i

SHORT TERM SURVIVAL OF PREMATURE INFANTS

ADMITTED TO THE NEW BORN UNIT AT MOI

TEACHING AND REFERRAL HOSPITAL, KENYA

BY

MAKOKHA FELICITAS OKWAKO MBChB (UoN)

SM/PGCHP/01/11

A Thesis Submitted in Partial Fulfillment of Requirements of Master of

Medicine (Child Health and Paediatrics) of School of Medicine, Moi

University

2014

i

DECLARATION

Student’s declaration

I declare that this research thesis is my original work and that it has not been presented for a

degree in any other university. No part of this thesis may be reproduced without the prior

written permission of the author and /or Moi University

Dr. Makokha Felicitas Okwako

SM/PGCHP/01/11

Sign………………………………...

Date…………………………………

Supervisors’ declaration

This research thesis has been submitted for examination with our approval as University

supervisors.

Prof. Winstone Nyandiko M.

Associate Professor and Pediatrician

Department of Child Health and Paediatrics, Moi University

Sign…………………………………

Date………………………………..

Dr. Eren Oyungu

Senior Lecturer and Pediatrician

Department of Medical Physiology, Moi University

Sign…………………………………

Date…………………………………

ii

DEDICATION

I would like to dedicate this thesis to my family: my husband Vincent and my sons Edgar,

Levinus and Brennan for being the pillar of my strength.

iii

ABSTRACT

Background: Prematurity is a major contributor to neonatal mortality globally and it

accounts for 28% of all neonatal deaths. Preterm infants are at an increased risk of neonatal

morbidity and mortality compared to full term infants. In order to achieve the fourth

Millennium Development Goal, there is need for reduction of neonatal deaths especially

those ascribed to prematurity. Data on hospital based survival rates for preterm infants is

important for decision making by obstetricians, neonatologists and hospital management in

predicting outcomes of care and development of interventions to improve outcomes of care.

Objective: To determine the proportion of premature infants admitted to the newborn unit at

Moi Teaching and Referral Hospital who survive to discharge.

Methodology: This was a prospective descriptive study conducted in the newborn unit at

Moi Teaching and Referral Hospital in Eldoret, Kenya. The study subjects were infants born

before 37 completed weeks of gestation. A minimum sample size of 175 premature infants

was required. Consecutive sampling was used to identify subjects. Data was collected using a

pretested structured questionnaire and analyzed using STATA version 10.0. Descriptive

statistics were used for continuous variables and frequency listing for categorical data. Cox

Proportional Hazards model was used to determine factors associated with survival and

Kaplan-Meier survival curves drawn.

Results: A total of 175 neonates were enrolled into the study and followed until discharge or

death. There were 82 (46.9%) male and 93 (53.1%) female infants. There were 27 (15.4%)

extremely preterm (less than 28 weeks), 54 (30.9%) very preterm (28 to less than 32 weeks)

and 94 (53.7%) moderate to late preterm (32 to less than 37 weeks) infants. Neonatal sepsis

(88.6%), hypothermia (67.4%) and respiratory distress syndrome (64.6%) were the main

diagnoses made. The overall survival to hospital discharge was 60.6% (95% CI 0.53-0.68).

The survival rate was 29.6% for infants born less than 28 weeks gestation, 50% for those

born at 28-31 weeks and 75.5% for those born at or above 32 weeks. Of the infants who did

not survive, 11 (15.9%) died within the first 24 hours while 56 (81.2%) died by the end of the

first week. Gestation age of 32 weeks (HR 0.39, 95% CI 0.18-0.8), birth weight >1000g (HR

0.27, 95% CI 0.20-0.78) and maternal antenatal care attendance (HR 0.52, 95% CI 0.3-0.9)

were associated with better survival. Caesarian section mode of delivery, versus spontaneous

vertex delivery, was associated with increased risk of death (HR 4.26, 95% CI 1.88-9.66).

Conclusions: Two thirds of premature infants admitted to MTRH new born unit survived to

discharge. The survival gestation age limit was 28 to less than 32 weeks category (50%

chance of survival as per WHO). Increasing gestation age, birth weight over 1000g and

maternal antenatal care clinic attendance were associated with better survival.

We therefore recommend that whenever possible preterm birth delivery should be delayed

until after 28 weeks gestation.

iv

TABLE OF CONTENTS

DECLARATION ........................................................................................................................................... i

DEDICATION .............................................................................................................................................. ii

ACKNOWLEDGEMENT ........................................................................................................................... vi

LIST OF TABLES ...................................................................................................................................... vii

LIST OF FIGURES ................................................................................................................................... viii

ACRONYMS/LIST OF ABRREVIATIONS .............................................................................................. ix

OPERATIONAL DEFINITIONS ................................................................................................................. x

CHAPTER ONE: INTRODUCTION ........................................................................................................... 1

1.1 Background Information ....................................................................................................... 1

1.2 Problem Statement ................................................................................................................ 4

CHAPTER TWO: LITERATURE REVIEW ............................................................................................... 6

2.1 Epidemiology of Prematurity ................................................................................................ 6

2.2 Factors associated with prematurity ...................................................................................... 6

2.3 Spectrum of disease and Management in Premature Infants ................................................ 8

2.4 Short term Survival of Premature infants ............................................................................ 17

2.5 Factors affecting Short term Survival of Premature infants ................................................ 18

CHAPTER THREE: RESEARCH QUESTION, OBJECTIVES AND JUSTIFICATION ....................... 21

3.1 Research Question ............................................................................................................... 21

3.2 Objectives ............................................................................................................................ 21

3.3 Justification of the Study ..................................................................................................... 21

CHAPTER FOUR: METHODOLOGY ...................................................................................................... 23

4.1 Study design ........................................................................................................................ 23

4.2 Study site ............................................................................................................................. 23

4.3 Study Population ................................................................................................................. 24

4.4 Sample Size ......................................................................................................................... 24

4.5 Sampling Technique ............................................................................................................ 25

4.6 Outcome measures .............................................................................................................. 25

4.7 Data Collection .................................................................................................................... 25

4.8 Study Execution .................................................................................................................. 26

4.9 Data management, data analysis and presentation .............................................................. 27

4.10: Ethical Considerations ..................................................................................................... 27

CHAPTER FIVE: RESULTS ..................................................................................................................... 29

5.1 Infant characteristics ........................................................................................................... 29

5.2 Maternal characteristics....................................................................................................... 30

v

5.3 Short term Survival ............................................................................................................. 31

5.3 Causes of morbidity ............................................................................................................ 32

5.4 Interventions ........................................................................................................................ 34

5.5 Correlates of Mortality ........................................................................................................ 34

CHAPTER SIX: DISCUSSION ................................................................................................................. 40

Study Limitations ...................................................................................................................... 48

CHAPTER SEVEN: CONCLUSIONS AND RECOMMENDATIONS ................................................... 49

REFERENCES ........................................................................................................................................... 50

APPENDICES ............................................................................................................................................ 55

vi

ACKNOWLEDGEMENT

I wish to thank my supervisors Prof. Winstone Nyandiko and Dr. Eren Oyungu for their

guidance and support; and Dr Ann Mwangi for guiding me through statistical aspects of this

study. I also acknowledge my research assistant and staff of the new born unit who helped me

to identify the study participants and treated them during their stay in the unit. Lastly, I thank

my family and entire pediatrics and child health fraternity for their moral, intellectual and

material support.

vii

LIST OF TABLES

Table 1: Infant Characteristics ................................................................................................. 29

Table 2: Maternal Characteristics ............................................................................................ 31

Table 3: Gestational age specific survival rate ........................................................................ 31

Table 4: Causes of Morbidity .................................................................................................. 33

Table 5: Correlates of Mortality (Causes of morbidity) .......................................................... 33

Table 6: Correlates of Mortality (Infant characteristics) ......................................................... 38

Table 7: Correlates of Mortality (Maternal characteristics) .................................................... 39

viii

LIST OF FIGURES

Figure 1: Place of delivery ....................................................................................................... 30

Figure 2: Admissions and Survivors ........................................................................................ 32

Figure 3: Overall Survival curve .............................................................................................. 35

Figure 4: Birth weight specific survival curves ....................................................................... 36

Figure 5: Gestational age specific survival curves .................................................................. 37

ix

ACRONYMS/LIST OF ABRREVIATIONS

ANC

CI

Antenatal care

Confidence interval

Cm Centimeter

CPAP Continuous positive airway pressure

ELBW Extremely low birth weight

g Gram

GA

GBS

Gestational age

Group B Streptococcus

HIV Human Immunodeficiency Virus

IMR Infant mortality rate

IREC Institutional Research and Ethics Committee

IVH Intraventricular hemorrhage

HR Hazard ratio

KDHS Kenya Demographic Health Survey

Kg Kilogram

KNH Kenyatta National Hospital

LBW Low birth weight

LGA Large for gestational age

MNCH Maternal Newborn and Child Health

MTRH Moi Teaching and Referral Hospital

NBU New born unit

NEC Necrotizing enterocolitis

NICU Neonatal intensive care unit

PDA Patent ductus arteriosus

RDS Respiratory distress syndrome

SGA Small for gestational age

UNICEF United Nations Children’s Fund

UN-MDGs United Nations Millennium Development Goals

VLBW Very low birth weight

WHO World Health Organization

x

OPERATIONAL DEFINITIONS

Neonate New born infant; less than 28 days old

Infant Child up to 12 months old

Prematurity A birth that occurs before 37 completed weeks of gestation.

Low birth weight Weight at birth less than 2500g

Very low birth weight Weight at birth between 1000g and 1499g

Extremely low birth

weight

Weight at birth less than 1000g

Neonatal mortality Death occurring within the first 28 days of life

Moderate to late preterm Gestational age of 32 to less than 37 weeks

Very preterm birth Gestational age of 28 weeks to less than 32 weeks

Extremely preterm birth Gestational age below 28 weeks

Infant mortality Death occurring before the first birth day

Perinatal mortality Death occurring between 28 weeks gestational age and within

28 days of life

Post neonatal mortality Death after 28 days to 12 months of life

Under five mortality Death between birth and the fifth birthday

Day 1 The day of birth

Small for gestational age Weight less than 10 th

percentile for age

Large for gestation Weight more than 90 th

percentile for age

Hospital discharge Act of allowing a patient to leave hospital after recovery or

attaining recommended weight if the neonate had low birth

weight

Hypothermia Axillary skin temperature less than 35.5 degrees Celsius

Survival limit Gestational age and birth weight at which a prematurely born

fetus/infant has a 50% chance of survival

1

CHAPTER ONE: INTRODUCTION

1.1 Background Information

Preterm birth is the leading cause of neonatal deaths and the second leading cause of death

after pneumonia in children under five years (1, 2). Mortality rates among premature

infants correlate with birth weight and gestational age with decreases in both associated

with poorer survival. Reliable data show that preterm births rates are increasing globally

(1, 2). An equivalent of 1 in every 10 children are born preterm and around one million

children die each year due to complications of preterm birth (1). In order to achieve the

fourth Millennium Development Goal (MDG) of reducing the under five mortality rate by

two thirds, there has to be a substantial reduction in neonatal deaths and especially those

ascribed to prematurity.

