Death within the first 28 days of life, or neonatal mortality, is still a major worldwide health issue, especially in low- and middle-income (LMIC) nations. [1] The World Health Organization (WHO) estimates that 2.4 million neonates die annually, making up almost half of all child fatalities under five. [2] Most of these deaths have place in LMICs, where timely access to newborn care is limited and healthcare infrastructure is frequently underfunded.  More than half of India's under-five mortality rate is attributable to neonatal mortality, which is still a serious problem in the nation.  This figure emphasizes how urgently efforts are needed to lower infant mortality, particularly in groups that are already at risk. [3]

Prematurity, low birth weight, congenital defects, infections, and labor and delivery problems are some of the variables that affect newborn mortality.  The most frequent causes of neonatal deaths among these are disorders like birth asphyxia, neonatal infection, and prematurity-related problems. [4] Because their organs and immune systems are still developing, premature and low-birth-weight babies are especially prone to infections and other problems.  Despite the fact that improvements in newborn care have greatly increased survival rates in high-income nations, systemic obstacles to timely and specialized healthcare

Have kept mortality rates in LMICs disproportionately high.  High infant mortality rates are further exacerbated by the fact that neonatal intensive care units (NICUs), which offer specialized care for critically ill newborns, are frequently in short supply or overburdened in LMICs.

In India newborn fatalities place a heavy load on the nation's healthcare system, with a newborn mortality rate of about 22 per 1,000 live births. [5] Tertiary-level facilities receive referrals from primary or peripheral healthcare centers for many of the infants in need of specialist care.  However, greater mortality rates are frequently the consequence of delays in diagnosis, transportation, and intervention, especially for outborn neonates—those born outside of tertiary care facilities. [6] Due to delayed availability to advanced medical treatment and the absence of early resuscitation procedures after birth, outborn neonates are generally at higher risk of bad outcomes than inborn neonates.  Higher death rates in this demographic are frequently the consequence of both the severity of their medical illnesses and the delay in receiving care.

Type of Study:

The study was an institution-based observational study with a prospective design.

Place of study and Duration: The study was conducted for 12 months, from November 2023 to December of 2024.

Sample Size: 120 patients in this study

Inclusion:

The inclusion criteria for the study population were

1. Outborn neonates referred to the level 3 NICU of the study institution
2. Those neonates whose parents/caregivers provided written informed consent to take part in the study.

Exclusion:

The exclusion criteria for the participants that were considered for the current study were as follows 1. Neonates with congenital anomalies

2. Neonates with any significant illness at birth

Statistical Analysis:

For statistical analysis, data were initially entered into a Microsoft Excel spreadsheet and then analyzed using SPSS (version 27.0; SPSS Inc., Chicago, IL, USA) and GraphPad Prism (version 5). Numerical variables were summarized using means and standard deviations, while categorical variables were described with counts and percentages. Two-sample t-tests, which compare the means of independent or unpaired samples, were used to assess differences between groups. Paired t-tests, which account for the correlation between paired observations, offer greater power than unpaired tests. Chi-square tests (χ² tests) were employed to evaluate hypotheses where the sampling distribution of the test statistic follows a chi-squared distribution under the null hypothesis; Pearson's chi-squared test is often referred to simply as the chi-squared test. For comparisons of unpaired proportions, either the chi-square test or Fisher’s exact test was used, depending on the context. To perform t-tests, the relevant formulae for test statistics, which either exactly follow or closely approximate a t-distribution under the null hypothesis, were applied, with specific degrees of freedom indicated for each test. P-values were determined from Student's t-distribution tables. A p-value ≤ 0.05 was considered statistically significant, leading to the rejection of the null hypothesis in favour of the alternative hypothesis.

Table 1. Distribution of participants according to their indication for NICU admission (n=120)

Table 2. Distribution of participants according to their duration of transport to study institution (n=120)

Table3. Distribution of participants according to their maternal characteristics (n=120)

