Validation of modified sick neonatal score, a simple clinical score for assessment of severity of illness and outcome in new-borns for resource poor settings

Background: India contributes to 25% of the neonatal deaths worldwide each year. Neonatal disease severity scoring systems are needed to make standardized comparison between performances of different units and also give prognostic information. Existing scoring systems are unsuitable for resource-limited settings which lack investigations like pH, pO2/FiO2 ratio and base excess. Modified sick neonatal score (MSNS) is based on eight routinely measured clinical variables in NICUs namely respiratory effort, heart rate, axillary temperature, capillary refill time, random blood sugar, pulse oximeter saturation, gestational age and birth weight found to be useful in resource poor settings. The aim of the study was to validate MSNS score for its clinical utility in predicting mortality.Methods: This was a cross sectional study done at NICU of Mamata Medical College Hospital. The parameters required for the score were recorded immediately at admission in NICU from 1 January 2020 to 1 January 2021 and scored using Modified sick neonatal score (MSNS). The total score was calculated and outcome was noted. The data collected were coded and analzed using SPSS Statistics for Windows, v21.0 Chi square test, Mann-Whitney U test and ROC analysis.Results: Total of 355 neonates got discharged, while 45 neonates expired. For a cutoff score of ≤10, sensitivity and specificity were 85.9% and 51.1%, respectively. Positive predictive value and negative predictive value were 93.3% and 31.5%, respectively. The Area under the curve (AUC) was 0.811 (95%CI: 0.788-0.835), which indicates the accuracy of 81.1%.Conclusions: MSNS is a better suited neonatal disease severity score for resource poor settings.

Design and Implementation of 16 Quality Indices for Neonatal Screening

Background: We aim to explore the realization of 16 quality control indices of neonatal disease screening in neonatal disease management information system, and to establish the multi-center management evaluation model under the framework of the whole province. Methods: Based on the network management system for neonatal screening originally used in our province, the quality management module of neonatal disease was developed and applied according to the quality control index. A quantitative evaluation method was established for the quality of all maternal and child health care institutions involved in neonatal disease screening related to neonatal disease screening, specimen collection and recall of suspected cases. Results: The quality management information was shared by all neonatal screening agencies. Comprehensive evaluation was given to the multi-center quality management measurements. Comprehensive evaluation was performed to quality among different medical institutions. Evaluation of applicability and timeliness of quality in practical work was conducted accordingly. Conclusions: The timeliness of self-improvement in centers and medical institutions were improved. The application of neonatal disease screening quality management index improved the requirements for neonatal screening management. The quality management module was adopted to conduct dynamic management of the whole management process, which greatly improved the overall management level of neonatal screening.

A QUALITATIVE LITERATURE REVIEW OF MACHINE LEARNING TECHNIQUES USED FOR DIAGNOSIS OF NEONATAL DISEASES IN HEALTHCARE

Background:There has been signicant growth in the use of Articial Intelligence (AI) for healthcare in the last decade. Aim: To identify effective AI techniques for the prediction & diagnosis of neonatal diseases and preventive measures & treatment plan for them. Neonates are newborn babies less than a month old. Methods:Research papers published in databases like IEEE Xplore, Medline, PUBMED and Elsevier were searched to nd publications reporting the application of AI for the prediction and prevention of neonatal diseases. The overall search strategy was to retrieve articles that included terms that were related to “NICU”, “Articial Intelligence”, “Neonatal diseases” and “Healthcare”. Results: Hundreds of papers were identied in initial search, out of which 13 publications met the evaluation criteria of related terms inclusion, AI for Neonatal Diseases in particular. These papers described application of AI techniques in neonatal healthcare for disease detection and were summarized for nal analysis. Most of the papers are focused on using supervised machine learning techniques for the prediction of diseases. Various other approaches in AI techniques used in neonatal disease diagnosis have been tested for related ndings, factors, methods, to address and document performance metrics. The comparative analysis of ML model evaluation parameters like AUC (Area under Curve), Specicity, Sensitivity, True Positive and False-negative Rates was done to develop the scope for improving performance of AI/MLtechniques. Conclusion: The systematic study and review of different AI techniques such as supervised machine learning; articial neural networks, data mining techniques used for neonatal disease diagnosis highlighted their role in disease prediction, management, and treatment plan. More studies are needed to improve the use of AI for timely prediction of neonatal diseases like respiratory distress syndrome, sepsis for increasing the survival chances in preterm or normal neonates. The supervised learning models like Support Vector Machines(SVM), Decision Trees, K nearest neighbors are found to be effective for neonatal disease detection and will be applied in future research.

