Comparitive Study Of Short-Term Outcomes Of Preterm Babies ≤ 30 Weeks In A Tertiary Care Hospital Between India And The U.K

Large numbers of preterm babies are born in middle income countries and neonatal care is improving in these countries. However, comparative clinical outcomes of preterm babies between a tertiary neonatal unit from middle income country and higher income country is limited. Objective: To compare the clinical prole and short-term outcomes of preterm babies ≤ 30 weeks gestation admitted to a tertiary care hospital in India and the U.K. Design: Retrospective cohort observational study using anonymised data from Electronic Patient Records. Participants: major congenital Intervention: This is an observational study and routine care was provided as per departmental standard protocols. of babies ≤ 28 weeks was lower in Indian centre. Incidence of ROP and sepsis was higher in Indian centre.


Introduction
Every year 15 million babies are born premature and one in every 10 babies is a preterm around the world [1,2]. Preterm birth rates have been reported to range from 5% to 7% of live births in developed countries but are estimated to be higher in developing countries which is as high as 18%. Over 60% of preterm births occur in Africa and South Asia. Striking inequalities exist between developed and developing countries in terms of the survival chances of a preterm infant. In many developing countries, infants weighing less than 2000 g corresponding to about 32 weeks of gestation in the absence of intrauterine growth restriction have little chance of survival. More than half of the babies with gestation ≤26 weeks survive in developed countries compared to low income countries where 90% of babies die [3,4]. However there has been drastic improvement in the survival and outcome of these babies over the past decade even in the developing countries which is solely attributed to the improvement in the level of neonatal care. A survey done recently in India showed that the median survival rate in extremely low birth babies (birth weight less than 1000gram) is as high as 60% and that in very low birth weight babies (less than 1500grams) is about 88% [5]. This study helps us to understand and compare the preterm birth related factors, level of care, their morbidities, outcomes and overall survival between level 3 units in the United Kingdom and India.

