Daily Ards Research Analysis
Three studies advance critical and perinatal care: a prospective study shows bedside point-of-care lung ultrasound (POC-LUS) within 2 hours of birth accurately predicts NICU admission in neonates with respiratory distress; a large multi-hospital analysis links hospital quality control to substantially lower in-hospital COVID-19 mortality; and a biomarker- and Doppler-based nomogram predicts adverse perinatal outcomes in fetal growth restriction with strong discrimination.
Summary
Three studies advance critical and perinatal care: a prospective study shows bedside point-of-care lung ultrasound (POC-LUS) within 2 hours of birth accurately predicts NICU admission in neonates with respiratory distress; a large multi-hospital analysis links hospital quality control to substantially lower in-hospital COVID-19 mortality; and a biomarker- and Doppler-based nomogram predicts adverse perinatal outcomes in fetal growth restriction with strong discrimination.
Research Themes
- Point-of-care imaging to triage neonatal respiratory distress
- Hospital quality systems and mortality in pandemics
- Biomarker-integrated risk prediction in fetal growth restriction
Selected Articles
1. Diagnostic Utility of Bedside "Point of Care Lung Ultrasound" in Predicting the Need For NICU Admission in Late Preterm and Term Newborns Having Respiratory Distress Soon After Birth in the Transition Period: A Prospective Observational Study.
In a prospective cohort of 97 late preterm and term neonates with respiratory distress, a POC-LUS score >5/18 within 2 hours of birth predicted NICU admission with AUC 0.903, sensitivity 64%, and specificity 98% (positive likelihood ratio 35). LUS scores correlated weakly with Silverman–Anderson scores, supporting LUS as an early triage tool.
Impact: Provides a rapid, non-invasive triage metric for early identification of neonates likely to require intensive care, with high specificity that can streamline resource allocation.
Clinical Implications: Implementing a POC-LUS score threshold (>5/18) within 2 hours of birth can prioritize NICU admission decisions and guide early respiratory support in resource-constrained settings.
Key Findings
- POC-LUS score >5/18 predicted NICU admission with AUC 0.903.
- Sensitivity 64% and specificity 98% (positive likelihood ratio 35; p<0.001).
- Weak positive correlation between LUS and Silverman–Anderson scores (r=0.325; p=0.001).
Methodological Strengths
- Prospective design with NICU admission decisions blinded to LUS findings.
- Use of a pre-validated LUS scoring system and ROC analysis.
Limitations
- Single-center study with a relatively small sample size (N=97).
- No external validation; sensitivity was moderate.
Future Directions: External validation across diverse centers and integration of LUS with clinical and biochemical markers could refine predictive performance and implementation pathways.
BACKGROUND: Point of care lung ultrasound (POC-LUS) is a rapid and simple method to evaluate infants with respiratory distress after birth. OBJECTIVES: The primary objective was to determine whether the POC-LUS score is a good predictor of NICU admission in late preterm and term infants born with respiratory distress when performed within the first 2 h of life. The secondary objective was to find a correlation between the LUS score and the clinical respiratory distress severity score. METHODS: A prospective observational study was carried out in a tertiary care neonatal unit (Level III) over 1 year on 97 late preterm and term infants having respiratory distress at birth. POC-LUS was performed in a transition nursery area within 2 h of birth, and LUS score was recorded as per a pre-validated LUS scoring system. The decision for NICU admission was independently taken by the medical team based on clinical criteria and blinded to the LUS findings. A receiver operating characteristic (ROC) curve was generated to predict NICU admission based on the LUS score. LUS score was also analyzed for correlation with clinical respiratory distress severity scoring, that is, Silverman-Anderson score (SA score). RESULTS: The mean gestational age of the infants in the study was 37.45 ± 1.88 weeks. Fourty-three percent of infants needed NICU admission. LUS score > 5/18 performed within 2 h after birth was an excellent predictor of NICU admission in late preterm and term infants with respiratory distress after birth (area under ROC curve 0.903, sensitivity 64%, specificity 98%, positive likelihood ratio 35, and p < 0.001). LUS score also had a weak positive correlation with the SA score (Pearson's correlation, r = 0.325; p = 0.001). CONCLUSION: A LUS score of > 5/18 is an excellent predictor of NICU admission in term and late-preterm infants with respiratory distress after birth.
2. Quality control of hospitals and its effect on hospitalized fatality rate of COVID-19.
Across 78,217 hospitalized adults with COVID-19 in 34 hospitals, attribute control charts identified hospital quality confirmation status, which was strongly associated with in-hospital fatality. Non–quality-confirmed hospitals had higher adjusted odds of death versus quality-confirmed hospitals at the first peak (aOR 3.48; 95% CI 2.38–5.10), with negative interactions indicating quality mitigated risk during surges.
Impact: Demonstrates, at scale, that systematic quality control is associated with lower mortality during a pandemic, informing health system preparedness and quality improvement strategies.
Clinical Implications: Implementing continuous quality monitoring (e.g., attribute control charts) and ensuring hospitals achieve quality confirmation can reduce in-hospital COVID-19 mortality, especially during peaks.
Key Findings
- 78,217 hospitalized COVID-19 patients across 34 hospitals were analyzed over four epidemic peaks.
- Non–quality-confirmed hospitals had higher adjusted odds of death versus quality-confirmed hospitals at the first peak (aOR 3.48; 95% CI 2.38–5.10).
- Quality interacted negatively with surge peaks, suggesting quality measures mitigated fatality risk; no significant interaction with age, SaO2, or ARDS.
Methodological Strengths
- Very large multi-hospital cohort with formal quality assessment via attribute control charts.
