Trends in Neonatal Mortality at a Tertiary Level Teaching Hospital Abstract

Introduction: Neonatal services at Tribhuvan University Teaching hospital (TUTH) was essentially up to level II till year 2008 and upgraded to level III care in later years. A 4 years retrospective study was carried out at TUTH, Kathmandu, Nepal to determine any change in the trend of neonatal mortality after the improvement in its services. Materials and Methods: Labor room record book, neonatal record book, perinatal audit data and neonatal record charts were used to collect the data. Results: During the study period, there were total of 15063 live births. The neonatal mortality ranges from 9.46 to 14.88 per 1000 live births per year. There was no significant fall in trend of neonatal mortality (x2 for linear trend=1.40, p=0.23). There was also no significant fall in trend in perinatal mortality rates over this period (x2 for linear trend=1.92, p=0.16).The number of neonates referred to other hospitals has been significantly reduced by 61 %.( x2 for linear trend=33.18, p<0.001). Majority of the neonatal deaths (72%) occurred within first 7 days of life and more than a third (39%) died within the first 24 hours of life. Respiratory distress syndrome, perinatal asphyxia and neonatal sepsis were three major causes of death. Deaths due to respiratory distress and perinatal asphyxia has not changed significantly over the years (p=0.4 and 0.25 respectively). Incidence of low birth weight ranges from 10.8 – 16.1% of total live births. 63% of neonatal mortality occurred in low birth weight babies. This trend has not changed in over the years (x2=1.03, p=0.31). Conclusion: With the improvement in the services, though neonatal mortality remained unchanged, referral rates and mortality due to respiratory distress syndrome of prematurity has decreased.

hospitals has been significantly reduced by 61 %.( x2 for linear trend=33.18,p<0.001).Majority of the neonatal deaths (72%) occurred within first 7 days of life and more than a third (39%) died within the first 24 hours of life.Respiratory distress syndrome, perinatal asphyxia and neonatal sepsis were three major causes of death.Deaths due to respiratory distress and perinatal asphyxia has not changed significantly over the years (p=0.4 and 0.25 respectively).Incidence of low birth weight ranges from 10.8 -16.1% of total live births.63% of neonatal mortality occurred in low birth weight babies.This trend has not changed in over the years (x 2 =1.03, p=0.31).Conclusion: With the improvement in the services, though neonatal mortality remained unchanged, referral rates and mortality due to respiratory distress syndrome of prematurity has decreased.

Introduction
A dvances in neonatal care have signifi cantly increased survival and decreased morbidity among infants admitted to neonatal intensive care units (NICUs) 1 .
There are, however, signifi cant variations in outcomes among NICUs 2 .Tribhuvan University Teaching hospital, the oldest medical college of Nepal is a tertiary care multidisciplinary hospital having 460 beds.It was established in the year 1972 and from its establishment it is providing obstetric services.It is a referral centre for obstetrics services where high-risk pregnant women are referred from throughout the country.However, neonatal service was started only in the year 1994 and neonatal care was essentially level II up to year 2008.With the increasing obstetrics referrals and large number of preterm deliveries, the existing neonatal services was upgraded to level III NICU.During this process in the year 2008, various equipments including monitors, syringe pumps, radiant warmer, incubators, CPAP, ventilators were added to the neonatal unit.The number of nursing staffs was increased to provide level III care from the year 2009.
At present, in neonatal unit there are three ventilators, several CPAP circuits, and three cardiothoracic monitors.Two paediatric residents are posted in rotation in neonatal ward throughout the year.There are all together 12 neonatal nursing staffs.Well babies and babies with minor problems like neonatal jaundice, suspected sepsis get treatment in postnatal ward.
As neonatal services were upgraded for last few years, it was time to analyze whether these measures have helped to improve neonatal care and survival.The neonatal mortality rates refl ect the effi ciency and effectiveness of health care services so this important indicator is used in this study.Indicators like neonatal mortality and perinatal mortality are also useful in planning for improved healthcare delivery 3 .It is important to review the pattern of neonatal morbidities and mortality at regular intervals so that neonatal care could be improved 4 .Therefore, this study was carried out with the objectives to analyze the trend of neonatal mortality and to see pattern of neonatal morbidities at neonatal unit of TUTH.

Materials and Methods
This is a retrospective descriptive study from April 2007 to April 2011 done at neonatal ward of TUTH.Labor room record book, neonatal record book, perinatal audit datas and neonatal record charts were used to collect the data.Data and fi gures of hospital deliveries, total live births, still births, gestational age, birth weight, fi rst week deaths, neonatal deaths and causes of death were recorded.The number of babies referred to other centers was also recorded.All these yearly data were compared using x 2 test with Stat direct software.