Methods
A retrospective Electronic Patient Records (EPR) review of preterm infants born at gestational age ≤30 weeks who were admitted to a tertiary care hospital in India and the U.K over a period of 5 years (2011-2015) was conducted. Babies with major congenital (e.g.: heart defects like Transposition of great vessels (TGA) or genetic abnormalities (e.g.: trisomies) were excluded. The Indian center is the Manipal Hospitals, a multispecialty private teaching hospital in Bangalore, which is an accredited level 3b Neonatal Intensive Care Unit (NICU) by the National Neonatal Forum of India (NNFI) [6] with 30 cots capacity (14 ITU,8 HDU & 8 SCBU). The nurse patient ratio is 1:1 for ventilated babies and 1:3 or 1:4 HDU/SCBU babies. The unit has support from various disciplines like cardiology, pediatric surgery, neurosurgery, ophthalmology, genetics, endocrinology and otorhinology (ENT). The hospital has a wellestablished fetal medicine and high risk obstetric unit. The unit provides care for inborn and outborn babies in equal numbers. Mothers of inborn babies receive appropriate antenatal care including fetal monitoring and antenatal steroid. The outborn babies are referred from different parts of Karnataka state and three other adjacent states (Andhra Pradesh, Telangana & Tamilnadu) with variable antenatal care and antenatal steroid uptake. The unit resuscitates babies from ≥24 weeks of gestation at birth as per ILCOR 2015 guidelines. The UK center is the Homerton University Hospital in London, which is a large tertiary NICU, and one of 4 NICUs in North East and North Central London Neonatal Operational Delivery Network, the unit provides regional medical intensive care services [7].The unit has 46 cots capacity (16 ITU,8 HDU & 22 SCBU). The network agreed pathway is that all high-risk pregnancies and impending preterm deliveries of less than 28 weeks' gestation transferred in-utero to Homerton hospital. The hospital has a regional fertility, fetal medicine and high risk obstetric units. Babies requiring surgery will be transferred to Royal London Hospital (pediatric surgery), Guy's and St Thomas' Hospital (cardiac) and Great Ormond Street Hospital (cardiac & all other sub-specialties). The unit practice was to resuscitate babies from ≥23 weeks of gestation at birth as per national guideline (since October 2019, the unit resuscitates babies born at 22 weeks' gestation after careful risk assessment) [8]. Almost all mothers receive full antenatal care and steroids uptake rate is more than 90%. The unit provide care to both inborn and outborn (ex-utero transfer) babies. There is a dedicated Neonatal Transfer Service (NTS) for London. Respiratory care provided in the both centers was as per European consensus [9] but in Indian center use of surfactant was physician's discretion due to cost factor. Use of CPAP, non-invasive and invasive ventilation, and indication for ventilation were similar in both centers as per consensus guidelines. Total Parenteral Nutrition and enteral feeding were given as per unit protocols prepared based on internationally accepted guidelines. Blood culture positive sepsis was de ned as per CDC de nition [10] in both centers.
Data was collected on antenatal factors including maternal risk factors, steroids, chorioamnionitis, etiology for preterm delivery (spontaneous and others), risk factors for sepsis (presence of ≥1 additional risk factors like maternal fever, chorioamnionitis, pre-labour rupture of membranes >18 hours, maternal Group-B streptococcus positive), mode of delivery, condition at birth, demographics of the baby, morbidities such as Intraventricular Hemorrhage, Patent Ductus Arteriosus (PDA), Necrotizing Enterocolitis (NEC), Retinopathy of Prematurity (ROP), Bronchopulmonary Dysplasia (BPD), blood culture proven infection, hypoglycemia, duration of neonatal unit admission, duration of ventilation, and nal outcome (discharge to home or death). Babies with birthweight <10 th centile was de ned as intrauterine growth restricted (IUGR) [11] from both centers. Hypoglycemia was de ned any episode of blood sugar level (glucometer) less than <45mg/dl (Indian center) or <2.6mmol/l (UK center) during NICU stay [12] , IVH was classi ed according JJ Volpe [13], NEC staging as per Modi ed Bell's staging [14] , ROP was classi ed according to International classi cation for retinopathy of prematurity [15] & BPD was de ned as oxygen dependency for 28 days and classi ed at 36weeks post conceptional age based on severity [16]. Discharge criteria was similar in both hospitals: baby should be minimum 35 weeks post menstrual age (PMA), weighing 1.8 kg and on full feeds by bottle/paladay or breastfeeding. Stable BPD babies on low ow nasal cannula oxygen otherwise meeting discharge criteria were discharged home with home oxygen support from both centers.
Data from both centers were collected after approval from the Research Ethics Committee (London -Westminster Research Ethics Committee (Ref: 18/LO/0339) and Manipal REC at respective centers.
Parents' consent was not obtained as this is a retrospective anonymized routinely collected clinical data study using Electronic Patient Records (EPR).
Patient and Public Involvement : Patient/Care takers were not involved in the study directly. This study was conducted using anonymised routinely collected clinical data from Electronic Patient Records (EPR).
Statistical analysis: Data was expressed as descriptive statistics using R software 3.2.3. Statistical analysis was done to compare outcomes of babies between two centres making adjustments for confounding factors. The categorical variables were analysed using Chi Square test and continuous variables using parametric or non-parametric tests as appropriate. Statistical signi cance was set at p value <0.05.

Results
A total of 740 babies from the UK centre and 294 babies from Indian centre were enrolled into the study. The mean (SD) gestational age of babies in the UK cohort 26.6 (2.16) weeks was signi cantly lower compared to Indian cohort 28.4 (1.66) weeks (p < 0.001). The mean (SD) birth weight of babies in the UK cohort 923.0 (292.14) grams was signi cantly lower compared to Indian cohort 1121.0 (320.75) grams (p < 0.001). There was no signi cant difference in gestation speci c birthweight between two cohorts. Sex distribution is similar in two centres with male predominance. A signi cantly higher percentage of mothers received antenatal steroid, and lower percentage of mothers had antenatal risk factors for preterm delivery and sepsis in the UK cohort (Table 1). A signi cantly higher number of babies were inborn (78.6%) in Indian centre compared to the UK centre (67.3%; p < 0.001). A signi cantly higher number of babies (73.5%) were delivered by c-section in Indian cohort (73.5%) compared to the UK cohort (45.5%; p < 0.001). Surfactant given in the delivery room and NICU in the UK cohort (86.9) compared to Indian cohort (49.3%; p < 0.001) is signi cantly higher. Other baseline characteristics are mentioned in Table 1. Similarly signi cantly higher number of babies in Indian centre developed hypoglycaemia, blood culture proven sepsis and pulmonary haemorrhage compared to the UK centre. The incidence of pneumothorax was comparable between two centres (Table 2). Higher number of babies were treated for PDA (32.3% v 10.7%; p < 0.001) and ROP (12.9% v 7.7%; p 0.002) in Indian centre. The incidence of BPD and IVH ≥ Grade 2 were signi cantly lower in babies managed at Indian centre. The incidence of NEC ≥ Stage 2 was comparable ( Table 2). The mean (SD) duration of ventilation days in UK cohort was 10.9 (15.73) and Indian cohort 5.4 (4.90) (p 0.378); similarly the mean (SD) oxygen dependency days was 32.6 (34.07) and 9.9 (9.26), (p < 0.001) in the UK and Indian cohort respectively (Fig. 1a,1b). The mean (SD) total duration of NNU days in UK cohort was 43.1 (34.07) and in Indian cohort was 34.9 (24.38), (p 0.005) (Fig. 2).    (Table 3).

