Editor’s Note

Yüklə 1,08 Mb.
ölçüsü1,08 Mb.
1   2   3   4   5   6   7   8   9   ...   15

Material and Methods
Clinical Facilities

Empangeni Neonatal unit is part of a child and Maternal Hospital, Lower Umfolozi Regional War Memorial Hospital (LURWMH) in North-East of Kwazulu-Natal, South Africa. This is the only maternity and neonatal regional referral centre for an area with a population estimated at 2.5 Million. According to census 2001, poverty rate and unemployment rates stand at 63.5% and 53.7% respectively. The proportion of households with access to safe water (32%) and sanitation (24%) are far below the national figures (79% and 62%, respectively). Fifty thousand (50 000) live births occur in the entire area annually including 10 000 at LURWMH .The neonatal service was introduced at the hospital in 1998 with a 15 unit beds without intensive care facilities. Between 1999 and 2008 the unit was expanded to 92 bed neonatal units: 16 NICU beds, 40 high care, 16 special care beds and a Kangaroo mother care (KMC) unit (20 beds). Nurse to patient ratio has been well below recommended figures (NICU: 1/3 to 1/4). During the study period the medical staff included 2 full time-paediatric consultants, fifteen medical officers and 2 interns. Because of the burden in neonatal admissions needing mechanical ventilation, our NICU unit is set up to allow only 2.8 meters square around every infant bed which is below the provincial norms of 5 meters squares. It is the Unit policy not to offer intermittent positive pressure ventilation (IPPV) to infants weighing less than 1000 g. VLBW infants 800-999 g are eligible to receive surfactant therapy and Nasal continuous positive airway pressure (NCPAP) with generally no back up option of intubation and ventilation for those infants for whom non-invasive ventilatory support failed. Depending on NICU bed availability infants weighing 900 -999 g without poor prognosis were considered for IPPV. Ethical approval for the study was obtained from the hospital Ethics Committee.

Data collections

Neonatal Data

A prospective, Electronic surveillance system of all neonatal admissions has been in place at LURWMH Neonatal Unit since 2000. Based on this system, monthly admission profile report is generated for routine hospital statistics, unit quality assurance projects and other various unit oral presentations. Routine processes leading to the database data collection is described elsewhere (18).

Using this database, retrospectively, we evaluated survival rates until hospital discharge among infants weighing 500 g to 1499 g admitted at LURWMH between January 2011 and December 2013. The GA (in completed weeks) recorded in the database was determined from routine obstetric assessment (based on dates, clinical examination and sometimes early ultrasound information). The validity of such assessment especially after 24 weeks gestation is questionable but this information reflects our routine practice reality with late booking as common antenatal challenges especially in rural settings. It will take time to reverse this stubble reality on the ground; we therefore decided to run analysis with this information on gestation being aware about the above reservation on assessment quality. The following determinants of infant survival were selected for analysis: infant birth weight and gestational age, antenatal care provision, the place of birth, mode of delivery, gender, surfactant therapy, ventilatory support, one minute Apgar score and antenatal steroid therapy. NCPAP or IPPV were classified as needing ventilation support for the purpose of analysis. Antenatal steroid information was unknown or not recorded for 70% (1193/1686) of the VLBW infants identified for this review. Among the 30% (493/1686) of infants with this information recorded, 49% (242/493) were born from mothers who received antenatal steroids. Because of the incomplete information in maternal files on antenatal steroid in the majority of the VLBW infants, we were unable to assess the impact of antenatal steroid on infant survival. Previous quality check (19) of the quality of Apgar information recorded in the neonatal database at LURWMH showed a tendency to Apgar score overriding. We did not assess Apgar score as independent variable for infant survival. The 10th percentile on the Fenton growth chart was used to determine whether the infant was small for gestation age (SGA). The cause of death was classified according to PPIP classification. Major categories include prematurity, asphyxia, infection and congenital anomalies. Each category is subdivided sub-categories; prematurity is subdivided into extreme immaturity, hyaline membrane disease (HMD), intra-ventricular haemorrhage (IVH), necrotizing enterocolitis (NEC), and pulmonary haemorrhage.

Maternity Data

Maternity data were retrieved from hospital PPIP data and from the Excel data summary compiled monthly using maternity clerk data forms.

