The last decade has witnessed a dramatic increase in the survival rates of extremely low birth weight (ELBW) infants in developed countries(1-6). Apart from aggressive resuscitation and management of tiny infants, other factors that have contributed to the improved survival are widespread use of antenatal steroids and liberal surfactant replacement therapy(1,7,8). Marked improve-ment in neonatal intensive care facilities has occurred in India during the last decade. However, there is paucity of published data on the survival rates, morbidity pattern and long-term outcomes of ELBW infants from India.

We present here, survival and morbidity data from a cohort of inborn ELBW infants over a seven-year period (without use of surfactant) born at our tertiary-level hospital.

All ELBW infants (birth weight <1000 g) born at the All India Institute of Medical Sciences (AIIMS) between 01 January 1994 and 31 December 2000 formed the study subjects. The neonatal intensive care unit at AIIMS is a ten-bedded unit with facilities to ventilate 4-6 neonates at one time. Approximately 2000 deliveries take place annually at our center. Level III care is provided to intramural neonates with round the clock support from pediatric surgery, ophthalmology, cardiology, radiology and other specialities. Maternal and neonatal details of infants admitted to the neonatal intensive care unit are entered in a high-risk register on a daily basis. Senior residents and consultants on a regular basis supervise this register entry. This register was used to obtain the demographic data for analysis. Additionally, medical records of all ELBW neonates were retrieved from the medical record section for complete details regarding morbidity. Relevant maternal and neonatal information was entered in forms made for this study.

Gestational age in completed weeks was best determined by information obtained from the last menstrual period, antenatal ultra-sonography and postnatal gestational assess-ment (using the Expanded New Ballard Score)(9). Standard criteria were used for definition of the common morbidities. All babies were regularly screened by ophthal-mologists for retinopathy of prematurity (ROP). A portable ultrasound machine was present in the unit round the clock. Residents and radiologists used this machine for screening and diagnosing intraventricular hemorrhage and patent ductus arteriosus. The neonatal deaths were discussed in the monthly perinatal meetings; the clinical and autopsy details were reviewed and a final diagnosis was made and recorded.

Various short-term outcomes were compared in the survivors and non-survivors using the chi square test for categorical variables. The relative risk and 95% confidence interval were calculated for all the important variables in the two groups.

Over the seven-year study period, a total of 12,807 live-born infants were delivered at AIIMS and 137 (1.07%) of these were ELBW infants. Year-wise, ELBW births constituted 1%, 1.19%,0.8%,1.16%,1.17%,0.98%, and 1.17% from 1994 to the year 2000 respectively.

A total of 67 infants (49%) survived to discharge with the yearly survival ranging from 14% to 60%. No infant less than 500 g birth weight (n=3) or less than 23 weeks gestation (n=4) survived. Survival rates with regards to categories of weight, gestation, sex, weight for gestational age, and perinatal asphyxia are presented in Table I.

Table I__Characteristics of Study Population (n = 137)

Condition

Survivors(67)
N (%)

Non-survivors(70)
N (%)

P value

RR
(95%CI)

Birth weight

  < 750 g

10 (23)

34 (77)

0.00002

0.37

  >= 750 g

57 (61)

36 (39)

(0.21-0.65)

Gestation

  <= 28 weeks

33 (37)

56 (63)

0.0001

0.52

  > 28 weeks

34 (71)

14 (29)

(0.38-0.73)

Sex

  Male

26 (41)

38 (59)

0.07

0.72

  Female

41 (56)

32 (44)

(0.51-1.04)

Weight for gestation

  AGA

38 (44)

49 (56)

0.10

0.75

  SGA

29 (58)

21 (42)

(0.54-1.05)

Perinatal asphyxia

22 (33)

36 (51)

0.028

0.64

(0.42-0.96)

AGA - appropriate for gestational age; SGA - small for gestational age.

Morbidity data could be retrieved from 120 (88%) of the 137 eligible case records (Table II). The most commonly encountered morbidities were respiratory distress and hyperbilirubinemia (65% each). Sepsis came a close second with 52% of the cohort being affected. The average length of stay for survivors was 49 days (S.D. ±15.9 days).

 

Table II__Morbidity Pattern in ELBW Infants (n=120)

Morbidity

Overall
Incidence
N (%)

Survivors
(n = 59)
n (%)

Expired
(n = 61)
n (%)

P 
value

RR
(95% CI)

Respiratory distress

78 (65)

36 (61)

42 (69)

0.37

0.89 (0.68-1.15)

RDS

35 (29)

14 (24)

21 (34)

0.19

0.69 (0.39-1.22)

Pneumonia

30 (25)

16 (27)

14 (23)

0.59

1.18 (0.63-2.20)

Air leak syndrome

8 (07)

0

8 (13)

0.004

0.0 (0.0-0.56)

Pulmonary hemorrhage

20 (17)

02 (03)

18 (30)

0.0001

0.11 (0.03-0.47)

CLD

11 (09)