It has been shown that deaths from preterm birth complications can be reduced by over 75%

even without neonatal intensive care services (2). However, low coverage, poor quality, and

inequities in the provision of essential antenatal interventions remain a challenge in many

sub-Saharan African countries (3). With an average of only 42% of births occurring in health

facilities, there is a coverage gap for obstetric and newborn care that needs to be addressed

(4).The burden on health systems imposed by care of preterm infants in high-income

countries is considerable and well recognized. Indeed it was estimated that the cost of care for

a single preterm birth in the USA was US$ 51 600 in the year 2005 (5). This cost is

unachievable in low-income countries but is actually of greater magnitude as preterm birth

rates are higher and the resources available fewer, characterized by understaffed hospitals

with ill equipped or non-existent neonatal care units which ultimately result in higher

neonatal mortality rates (6).

Over 90% of babies born in low-resource settings before 28 weeks gestation die in the first

few days of life (< 10% die in high-income nations), a 10:90 survival gap. In developed

2

countries, 50% of babies born at 24 weeks survive, whereas in low-resource nations, this

survival rate is not achieved until 32 weeks of gestation (2).

Most published trials of neonatal care focus on incremental gains with high-technology care

for example changes in ventilation methods which in fact has limited relevance to the

resource limited settings where 99% of neonatal deaths occur (7).

Over 60% of preterm deliveries worldwide occur in Sub-Saharan Africa and South Asia

regions where resources for neonatal care are limited and progress to reduce neonatal

mortality has been slow (2, 7, 8).

It has also been shown that strengthening existing programs within health facilities could

prevent many deaths, even without high-tech equipment and supplies. Many newborn deaths

could be prevented with facility-based interventions such as neonatal resuscitation, thermal

care around the time of birth for all neonates, early and exclusive breastfeeding, baby hats,

blankets, infection prevention (basic hand washing with soap and clean environment), and

continuous positive airway pressure as well as antenatal steroids and Kangaroo Mother Care

for preterm babies (9).

Frieberg et al (10) argues that since low resource countries cannot be expected to scale up all

essential Maternal Neonatal and Child Health (MNCH) interventions simultaneously,

prioritization and phasing are required in order to generate success that will lead to increased

investment and trust in health systems. Using coverage of skilled attendance at birth as a

marker of health system access and equity of service delivery, Kenya was categorized in

middle health system context (skilled birth attendance 30-60%) which provided a framework

for assessment of priority MNCH interventions. In their model they demonstrated that

increasing coverage for essential MNCH interventions could lead to 85% reduction in

neonatal mortality in resource limited countries like Kenya. This study sought to determine

short term survival of preterm infants as baseline information that would be used to inform

Moi Teaching and Referral Hospital (MTRH) management, obstetricians, pediatricians,

3

neonatologists and other stakeholders on the current hospital based outcomes of preterm

infants. This information will be important for decision making on the new born unit’s

survival limit and expected outcomes of care.

Previous studies done Kenyatta National Hospital (KNH) and MTRH have focused on

neonatal survival rates of low birth weight infants. Although concordance exists between the

three low birth weight categories and the three prematurity categories, they are not

interchangeable since not all low birth weight infants are preterm hence need for this study

that specifically focuses on preterm infants.

Previous studies done in Kenya and other Sub Saharan countries have demonstrated high

neonatal mortality rates with extremely and very low birth weight babies having the lowest

survival rates. A prospective study done at KNH in 1996 showed an overall neonatal survival

rate of infants less than 2000 grams to be 62.2%. None of the 23 infants born weighing less

than 1000g survived to hospital discharge (11). In a prospective study done in MTRH in

1999, the seven day mortality rate of infants admitted to the special care nursery was reported

to be 19.7%. Fifteen percent of the neonates admitted to the special care unit at the time of

that study were preterm (12).

An unpublished study done in MTRH in the same special care unit in the year 2006 showed

that the neonatal mortality of low birth weight infants was 51.6%. This was at a time when

there was no continuous positive airway pressure (CPAP) for respiratory support. Similarly,

none of the infants born weighing less than 1000g survived to hospital discharge (13).

A retrospective study done in Nigeria reviewing records of infants born between 1998 and

2001 in a tertiary hospital without CPAP showed an overall neonatal mortality of 19.4% with

31.9% of the mortality being attributed to prematurity (14).

It has been shown that those countries with low neonatal mortality rate achieved it even

before the technological advancements seen today (15). Since preterm infants are more

vulnerable compared to term infants, determining their hospital based survival rate will

4

provide useful baseline data which future development of strategies to improve neonatal care

outcomes in MTRH would refer to.

1.2 Problem Statement

Globally 15 million babies are born prematurely and neonatal deaths account for forty

percent of all deaths occurring in children under five years of age (1, 2).

According to the Kenya Demographic Health Survey (KDHS) of 2008-09, Kenya has a high

neonatal mortality rate of 31/1000 live births with neonatal deaths accounting for 60% of the

infant mortality rate (16).

Some of the effective high impact interventions that have reduced child, infant and neonatal

mortality in developed countries are being implemented low income countries. But lack of

data and contextual information for priority setting in the resource-constrained settings has

hampered similar progress in developing countries.

MTRH is the second largest public hospital in Kenya, has the second largest public new born

unit in the country and serves as a referral centre for all health facilities in the Western part of

Kenya. The number of premature infants admitted to the new born unit has increased over the

last five years and according to a clinical audit done in the year 2011, in which third of the

infants admitted to the new born unit every month were preterm (12). In MTRH, there has

been improvement in staff training on neonatal resuscitation, infection prevention by hand

washing with soap and general hygiene, appropriate antibiotic treatment for neonatal sepsis,

Kangaroo mother care, thermal provision around the time of delivery and thereafter. In

addition CPAP respiratory support is now available for use in neonates with respiratory

distress syndrome (RDS). Those improvements coupled with additional space and human

resource were expected to improve short term neonatal survival in variance with the higher

neonatal mortality among low birth weight infants reported in the same hospital in the 2006

study (13). This study therefore sought to determine survival to hospital discharge of

5

premature infants admitted to the new born unit at MTRH following the expansion and

increase in the number of premature infants admitted to the unit.

6

CHAPTER TWO: LITERATURE REVIEW

2.1 Epidemiology of Prematurity

Birth weight and gestation age have traditionally been used as strong indicators of the risk of

neonatal death. World Health Organization (WHO) reported that 9.6% of all births were

preterm in 2005, which translated to about 12.9 million births. Approximately 85% of this

burden was concentrated in Africa and Asia, where 10.9 million births were preterm. About a

half a million preterm births occurred in Europe and the same number occurred in North

America (17).

Over the past two decades, the LBW rate has increased primarily because of increased

number of preterm births. In the United states, in 2008, about 12.3% of all live births were

less than 37 completed weeks gestational age and low birth weight infants accounted for

8.2% of live births in both 2007 and 2008 (18).

The KDHS report does not collect data on preterm births but reports data on low birth weight.

According to the KDHS 2008/09, it was reported that 6% of the live births weighed less than

2500g (16). MTRH new born unit offers neonatal care to premature infants born in the

hospital, those born at home and those referred from lower level health facilities.

2.2 Factors associated with prematurity

About two-thirds of preterm births are spontaneous; these births follow preterm premature

ruptured membranes and preterm labour or related diagnoses, such as cervical insufficiency.

Even though the primary and major determinant of birth weight is gestational age, there are

other secondary factors that either directly or indirectly determine the weight of a baby at

birth. It is difficult to separate completely factors associated with prematurity from those

associated with low birth weight.

In a retrospective study done in Northern Tanzania to determine risk factors associated with

low birth weight, it was found that gestational age below 37 weeks was strongly associated

7

with low birth weight. A strong positive correlation existed between preterm birth,

intrauterine growth restriction and low socioeconomic status (19).

Pregnant women from families with low socioeconomic status have been shown to have

higher rates of maternal under-nutrition, poor utilization of antenatal services, lower maternal

age, short interval pregnancies, pregnancy related illnesses and lower maternal education

level; most of which have been associated with preterm birth and low birth weight (19, 20).

In a retrospective review of 2,216 deliveries at the labour ward of the University of Nigeria,

Enugu for two years, maternal age <20 years and ≥ 35 years was associated with relatively

high incidence of low birth weight. Other factors identified as risk factors for delivery of

LBW infants in the same study included lack of antenatal care, female gender, grand

multiparty and multiple gestation (20).

Additionally, in a retrospective study done in Calabar, Nigeria to determine factors that

influenced the incidence of preterm births; previous induced abortion, nulliparity, out of

wedlock birth and lack of antenatal care were found to significantly increase the incidence of

preterm delivery. Women with multiple pregnancy or previous preterm birth were at an

increased risk of preterm delivery. Antenatal complications particularly malaria and anemia

were also noted to be risk factors (21).

In a South African study done to determine obstetric causes of very low birth weight

(VLBW) found that hypertension disease was present in 44.7% of the deliveries, spontaneous

preterm labour in 28.8% of the cases, preterm premature rupture of membranes and

congenital anomalies in 9.5% and 1.3% respectively (22).

Utilization of antenatal services in Kenya has been good with the proportion of women

seeking ANC services from a trained medical provider in their most recent birth rising from

88% to 92% in the past five years. But only 15% of the mothers received ANC care during

the first trimester (16).

8

In a study done in rural South Africa, maternal human immunodeficiency virus (HIV)

infection was shown to cause small for gestation infants but not preterm births. However,

infants of HIV positive mothers in this study had a 3-fold significantly increased hazard to

infant death (23).

However, in a study done in Rural Mozambique between 2003 and 2006, there was no

statistical difference observed in adverse pregnancy outcomes such as low birth weight,

spontaneous abortion, preterm birth and still birth between HIV positive and negative

mothers (24).

2.3 Spectrum of disease and Management in Premature Infants

Premature infants are at risk for developing short-term complications that result from

anatomic or functional immaturity during the neonatal period.

The risk of complications increases with decreasing gestational age with serious morbidities

occurring in the extremely premature infants. Immature organs of a premature infant have

functional limitations leading to many physiologic challenges when adapting to the extra-

uterine environment.

2.3.1 Respiratory

Respiratory distress syndrome (RDS) – occurs in 60-80% of premature infants less than 28

weeks gestation, 15-30% of those between 34-36 weeks and 5% of term babies (25).

Deficient synthesis or release of surfactant, together with small respiratory units and a

compliant chest wall in preterm babies, produces atelectasis and results in perfused but not

ventilated alveoli, which causes hypoxia.

In a study done in KNH new born unit in 1996 to determine survival of infants weighing less

than 2000g at birth, it was reported that 43% developed RDS (11).

9

Another study conducted in the same unit, four years later, to quantify morbidity and

mortality of low birth weight infants showed that the leading diagnosis on admission or

discharge of low birth weight infants was respiratory distress syndrome at 69% (26).

For these reasons, preterm birth deliveries in MTRH are attended by clinicians experienced

in neonatal resuscitation, mostly pediatrics residents.

Most effective prevention of RDS is prolongation of gestation and prevention of extreme

prematurity whenever possible.

The management of RDS involves use of antenatal steroids, postnatal supplemental oxygen,

exogenous surfactant and ventilatory support. Surfactant therapy for preterm infants with or

at risk of RDS improves survival and reduces risk of pneumothorax (25).

Exogenous surfactant and mechanical ventilatory support are the two interventions that are

not available in MTRH newborn unit.

Nasal bubble CPAP is the main respiratory support modality for premature neonates with

RDS in MTRH new born unit.

Apnea is a common problem in preterm infants that may be due to immaturity of the

respiratory centre or an associated illness. Apneic spells are considered clinically significant

if the episodes are greater than twenty seconds duration or when shorter episodes are

accompanied by hypoxia and/or bradycardia. Several theories exist regarding the

pathogenesis of apnea of prematurity, none of which have been confirmed as being the single

cause to date. One theory describes the role of adenosine as a central respiratory inhibitor.