- Adjusted regression modeling (Poisson and binomial) accounting for patient characteristics.
Limitations
- Retrospective observational design with potential residual confounding and hospital-level unmeasured factors.
- Generalizability may be limited to one province and specific timeframes of the pandemic.
Future Directions: Prospective implementation studies testing real-time control chart–driven interventions and cross-country validation of quality metrics on outcomes during respiratory pandemics.
The reported fatality rate of COVID-19 was significantly differ between different countries, provinces in a country, and hospitals in a province. It is important to find and analysed the source of variation in fatality rates to control future epidemics of infection disease. This study propose an approach to investigates the hospital quality of care and its impact on the in-hospital fatality rates of COVID-19 among over 30 years old patients in Iran. The study included 78,217 COVID-19 hospitalized patients over 30 years old from 34 hospitals in Razavi Khorasan province between January 20, 2020, and June 20, 2021. Attribute control charts were employed to evaluate the quality of care in hospitals during four peaks of COVID-19. To account for the impact of hospital quality on in-hospital fatality rates by adjusting on patient characteristics, two predictive models; Poisson and Binomial regression were utilized. The adjusted odds ratio of COVID-19 fatality in no quality confirmed (NQC) hospitals to the quality confirmed (QC) hospitals is 3.48 C.I.95% (2.38, 5.10) at the first peak among patients without other risk factors. The significant negative interactions with QC and peaks indicates that the provision of quality care serves to mitigate the risk of fatality in NQC hospitals during these peaks. The significant negative interaction between diabetes and QC demonstrated that the relative risk in NQC hospitals for no diabetes patient is higher than diabetic patients. There is no significant interaction between age, SaO2, and ARDS with QC; it means that the odds of fatality in NQC hospitals is not depends to these factors. The use of attribute control charts facilitated timely identification of trends and outliers, enabling proactive interventions to enhance patient care. This research underscores the critical role of hospital quality in managing patient outcomes during the COVID-19 pandemic and emphasizes the necessity for continuous monitoring and quality improvement initiatives in healthcare settings. The findings advocate for collaborative efforts among healthcare providers to implement best practices, particularly during surges in cases, to optimize resource allocation and improve overall quality of care.
3. A nomogram for predicting adverse perinatal outcome with fetal growth restriction: a prospective observational study.
A prospective cohort of 122 FGR pregnancies yielded a five-variable nomogram (gestational age at diagnosis, umbilical and uterine artery Dopplers, PlGF MoM, sFlt-1 MoM) predicting adverse perinatal outcomes with strong discrimination (AUC 0.87 training; 0.86 validation) and favorable calibration and decision-curve utility.
Impact: Integrates Doppler hemodynamics with placental biomarkers to deliver a clinically usable risk tool for FGR, informing timing of delivery and neonatal preparedness.
Clinical Implications: The nomogram can guide surveillance intensity and delivery planning for FGR, potentially reducing complications such as neonatal respiratory distress syndrome and prolonged NICU stay.
Key Findings
- Five predictors (GA at diagnosis, abnormal UA/UtA Dopplers, PlGF MoM, sFlt-1 MoM) formed the nomogram.
- Discrimination was strong: AUC 0.87 (training) and 0.86 (validation).
- Calibration and decision curve analyses supported clinical applicability.
Methodological Strengths
- Prospective cohort with predefined inclusion and internal validation.
- Variable selection via LASSO integrating clinical, ultrasound, and biomarker data.
Limitations
- Single-center study with a modest sample size (N=122).
- Lack of external validation; generalizability remains to be tested.
Future Directions: External, multicenter validation and evaluation of how the nomogram changes management decisions and neonatal outcomes in pragmatic trials.
BACKGROUND: Fetal growth restriction (FGR) is a major determinant of perinatal morbidity and mortality. Our study aimed to develop a prediction model for the risk of FGR developing adverse perinatal outcome (APO) and evaluate its performance. METHODS: This was a prospective observational cohort study of consecutive singleton gestations meeting the ACOG-endorsed criteria for FGR from January 2022 to June 2023 at Obstetrics and Gynecology Hospital of Fudan University. Clinical information, ultrasound indicators and serum biomarkers were collected. The primary composite APO comprised one or more of: perinatal death, intrauterine demise, intraventricular hemorrhage, periventricular leukomalacia, seizures, necrotizing enterocolitis, neonatal respiratory distress syndrome, sepsis and the length of stay in the neonatal intensive care unit > 7 days. Least absolute shrinkage and selection operator regression was used to screen variables for nomogram model construction. The discrimination, calibration and clinical effectiveness of the nomogram were evaluated using receiver operating characteristic curve, calibration plots and decision curve analysis in training and validation cohorts. RESULTS: A total of 122 pregnancies were enrolled in the final statistical analysis. Five variables were identified to establish a nomogram, including gestational weeks at diagnosis, abnormal umbilical artery Doppler, abnormal uterine artery Doppler, and multiples of the median values of placental growth factor and soluble fms-like tyrosine kinase-1. The area under the receiver-operating-characteristics curve of 0.87 (95% CI, 0.75-0.99) and 0.86 (95% CI, 0.74-0.98) in the training and validation cohort respectively, indicated satisfactory discriminative ability of the nomogram. The calibration plots showed favorable consistency between the nomogram's predictions and actual observations. Decision curve analysis supported its practical value in a clinical setting. CONCLUSIONS: A nomogram was developed and validated to possess the promising capacity of predicting APO in FGR-afflicted neonates, and may prove useful in counseling and management of pregnancies complicated by FGR.