06 (10)

05 (08)

0.70

1.24 (4.0-3.85)

Sepsis

62 (52)

29 (49)

33 (54)

0.58

0.91 (0.64-1.29)

  Culture positive

33 (28)

17 (29)

16 (26)

0.75

1.10 (0.61-1.96)

Sclerema

9 (08)

-

9(15)

0.002

0.0 (0.0-0.48)

Meningitis

9 (08)

04 (07)

05 (8)

0.76

0.83 (0.23-2.93)

NEC

16 (13)

08 (14)

08 (13)

0.94

1.03 (0.42-2.57)

Seizures

13 (11)

01 (02)

12(20)

0.001

0.09 (0.01-0.64)

IVH

35 (29)

14 (24)

21(34)

0.20

0.69 (0.39-1.22)

ROP*

17 (14)

16 (27)

1 (2)

-

-

Hyperbilirubinemia*

78(65)

58 (99)

20 (33)

-

-

Hypoglycemia

43 (36)

17 (29)

26 (43)

0.11

0.68 (0.41-1.11)

PDA

22 (18)

11 (19)

11 (18)

0.93

1.03 (0.49-2.2)

Acute renal failure

11 (09)

0

11 (18)

0.0006

0.0 (0.12-0.90)

RDS - Respiratory distress syndrome; CLD-Chronic lung disease; NEC-Necrotizing enterocolitis;
IVH - Intraventricular hemorrhage; PDA-Patent ductus arteriosus; ROP-Retinopathy of prematurity.
* Not compared because non-survivors died before this morbidity could develop.

The mortality rate was 51 % (n=70) and the most common cause of death was sepsis accounting for 41% of all deaths. Other causes of mortality included prematurity(24%), asphyxia(13%), malformations(4%) and others(16%). Of the infants who expired, 27% (n = 19) died within 24 hours of birth. Forty six percent (n = 32) died within 72 hours of age. Of the 38 infants who survived beyond 72 hours age but finally died, 30 infants (79%) died of sepsis.

The overall survival rate of 49% in our ELBW infants compares well with most data from developed countries in the pre-surfactant era. Narang et al(10), in a recent study on 122 ELBW infants, reported a survival of 22%. Saigal and co-workers(11) from Canada quoted a survival rate of 46% between 1977-80 and 48% between 1981-84. Kitchen et al(12) reported survival rates of 35% in their 1977-82 cohort, which increased to 47% in their 1985-87 cohort. With the availability of surfactant therapy, survival of ELBW infants increased to 60% in 1989-90(13) and 79% in the mid-nineties(14).

Our survival rate of 61 % for infants more than 750 grams compares well with the range of 55% to 68% reported for this birth weight group without the use of surfactant(1,2,7). This result is re-assuring for a developing ‘nation, where expensive modalities like surfactant may not be available for all preterm neonates and more than 50% neonates in this category can be salvaged without the use of surfactant. In the context of a developing country, infants less than 750 g continue to remain a high-risk group with barely one-fourth of them surviving to discharge despite the same effort and resources being utilized as in infants more than 750 g.

Gestation specific survival data from our study showed better survival in infants above 28 weeks of gestation. Survival rates of those less than 28 weeks gestation is lower than those in a couple of other studies(15,16). If resources are to be optimized, then further improvement in the care of infants more than 28 weeks gestation has to be achieved before turning our focus to those less than 28 weeks.

Our study population consisted of 36% SGA infants and this is known to influence outcome in improving survival(1). Analysis of our data did not reveal a significant improved survival for SGA infants and this is in conformity with the recent findings of Teberg et al.(14).

Respiratory distress, sepsis and jaundice were the most common morbidities reported in our study. Almost two-thirds of all babies required ventilation. This would have significant implications for units planning to take care of this group of babies. Adequate ventilation facilities and infection control measures would be necessary to achieve reasonable short-term outcomes in these high-risk neonates. Sepsis and immaturity were the commonest causes of death in our study. In the study by Narang et al.(10), the three most important causes of death were septi-cemia (52%), pulmonary hemorrhage (17%) and respiratory distress syndrome (11%). Evidence from both these studies show that sepsis is probably the most important cause of death in ELBW neonates. Among the non-survivors, sepsis was responsible for 79% of the mortality after 72 hours of age. This would imply that despite successful management in the initial 72 hours, babies expired due to a potentially preventable cause. Adequate infection control measures would be required to achieve better survival rates for this group of babies.

In a developing country where the cost of neonatal care is still prohibitive, reasonable survival rates (>60%) can be achieved in babies weighing more than 750 grams at birth without the use of surfactant. Sepsis is an important preventable cause of morbidity and mortality in these neonates.

Contributors: All the authors were involved in the case management of these neonates. SN, AU and RA were involved in the data collection and analysis. SN and RA were responsible for preparation of the manuscript. AKD, VKP and MBS reviewed the manuscript. RA will act as the guarantor of the article.

Funding: None.

Competing interests: None stated.