Adenosine is a nucleoside component of compounds such as adenosine triphosphate and

cyclic adenosine monophosphate, which are crucial to numerous biochemical processes. In

term infants, apnea is always worrisome and demands immediate diagnostic evaluation.

Periodic breathing must be distinguished from prolonged apneic pauses because the latter

may be associated with serious illnesses. Apnea is a feature of many primary diseases that

affect neonates. These disorders produce apnea by direct depression of the central nervous

10

system's control of respiration (hypoglycemia, meningitis, drugs, hemorrhage, seizures),

disturbances in oxygen delivery (shock, sepsis, anemia), or ventilation defects (pneumonia,

RDS, persistent pulmonary hypertension of the newborn , muscle weakness).

In a retrospective study in KNH using records of low birth weight infants admitted from

January to December 2000, 42% of the infants experienced apneic attacks during the

admission period (26).

Because of limited availability of pharmacotherapeutic agents such as caffeine in MRTH new

born unit, apnea is managed using menthylxanthine, aminophyline.

2.3.2 Cardiovascular

Persistent patency of the ductus arteriosus is a major cause of morbidity and mortality in

premature infants. Prematurity has been identified as the major predisposing factor to patent

ductus arteriosus (PDA). In a prospective cohort study done at the University College

Hospital, Ibadan, 35% of the preterm babies were found to have PDA compared to the overall

incidence of 24.5% amongst all admissions (27).

In infants born prior to 28 weeks of gestation, a haemodynamically significant PDA can

cause cardiovascular instability, exacerbate respiratory distress syndrome, prolong the need

for assisted ventilation and increase the risk of bronchopulmonary dysplasia, intraventricular

hemorrhage, renal dysfunction, cerebral palsy and mortality (25).

Cyclo-oxygenase inhibitors such as indomethacin and ibuprofen remain the mainstay of

medical therapy for PDA, and can be used both for prophylaxis as well as for rescue therapy

to achieve PDA closure. Surgical ligation is also effective and is used in infants who do not

respond to medical management. Although both medical and surgical treatment have proven

efficacy in closing the ductus, both modalities are associated with some adverse effects.

Because the ductus does undergo spontaneous closure in some premature infants, early

identification of infants with a symptomatic PDA in MTRH new born unit is done through

11

clinical examination and echocardiography. Both medical management by clinicians and

ligation of PDA by cardiothoracic surgeon is available at MTRH.

2.3.3 Hematologic

Anemia of prematurity occurs in low birth weight infants 1–3 mo after birth, is associated

with hemoglobin levels below 7–10 g/dL, and is clinically manifested as pallor, poor weight

gain, decreased activity, tachypnea, tachycardia, and feeding problems. Repeated phlebotomy

for blood tests, shortened red blood cell survival, rapid growth, and the physiologic effects of

the transition from fetal (low PaO2 and hemoglobin saturation) to neonatal life (high partial

pressure of oxygen and hemoglobin saturation) contribute to anemia of prematurity. The

oxygen available to neonatal tissue is lower than that in adults, but a neonate's erythropoietin

response is attenuated for the degree of anemia and, as a result, hemoglobin and reticulocyte

count levels are low. In MTRH, anemia of prematurity is managed by transfusing packed red

blood cells based on the available transfusion guidelines.

Folic acid and iron supplementation is started after two weeks and four weeks respectively to

prevent anemia of prematurity. Transfusion is thought to improve oxygen transport and

cardio-respiratory function thus reducing need for oxygen use and ventilatory support.

However packed red cell transfusion in preterm infants has been shown to increase the risk of

necrotizing enterocolitis. In a retrospective study in the USA, the infants who received

packed red cell transfusion had increased adjusted odds of developing NEC compared with

infants who did not receive a transfusion (28).

2.3.4 Gastrointestinal

Hyperbilirubinemia –Preterm infants are at a greater risk of developing jaundice compared

to term or normal birth weight infants.

Premature babies are prone to hyperbilirubinemia due to increased load of bilirubin to be

metabolized by the liver (polycythemia, shortened red cell life as a result of immaturity or

12

transfused cells, increased enterohepatic circulation due to slow gastric movement, infection)

and reduced ability to conjugate bilirubin due activity of the liver transferases due to

prematurity.

In a Swedish population based study conducted to determine morbidity in moderately preterm

infants, that is, infants born at 30 to 34 completed gestational weeks, 59% of the infants had

hyperbilirubinemia (29).

A systematic review of nine studies to evaluate the efficacy of prophylactic phototherapy in

preterm infants found that it helped to maintain a lower serum bilirubin concentration and

may have an effect on the rate of exchange transfusion and the risk of neurodevelopmental

impairment. However, it was recommended further well-designed studies that could enable to

determine the efficacy and safety of prophylactic phototherapy on long-term outcomes

including neurodevelopmental outcomes be done (30).

In MTRH new born unit, pathological neonatal jaundice is treated using phototherapy and

exchange transfusion depending on the bilirubin levels and the clinical characteristics of the

patient in reference to standard normograms.

Necrotizing Enterocolitis

Necrotizing enterocolitis (NEC) is the most common cause of gastrointestinal-related

morbidity and mortality in the neonatal intensive care unit (NICU). The incidence of NEC is

1–5% of infants in neonatal intensive care units (25).

NEC is rare in term infant whereas in preterm infants it begins at 10-15 days after birth. The

cause of NEC remains unclear but is most likely multifactorial. The risk factors for

developing NEC are prematurity, enteral feeding, ischemia, infective agents and bacterial

colonization of the gut. In a retrospective study done in Tel Aviv, Israel to determine short-

term effects in human milk fed versus formula fed preterm infants; it was found that the rates

of NEC were lower in the group of infants that were fed on human milk (31).

13

Preterm infants receiving certain treatments have been shown to have increased incidence of

developing NEC. In a retrospective controlled cohort study conducted in the USA, preterm

infants from 23 to less than 31 weeks who had received more packed red cell transfusions and

more weeks of antibiotic therapy for nosocomial infection developed NEC than did infants

who had not (32). Almost all preterm infants in MTRH newborn unit are fed on breast milk

and whenever NEC is suspected, enteral feeding is stopped and the infant put on parenteral

antibiotics. Complicated cases of NEC are managed by pediatric surgeons.

Nutritional problems

Preterm nutrition contributes significantly to their short term and long term outcomes.

Premature infants born weighing less than 1500 grams are not able to coordinate sucking,

swallowing, and breathing. Feeding into the gastrointestinal tract (enteral feeding) helps with

gastrointestinal tract development and growth (25).

The introduction of enteral feeds for very low birth weight (VLBW) infants may however be

delayed due to severe respiratory distress or concern that early introduction may not be

tolerated and may increase the risk of necrotizing enterocolitis.

Delay in enteral feeding may diminish the functional adaptation of the immature

gastrointestinal tract and prolong the need for total parenteral nutrition which is not available

in most resource limited new born units like MTRH. Early trophic feeding by giving infants

very small volumes of breast milk during the first week after birth, may promote intestinal

maturation, enhance feeding tolerance and decrease time to reach full enteral feeding.

According to WHO guidelines on optimal feeding of low birth weight infants in low and

middle income countries, LBW infants including those with VLBW should be fed on their

mother’s own milk and that is what is being implemented in MTRH new born unit (33).

14

2.3.5 Metabolic

Hypothermia — Premature infants are at greatest risk for hypothermia immediately after

birth in the delivery room and even during admission in the NBU.

Rapid heat loss occurs in premature infants because of their relatively large body surface

area, thin skin and inability to produce enough heat. Heat is lost by conduction, convection,

radiation, and evaporation. Hypothermia may contribute to metabolic disorders such as

hypoglycemia or acidosis. In extremely premature infants of less than 26 weeks gestation,

hypothermia is associated with increased mortality and, in survivors, pulmonary

insufficiency.

In a population-based cohort study of 8782 very low birth weight infants born in California

neonatal intensive care units in 2006 and 2007, 56.2% of the infants were found to be

hypothermic. Low birth weight, cesarean delivery and a low Apgar score were associated

with hypothermia in that study (34).

Standard newborn care in the delivery room to prevent hypothermia in MTRH new born unit

is achieved by maintaining the room temperature at a minimum of 25ºC, drying the baby

thoroughly immediately after birth then removal of any wet blankets and use of pre-warmed

radiant heaters if resuscitation is needed.

Premature low birth weight neonates <1500g are nursed in incubators warmer to avoid

hypothermia in MTRH new born unit. Since the incubators are limited, when available

patients exceed the incubator numbers, Kangaroo mother care is utilized. Triaging is done

based on the gestation age, birth weight and other clinical symptoms.

Hypoglycemia

Premature babies are prone to hypoglycemia due to inadequate glycogen stores. Since

glycogen is deposited during the third trimester of pregnancy, infants born prematurely have

diminished reserves. There is evidence that hypoxemia and ischemia may potentiate the role

of hypoglycemia in causing permanent brain damage (25).

15

In MTRH newborn unit, premature neonates born before arrival have a random blood sugar

level taken and appropriate interventions in form of intravenous dextrose solution or feeding

started as soon as possible.

2.3.6 Renal

Electrolyte imbalance

Premature infants are prone to developing hyponatremia due to urinary losses of sodium

disproportionate to the intake and daily requirement.

In a comparative cohort study conducted in KNH in 1998 to evaluate the impact of early

neonatal morbidity on serum sodium levels, sick preterm infants were found to develop

significant hyponatremia more often than their healthy counterparts (35).

This justifies the analysis of serum electrolytes and other renal function markers in preterm

infants during admission.

The smallest sickest infants are at greatest risk of metabolic bone disease due to

hypocalcaemia. Progressive osteopenia with demineralized bones and occasionally

pathologic fractures may develop. The major cause is inadequate intake of calcium to meet

the requirements for growth. Poor intake of phosphorus and vitamin D are additional risk

factors. Contributing factors include prolonged parenteral nutrition, vitamin D and calcium

malabsorption, intake of unsupplemented human milk, immobilization, and urinary calcium

losses from chronic diuretic use (25).

In MTRH new born unit, supplementation of vitamin D is done using multivitamin

formulation containing vitamin D that is given from two weeks of age helps to prevent rickets

of prematurity.

16

2.3.7 Central Nervous System

Intraventricular hemorrhage

Intraventricular hemorrhage (IVH) is a major complication of prematurity. IVH typically

occurs in the germinal matrix, which is a richly vascularized collection of neuronal-glial

precursor cells in the developing brain. It increases in frequency with decreasing gestational

age and birth weight. The etiology of IVH is multifactorial and is primarily attributed to the

intrinsic fragility of the germinal matrix vasculature and the disturbance in the cerebral blood

flow. Severe intraventricular hemorrhage is noted in approximately 25% of infants 501–750

g; in 12% between 751 and 1,000 g; in 8% between 1,001 and 1,250 g; and in 3% between

1,251 and 1,500 g (25).

In a prospective study conducted in Nigeria between 1992 and 1994, transfontanelle

ultrasound scans were performed on 93 very low birth weight neonates. Twenty two percent

of the neonates had mild IVH, whereas 7.5% had moderate to severe IVH (36). There is no

routine screening for IVH using cranial ultrasound in MTRH new born unit as it done in

developed countries.

