100 000 live births [1, 2, 3]. Pregnancy in women with OI carries increased risks such as antepartum hemorrhage, placental abruption, maternal fractures, gestational diabetes, and neonatal morbidity [4, 5]. Fertility is generally preserved in mild cases, with pregnancy often proceeding without major complications [6]. Contrastingly, reports of pregnancies complicated by severe OI, including type III, are extremely limited [2, 3], most focusing on anesthetic management during delivery. In early pregnancy, only one case has been reported, describing termination with methotrexate at 13 weeks [7].

Chronic abruption‐oligohydramnios sequence (CAOS) is a rare obstetric condition characterized by persistent vaginal bleeding followed by oligohydramnios, without membrane rupture and in the absence of an identifiable bleeding source, such as placenta previa [8]. It is often associated with poor fetal outcomes, particularly when bleeding begins early in pregnancy, which tends to result in a more severe clinical course [9].

Herein, we describe a pregnancy complicated by type III OI, with persistent vaginal bleeding beginning in early gestation, followed by CAOS diagnosis at 19 weeks, ultimately leading to pregnancy termination via cesarean section at 20 weeks. Such coexistence has not previously been reported. We also discuss potential pathophysiological mechanisms underlying CAOS, focusing on the possible role of type I collagen abnormalities associated with OI.

2. Case Presentation

The patient was a 24‐year‐old woman, 110 cm tall and weighing 26.7 kg, gravida 2 para 0. She had sustained fractures at birth and experienced > 20 fractures during early childhood. There was no notable family history. She had been wheelchair‐bound since the age of 3 years. Radiographic imaging at age 20 years (Figure 1a–c) revealed severe scoliosis, a deformed, narrow thoracic cage, and a contracted pelvis. Clinical features were consistent with OI, including blue sclerae and dentinogenesis imperfecta. After undergoing orthopedic surgery for a fracture at age 20 years, she had not received routine medical follow‐up.

FIGURE 1.

Radiographic images at 20 years of age. (a) Full‐body view of the trunk. (b) Chest radiograph showing marked scoliosis and a severely deformed, narrowed thoracic cage. (c) Pelvic radiograph demonstrating severe pelvic deformity.

Her first pregnancy ended in a spontaneous miscarriage during early gestation, with natural expulsion of the gestational sac. In her second spontaneous pregnancy, persistent vaginal bleeding began early. At 13 weeks' gestation, she was referred to our hospital due to relocation. Initial ultrasonography revealed a circumferential subchorionic hematoma (SCH) along the chorion laeve, sparing the placental attachment site; the amniotic cavity was preserved (Figure 2a,b). A detailed early pregnancy scan showed no apparent fetal structural abnormalities.

FIGURE 2.

Ultrasonographic and magnetic resonance imaging findings during pregnancy. (a) At 13 weeks' gestation: A large subchorionic hematoma (SCH) is delineated by a dashed line. (b) At 13 weeks: The maximum vertical pocket (MVP) measures 28 mm. (c) At 19 weeks: A large SCH persists. (d) At 19 weeks: Amniotic fluid is nearly absent, and fetal growth restriction (FGR) is evident. (e) T2‐weighted sagittal image showing a severely deformed and extremely contracted pelvis, with an anteroposterior diameter of approximately 3 cm between the sacral promontory and pubic symphysis (double‐headed arrow). The fetal head is visible within the pelvic cavity. (f) T2‐weighted coronal image demonstrating a subchorionic hematoma (SCH; dashed line), the placenta (*), and the fetal lower leg (arrow).

The patient had not undergone prior genetic testing and wished to understand potential fetal implications. She and her partner received genetic counseling and consented to testing for OI. Genetic analysis identified a novel heterozygous variant in the COL1A1 gene: Chr17:50192695 c.1876‐2A>C (splice acceptor variant).

During subsequent prenatal visits, the SCH persisted and vaginal bleeding continued. The amniotic fluid volume gradually decreased, and fetal growth restriction (FGR) developed. At 16 weeks, the patient's blood pressure began to rise, prompting the initiation of antihypertensive therapy. By 19 weeks, the amniotic cavity was nearly obliterated (Figure 2c), leading to CAOS diagnosis. Concurrent FGR was noted (Figure 2d), and the patient was admitted for inpatient management.

Magnetic resonance imaging was performed to assess the extent of hematoma and pelvic anatomy. The obstetric conjugate diameter measured only 3 cm (Figure 2e), and the hematoma extended circumferentially around the placental margin (Figure 2f). Following counseling, we recommended amnioinfusion to allow fetal morphological evaluation and assess the feasibility of continuing the pregnancy, and consent was obtained. The procedure temporarily restored the amniotic cavity, permitting fetal visualization. Mid‐trimester ultrasonography revealed no structural anomalies. However, by the subsequent day, the amniotic cavity had collapsed again, and the patient reported fluid leakage from the vagina. She subsequently developed gestational hypertensive nephropathy and worsening anemia. Given the maternal and fetal conditions, pregnancy continuation was deemed unfeasible, and elective termination was chosen.

With multidisciplinary support from anesthesiology, nephrology, cardiology, orthopedics, and psychiatry, an elective cesarean section was performed at 20 weeks and 6 days under general anesthesia. Preoperative imaging showed a markedly deformed pelvis with an extremely narrow birth canal, while the fetal biparietal diameter was approximately 4 cm, rendering vaginal delivery physically impossible. No intraoperative or postoperative complications occurred. The fetus showed no overt phenotypic abnormalities, including fractures. The placenta had a circummarginate configuration (Figure 3a), and histopathology revealed hemosiderin (Figure 3b) and prominent fibrin deposition (Figure 3c). The postoperative course was uneventful, and the patient was discharged on postoperative day 6 with continued oral antihypertensive therapy.