The most common indications for NICU admission were respiratory distress syndrome (76.7%, n=92) and early-onset sepsis (69.2%, n=83). Other notable conditions included disseminated intravascular coagulation (DIC) and shock, each at 10% (n=12), followed by meconium-stained liquor (10%, n=12). Neonatal hyperbilirubinemia and acute kidney injury were present in 9.2% (n=11) of cases each. Hypoxic ischemic encephalopathy (HIE) occurred in 7.5% (n=9), and pulmonary hypertension (PPHN) was reported in 4.2% (n=5). Most neonates (54.2%, n=65) took 1-2 hours to reach the study institution, followed by 33.3% (n=40) transported within 1 hour. A smaller proportion (12.5%, n=15) had transport durations exceeding 2 hours. The mean maternal age was 25.1 years, and the mean gestational age was 34.9 weeks. The majority of mothers were multipara (86.7%, n=104). Small for gestational age (SGA) neonates accounted for 7.3% (n=9), while large for gestational age (LGA) neonates were only 0.8% (n=1). Complications like Rh negativity (3.3%, n=4) and twin pregnancies (5.8%, n=7) were noted. The mean hospital stay was 6.9 ± 7.2 days. Out of 120 neonates, 55% (n=66) were discharged, while 44.2% (n=53) died. One neonate (0.8%) left against medical advice. The high mortality rate highlights the critical condition of referred neonates. There was a strong negative correlation observed between the post-transport TOPS and the SNAP PE II scores in the participants (p-value <0.001). It was seen that at a cut-off of 5.5, pre-transport TOPS score had a sensitivity of 89.6% and specificity of 84.9% in the prediction of death in the participants (AUC 0.919, p-value <0.001).

The purpose of the current study was to assess the predictive value of the SNAP-PE II (Score for Neonatal Acute Physiology II with Perinatal Extension) and TOPS (Temperature, Oxygen Saturation, Perfusion, and Sugar) scores in predicting neonatal mortality among outborn neonates admitted to a tertiary-level NICU.  By comparing the TOPS scores before and after transfer, the study also investigated how transit affected the physiological stability of the newborn.  According to the findings, 81.7% of newborns were admitted during the first twelve hours of their lives. As studies by Parekh et al. have shown, early admission after transport is a significant factor impacting newborn outcomes. [7] (2018) and Hapani et al. [8] (2019). Parekh et al. [7] highlighted the importance of pre-referral stabilization and timely transport to prevent mortality, whereas Hapani et al. [8] emphasized that One important indicator of mortality, hypoxemia, is made worse by delays.  In line with Verma et al. [9] (2017), who noted comparable transit durations and higher mortality among infants with extended delays, a sizable fraction of the neonates in the current research were transported for one to two hours. These findings underscore the need for efficient transport systems, as delayed interventions following transport can exacerbate neonatal morbidity. The current study identified that a majority of neonates had low birth weights (mean 2.1 kg) and were born to multiparous mothers. Similar findings were reported by Lima et al. [10] (2020), who observed a high prevalence of low birth weight among non survivors. The current study's mean gestational age of 34.9 weeks emphasizes the preterm neonates' fragility, which is consistent with Meshram et al.'s findings. [11] (2023), where prematurity and low birth weight were significant predictors of mortality. Respiratory distress syndrome (76.7%) and early-onset sepsis (69.2%) were the most common indications for NICU admission. These results are consistent with research by Verma et al. [12] (2017), who reported similar morbidity patterns among outborn neonates. Early onset sepsis remains a significant concern, particularly among transported neonates, as it often exacerbates hypoxia and hemodynamic instability, further increasing mortality risks. A statistically significant deterioration in the TOPS score was observed during transport, as indicated by a mean increase from 6.1 ± 1.2 pre-transport to 6.6 ± 1.1 post-transport (p=0.001). This deterioration aligns with findings by Asari et al. [13], who observed that during unplanned travel, infant physiological parameters—in particular, temperature and oxygen saturation—frequently deteriorate.. Similarly, Shah et al. [14] (2023) reported that Pre-transport stabilization considerably decreased mortality, particularly when it came to treating hypoxia and hypothermia.  The results of the current study confirm the vital significance of pre-transport stabilization, especially for treating hypothermia, an often reported problem. in Cavallin et al. (2022).[15]

 The ability to predict mortality among outborn neonates admitted to the NICU is critical for prioritizing care and improving outcomes. The accuracy of the TOPS and SNAP-PE II scores in forecasting newborn fatalities was assessed in the current study.  The sensitivity and specificity were 76.1% and 94.3%, respectively, with an AUC of 0.918 (p<0.001) at a post-transport TOPS score cut-off of 5.5.  With an AUC of 0.903 (p<0.001), a cut-off of 21 for the SNAP-PE II score produced a sensitivity of 94.2% and a specificity of 79.1%.  These findings demonstrate both scores' great predictive power, despite their differing degrees of sensitivity and specificity. TOPS's practical utility in emergency situations, where quick mortality risk assessment is essential for resource allocation, is highlighted by its capacity to provide such precision with a small set of parameters.  The SNAP-PE II score, on the other hand, showed more sensitivity and was able to capture a greater range of illness severity due to its broader physiological parameters.