2.3.8 Infectious disease

Neonatal sepsis

Premature babies are prone to sepsis due to immature immune system and under developed

natural barrier mechanisms. Two patterns of neonatal sepsis have been described: early-onset

disease, which presents at <3 days of age, and late-onset disease, which presents at 3 days of

age or later. In the 1990s, widespread implementation of maternal chemoprophylaxis led to a

65% decrease in the incidence of early-onset neonatal Group B Streptococcus (GBS) disease

in the USA; from 1.7/1,000 live births to 0.6/1,000 live births, whereas the incidence of late-

onset disease remained stable at 0.4/1,000 (25). Clinical presentation of sepsis in premature

infants is subtle and non-specific; therefore a high index of suspicion is very important. Signs

17

and symptoms of sepsis include temperature instability, respiratory distress, apnoea, feed

intolerance, jaundice, lethargy or irritability. Clinical diagnosis of suspected neonatal sepsis

as a cause of morbidity is done in Kenya as demonstrated in a study done in KNH where 41%

of the infants had a clinical diagnosis of neonatal infection (11).

Similarly, in a retrospective conducted in KNH to quantify morbidity and mortality among

low birth weight infants admitted in the year 2000, 37% had suspected sepsis out of which

only 14% had a blood culture done to confirm the diagnosis (31).

The first line treatment of infants with suspected sepsis is broad-spectrum antimicrobial

agents; Benzyl penicillin and an aminoglycoside. Once a pathogen is identified, after culture

and sensitivity results, antimicrobial therapy is narrowed.

Prevention strategies include strict compliance with hand washing and universal precautions,

limiting nurse-to-patient ratios and avoiding crowding, minimizing the risk of catheter

contamination, meticulous skin care, encouraging early appropriate advancement of enteral

feeding, education and feedback to staff, and surveillance of nosocomial infection rates in the

new born unit (25).

2.4 Short term Survival of Premature infants

Estimates of the probability of survival of very preterm infants admitted to NICU care are

vital for counseling parents on expected outcomes of care and in planning to improve

neonatal services. Survival rates of preterm infants have improved over the last five decades especially in developed countries and trends well documented (2).

Developed countries have higher survival for newborns with birth weights above 1000g at

94% (20). In developed countries there were major changes in both obstetric and neonatal

care during the 1990s. These changes were associated with decreases in mortality and

morbidity for VLBW infants. In these such countries, they are currently concentrating on

18

improving outcome of babies with birth weight 500g to 1000g and they are already reporting

good survival as exemplified by survival rates of 55% and 88% for neonates between 501-

750 g and 751-1000g surviving respectively in the USA (25).

In Kenya and other Sub-Saharan countries, survival rates of very low birth weight and who

mostly are premature infants are still low. A prospective study done at KNH in 1996 showed

overall survival to hospital discharge for infants less than 2000 grams to be 62.2%. None of

the 23 infants born less than 1000g survived the neonatal period in that study (11). In a

prospective study done in MTRH in 1999, the perinatal mortality of neonates admitted to the

special care nursery was 19.7% (12). An unpublished study done in MTRH in the same

special care nursery in 2006 showed overall survival for neonates below 2000g was 48.4%

(13). A retrospective study in Nigeria reviewing records of children born between 1998 and

April 2001 in a tertiary hospital without CPAP had an overall mortality of 19.4% with 31.9%

of the mortality attributed to prematurity (14).

All these studies demonstrated low survival rates for premature infants especially the

extremely preterm babies.

2.5 Factors affecting Short term Survival of Premature infants

2.5.1 Fetal factors

Gestational age and birth weight are the most important determinants of premature infant

short term survival. Mortality rates amongst premature infants correlate with birth weight and

gestational age with decreases in both associated with poorer survival (25). Thus infants born

with the lowest gestational age and birth weight have the largest impact on infant mortality

because they have the greatest risk of death. In two different studies conducted in KNH and

MTRH new born units, the neonatal mortality rate among infants weighing less than 1000g

was 100% (11, 13). The gestation age specific survival rate in the KNH study was low for

the extremely preterm infants with only 9% of those born less than 28 weeks gestation

19

surviving (11). Male gender and low Apgar score at 5 minute have also been associated with

poor survival.

In a study done by in South Africa to determine survival of very low birth weight infants,

male gender and low Apgar score at 5 minutes, were associated with poor survival (37).

Birth weight-specific neonatal diseases such as grade IV intraventricular hemorrhage,

neonatal sepsis (severe group B streptococcal pneumonia or meningitis), and neonatal

malformations e.g. pulmonary hypoplasia also contribute to a poor outcome (25).

2.5.2 Maternal factors

Lack of antenatal care clinic attendance by pregnant mothers has been associated with poor

preterm infant survival. In a study done to determine survival of very low birth weight infants

in a public hospital in South Africa, infants whose mothers did not attend ANC had poor

survival (37). Mode of delivery has been shown to affect survival of infants with spontaneous

vertex delivery having higher mortality compared to Caesarean section as reported in studies

done in KNH, MTRH and South Africa (26, 37).

Maternal antenatal use of steroids by pregnant women has been associated with lower

incidence of RDS and better survival (25).

2.5.3 Level of care

Premature infants are more likely to survive if they are born in high level neonatal intensive

care than in lower level hospitals with limited facilities.

The new American Association of Pediatrics neonatal levels of care consist of four levels

with no subdivisions:

Level I: Provides basic neonatal care for term babies; can perform neonatal resuscitation at

every delivery and care for infants born at 35-37 weeks gestation who remain physiologically

stable.

20

Level II: Provides specialty care for newborns at 32 weeks’ gestation or more and weighing

1,500 grams or more with problems expected to resolve rapidly or who are convalescing from

a higher level of intensive care.

Level III: Provides sub-specialty care for high-risk newborns needing continuous life support

and comprehensive care for critical illnesses. This includes infants weighing less than 1,500

grams or who were less than 32 weeks’ gestation at birth.

Level IV: Includes capabilities of a level III neonatal intensive care unit (NICU) as well as

the ability to provide on-site pediatric medical and surgical subspecialists to care for infants

with complex congenital or acquired conditions, coordinate transport systems, and provide

outreach education within their catchment area. Level IV intensive care units have the highest

neonatal survival rates. The relative risk of neonatal mortality for infants born with very low

birth weights was twofold higher in Level II centers than in Level III centers (5).

Improvements in newborn intensive care, including the use of surfactant treatment and

antenatal steroid therapy to prevent and treat neonatal respiratory distress syndrome, have

resulted in decreased mortality rates of preterm infants, except in those who are at the limit of

viability (17).

In a retrospective cohort study conducted in a South African tertiary public hospital, in a level

II NICU focusing on deliveries conducted from the year 2000 to 2002, low survival to

hospital discharge of the extremely preterm infants was attributed to the lack or limited

availability of exogenous surfactant and mechanical ventilation. (37). The level of care

provided at MTRH new born unit given the available facilities is at level II since there is no

exogenous surfactant and no facilities for continuous ventilatory support like conventional

mechanical ventilation.

21

CHAPTER THREE: RESEARCH QUESTION, OBJECTIVES AND

JUSTIFICATION

3.1 Research Question

What proportion of premature infants admitted to the new born unit at Moi Teaching and

Referral Hospital survive to discharge?

3.2 Objectives

3.2.1 Primary Objective

To determine the proportion of premature infants admitted to the newborn unit at Moi

Teaching and Referral Hospital who survive to discharge.

3.2.2: Secondary Objectives

1. To determine the survival in the different gestational age categories; less than 28

weeks, 28 to less than 32 weeks and 32 to less than 37 weeks.

2. To describe the factors that are associated with short term survival of premature

infants admitted to the new born unit at MRTH.

3.3 Justification of the Study

Neonatal survival data has been utilized by countries and NICUs to demonstrate trends in

preterm birth outcomes over the years (2). It is difficult to formulate appropriate interventions

to improve neonatal care and outcomes without reliable data on survival rate.

In KNH and MTH earlier studies done focusing on survival of low birth weight infants

reported poor survival of very low birth weight and extremely low birth weight infants most

of whom are known to be preterm (11,13).

In MTRH, there has been significant increase in the number of preterm infants admitted to

the new born unit following the expansion of the unit. However, both the short term and long

term survival of preterm infants admitted to the new born unit had not been documented (12).

22

Facility specific data on survival of preterm infant and causes of morbidity and mortality

would help the MTRH management to develop interventional strategies aimed at improving

outcomes of this vulnerable group of patients. The findings will be useful to obstetricians,

paediatricians and neonatologists when making decisions on delivery and care of these

preterm infants and when counseling parents on expected outcomes. Findings of this study

may also be used for advocacy to improve neonatal services at MTRH and other public

facilities in the country taking care of preterm infants.

Lastly, this information will help advice the peripheral facilities on appropriate referral of

mothers with preterm labour to MTRH given the viability limit findings from this study.

We projected that at the current rate of preterm birth deaths and with few years remaining,

Kenya is unlikely to achieve the fourth MDG target of reducing under five mortality rate by

the year 2015 and even beyond 2015 (16).

23

CHAPTER FOUR: METHODOLOGY

4.1 Study design

This was a prospective descriptive study. Study participants were recruited at admission by

the principal investigator and followed during their stay in the unit, taking note of all

significant clinical events until either discharge home or death. There was no intervention by

research team.

4.2 Study site

The study was conducted in the newborn unit of MTRH which is located in Eldoret town,

about 300km from Nairobi, in Uasin Gishu County, Kenya. The hospital is an 800 bed

capacity tertiary hospital that serves as a referral hospital for the western part of Kenya, with

a catchment population of about 13 million people - 33% of Kenyan population. The hospital

provides various services ranging from primary to specialized care and serves urban, peri-

urban and rural populations from near and far counties. The hospital also serves patients from

neighboring countries; Uganda, Sudan, South Sudan and Rwanda.

The hospital’s new born unit is located in the Riley Mother and Baby hospital wing, a new

extension of the hospital that was opened in 2009. The NBU has a capacity of fourteen

incubators, forty eight cots and is able to provide basic neonatal services and non invasive

respiratory support using nasal bubble CPAP. The unit does not have a neonatal intensive

care unit (NICU) and does not provide mechanical ventilation and exogenous surfactant for

preterm babies.

The staffs allocated to the unit include two neonatologists, three pediatricians, thirty five

nurses, a nutritionist, paediatrics resident doctors, medical officer interns and clinical officer

interns.

All premature low birth weight neonates weighing less than 1700g, born in the MTRH labour

ward, born at home or referred from lower level health facilities in the catchment area are

24

admitted to the MTRH new born unit. They are managed using the basic Pediatric protocol in

Kenya which has been adopted from the WHO guidelines.

4.3 Study Population

Premature neonates admitted to the MTRH new born unit from December 2012 to August

2013 who met the inclusion criteria were recruited. This included premature neonates born in

MTRH labour ward, those referred from other health facilities and those born at home.

Inclusion Criteria

a) Neonates who were born at less than 37 completed weeks gestational age.

b) Preterm neonates whose mothers gave informed consent to have their infants enrolled

in the study.

Exclusion Criteria

a) Preterm infants with congenital malformations not compatible with life.

4.4 Sample Size

The Fischer’s formula for calculating the sample size for a simple random sample without

replacement was used as follows; N= Z 2

α P (1-P) W 2

Where:

Z α is the standard normal deviate and =1.96 for a 95% confidence level

P is the expected proportion of premature infants who survive to discharge in

MTRH, 50%. In studies done in similar resource-poor settings birth weight

(not gestational age) has often been used as a proxy measure for maturity, thus

getting short term survival proportion as an outcome measure is difficult. This

study therefore used a median value of 50% since the exact outcome

proportion is unknown.

