Use of artificial intelligence in obstetric and gynaecological diagnostics: a protocol for a systematic review and meta-analysis.

INTRODUCTION: Emerging developments in applications of artificial intelligence (AI) in healthcare offer the opportunity to improve diagnostic capabilities in obstetrics and gynaecology (O&G), ensuring early detection of pathology, optimal management and improving survival. Consensus on a robust AI healthcare framework is crucial for standardising protocols that promote data privacy and transparency, minimise bias, and ensure patient safety. Here, we describe the study protocol for a systematic review and meta-analysis to evaluate current applications of AI in O&G diagnostics with consideration of reporting standards used and their ethical implications. This protocol is written following the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 checklist. METHODS AND ANALYSIS: The study objective is to explore the current application of AI in O&G diagnostics and assess the reporting standards used in these studies. Electronic bibliographic databases MEDLINE, EMBASE and Cochrane will be searched. Study selection, data extraction and subsequent narrative synthesis and meta-analyses will be carried out following the PRISMA-P guidelines. Included papers will be English-language full-text articles from May 2015 to March 2024, which provide original data, as AI has been redefined in recent literature. Papers must use AI as the predictive method, focusing on improving O&G diagnostic outcomes.We will evaluate the reporting standards including the risk of bias, lack of transparency and consider the ethical implications and potential harm to patients. Outcome measures will involve assessing the included studies against gold-standard criteria for robustness of model development (Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis, model predictive performance, model risk of bias and applicability (Prediction model Risk Of Bias Assessment Tool and study reporting (Consolidated Standards of Reporting Trials-AI) guidance. ETHICS AND DISSEMINATION: Ethical approval is not required for this systematic review. Findings will be shared through peer-reviewed publications. There will be no patient or public involvement in this study. PROSPERO REGISTRATION NUMBER: CRD42022357024 .

Cardiogenic shock in pregnancy.

Cardiac disease complicates 1%-4% of pregnancies globally, with a predominance in low and middle-income countries (LMICs). Increasing maternal age, rates of obesity, cardiovascular comorbidities, pre-eclampsia and gestational diabetes all contribute to acquired cardiovascular disease in pregnancy. Additionally, improved survival in congenital heart disease (CHD) has led to increasing numbers of women with CHD undergoing pregnancy. Implementation of individualised care plans formulated through pre-conception counselling and based on national and international guidance have contributed to improved clinical outcomes. However, there remains a significant proportion of women of reproductive age with no apparent comorbidities or risk factors that develop heart disease during pregnancy, with no indication for pre-conception counselling. The most extreme manifestation of cardiac disease is cardiogenic shock (CS), where the primary cardiac pathology results in inadequate cardiac output and hypoperfusion, and is associated with significant mortality and morbidity. Key to management is early recognition, intervention to treat any potentially reversible underlying pathology and supportive measures, up to and including mechanical circulatory support (MCS). In this narrative review we discuss recent developments in the classification of CS, and how these may be adapted to improve outcomes of pregnant women with, or at risk of developing, this potentially lethal condition.

New-onset movement disorder in the immediate postpartum period.

This case report describes a woman who developed involuntary, uncoordinated movements of her face and limbs following a spontaneous vaginal delivery, complicated by postpartum haemorrhage. Using systematic assessment with multidisciplinary team input, a differential diagnosis was proposed and relevant investigations were undertaken. Atypical eclamptic or generalised seizures were excluded clinically and neuroimaging ruled out an intracranial vascular event such as stroke or venous sinus thrombosis. Local anaesthetic systemic toxicity was managed empirically with intravenous lipid emulsion and intravenous fluids. A diagnosis of drug-induced dyskinesia was made, most likely secondary to ondansetron, with which dystonias and myoclonus have been described. This woman's symptoms were transient and resolved within 2 hours.This case presented a complex differential diagnosis, highlighting the paucity of guidance available. We propose a diagnostic algorithm to aid in the identification of acute involuntary movements in pregnancy and the puerperium.

Number and function of uterine natural killer cells in recurrent miscarriage and implantation failure: a systematic review and meta-analysis.