Richardson et al. (2001),[16] who validated SNAP-PE II, found an AUC of 0.91 ± 0.01, establishing its role as a robust mortality predictor. Similarly, Lima et al. [17] (2020) identified SNAP-PE II as highly correlated with neonatal outcomes, noting that parameters such as low birth weight, hypothermia, and respiratory compromise were key contributors to mortality. In the present study, SNAP-PE II’s higher sensitivity makes it particularly valuable for detecting subtle but significant clinical deterioration, which may otherwise be missed with simpler tools like TOPS. The importance of hypothermia, hypoxia, and poor perfusion—the core components of the TOPS score—was particularly evident in this study. Hypoxia, identified in a large subset of non-survivors, has consistently been reported as a major contributor to neonatal mortality. For example, Pathak et al. [18] (2019) found that hypoxia had a sensitivity of 92.2% for predicting adverse outcomes, aligning with the present findings. The observed specificity of TOPS in this study underscores its practical use in identifying neonates at immediate risk of death. However, the SNAP-PE II score, despite lower specificity, provided superior sensitivity. This dual finding emphasizes the complementary roles of the two tools: TOPS for rapid identification of critical cases in resource-constrained environments and SNAP-PE II for comprehensive risk stratification where advanced diagnostics are available.

Ultimately, both scores demonstrated strong predictive power, reflecting their utility in mortality assessment. The TOPS score, with its simplicity, offers an accessible tool for bedside evaluation, as emphasized by Hapani et al. [19] (2019). On the other hand, SNAP-PE II remains invaluable in tertiary settings, where its higher sensitivity allows the simplicity of the TOPS score—relying on four easily measurable parameters— remains its greatest advantage. Similarly, Hapani et al. [19] (2019) demonstrated that TOPS parameters such as hypoxia and hypothermia were highly predictive of mortality, with sensitivity and specificity values approaching those of SNAP-II. These findings reinforce the utility of TOPS in settings where advanced monitoring or laboratory support is unavailable. On the other hand, the SNAP-PE II score, though more resource-intensive, provides a comprehensive assessment of illness severity. By incorporating multiple physiological variables and perinatal factors, SNAP-PE II enables a more thorough risk stratification. The present study similarly demonstrated the superior sensitivity of SNAP-PE II, making it particularly valuable for identifying neonates with subtle yet life-threatening physiological derangements. In contrast, the TOPS score, with its focus on basic clinical parameters, is both feasible and effective in such settings.

The study's post-transport TOPS and SNAP-PE II scores showed a substantial negative association (r = -0.725, p<0.001), suggesting that both instruments accurately measure mortality risk, albeit in different ways.  While SNAP-PE II provides a more thorough, systemic assessment, TOPS concentrates on immediate clinical indicators, making it perfect for quick triage.  Neonatal mortality was significantly influenced by hypoxia and inadequate perfusion, according to the analysis of the individual TOPS components. Hypothermia was another prevalent parameter, consistent with findings by Shah et al.[14] (2023), both of which emphasized the association between hypothermia and adverse outcomes. Shah et al. [14] particularly noted that effective pre-transport stabilization to address hypothermia significantly improved survival rates.

We concluded that highlights the utility of both TOPS and SNAP-PE II scores in predicting neonatal mortality among outborn NICU admissions. TOPS, with its simplicity, is ideal for rapid assessments in resource-limited settings, while SNAP-PE II provides comprehensive insights in advanced care environments. Both tools demonstrated strong predictive accuracy, supporting their integration into neonatal transport protocols to enhance clinical decision-making. The findings emphasize the importance of pre-transport stabilization, particularly addressing hypothermia and hypoxia, to improve neonatal outcomes. Implementing standardized scoring systems could optimize resource allocation and reduce neonatal mortality in tertiary care settings.