W is the desired width of the confidence interval and = 0.05

25

Thus: Replacing these values in the above formulae:

=384

Adjusting for finite population for premature infants admitted to MTRH new born unit based

on the 2011 hospital medical records where an average of 30 premature neonates were

admitted per month, for 9 months: 270

Thus:

; N=270 (N= the population size while nf = is the finite sample size)

= 158.6; ≈ 159

Adjusting for transfer outs, an additional 10% more were recruited giving a final sample size

of 175 premature infants.

4.5 Sampling Technique

Consecutive sampling of premature neonates admitted to MTRH new born unit was done.

Every next study subject meeting the inclusion criteria was recruited until the desired

minimal sample size of 175 was attained.

4.6 Outcome measures

The primary outcome measure was the proportion of premature infants who survived to

hospital discharge. The secondary outcome was the length of hospital stay.

4.7 Data Collection

Data was collected by the principal investigator using a pretested standard questionnaire and

follow up data collection form. The demographic data, neonatal and maternal characteristics

were entered in the data collection form at admission. Gestation age was calculated using two

26

methods; the last menstrual period (LMP) as an entry point and then New Ballard score

method for analysis. For most of the infants, there was no significant difference between

gestational age by LMP and that by new Ballard score. Infants’ anthropometric

measurements; weight, length and head circumference were taken by the PI or research

assistant. Any missing maternal data was obtained through maternal interview and by

checking the ANC attendance booklet.

4.8 Study Execution

The clinicians working in the new born unit were sensitized about the study to enable them

inform the principal investigator and research assistant, a NICU nurse, whenever preterm

infants were admitted in the new born unit. The research assistant identified infants born at

less than 37 completed weeks by LMP and informed the principal investigator. Mothers’ of

premature infants were identified in the postnatal ward or hostel and informed written

consent was obtained from those whose babies met the inclusion criteria.

The gestational age was determined by calculating the number of weeks from the first day of

the last menstrual period and confirmed using the new Ballard score within 48 hours of

admission (38).The infant’s demographic data, maternal antenatal and delivery data and the

clinical characteristics were collected by the principal investigator at admission through an

interview of the mother and physical examination of the baby.

Information on whether the mother had started attending ANC and their HIV serology status

was obtained. Presence of any antenatal and perinatal maternal morbidity was noted.

Anthropometric measurements taken at admission were birth weight in grams (g), length and

head circumference in centimeters. Weight was measured using electronic digital weighing

scale, SECA model 728 to the nearest 1g.

The diagnosis made and interventions started at admission were recorded. The participants

received standard newborn care based on the diagnosis made as per the new born care

27

protocol. Daily ward rounds were done by the health care providers in NBU and appropriate

investigations done to diagnose any new causes of morbidity suspected.

The principal investigator followed up the study subjects and collected information on the

significant events; causes of morbidity and interventions received during hospitalization and

updated the information in the follow up data collection form until discharge or death. . The

length of stay and final outcome were noted.

Discharge was decided by the clinicians taking care of the subjects in the new born unit upon

recovery or achieving the recommended weight for discharge.

4.9 Data management, data analysis and presentation

Data collected was coded to maintain confidentiality and then entered into a Microsoft access

data base. Data was then checked for consistency by providing validation checks in Microsoft

access. Data was then exported to STATA version 10.0 for analysis.

The data was analyzed at 95% level of confidence. Descriptive statistics such as mean and

median were used for continuous variables. Frequency listings and percentages were used to

describe categorical variables. Survival analysis was done using Cox Proportional Hazards

model was used to determine factors associated with survival and Kaplan-Meier Survival

curves drawn. Significant correlation was when the confidence interval did not contain 1 and

p-value was less than 0.05. Data was presented in prose, tables, figures and curves.

4.10 Ethical Considerations

Approval to conduct the study was granted by Institutional Research and Ethics Committee

(IREC) and the management of MTRH. Patients details obtained were handled confidentially

by use of computer password known by the principal investigator only. All patients received

the necessary standard treatment regardless of their willingness or unwillingness to

participate in the study. No incentives or inducements were used to convince mothers to

28

allow their babies to participate in the study. Informed and written consent was emphasized.

Findings of this study will be shared with MTRH management.

29

CHAPTER FIVE: RESULTS

A total of 175 preterm infants were studied from December 2012 to August 2013. There were

82 males and 93 female infants giving a male to female ratio of about 1:1.13.

5.1 Infant characteristics

The mean gestation age was 31.94 weeks (±3.06) and the mean birth weight was 1342g

(± 355.5). Among the infants recruited 14.9% (26) were extremely low birth weight, 46.3%

(81) were very low birth weight whereas 38.9% (68) were low birth weight.

Table 1: Infant Characteristics

Characteristic Frequency (n=175) %

Gender: Male 82 46.9

Female 93 53.9

Gestation age: <28weeks 27 15.4

28-31weeks 54 30.9

32-36 weeks 94 53.7

Mode of delivery: SVD 138 78.8

EMCS 22 12.6

SBD 15 8.6

30

Figure 1: Place of delivery

5.2 Maternal characteristics

The median maternal age was 24 years (range from 14 to 42 years). Twenty percent of the

mothers were teenagers. The majority of mothers, 117 (67.2%) were married while 56

(32.2%) were single. Most of the mothers whose infants were studied were primigravidae, 85

(48.6%). Sixty four percent of the mothers had attended antenatal clinic at least once. All

mothers had their HIV status known with the most of them, 164 (93.7%) being HIV negative.

Sixty nine percent of the mothers reported to have developed spontaneous preterm labour.

Mothers who had obstetric complications prior to delivery included 24 with preeclampsia, 13

with Ante partum hemorrhage and 8 with premature rupture of membrane. Most mothers who

delivered through caesarian section had antenatal complications with 14 (63.6%) having

preeclampsia.

111, 64%

37, 21%

27, 15%

MTRH

OTHER FACILITY

HOME

31

Table 2: Maternal Characteristics

Characteristic Frequency (n=175) %

Age (Years): <20 36 20.6

20-25 73 41.7

26-30 34 19.4

31-35 15 8.6

>35 17 9.7

Employment status: Unemployed 113 64.6

Self employment 49 28.0

Formal employment 13 7.4

5.3 Short term Survival

The overall proportion who survived to hospital discharge was 60.6% (95% CI 0.53-0.68).

Of the babies who did not survive, 11 (15.9%) died within the first 24 hours with 56 (81.1%)

dying by the end of the first week. Only two infants died after the end of the neonatal period.

The mean length of stay for the infants who survived to hospital discharge was 25.8days

(±16.1) whereas for those who died it was 6.2 days (±7.6).

Table 3: Gestational age specific survival

Gestation age

in weeks

Admissions Survivors at

discharge

Survival %

< 28 27 8 29.6

28-31 54 27 50.0

32-36 94 71 75.5

Total 175 106 60.6

32

Figure 2: Admissions and Survivors

5.3 Causes of morbidity

The diagnoses made at admission for majority of the study participants were prematurity, low

birth weight and respiratory distress syndrome. Various clinical diagnoses made by the

clinicians during the infant stay in the new born unit were noted.

Although 155 (88.6%) infants were suspected to have neonatal sepsis, only 24 (15.5%) had

blood cultures done. The organisms species that were isolated included: Klebsiella (10),

Coagulase negative staphylococcus (4), E. coli (1), Enterococcus (1) and Citrobacter (1). Two

cultures had no growth obtained and five had no results back by the time of the patient death

or discharge home. Hypothermia (HR 2.87, 95% CI 1.06-7.73, p-value 0.00), apnoea (HR

26

81

68

3

48

55

Less than 1000g 1000-1499g 1500-2449g

Admissions Survivors

33

5.3, 95% CI 2.58-10.98, p-value 0.04) and respiratory distress syndrome (HR 2.92, 95% CI

1.29-6.60, P-value 0.01) were significantly associated with higher hazard of dying.

Table 4: Causes of Morbidity

Diagnosis Frequency %

Suspected Neonatal sepsis 155 88.6

Hypothermia 118 67.4

Neonatal jaundice 115 65.7

Respiratory distress syndrome 113 64.6

Apnoea 76 43.4

Anemia 71 40.6

Necrotizing enterocolitis 13 7.4

Hypoglycemia 8 4.6

Table 5: Correlates of Mortality (Causes of morbidity)

Variable HR 95%CI p-value

Neonatal Sepsis 0.46 0.15 1.43 0.18

Anemia 0.86 0.51 1.44 0.57

Neonatal Jaundice 0.35 0.21 0.58 0.00

Hypothermia 2.87 1.06 7.73 0.04

Hypoglycemia 2.31 0.88 6.06 0.09

Neonatal Enterocolitis 1.19 0.59 2.37 0.63

Apnoea 5.30 2.58 10.89 0.00

RDS 2.92 1.29 6.60 0.01

34

5.4 Interventions

All preterm babies were started on IV fluids, intramuscular vitamin K injection and

tetracycline eye ointment at admission. Antibiotics were used in 98.9% of the study subjects.

The first line antibiotics were Benzyl penicillin and gentamicin, second line was a third

generation cephalosporin and amikacin and third line was a fourth generation cephalosporin

or meropenem and vancomycin. A total of 71(40.6%) infants were diagnosed to have anemia,

42 (73.2%) of those received blood transfusion.

Babies weighing less than 1500g were nursed in incubators. Eighty infants (45.7%) were

nursed in incubator during their stay in the newborn unit.

5.5 Correlates of Mortality

Infant characteristics

The proportion of infants who survived to discharge increased with increasing gestation age

and birth weight with those born weighing over 1500g having the highest survival proportion

of 80.9%. The hazard of dying was low among low birth weight infants compared to

extremely low birth weight infants (HR 0.4, 95% CI 0.20-0.78, p-value 0.003).

The hazard of dying was low for preterm infants born after 31 weeks gestation compared to

those born at or less than 28 weeks gestation (HR 0.39, 95% CI 0.18-0.82, p-value=0.013).

The hazard of dying was higher for premature infants born through cesarean section

compared to those born via spontaneous vertex delivery (HR 4.25 95% CI 1.875-8.662; p-

value=0.001)

35

Figure 3: Overall Survival curve

0 .0

0 0

.2 5

0 .5

0 0

.7 5

1 .0

0

P ro

p o

rt io

n s

u rv

iv in

g

0 20 40 60 80 Length of stay (days)

Kaplan-Meier estimate

36

Figure 4: Birth weight specific survival curves

0 .0

0 0

.2 5

0 .5

0 0

.7 5

1 .0

0

P ro

p o

rt io

n s

u rv

iv in

g

0 20 40 60 80 Length of stay (days)

Below 1000

1000-1499

1500-2500

Kaplan-Meier estimate by birthweight

Log Rank Test (p-value=0.000)

37

Figure 5: Gestational age specific survival curves

0 .0

0 0

.2 5

0 .5

0 0

.7 5

1 .0

0

P r o

p o

r ti

o n

s u

r v

iv in

g

0 20 40 60 80 Length of stay (days)

Below 28 weeks

28-31 weeks

32-36 weeks

Kaplan-Meier estimate by gestation age

Log Rank Test (p-value=0.000)

38

Table 6: Correlates of Mortality (Infant characteristics)

Variable HR 95%CI P-value

Place of delivery

Facility vs MTRH 1.91 0.72 5.10 0.20

Home vs MTRH 2.48 0.87 7.11 0.09

GA in weeks

28-31 vs <28 0.74 0.37 1.49 0.4

32-36 vs <28 0.39 0.18 0.82 0.013

Sex: Male vs Female 1.50 0.92 2.47 0.108

Mode of delivery

SBD vs SVD 1.62 0.76 3.45 0.209

EMCS vs SVD 4.26 1.88 9.66 0.001

Birth weight in g

1000-1499 vs <1000 0.40 0.20 0.78 0.008

1500-2500 vs <1000 0.27 0.11 0.64 0.003

Age at admission in hours

1-6 vs <=1 0.59 0.24 1.42 0.238

6-12 vs <=1 1.99 0.23 17.22 0.532

12-24 vs <=1 0.87 0.22 3.51 0.849

>24 vs <=2 0.24 0.03 1.93 0.18

Apgar score

4-5 vs 3 2.29 0.77 6.80 0.137

6-7 vs 3 0.78 0.29 2.15 0.636

8-10 vs 3 0.50 0.24 1.04 0.065

39

Maternal characteristics

The hazard of dying was lower for premature infants whose mothers attended ANC compared

to those whose mothers didn’t (HR 0.52 95% CI 0.3-0.9; P-value=0.02).