Discussion
The survival of preterm babies had increased over the past 2 decades especially for babies less than 28 weeks gestational age in developed countries [4] however the survival in developing countries at similar gestational age is still catching up. Recent studies have revealed that median survival of VLBW babies in India is as high as 88% [5].There is no comparative study of preterm survival between a level 3 centre in the U.K(country with advanced neonatal care) and comparable tertiary care centre in India. This study was intended to make such comparison. In our study the overall survival rate is 87.2% and 83% (p = 0.091) in U.K and Indian cohort respectively which perhaps is not the true re ection between two cohorts.
When the survival according to gestational age was analysed there is a wide discrepancy in survival rate of babies ≤ 28 weeks.  [17] (p < 0.001) (Table 4). Similarly when the Indian study centre was compared to a similar centre data [18] (due to paucity of overall national published data for gestation speci c mortality) from India, the outcomes were better in the Indian study centre with statistical signi cance (p < 0.001) at 28, 29 & 30 weeks' gestation. There is obvious signi cant difference in survival outcomes for babies less than 28 weeks of gestational age between the two study centres studied from India and the UK. Lower rate of antenatal steroid cover in Indian centre may be an important factor for increased mortality. In India resuscitation guidelines for babies less than 28 weeks is an ethical conundrum. There are no strict ethical guidelines, poor parental acceptability, nancial constraints, lack of robust support post discharge from the state, leave against medical advice (LAMA) before discharge are known factors [19]. In summary many non-medical factors will in uence the outcome in Indian scenario. Babies managed in the UK centre were more immature compared to babies treated in Indian centre. A signi cantly higher number of babies were IUGR in the UK centre. The catchment area of Homerton hospital has a large ethnic minority population, signi cant deprivation and maternal co-morbidities such as diabetes and hypertension. Networking hospitals also transfer women with pregnancy induced hypertension and fetal growth restriction for specialist care to Homerton, which could explain higher number of IUGR babies in the UK centre. When the major preterm morbidities were analysed the rates of BPD was higher in the U.K cohort (30% vs 16%) due to increased survival of babies ≤26 weeks of gestation compared to Indian cohort. EPICure 2 study [4] showed nearly 60% of babies ≤26 weeks had BPD. Studies from India [20,21] reported an incidence of BPD of 28.7%, 10.7% & 11.2% in infants less than 28 weeks, 29-30 weeks and 31-32 weeks respectively The BPD rates from Indian centre in this study is lower than reported studies from India. Necrotising enterocolitis ≥stage 2 is 11.5% and 12.6% in U.K and Indian cohort, results are comparable to the previous studies [23].Number of babies with severe Retinopathy of Prematurity (ROP) and number of ROP needing treatment with laser or Avastin was signi cantly higher in Indian centre (12.9% vs 7.7%) despite higher gestational age of cohort than the UK centre possibly explained by difference in practices or higher risk factors. EPICure study [4] from the UK has shown an increased incidence of ROP needing treatment (Laser/Avastin) from 13% to 22%, however data from India for babies with ROP needing treatment (laser/Avastin) is between 16.6-20.7% [23,24]. Intraventricular haemorrhage grade 2 and above was 15.6% and 8.2% (p < 0.001) in U.K and India respectively which is slightly less than the previous studies [4,25].
Signi cantly higher number of babies from Indian centre (32.3% vs 10.7%) received treatment (medical and surgical) for Patent Ductus Arteriosus (PDA). EPICure 2 study [4] revealed about 51% of babies ≤26 weeks received treatment (medical/ surgical) during their stay in neonatal intensive care unit. The percentage of babies treated for PDA in Indian cohort is high due to low threshold for PDA closure. Both the centres used different criteria for PDA treatment decision. At Homerton hospital (UK centre), only ECHO con rmed symptomatic PDA in a ventilated baby received PDA treatment. In Indian centre, ECHO con rmed symptomatic PDA and needing oxygen >30% criteria was used. From various previous studies we know that PDA treatment rates depend on the individual centres, treating neonatologist and many other factors [26] The proportion of mothers who received at least one complete course of antenatal steroids was 73.4% in the U.K cohort as compared to only 37.4% in Indian cohort. According to NNAP U.K 2016 data around 86% of eligible mothers received antenatal steroids [27] and in India there is wide variation in steroid uptake rate ranging from 48% to 74% [24,28] from previous studies . The low rate in Indian cohort perhaps presumed to be due to more high risk mother where there was not enough time to give antenatal steroid. There is clear evidence that antenatal steroid helps in reducing mortality and complications of prematurity in developed and developing world [29]. A signi cantly lower number of babies received surfactant during resuscitation at birth or any time during rst 72 hours (86.9% versus 49.3%) in Indian centre; this may be because of babies in Indian cohort were of higher gestation and did not require surfactant, and arguably cost is also a factor. A study form India [30] reported around 43% of babies less than 32 weeks received surfactant. The risk factor of sepsis prior to delivery like preterm pre-labour rupture of membranes (pPROM) >24hours, foul smelling liquor, chorioamnionitis, maternal fever was found to be 11% in U.K cohort as compared to signi cant 43.6% (p<0.001) in Indian cohort. This corroborates with a signi cant postnatal blood culture proven sepsis (1.7% v 32.4%) in Indian cohort, it is known from previous studies the increased incidence of sepsis in Indian context [31].
In U.K due to regionalisation of neonatal care the proportion of babies delivered at specialist care centres improved from 18% to 49% and the survival of premature babies has improved from 88% to 93% [32]. On the contrary, in India there is no regionalisation of neonatal care. There are only few centres of excellence in public sector where majority of babies are born. The lack of robust transport facility or collaboration [33] between public and private sectors, inability to bear the high expenses of preterm babies, lack of universal insurance coverage compels the health care providers and parents to go for withdrawal of life supporting care or LAMA when faced with dilemma of treating babies less than 26weeks. The Government of India (GOI) through its National Health Mission (NHM) programme is providing many rural and urban districts now with special newborn care units (SNCUs) with provision for at least secondary level care along with providing trained manpower with the aim to reduce neonatal mortality rate (NMR) from the current 29 per 1000 live births to <10 per 1000 live births across the country by 2030 [34]. Since the start of NHM, 51% of pregnant women are receiving 4 or more antenatal visits and this has rose form 37% over past decade [35]. Improvement in coverage of antenatal check-up under NHM and uptake of antenatal steroids could perhaps leads to decrease in NMR as prematurity contributes to about 35% of neonatal mortality. Whether these strategies are enough or more robust strategies are needed like investing in advanced neonatal care (surfactant, CPAP, ventilators, parenteral nutrition) for extreme preterm babies is a question to reckon about.