Result
From 2007 April to 2011 April, total of 15063 live babies were delivered at TUTH.Thirteen percent of these newborns required admission in neonatal unit.The yearly distribution of number of admitted newborns and referred newborn is shown in Table 1.There was no signifi cant increase in number of neonatal admission in neonatal unit (x 2 for linear trend=1.40,p=0.23) over these years.
The yearly distribution of birth as well as total number of neonatal death is shown in table 2.
The neonatal mortality ranges from 9.45 to 14.88 per 1000 live births per year (Figure 1).
There was no signifi cant fall in trend in neonatal mortality rates over this 4 year period (x 2 for linear trend=1.46,p=0.23).
The perinatal mortality rate over this period ranges from 15.59 to 20.24 per 1000 total births (table 3 and fi gure 2).There was also no signifi cant fall in trend in perinatal mortality rates.(x 2 for linear trend= 1.92, p= 0.16).
The number of neonates referred to other hospitals has been signifi cantly reduced by 61%.( x2 for linear trend=33.18,p < 0.001).Majority of the neonatal deaths (72%) occurred within fi rst 7 days of life and more than a third (39%) died within the fi rst 24 hours of life (fi gure 3).
Yearly distribution of 1 st day neonatal death as proportion of early neonatal death is illustrated in fi gure 3.
Respiratory distress syndrome, perinatal asphyxia and sepsis were three major causes of neonatal deaths (fi g 4).In recent years, early deaths due to respiratory distress syndrome and prematurity have gradually decreased from 33% to 25% whereas neonatal mortality due to perinatal asphyxia has increased from 41% to 51%.The trend for these mortalities have not changed signifi cantly over 4 year period (x 2 for respiratory distress syndrome= 1.07, p=0.2, x 2 for perinatal asphyxia=1.47 and p=0.22).Ninety percent of mortality after 1 st week occurred due to neonatal sepsis.
Incidence of low birth weight babies ranges from 10.8 -16.1% of total live birth (table 4).63% of early neonatal mortality occurred in low birth weight babies and the mortality trend of low birth weight babies has not changed over these years (x 2 =1.03, p=0.31).

Discussion
The present study describes the trend in neonatal mortality at TU Teaching Hospital over a period of four years.Neonatal mortality of Nepal in 2006 is 33 per 1000 live births 5 .During the study period, neonatal mortality rate of TUTH range from 6.07-10.13/1000live births which is comparable with the other hospitals of Kathmandu valley during same time 6,7,8 .The global estimation of PMR is 10 per 1000 births in developed countries, 50 per 1000 births in developing countries and 60 per 1000 births in least developed countries 9 .At Tribhuvan University Teaching Hospital (TUTH) during the study period PMR ranged from 15-20 per 1000 births which has remained unchanged as over previous years 10 .PMR in different hospitals of the Kathmandu valley during these years ranges from 31.3 to 14.4 per 100 births 6,7,8,11 .
Prematurity, asphyxia and septicemia were three main causes of early neonatal deaths.During the study period, 41-51% of early neonatal mortality occurred due to perinatal asphyxia and 25-33% of early neonatal mortality occurred due to respiratory distress syndrome of prematurity.This fi nding is comparable to the similar study 12 done in College of Medical Science, Bharatpur which showed 48.0% of neonatal death is due to birth asphyxia, 29.0% due to neonatal sepsis and 22.0% due to prematurity.A population based cohort study 13 done in Southern Nepal showed 30.0% of NND is due to birth asphyxia.Similar study 14 done at Patan Hospital revealed 30.0% on early neonatal death was due to respiratory distress syndrome, 25.0% due to neonatal sepsis, 16.0% due to congenital anomalies and 13.0% due to birth asphyxia.This indicates that mortality due to perinatal asphyxia is more frequent in TUTH than in Patan hospital.According to the World  Health Organization (WHO), between four and nine million newborns develop birth asphyxia each year.Of those, an estimated 1.2 million die and at least the same number develop severe consequences, such as epilepsy, cerebral palsy, and developmental delay 15 .The proportion of perinatal asphyxia is quite high for tertiary level hospital and its increasing trend warrants improvement in intrapartum monitoring and obstetric services at TUTH.The mortality trend of low birth weight babies has not changed over these years (x 2 =1.03, p=0.31), though the deaths due to respiratory distress have decreased.There is signifi cant reduction in number of referrals too.Increased use of antenatal corticoids and advancements in the fi eld of assisted ventilation played a role to decrease neonatal mortality due to respiratory distress.Neonatal sepsis was main contributor for late neonatal death which necessitates strict adherence to infection control measure.
Advancing intensive care technologies is closely associated with improve survival in low birth weights. 16ith the easy access to surfactant and advancements in the fi eld of assisted ventilation, increased survival of low birth weight babies is expected at TUTH in future.

Conclusion
Even with improvement in the services, the neonatal mortality rate remained unchanged.The referral rate has been signifi cantly reduced and mortality due to respiratory distress syndrome of prematurity has decreased.Perinatal asphyxia and neonatal sepsis are still the major causes of neonatal mortalities.Neonatal services have to be further improved to decrease of neonatal mortalities signifi cantly.

Fig 3 :Fig 4 :
Fig 3: First week and fi rst day neonatal deaths

Table 4 :
Total LBW and preterm LBW babies