FIGURE 3.

Placental findings in a chronic abruption‐oligohydramnios sequence (CAOS). (a) Macroscopically, the placenta exhibits a circumvallate configuration. The fetal surface shows yellow‐brown discoloration. The placental parenchyma appears dark red with heterogeneous coloration and was globally friable, suggestive of chronic hemorrhagic changes. (b) Histological examination of the chorionic plate reveals hemosiderin deposition (arrowhead) and phagocytosis by macrophages. (c) Histological examination of the placental marginal area shows prominent fibrin deposition.

3. Discussion

To our knowledge, no previous reports have described CAOS complicating an OI type III pregnancy. In the present case, genetic testing revealed a heterozygous COL1A1 variant, Chr17:50192695 c.1876‐2A>C, classified as a splice acceptor variant and considered a novel pathogenic mutation. A mild form of OI associated with a different mutation (c.1876‐2delA) at the same splice acceptor site has been reported [10]. In the present case, variant interpretation by the Japanese Consortium for Research on Skeletal Dysplasia predicted a relatively mild phenotype, based on the assumption that the splicing abnormality would lead to nonsense‐mediated mRNA decay (NMD) and haploinsufficiency.

However, the patient exhibited a severe clinical phenotype consistent with OI type III. This discrepancy may reflect variable splicing outcomes, whereby exon skipping may allow some aberrant transcripts to escape NMD and potentially result in dominant‐negative effects. Although the precise splicing outcome could not be confirmed owing to the lack of RNA‐level analyses, this case highlights the complexity of genotype–phenotype correlations in OI, even among variants affecting the same splice‐site region.

In this case, there was no evidence of abnormal placental location, such as placenta previa. However, persistent vaginal bleeding began early in the pregnancy, accompanied by an enlarging SCH extending into the second trimester. Although the amniotic cavity was initially intact, it gradually diminished and eventually collapsed without clear evidence of membrane rupture. While iatrogenic premature rupture of membranes cannot be completely excluded, clinical features consistent with CAOS were present before amnioinfusion in this case. Accordingly, this clinical course is more compatible with marked membrane fragility associated with CAOS. Histopathological examination of the placenta revealed hemosiderin deposition, consistent with CAOS.

Type I collagen is abundant in the myometrium, decidua, and amnion and is essential for pregnancy maintenance [11, 12, 13, 14]. Furthermore, patients with OI have a recognized bleeding tendency, which may be partially due to vascular endothelial dysfunction [15]. In this case, reduced myometrium elasticity, vascular fragility possibly associated with type I collagen abnormalities, together with structural factors such as a contracted pelvis and narrow birth canal hindering hematoma drainage, likely contributed to CAOS development.

Although this case demonstrates a temporal and pathophysiological association between severe OI and CAOS development, no molecular or histologic evidence directly demonstrates causality. Moreover, the clinical definition and diagnostic boundaries of CAOS remain incompletely established, particularly in differentiating it from related conditions such as preterm premature rupture of membranes. In this case, the clinical features were considered consistent with CAOS based on the presence of persistent vaginal bleeding without an identifiable placental source, progressive oligohydramnios in the absence of clinical evidence of membrane rupture before intervention, and supportive imaging and pathological findings. This reflects a practical diagnostic approach used in clinical decision‐making rather than a formally established diagnostic criterion. Nevertheless, the combination of persistent subchorionic hematoma, membrane collapse after amnioinfusion, and histopathologic evidence of chronic hemorrhage represents a biologically plausible consequence of collagen dysfunction. Antepartum hemorrhage is a well‐recognized risk in OI pregnancies; particularly, when persistent bleeding occurs in severe OI, clinicians should consider the possibility of CAOS and monitor carefully.

This case highlights that connective‐tissue fragility in OI may contribute to the pathogenesis of obstetric complications such as CAOS. It should be regarded as hypothesis‐generating, underscoring the need for nationwide or multicenter case accumulation and further clinical and molecular studies to refine diagnostic frameworks and clarify the mechanisms underlying CAOS.

Author Contributions

A.O. managed the patient and drafted the manuscript. T.I., M.G., and K.S. were involved in diagnosis and treatment planning and contributed to the clinical management. M.T. was responsible for genetic counseling and contributed to the interpretation of the genetic findings. K.I. supervised the case management and critically revised the manuscript. All authors have read and approved the final version of the manuscript.

Funding

The authors have nothing to report.

Disclosure

An earlier version of this case was presented at the 44th Annual Meeting of the Japan Society of Maternal and Fetal Medicine, which was held in Sendai, Japan, on 2–4 September 2022.

Ethics Statement

This study was approved by the Institutional Review Board of Showa Medical University, approval number CR2025017.

Consent

Written informed consent was obtained from the patient for participation in the study and for publication of the case details and accompanying images.

Conflicts of Interest

The authors declare no conflicts of interest.

Okuyama A., Izdebski T., Goto M., Seo K., Tominaga M., and Ichizuka K., “Co‐Occurrence of Osteogenesis Imperfecta Type III and Chronic Abruption‐Oligohydramnios Sequence: A Case Report Suggesting a Possible Role of Type I Collagen Fragility,” Journal of Obstetrics and Gynaecology Research 52, no. 2 (2026): e70210, 10.1111/jog.70210.

Data Availability Statement

The data supporting the findings of this study consist of detailed individual‐level clinical information, imaging findings, histopathological results, and genetic data from a single patient. Due to the rare nature of the condition and the inclusion of potentially identifiable clinical and genetic information, these data cannot be publicly shared, even in anonymized form, in order to protect patient privacy and to comply with ethical and legal requirements. The study was conducted with approval from the institutional review board, and written informed consent was obtained for publication of the case details and accompanying images. Accordingly, data access is restricted.