1. Phukan D, Ranjan M, Dwivedi LK. Impact of timing of breastfeeding initiation on neonatal mortality in India. International breastfeeding journal. 2018 Dec;13:1-0.
2. Dandona R, Kumar GA, Henry NJ, Joshua V, Ramji S, Gupta SS, Agrawal D, Kumar R, Lodha R, Mathai M, Kassebaum NJ. Subnational mapping of under-5 and neonatal mortality trends in India: the Global Burden of Disease Study 2000–17. The Lancet. 2020 May 23;395(10237):1640-58.
3. Houweling TA, van Klaveren D, Das S, Azad K, Tripathy P, Manandhar D, Neuman M, de Jonge E, Been JV, Steyerberg E, Costello A. A prediction model for neonatal mortality in low-and middle-income countries: an analysis of data from population surveillance sites in India, Nepal and Bangladesh. International journal of epidemiology. 2019 Feb 1;48(1):186-98.
4. Kumar GA, Dandona R, Chaman P, Singh P, Dandona L. A population-based study of neonatal mortality and maternal care utilization in the Indian state of Bihar. BMC pregnancy and childbirth. 2014 Dec;14:1-0.
5. Sankar MJ, Neogi SB, Sharma J, Chauhan M, Srivastava R, Prabhakar PK, Khera A, Kumar R, Zodpey S, Paul VK. State of newborn health in India. Journal of perinatology. 2016 Dec;36(3):S3-8.
6. Fadel SA, Rasaily R, Awasthi S, Begum R, Black RE, Gelband H, Gerland P, Kumar R, Liu L, Mathers C, Morris SK. Changes in cause-specific neonatal and 1–59-month child mortality in India from 2000 to 2015: a nationally representative survey. The Lancet. 2017 Oct 28;390(10106):1972-80.
7. Al-Momani MM. Admission patterns and risk factors linked with neonatal mortality: A hospital-based retrospective study. Pakistan journal of medical sciences. 2020 Sep;36(6):1371.
8. Chheda A, Khadse S, Valvi C, Kulkarni R, Hiremath A. Importance of Temperature, Oxygen Saturation, Perfusion, Sugar (TOPS) Parameters and the concept of TOPS score for neonatal transport in India-A pilot Project. Pediatric Oncall Journal. 2018 Oct 23;15(3):69-72.
9. Morse S, Groer M, Shelton MM, Maguire D, Ashmeade T. A systematic review: the utility of the revised version of the score for neonatal acute physiology among critically ill neonates. The Journal of perinatal & neonatal nursing. 2015 Oct 1;29(4):315-44.
10. Harsha SS, Archana BR. SNAPPE-II (score for neonatal acute physiology with perinatal extension-II) in predicting mortality and morbidity in NICU. Journal of clinical and diagnostic research: JCDR. 2015 Oct 1;9(10):SC10.
11. Asker HS, Satar M, Yıldızdaş HY, Mutlu B, Özyurt BM, İpek MŞ, et al. Evaluation of score for neonatal acute physiology and perinatal extension II and clinical risk index for babies with additional parameters. Pediatr Int. 2016;58(10):984–7
12. Verma SK, Tiwari AK, Chauhan DS. Experimental evaluation of flat plate solar collector using nanofluids. Energy conversion and Management. 2017 Feb 15;134:103-15.
13. Asari A, Widyartono D, Shah NA, Dharma BA. The effect of knowledge management integration on e-learning system toward lecturers’ performance at university institutions. Desidoc journal of library & information technology. 2020 Nov 1;40(6):334-8.
14. Shah A, Paturi S, Chava S. Trillion dollar words: A new financial dataset, task & market analysis. arXiv preprint arXiv:2305.07972. 2023 May 13.
15. Cavallin E. Interaction between designers and AI: literature review of research to highlight the characteristics of co-creation and methods for analyzing it. AHFE INTERNATIONAL. 2024;119.
16. Richardson AE, Hadobas PA, Hayes JE. Extracellular secretion of Aspergillus phytase from Arabidopsis roots enables plants to obtain phosphorus from phytate. The Plant Journal. 2001 Mar;25(6):641-9.
17. Lima CK, de Medeiros Carvalho PM, Lima ID, de Oliveira Nunes JV, Saraiva JS, de Souza RI, da Silva CG, Neto ML. The emotional impact of Coronavirus 2019-nCoV (new Coronavirus disease). Psychiatry research. 2020 May 1;287:112915.
18. Pathak R, Negi VS, Rawal RS, Bhatt ID. Alien plant invasion in the Indian Himalayan Region: state of knowledge and research priorities. Biodiversity and Conservation. 2019 Oct;28(12):3073-102.
19. Hapani S, Sher A, Chu D, Wu S. Increased risk of serious hemorrhage with bevacizumab in cancer patients: a meta-analysis. Oncology. 2010 Nov 3;79(1-2):27-38.