Limitations
A large difference in sample size and signi cant difference in gestational age of babies managed between Indian and the UK centres. Details of antenatal care, day to day management of babies in NICU, details of early onset and late onset sepsis, death within 24 hours or labour ward deaths, and limitation of life supporting treatment (withdrawal, withholding or Do Not Resuscitate Order), leave against medical advice (LAMA), discharge weight and gestation were not studied. The other contributing factors such as family demographics, hospital settings including staff ratio to number of patients and level of staff training in neonatal medicine were not investigated.

Conclusion
The survival of babies born at more than 28 weeks of gestation treated in a tertiary neonatal unit from India and the UK is comparable. However, survival of babies born at ≤28 weeks of gestation is still a challenge in best of centres in India. Only 37% mothers received antenatal steroid from Indian centre.
Signi cantly higher number of babies developed severe ROP and received treatment for PDA from Indian centre despite treating more mature babies compared to the UK centre. The overall survival and morbidities of preterm infants in middle income countries could be improved by increasing the antenatal steroid provision to all eligible pregnant women. Further studies including details of antenatal care are required to draw rm conclusion.

Declarations
Funding: This research did not receive any speci c grant from funding agencies in the public, commercial or not-for-pro t sectors.
Contributors: NA, SD and KN developed the study design. SD: Conceived the idea, collected and sorted the data from two centres, drafted the initial manuscript. NA: Supervised the data collection from the UK and also the nal manuscript. KN: Supervised data collection from India and nal manuscript. AK conducted statistical analysis of data. SD, NA, KN and AK approved the nal manuscript.