BACKGROUND: Uterine natural killer cells (uNK) are the most abundant lymphocytes found in the decidua during implantation and in first trimester pregnancy. They are important for early placental development, especially trophoblast invasion and transformation of the spiral arteries. However, inappropriate uNK function has been implicated in reproductive failure, such as recurrent miscarriage (RM) or recurrent implantation failure (RIF). Previous studies have mainly focussed on peripheral NK cells (pNK), despite the well-documented differences in pNK and uNK phenotype and function. In recent years, there has been an explosion of studies conducted on uNK, providing a more suitable representation of the immune environment at the maternal-foetal interface. Here, we summarize the evidence from studies published on uNK in women with RM/RIF compared with controls. OBJECTIVE AND RATIONALE: The objectives of this systematic review and meta-analysis are to evaluate: differences in uNK level in women with RM/RIF compared with controls; pregnancy outcome in women with RM/RIF stratified by high and normal uNK levels; correlation between uNK and pNK in women with RM/RIF; and differences in uNK activity in women with RM/RIF compared with controls. SEARCH METHODS: MEDLINE, EMBASE, Web of Science and Cochrane Trials Registry were searched from inception up to December 2020 and studies were selected in accordance with PRISMA guidelines. Meta-analyses were performed for uNK level, pregnancy outcome and uNK/pNK correlation. Narrative synthesis was conducted for uNK activity. Risk of bias was assessed by ROBINS-I and publication bias by Egger's test. OUTCOMES: Our initial search yielded 4636 articles, of which 60 articles were included in our systematic review. Meta-analysis of CD56+ uNK level in women with RM compared with controls showed significantly higher levels in women with RM in subgroup analysis of endometrial samples (standardized mean difference (SMD) 0.49, CI 0.08, 0.90; P = 0.02; I2 88%; 1100 women). Meta-analysis of CD56+ uNK level in endometrium of women with RIF compared with controls showed significantly higher levels in women with RIF (SMD 0.49, CI 0.01, 0.98; P = 0.046; I2 84%; 604 women). There was no difference in pregnancy outcome in women with RM/RIF stratified by uNK level, and no significant correlation between pNK and uNK levels in women with RM/RIF. There was wide variation in studies conducted on uNK activity, which can be broadly divided into regulation and receptors, uNK cytotoxicity, cytokine secretion and effect of uNK on angiogenesis. These studies were largely equivocal in their results on cytokine secretion, but most studies found lower expression of inhibitory receptors and increased expression of angiogenic factors in women with RM. WIDER IMPLICATIONS: The observation of significantly increased uNK level in endometrium of women with RM and RIF may point to an underlying disturbance of the immune milieu culminating in implantation and/or placentation failure. Further research is warranted to elucidate the underlying pathophysiology. The evidence for measuring pNK as an indicator of uNK behaviour is sparse, and of limited clinical use. Measurement of uNK level/activity may be more useful as a diagnostic tool, however, a standardized reference range must be established before this can be of clinical use.

Hyponatraemia compounding pre-eclamptic toxaemia in a patient with type 1 diabetes.

We report a case of profound, symptomatic hyponatraemia in association with pre-eclamptic toxaemia (PET) in a 38-year-old nulliparous woman with type 1 diabetes mellitus. This patient developed hypertension and proteinuria at 31+6 weeks' gestation and was admitted for management of pre-eclampsia. Severe headache, visual disturbance and nausea were associated with a hyponatraemia of 115 mmol/L followed by ketoacidosis. This was reversed through fluid restriction, supplementation with 1.8%-3.0% hypertonic saline and a volume-reduced variable-rate insulin infusion. Clinical stability was achieved and she was subsequently worked up for an induction of labour for worsening pre-eclampsia. Hyponatraemia in the context of PET has been previously reported as rare. However, it has complications that may significantly compound the sequelae of severe PET. We propose that specific and focused monitoring of serum sodium levels should become common practice in the management of women with this condition to allow for timely, measured correction of abnormalities.

Is the Host Viral Response and the Immunogenicity of Vaccines Altered in Pregnancy?

The intricacy of the maternal immune system arises from its ability to prevent a maternal immune response against a semi-allogenic fetus, while protecting the mother against harmful pathogens. However, these immunological adaptations may also make pregnant women vulnerable to developing adverse complications from respiratory viral infections. While the influenza and SARS pandemics support this theory, there is less certainty regarding the clinical impact of SARS-CoV-2 in pregnancy. In the current COVID-19 pandemic, vaccine development is key to public preventative strategies. Whilst most viral vaccines are able to induce a seroprotective antibody response, in some high-risk individuals this may not correlate with clinical protection. Some studies have shown that factors such as age, gender, and chronic illnesses can reduce their effectiveness and in this review, we discuss how pregnancy may affect the efficacy and immunogenicity of vaccines. We present literature to support the hypothesis that pregnant women are more susceptible to respiratory viral infections and may not respond to vaccines as effectively. In particular, we focus on the clinical implications of important respiratory viral infections such as influenza during pregnancy, and the pregnancy induced alterations in important leukocytes such as TFH, cTFH and B cells, which play an important role in generating long-lasting and high-affinity antibodies. Finally, we review how this may affect the efficacy of vaccines against influenza in pregnancy and highlight areas that require further research.

Sepsis: Precision-Based Medicine for Pregnancy and the Puerperium.

Sepsis contributes significantly to global morbidity and mortality, particularly in vulnerable populations. Pregnant and recently pregnant women are particularly prone to rapid progression to sepsis and septic shock, with 11% of maternal deaths worldwide being attributed to sepsis. The impact on the neonate is considerable, with 1 million neonatal deaths annually attributed to maternal infection or sepsis. Pregnancy specific physiological and immunological adaptations are likely to contribute to a greater impact of infection, but current approaches to the management of sepsis are based on those developed for the non-pregnant population. Pregnancy-specific strategies are required to optimise recognition and management of these patients. We review current knowledge of the physiology and immunology of pregnancy and propose areas of research, which may advance the development of pregnancy-specific diagnostic and therapeutic approaches to optimise the care of pregnant women and their babies.