Maternal age, education level, marital status and HIV status were not significantly associated

with short term survival as shown in table 7 below.

Table 7: Correlates of Mortality (Maternal characteristics)

Variable HR 95% CI P-value

Age 20-25 vs <20 1.13 0.59 2.17 0.716

25-30 vs <20 1.12 0.50 2.48 0.786

30-35 vs <20 0.99 0.34 2.91 0.981

>35 vs <20 0.32 0.09 1.18 0.087

Education: Post primary vs primary 1.25 0.71 2.20 0.440

Marital status- Married vs Single 0.89 0.48 1.64 0.704

ANC attendance-Yes vs No) 0.52 0.30 0.90 0.020

Parity >0 vs 0 1.18 0.67 2.06 0.571

HIV status-Pos vs Neg 1.57 0.55 4.54 0.403

40

CHAPTER SIX: DISCUSSION

Demographic data

In this study the male and female subjects were basically the same which is similar to the

findings of Kramer in his 1987 meta analysis which found no overall difference in sex

distribution in low birth weight infants (39). There was a slight female preponderance among

infants surviving to hospital discharge however the difference was not statistically significant.

Simiyu in a study done in KNH found a similar gender difference in survival of low birth

weight infants which was explained by death of more male newborns but the difference was

also not statistically significant (26). A study by Velaphi et al on survival of very low birth

weight infants in South Africa showed that male gender was significantly associated with

poor neonatal survival (37). The finding in this study suggests that gender has no role in

determination of survival; however this question seems not to be conclusively answered

because other studies have reported significant difference.

Short term Survival

Preterm newborns are known to have greater risk of both morbidity and mortality compared

to term neonates. There are not many studies in Africa that have looked specifically at

mortality among preterm infants. Majority of the studies looked at overall neonatal mortality

and prematurity was reported a leading cause of neonatal mortality in most of the studies. In

this study, two thirds of preterm infants survived to hospital discharge which is similar to

findings by Were et al in a prospective study that looked at 163 infants weighing less than

2000g and found a 62.2% survival to hospital discharge (11). The survival rate in this study is

slightly higher than what Simiyu in a study done in the year 2000 which reported neonatal

mortality low birth weight infants to be 57.4% in the new born unit of KNH (26). That was a

retrospective study in a setting where 23 % of the files were missing and hence this could

have introduced bias.

41

Kasirye-Bainda et al reported an overall neonatal mortality rate of 24.6% in KNH and

reported that LBW and prematurity accounted for 95.6% of the mortality (40).

In a Tanzanian tertiary referral hospital, neonatal mortality rate of 19% was reported by

Klingenberg et al and gestation less than 31 weeks accounted for 67% of the mortality (41).

Survival was different for the three birth weight categories with the low birth weight having

the best outcome and extremely low birth weight the worst outcome.

This is similar to findings by Kasirye- Bainda et al who reported survival of 48.7% among

babies with birth weight less than 1500g while Were et al in the same hospital reported zero

survival of newborns less than 1000g and two thirds survival for bigger ones (11, 40).

Increasing survival proportion with increasing birth weight is also true of gestation age as

shown in this study finding that the hazard of dying was significantly lower for infants with

gestational age 32 to less than 37 weeks compared to those below 28 weeks.

Were et al reported 69% survival among those with gestation 32 to 35 weeks and only 9% for

those less than 28weeks gestation (11).

Increasing gestational age and birth weight is associated with better respiratory maturity

which enables preterm infants to adapt better to extra-uterine life.

The survival proportion observed in this study is much lower than the one reported in

countries with advanced neonatal care. In South Africa, Velaphi et al reported survival to

hospital discharge of 32% among premature infants born at gestation below 28 weeks (37).

However, the best survival has been reported in the developed countries where survival for

newborns with birth weights above 1000g is above 94% (42, 43). In deed research in

developed countries is currently concentrating on improving outcome of babies with birth

weight 500g to 1000g and they are already reporting good survival as exemplified by survival

rates of 55% and 88% for neonates between 501-750 g and 751-1000g surviving respectively

reported in the USA (44). Changes in obstetric care and availability of advance neonatal

intensive care facilities have led to better outcomes in developed countries.

42

The risk of developing RDS in infants born less than 28 weeks gestation age is 60-80% (25).

Therefore in a resource limited new born unit without exogenous surfactant and ventilatory

support like in this study site, survival of infants born at less than 28 weeks remains low.

It has been shown that 25 – 45% of neonatal mortality occurs within the first 24 hours mainly

as a result of birth asphyxia and acute complications of prematurity. Of the infants who did

not survive in his study, a tenth died within the first 24 hours while cumulatively four fifths

died during the first seven days of life. Other studies reported higher deaths within the first 24

hours, Simiyu (36%), Were et al (28%) and Ezechukwu et al (64.5%) in tertiary hospitals in

Kenya and Nigeria respectively (11, 14, 26). Similar findings were reported by Kasirye-

Bainda et al where 86.8% of the neonatal deaths in a study done in KNH occurred within the

first week of life (40). Notably those studies were done in hospitals without continuous

positive airway pressure (CPAP) facilities and in an era where were exogenous surfactant and

antenatal steroids were not widely used; factors which may explain the current finding of a

lower proportion that died in MTRH in the first 24 hours with availability of CPAP and use

of antenatal steroids. The higher numbers of preterm infants dying during the first week of

life could be due to acute complications of prematurity occurring in a setting with limited

neonatal intensive facilities to support the preterm infants especially those with RDS and

early onset neonatal sepsis.

Extremely low birth weight infants who survived to hospital discharge had long length of

hospital stay. This is similar to findings by Simiyu et al in a study to quantify the morbidity

and mortality of low birth infants in KNH (26). This could be attributed to morbidity

associated with complications of extreme prematurity and time taken to gain recommended

weight before discharge more so in a setting where there is no total parenteral nutrition for

these infants for infants who can’t tolerate enteral feeds.

This study showed that the limit of viability in MTRH newborn unit was 28 to less than 32

weeks gestation category. Although the WHO has established the upper limit of viability at

43

37 completed weeks of gestation they have not set the lower limit. The lower limit is defined

by fetal organ maturity and advances in high risk obstetrics care and neonatal intensive care.

The USA currently defines this lower limit as about 25 weeks or weight above 500g (43).

Compared the developed countries, our viability limit is still high, a finding that could be

explained by limitation in high risk obstetric care and neonatal intensive care.

The mortality rate of preterm infants admitted in MTRH new born unit is still high especially

for the very preterm and extremely preterm infants.

Factors associated with survival:

Infant characteristics

In this study, there was a significant improvement in the proportion of premature infants

surviving to hospital in the gestation age 32 to 36 weeks category compared to less than 28

weeks or 28 – 31 weeks gestation category. Studies done in KNH and MTRH before reported

similar findings, showing a positive correlation between gestation age and survival (11, 13).

Indeed it has been stated that the most significant public health intervention to reduce

neonatal mortality is reduction in rate of preterm deliveries by prolongation of gestation age

whenever possible. This study suggests that efforts should not only aim at reducing preterm

births but at increasing gestation age by maximum possible days. This approach is likely to

bear more benefit in low resource settings where neonatal intensive care is often inadequate

or absent.

The finding that Caesarian section was associated with significant higher hazard of dying was

unexpected. This finding was contrary to other studies where caesarian section was

associated with better survival (11, 12, 37). In a retrospective audit of births that occurred in

MTRH labour ward between 2004 and 2011, Yego et al reported that majority of early

neonatal deaths followed vaginal deliveries (45).

44

We postulated that this finding could partly be explained by the fact that two thirds of

mothers who delivered through cesarean section in this study had severe preeclampsia which

could have increased the risk of poor outcome. However there could be other reasons and

therefore the area needs to be explored in future studies.

Infants born in MTRH had better probability of survival compared to those born either in

outside facility or at home, however after adjusting for other factors, difference was not

statistically significant. This finding is in agreement with those by Simiyu in KNH, by

Welbeck et al in a study done in Accra Ghana. However, it is different from findings from a

study done by Udo et al in Nigeria where survival of neonates born in the tertiary hospital

facility had better survival than those born in outside the facility (26, 46, 47). The poor

outcome in the infants born before arrival in the Ghana study was attributed to delay in

transportation of neonates to the tertiary hospital and inappropriate transportation which

adversely affected the neonates.

In this study, two thirds of the infants were born in MTRH and whereas those born at lower

level facility or at home were fewer likely contributing to statistical insignificance. However

studies that look at how babies born outside are transported to our facility and cared for

during transport may offer insights into practices that could have improved survival.

Babies with Apgar score above 3 at five minutes were more likely to survive compared to

those with lower and in addition this probability was directly proportional to the score with 8

to 10 having the best survival; however the benefit was not statistically significant after

adjusting for other factors. In South Africa, Velaphi et al reported that Apgar score less than 6

at 5 minutes was associated with poor survival. Increase in Apgar scores was reported to have

better survival to hospital discharge in preterm infants, which is in agreement with findings in

our study (37). Birth asphyxia in premature infants worsens the respiratory insufficiency due

to RDS and makes short term survival poor.

45

Maternal characteristics

Two thirds of mothers of the infants who were studied had attended ANC clinic at least once

before giving birth. This proportion is lower than 92% reported in KDHS of 2008-09 and

could be explained by the fact that a significant number of mothers start attending ANC late

in the 2 nd

and 3 rd

trimester thus those who had extremely and very preterm deliveries had not

yet started attending ANC (16).

Babies whose mothers attended at least one ANC had a lower hazard of dying compared to

those whose mothers didn’t. This is similar to findings by Velaphi et al in a study where

infants whose mothers attended ANC were one and half times more likely to survive (37).

Antenatal clinic attendance has been shown to be one of the major contributors to improved

neonatal survival in the USA (41). Antenatal care is useful in reducing preterm births by

aiding identification of mothers at risk for preterm delivery; offers an opportunity for timely

monitoring, intervention and admission to the new born unit. This study did not however look

at the optimum number of antenatal visits that were associated with improved survival. It has

been estimated that improving the quality of prenatal care would decrease prematurity

associated deaths by 75% (2).