Tetanus in a rural low-income intensive care unit setting

Tetanus is a potentially severe but preventable infection. In resource-rich settings, vaccination programmes have reduced tetanus to a rare disease, though still carrying an overall mortality of some 13%. However, in low income settings, tetanus remains common, and is a significant cause of mortality—though major World Health Organisation programmes are successfully targeting neonatal disease. Data concerning the frequency and outcomes of non-neonatal tetanus in low income settings are very sparse. We aimed to utilize a unique intensive care unit-based dataset to elicit clinical and demographic features and mortality in a large cohort of tetanus patients admitted over an eleven year period to a single hospital centre in a rural low-income setting in northern Uganda. A total of 268 patients with tetanus were admitted to the Intensive Care Unit at St Mary’s Hospital, Lacor between 2005 and 2015; the records of 190 were retrievable and had sufficient information to be assessed. 29 were neonates (median age 7 days, IQR 0), 52 children (1-16yrs; median age 11 years, IQR 4.5) and 109 were adults (median age 42 years, IQR 23). There was no seasonal pattern in the frequency of admissions. Of the 190 patients, 69 had endotracheal intubation with intermitent positive pressure ventilation, and 57 patients had central line placement. The overall mortality was 51.5% – 72.4% in neonatal disease, 25% in children and 57.8% in adults. The requirements for neither central line insertion, nor endotracheal intubation, nor intermittent positive pressure ventilation were independently linked to mortality rates. By contrast with neonatal and childhood disease, there was a marked male preponderance in adult tetanus—94 males and 15 females (gender difference p < 0.001) – and although year-on-year breakdown suggested no obvious upward or downward trend over the span of our study in total numbers of tetanus admissions, a trend towards an increasing incidence of adult tetanus was apparent. These findings confirm that adult tetanus remains a major problem in rural low income settings, particularly in males, and suggests that more resources should be devoted to vaccination programmes targeting men.

Disease burden due to Group B Streptococcus in the Indian population and the need for a vaccine – a narrative review

Streptococcus agalactiae, a Gram-positive bacterium, causes invasive infection known as Group B streptococcal disease (GBS). It is a leading cause of neonatal death and complications prior to delivery. The burden of GBS is unknown in India despite the high incidence of preterm and stillbirths. In this study, we performed a narrative review of the available literature (published in the last 10 years) on the epidemiology of GBS, using PubMed and Google Scholar, to understand its impact in India and evaluate potential strategies to prevent the disease in the high-risk population, that is, neonates. The review showed that the incidence of early- and late-onset GBS in neonates (per 1000 live births) was in the ranges of 0.090–0.68 and 0.0–0.07 respectively. The overall case fatality rate reported in only one study was 0.63. In pregnant women, the prevalence of GBS colonization was 2–62% and its transmission to their newborns varied from 6.7% to 11.1%. The serotype distribution of GBS is unclear, but some studies reported the distribution of types Ia, Ib, II, III, V, VII among pregnant women in India. The associated risk factors for GBS colonization in pregnant women are unclear but a few studies suggest the role of age and multigravida, while the risk factors in neonates include preterm birth, prolonged rupture of membrane (⩾18 h), maternal fever, obstetric complications, and prolonged labor >18 h. Screening of GBS is not a routine practice in India and intrapartum antibiotics prophylaxis is limited to only in risk conditions to prevent neonatal disease transmission. A few studies also suggest that high birth rate, poor detection methods, and financial constraints limit routine GBS screening in a developing country such as India. Hence, maternal vaccination is the most promising strategy to prevent neonatal GBS and Pfizer’s hexavalent GBS conjugate vaccine (GBS6) is being developed for GBS neonatal disease.

Genomic and phenotypic characterisation of invasive neonatal and colonising group B Streptococcus isolates from Slovenia, 2001–2018

Background Group B Streptococcus (GBS) is the leading cause of invasive neonatal disease in the industrialized world. We aimed to genomically and phenotypically characterise invasive GBS isolates in Slovenia from 2001 to 2018 and contemporary colonising GBS isolates from screening cultures in 2018. Methods GBS isolates from 101 patients (invasive isolates) and 70 pregnant women (colonising isolates) were analysed. Basic clinical characteristics of the patients were collected from medical records. Antimicrobial susceptibility and phenotypic capsular serotype were determined. Whole-genome sequencing was performed to assign multilocus sequence types (STs), clonal complexes (CCs), pathogenicity/virulence factors, including capsular genotypes, and genome-based phylogeny. Results Among invasive neonatal disease patients, 42.6% (n = 43) were females, 41.5% (n = 39/94) were from preterm deliveries (< 37 weeks gestation), and 41.6% (n = 42) had early-onset disease (EOD). All isolates were susceptible to benzylpenicillin with low minimum inhibitory concentrations (MICs; ≤0.125 mg/L). Overall, 7 serotypes were identified (Ia, Ib, II-V and VIII); serotype III being the most prevalent (59.6%). Twenty-eight MLST STs were detected that clustered into 6 CCs. CC-17 was the most common CC overall (53.2%), as well as among invasive (67.3%) and non-invasive (32.9%) isolates (p < 0.001). CC-17 was more common among patients with late-onset disease (LOD) (81.4%) compared to EOD (47.6%) (p < 0.001). The prevalence of other CCs was 12.9% (CC-23), 11.1% (CC-12), 10.5% (CC-1), 8.2% (CC-19), and 1.8% (CC-498). Of all isolates, 2.3% were singletons. Conclusions A high prevalence of hypervirulent CC-17 isolates, with low genomic diversity and characteristic profile of pathogenicity/virulence factors, was detected among invasive neonatal and colonising GBS isolates from pregnant women in Slovenia. This is the first genomic characterisation of GBS isolates in Slovenia and provides valuable microbiological and genomic baseline data regarding the invasive and colonising GBS population nationally. Continuous genomic surveillance of GBS infections is crucial to analyse the impact of IND prevention strategies on the population structure of GBS locally, nationally, and internationally.