Maternal age and parity were not found to be statistically significant in contributing to

neonatal survival. This is similar to findings by Welbeck et al in a study done in a tertiary

teaching hospital in Accra Ghana where maternal parity and age were not found to

significantly contribute to survival of at risk neonates (46). However, Yego et al reported that

a significantly higher number of early neonatal deaths occurred for infants whose mothers’

were in the age group of 15 to 24 years (45). This finding was due to by high rate of preterm

labour and late admission to hospital among young mothers. The difference in findings

between this study and our study could be attributed to the sample selection because our

study excluded term newborns. Those with preterm labour tend not to delay in seeking care

as observed in term labour. Additionally, this could be explained by the fact that after

46

gestational age and birth weight, the next important determinant of short term survival of

preterm infants is the level of neonatal care service in the unit which is independent of

maternal demographic characteristics.

Maternal HIV status was not significantly associated with survival to hospital discharge in

this study. In studies done in rural Mozambique and South Africa by Ndirangu et al and

Nianiche et al respectively, maternal human immunodeficiency virus (HIV) infection was

shown to cause small for gestation infants but not preterm births (23, 24). The findings in our

study similar to those two studies suggest that HIV infection does not worsen the neonatal

survival. This could be attributed to the natural history of vertically transmitted HIV whereby

there is no significant immunosuppression in the first month of life and the neonate has

maternal antibodies.

It is important to note that in this study the number of HIV positive mothers was few which

was comparable to the general prevalence of HIV amongst pregnant women (6.4%) reported

in the Rift valley region in the KDHS report (16). Implementation of prevention of mother to

child transmission of HIV interventions has also reduced the risk of vertical transmission and

hence reduced morbidity and mortality for sero-exposed infants.

Causes of morbidity and Interventions

In this study, respiratory distress syndrome was one of the most common diagnoses made and

it was associated with higher hazard of dying. In a study done in KNH new born unit in 1996

to determine birth weight specific survival of infants weighing less than 2000g at birth, 43%

developed RDS (11). Another study conducted in the same unit, four years later, to quantify

morbidity and mortality of low birth weight infants showed that the leading diagnosis on

admission or discharge of low birth weight infants was respiratory distress syndrome at 69%

(26). The high rates of respiratory distress syndrome were expected as a consequence of lung

47

immaturity and perhaps lack of timely prenatal steroid intervention and lack of exogenous

surfactant in MTRH new born unit.

This study found that although majority of the infants were suspected to have sepsis and

started on antibiotics, most of them did not have blood cultures done. We observed that it was

not routine to have all premature infants suspected to have neonatal sepsis have blood

cultures done. This was partly due to stock out of blood culture specimen collection bottles

and also due to clinicians prioritizing only infants who were very sick and or not responding

to initial antibiotic therapy. Simiyu, in two retrospective studies conducted in KNH on

newborn babies admitted to the new born unit and the general pediatric wards reported that

37% and 71%had suspected sepsis yet only 14% and 8.4% had confirm sepsis diagnosis

respectively (26, 48). One common finding in these studies was use of antibiotics without

laboratory confirmation of sepsis. This posed a risk for development of antibiotic resistance.

Meticulous infection control and treatment interventions are required to reduce morbidity due

to sepsis. It is possible that improvement in newborn intensive care technology will not

necessary improve short term outcomes unless there is reduction in the risk of sepsis.

In comparison, in the United States, morbidity due to sepsis has decreased remarkably and

currently the main causes of morbidity and mortality in preterm infants are respiratory

distress syndrome and intraventricular hemorrhage (41). In this study we did not find any

diagnosis of intraventricular hemorrhage which could be explained by the low index of

suspicion among clinicians and lack of routine IVH screening in MTRH new born unit.

Apnoea and hypothermia were associated with increased hazard of dying. These are acute

complications of prematurity which occur more often in critically ill infants. This finding was

similar to findings by Simiyu in a study done in KNH where apnoea was occurred in 29% of

the patients, it was managed with rectal aminophylline but only 12% of these infants survived

(26). Lack of conventional drugs such as caffeine citrate in the unit could have contributed to

higher mortality in neonates with apnoeic attacks.

48

Hypothermia occurred in two thirds of the study participants. This proportion is higher than

that reported by Simiyu in a KNH study where hypothermia occurred in 27% of the patients

(26). The higher incidence could be attributed to the limited number of incubators available

in the unit leading to nursing of premature infants in open cots with space heaters where

thermal control was poor. Implementation of Kangaroo mother care in the unit was not

routinely performed and therefore, from the findings of this study, there is need to scale up

and formally implement KMC in the unit. Infants born before arrival were also noted to have

hypothermia at admission, especially those transported by private means. Referring health

facilities and the general public too need to be sensitized on use of KMC when referring

preterm infants to higher level facilities like MTRH new born unit and that could reduce the

incidence of hypothermia.

Study Limitations

 Assumption that survival probabilities are the same for study subjects recruited into

the study at different times during the study period.

 Estimating the gestational age: use of last menstrual period and New Ballard score,

without first trimester ultrasound; each of them is not ideal but combining the two

improves accuracy.

49

CHAPTER SEVEN: CONCLUSIONS AND RECOMMENDATIONS

CONCLUSIONS

 Two thirds of premature infants admitted to MTRH new born unit survived to

discharge.

 Majority of the infants died within the first seven days.

 The survival limit was the 28 to less than 32 weeks gestational age category.

 Increasing gestation age, birth weight over 1000g and maternal antenatal care clinic

attendance were associated with better survival.

 Caesarean section mode of delivery was associated with poor survival.

RECOMMENDATIONS

1. Whenever possible preterm birth delivery should be delayed until after 28 weeks

gestation.

2. More effort should be put in increasing early antenatal care clinic attendance.

3. Further studies are needed to evaluate why caesarean mode of delivery of preterm

infants was associated with poor survival and audit the efficacy of current

interventions being implemented in the unit.

4. A study that follows premature infants after discharge to determine intermediate and

long term survival rates.

50

REFERENCES

1. Preterm birth, Fact Sheet, World Health Organization, 2013 November; 363.

2. Althabe F, Bhutta Z, Blencowe H, Chandra-Mouli V, Chou D, et al. Born Too Soon:

The Global Action Report on Preterm Birth. May 2012.

3. Bryce J, Daelmans B, Dwivedi A, Fauveau V, Lawn JE, et al. Countdown to 2015 for

maternal, newborn, and child survival: the 2008 report on tracking coverage of

interventions. Lancet. 2008; 371:1247–1258.

4. UNICEF. State of the World's Children 2010.

5. Institute of Medicine. Preterm Birth. Causes, Consequences, and Prevention.

Washington, DC: National Academies Press; 2007.

6. Ruiz-Peláez JG, Charpak N, Cuervo LG. Kangaroo Mother Care, an example to

follow from developing countries. BMJ. 2004; 329:1179–81.

7. Kinney MV, Kerber KJ, Black RE, Cohen B, Nkrumah F et al. Science in Action:

Saving the lives of Africa's Mothers, Newborns, and Children working group,

Axelson H, et al. PLoS Med. 2010 Jun 21; 7(6):e1000294.

8. Lawn JE, Cousens S, Zupan J. 4 million neonatal deaths: When? Where? Why?

Lancet. 2005; 365:891–900.

9. Lawn JE, Kerber K, Enweronu-Laryea C, Cousens S. 3.6 million neonatal deaths-

what is progressing and what is not? Semin Perinatol. 2010; 34:371–386.

10. Frieberg IK, Kinney MV, Lawn JE, Kerber KJ, Odubanjo MO, et al. Sub-Saharan

Africa’s mothers, newborns, and children: how many lives could be saved with

targeted health interventions? PLoS Med 2010; 7: e1000295.

11. Were FN, Mukhwana B.O, Musoke RN. Neonatal Survival of Infants less than 2000

grams born at Kenyatta National Hospital, EAMJ 2002 Feb; 79: 77-79.

51

12. Ayaya S.O, Esamai F O, Rotich J, Liechty E. Perinatal Mortality in the Special Care

Unit of Moi Teaching and Referral Hospital, Eldoret Kenya. EAMJ 2004 Nov;

81(11): 555-556.

13. Njuguna F, Nyandiko W, Kiptoon P. Mortality among Low Birth Weight neonates

admitted to the New born Unit at Moi Teaching and Referral Hospital. Unpublished

MMed. Thesis, Moi University 2008.

14. Ezechukwu C.C, Ugochuku E.F, Egbuonu I, Chukwuku J.O. Risk Factors for

Neonatal Mortality in a Regional Tertiary Hospital in Nigeria. Nigeria Journal of

Clinical Practice 2004; 7(2): 50-52.

15. Wilson A.L. Survival of low birth weight infants. Pediatrics 1982; 70: 587-592.

16. Kenya Demographic and Health Survey 2008-09. Kenya National Bureau of Statistics

(KNBS) and ICF Macro. 2010; 103-110.

17. Beck S, Wojdyla D, Say L, Betran AP, Merialdi M, et al. The worldwide incidence of

preterm birth: a systematic review of maternal mortality and morbidity. Bull World

Health Organ. 2010; 88:31–38.

18. WHO, UNICEF. Low birth weight: Country, Regional and Global estimates, Geneva,

UNICEF and WHO, 2004.

19. SIZA JE. Risk factors associated with low birth weight of neonates among pregnant

women attending a referral hospital in northern Tanzania, Tanzania Journal of Health

research 2008; 10(1).

20. Chukwudi NK, Ejike O, Adimora GN, Ibe BC. Influence of Biosocial Factors on

Incidence of Low Birth Weight babies in Enugu. Nigerian Journal of Pediatrics 2002;

29(4): 99-102.

21. Etuk SJ, Etuk IS, Oyo-Ita AE. Factors Influencing the Incidence of Pre-term birth in

Calabar, Nigeria. Nigerian Journal of Physiological Sciences; 20: 63-68.

52

22. Odendaal ES, Steyn DW, Odendaal HJ. Obstetric causes for delivery of very low birth

weight babies at Tygerberg Hospital, S Afr Med 2003; 93:61-64.

23. Ndirangu J, Newell ML, Bland RM, Thorne C. Maternal HIV infection associated

with small-for-gestation infants but not preterm births: evidence from rural South

Africa. Hum Reprod. 2012 June; 27(6):1846-1856.

24. Naniche et al. Impact of Maternal Human Immunodeficiency Virus Infection on Birth

Outcomes and Infant survival in Rural Mozambique. Am J Trop Med Hyg May 2009;

80(5): 870– 876

25. Robert M Kliegman et al. Nelson Text Book of Pediatrics, 19th Edition: WB Saunders

2011; Chapter 91: 91.2

26. Simiyu DE. Morbidity and Mortality of low birth weight infants in the new born unit

of Kenyatta National Hospital, Kenya EAMJ 2004 Jul; 81(7): 367-374

27. Omokhodion SI, Baiyeroju-Agbeja AM, Adegboye VO,Adeyemo A, Lagunju IA.

Patency of the Ductus Arteriosus in Newborn: Experience in a Special Care Baby

Unit. Nigerian Journal of Paediatrics 2001; 28(3): 61-67.

28. Paul DA, Mackley A, Novitsky A, Zhao Y, Brooks A, Locke RG. Increased odds of

necrotizing enterocolitis after transfusion of red blood cells in premature infants.

Pediatrics. 2011 Sep; 128(3): 593-4.

29. Altman M, Vanpée M, Cnattingius S, Norman M. Neonatal morbidity in moderately

preterm infants: a Swedish national population-based study. Pediatrics 2011;

158(2):239-44.

30. Okwundu CI, Okoromah CA, Shah PS. Prophylactic phototherapy for preventing

jaundice in preterm or low birth weight infants. Cochrane Database Syst Rev 2012 Jan

18; 1:CD007966.

53

31. Maayan-Metzger A, Avivi S, Schushan-Eisen I, Kuint J. Human milk versus formula

feeding among preterm infants: short-term outcomes. American Journal of

Perinatology 2012 Feb; 29(2): 121-6.

32. Carter BM, Holditch-Davis D, Tanaka D, Schwartz TA. Relationship of neonatal

treatments with the development of necrotizing enterocolitis in preterm infants.

Nursing research. 2012 Mar-Apr; 61(2): 96-102.

33. WHO Guidelines on Optimal feeding of low birth weight infants in low and middle-

income countries 2011:16.

34. Miller S.S, Lee HC, Gould J.B. Hypothermia in very low birth weight infants:

distribution, risk factors and outcomes. Journal of Perinatology 2011; 31: 49-56.

35. Ngwiga DN, Were FN, Musoke RN. Hyponatremia in very low birth weight infants.

EAMJ 2002; 79(3):120-123.

36. Ajayi O and Nzeh DA. Intraventricular haemorrhage and Periventricular leucomalacia

in Nigerian infants of very low birth weight. West Afr. J of Medicine Apr-Jun 2003;

2(2): 164-166.

37. Velaphi SC, Mokhachane M, Mphahlele RM, Beckh-Arnold E, Kuwand ML, et al.

Survival of very low birth weight infants according to birth weight and gestational age

in a public hospital. S Afr Med J 2005; 95: 504-509.

38. Ballard JL, Khoury JC, Wedig K, Wang L, Eilers-Walsman BL, et al. New Ballard

score, expanded to include extremely premature infants. J Pediatr 1991; 119: 417-423.

39. Kramer, M.S. Determinants of low birth weight: methodological assessment and meta

analysis> Bull. WHO 1987; 65: 563-737.

40. Kasirye-Bainda E, Musoke F. N. Neonatal morbidity and mortality at Kenyatta

National Hospital New Born Unit. EAMJ 1992; 69(7): 360 – 365.

54

41. Klingenberg C, Ohanu R, Oneko M, Sam N, Langalad N. Neonatal morbidity and

mortality in a Tanzanian tertiary care referral hospital. Annals of Tropical Paediatrics.

Dec 2003; 23: 293 -299.

42. Horbar J.D, Badger C.J, Carpenter J.H et al. Trends in mortality and morbidity for

very low birth weight babies 1990 – 1999. Pediatrics 2002; 110: 145 – 151.

43. March of Dimes Fact sheet: Preterm Birth. 2007.

44. Lemons J, Bauer C. R, William OK et al. Very low birth weight outcomes of National

Institute of Child Health and Human Development, Neonatal Research Network

January 1995 through December 1996. Pediatrics Jan 2001; 107. Doi:

10.1542/paed.107.l.e

45. Yego F, Stewart J.W, Byles J., Nyongesa P, Aruasa W, D’Este C. Retrospective

analysis of maternal and neonatal mortality at a teaching and Referral hospital in

Kenya. Reproductive Health Journal 2013; 10:13.

46. Welbeck J, Biritwum R.B., Mensah G. Factors affecting survival of most at risk

newborn at Korle Bu Teachinh Hospital, Accra, Ghana. WAJM 2003; 11(1).

47. Udo J. J, Anah M. U, Ochigbo S. O, Etuck I.S, Ekanem A. D. Neonatal morbidity and

mortality in Calabar, Nigeria: A Hospital Based Study. Nigerian Journal of Clinical

Practice. 2003; 11(3):285 - 289.

48. Simiyu D.E. Morbidity and mortality of neonates admitted in general Paediatrics

wards at Kenyatta National Hospital. EAMJ 2003; 80(12): 611-6.

55

APPENDICES

APPENDIX 1: DATA COLLECTION FORM

Demographic data:

Serial No……………………… IP No………………………..

Date of admission……/……/……. Date of discharge/death…/……/

Residence district……………… Place of delivery: MTRH…..H/Facility… Home….

Infant characteristics

Gestational age--------wks by LMP/-------wks by Ballard Sex: Male….Female……..

Mode of delivery: SVD…..SBD….AVD……..EMCS………ELCS…………

Birth weight…………g Age at the time of admission……..Hrs

Apgar score at 5min------/10

Maternal characteristics

Age:………….Years Level of education: None…Primary……….Post primary…….

Employment status: Unemployed…..Self employment….Formal employment……….

Parity: PARA……+…… Interval between this last birth and Conception………Months

ANC attendance: Yes/No HIV Serology Neg/Pos

Antenatal Complications Yes/No, If Yes specify……………………………………………..

Clinical characteristics

Diagnosis on admission……………………………………

Causes of Morbidity during admission Interventions on admission

o Neonatal Sepsis IV fluids

o Anemia IV antibiotics

o Neonatal jaundice Blood product transfusion

o Hypothermia Phototherapy

o Hypoglycemia Anticonvusants

o Necrotizing enterocolitis Oxygen

o Apnoea CPAP

Outcome: 1) Discharge home 2) Death

Length of stay…………days

56

APPENDIX 2: NEW BALLARD SCORE FOR ASSESSMENT OF GESTATIONAL

AGE AT BIRTH.

PHYSICAL MATURITY

SIGN SCORE SIGN

SCORE -1 0 1 2 3 4 5

Skin

Sticky,

friable,

transparent

gelatinous,

red,

translucent

smooth

pink,

visible

veins

superficial

peeling

&/or rash,

few veins

cracking,

pale

areas,

rare

veins

parchment,

deep

cracking,

no vessels

leathery,

cracked,

wrinkled

Lanugo

None sparse abundant thinning bald

areas

mostly

bald

Plantar

Surface

heel-toe

40-50mm: -1

<40mm: -2

>50 mm

no crease

faint red

marks

anterior

transverse

crease only

creases

ant. 2/3

creases

over entire

sole

Breast Imperceptible barely

perceptible

flat areola

no bud

stippled

areola

1-2 mm

bud

raised

areola

3-4 mm

bud

full areola

5-10 mm

bud

Eye /

Ear

lids fused

loosely: -1

tightly: -2

lids open

pinna flat

stays

folded

sl. curved

pinna;

soft; slow

recoil

well-

curved

pinna; soft

but ready

recoil

formed

& firm

instant

recoil

thick

cartilage

ear stiff

Genitals

(Male)

scrotum flat,

smooth

scrotum

empty,

faint rugae

testes in

upper

canal,

rare rugae

testes

descending,

few rugae

testes

down,

good

rugae

testes

pendulous,

deep rugae

Genitals

(Female)

clitoris

prominent &

labia flat

prominent

clitoris &

small labia

minora

prominent

clitoris &

enlarging

minora

majora &

minora

equally

prominent

majora

large,

minora

small

majora

cover

clitoris &

minora

TOTAL PHYSICAL MATURITY SCORE

57

NEUROMUSCULAR MATURITY

SIGN SCORE SIGN

SCORE -1 0 1 2 3 4 5

Posture

Square

Window

Arm

Recoil

Popliteal

Angle

Scarf

Sign

Heel To

Ear

TOTAL NEUROMUSCULAR SCORE

Maturity Rating

Score Weeks

-10 20

-5 22

0 24

5 26

10 28

15 30

20 32

25 34

30 36

35 38

40 40

45 42

50 44

58

APPENDIX 3: FOLLOW UP FORM

At

birth

D2 D3 D4 D5 D6 D7 WK

2

WK

3

WK

4

WK

5

Weight (g)

Minimum wt (g)

No. of days to

minimum wt

No. of days to

recover birth wt

Length (cm)

Head circumference

Morbidity

Respiratory distress

syndrome

Neonatal sepsis -

confirmed

Neonatal sepsis -

suspected

Anemia

Physiologic

Neonatal Jaundice

Pathologic NNJ

Hypoglycemia

Apnoea

Necrotizing

enterocolitis

Hypothermia

Interventions

1 st line antibiotics

2 nd

line antibiotics

3 rd

line antibiotics

Lab investigations

Continuous positive

airway pressure

Blood transfusion

Imaging

Surgery

59

APPENDIX 4: CONSENT TO PARTICIPATE IN THE STUDY

SERIAL NUMBER ………..

Background

You are being asked to participate in a research study. Before you decide, it is important for

you to understand why the research is being done and what it will involve. Read the

following information carefully and ask us if there is anything that is not clear or if you

would like more information. Please take time to decide whether you want to take part in this

study. The

purpose of this study is to determine the proportion of premature infants surviving to

discharge at MTRH and the factors that influence their survival. Our study is for research

purposes but we hope that the information obtained will be used to inform the hospital and

other policy formulators which will result in improved healthcare service delivery.

Study Procedure

The study involves filling out a questionnaire capturing you and your baby’s biodata and the

presenting symptoms and signs at admission in the NBU. A record will be kept of the

anthropometric measurements and causes of morbidity during hospitalization. The findings

during subsequent assessments cannot be linked to your baby and are completely anonymous

and confidential since we shall be using serial numbers.

Risks

There are no risks involved in this study. This study will be anonymous. The baby will

receive treatment as per the diagnosis made based on the hospital and Ministry neonatal

protocols.

Benefits

There are no direct medical benefits to your child for participating in this study. A potential

benefit of the study will be improved healthcare service delivery based on the

recommendations of this study.

Alternative Procedures

You may choose your baby not to participate in this study

Confidentiality

This research will be conducted in accordance with all the Kenyan laws and regulations that

protect rights of human research subjects. All records and other information obtained will be

kept strictly confidential and your baby’s protected health information will not be used

without permission. All data collection tools will be identified by number or otherwise coded

to protect any information that could be used to identify your baby. Results of this study may

be published, but no names or other identifying information will be released.

Person to Contact

If you have questions, complaints or concerns about this study, you can contact the

investigator from Moi University, School of Medicine, department of Child Health and

Paediatrics, Postgraduate progarmme ; Dr. Makokha Felicitas Okwako +254722622651

email drmakfelis@gmail.com

Institutional Review Board

This study has been approved by the Institutional Research and Ethics Committee (IREC) of

Moi University/Moi Teaching and Referral Hospital. Contact IREC if you have questions

60

regarding your child’s right as a participant, and also if you have complaints or concerns

which you do not feel you can discuss with the investigator.

Contact IREC using the address ; The Chairman IREC, Moi Teaching and Referral Hospital,

PO BOX 3, Eldoret, Kenya. Tel. 33471/2/3

Voluntary Participation

It is up to you to decide whether your baby takes part in this study. Refusal to participate or

the decision to withdraw from this research will involve no penalty or loss of benefits to

which your child is otherwise entitled. This will not affect your relationship with the

investigators.

Right of investigator to withdraw

The investigator can withdraw your baby form the research without your approval.

Costs and Compensation to participants

There is no cost to you, and there is no compensation to subjects for participation in this

study.

Number of Participants: 175 babies

Authorization for use of your protected health information

This study that does not entail the use of your baby’s protected health information.

Thank you for your baby’s participation in this research and we truly appreciate your help.

CONSENT

By signing this consent form, I confirm I have read the information in this consent form and

have had the opportunity to ask questions. I will be given a signed copy of this consent form.

I voluntarily agree to take part in this study.

Name of Caregiver ………………………………Signature/Mark…………..

Date…………..

Name of Investigator ……………………………Signature………………….

Date…………..

61

APPENDIX 5: IREC APRROVAL

62

APPENDIX 6: HOSPITAL APPROVAL