Statistical Analysis

Capturing and analysis were carried out using CDC Epi-Info-3_5_1 programme. This includes simple frequency analysis, Stratified analysis to determine stratum specific odds ratio, Chi-square associations to determine odds ratios and confidence intervals, Summary odds ratio and parametric& non parametric one way analysis of variance test for comparing means. The primary outcome measures were death and survivor at hospital discharge. For the purposes of analysis infants transferred out and those directly discharged home were regarded as survivors. Cross-tabulations of categorical variables (antenatal clinic, place of birth, delivery mode, gender, ventilatory need, surfactant therapy) with survival were produced. Infant birth weight was recoded into two categorical groups (500- 799 g; 800- 1499 g). Statistical associations between these categorical variables and survival were assessed using the Chi-Square of association. A 5% level of significance was used. Those variables found significant in the univariate analysis were entered into a multiple unconditional regression analysis with intercept. A separate model was used to examine the influence of infant birth weight groups (in increment of 100 g) on survival, using the birth weight group of 1000-1099 g as a reference mark.


There were 29285 live births recorded at LURWMH during the 3 year-period (Table 1) and thirty one percent (9238) were admitted to the Neonatal Unit, among them 1686 VLBW infants (18.2%; 5.7% of neonatal admissions and total live births, respectively). Their BW and GA mean were 1.127 g (range: 500-1499 g) and 29.3 weeks (range: 23-41 weeks) respectively. The majority of VLBW infants (97.1%) had an estimated GA less than 34 weeks and only 21.1% (358/1686) were assessed as SGA. There were 465 deaths (27.5%) and 1221 survivors (72.5%). The VLBW deaths constituted 60% of the total neonatal deaths. There was significant difference with regard to mean BW (1205 V. 980; p<0.000) and mean GA (30 V. 27.8; P<0.000) between the infants who survived and those who died.

Perinatal characteristics for VLBW infants at LURWMH (Table 2)

Antenatal clinic attendance

Antenatal clinic coverage was 91.8 % among mothers of the VLBW infants during the 3 year period if we define the standard as being at least one clinic attendance.

Place of birth

Seventy five per cent of infants were born at LURWMH and the rest were referred from clinics, district hospitals or born at home or born before arrival. The place of birth was not associated significantly to survival (OR =0.8; 95% CI: 0.628- 1.0596).

Mode of delivery and Gender

Overall, 50.9% of VLBW infants were delivered by caesarean section (C/S). The proportion Of C/S was lower among infant weighing 800 g (33.5% v. 53.5%; p< 0.00001). Fifty one per cent (51%) of infants were female. The female proportion was higher in the BW group 500-999 g (52%) than in the category of infants weighing 1000 g or more (50%) but this was not statistically significant (p=0.4593).

Ventilation support and surfactant therapy

Overall, 43.5% of infants received ventilation support and 48.1% received surfactant therapy. The proportions of both ventilation support and surfactant therapy were higher in the BW category 800-999 g (63.6% and 90.8% respectively).

Causes of death

Sixty one per cent (61.5%), of the deaths were code as prematurity related (Table 3). Among them almost a third (31.8%) was classified as immaturity. Infection was second on the list (17.5%), followed by perinatal asphyxia (13.1%).

Survival by birth weight and gestational age

The survival to discharge was 40.6% (95% CI 37.2%-45.2%) for infants weighing < 1000 g, and 85.1% (95% CI 82.9%-87%) for those weighing 1000-1499 g. The rates of survival as expected, increased with increasing BW from 58.3 % for those in the BW group 800-899 g to 64.90% in the group 900-999 g , escalating to 88.8% and 93.8% in the BW groups of 1200-1299 g , 1400-1499 g, respectively (Figure 1). Survival rates according to GA showed similar pattern with survival rates at 26, 28, 30, and 32 weeks gestation were 40.5%, 74.0%, 84.5% and 92.7%, respectively (Figure 2).

Factors associated with VLBW infant’s survival

On univariate analysis and after adjusting odds ratio according to BW, caesarean section, female gender and ventilation support were found to be associated with VLBW infant’s survivals (Table 4). These variables and the BW were entered into the multivariate logistic model which result is shown in Table 5 & 6). The final model showed that BW (1000 g or more), female gender and C/S were significantly associated with survival in this VLBW infant population. The Odds ratios for deaths for BW < 1000 g and BW < 800 g were 8.0653 (95% CI: 6.3003-10.3003) and 23.3684 (95% CI: 15.5569-35.1022), respectively. Provision of ventilation support was not associated with improved survival (OR 1.1; 95% CI: 0.8667-1.4211) in the model where the BW dichotomy was 1000 g or more vs. less 1000 g. when the BW dichotomy was shifted to 800 g or more vs. less 800 g, the need for ventilation support was a poor prognosis for survival (OR 0.5772; 95% CI: 0.4433-0.7515). BW was the single most important predictor of survival or mortality. When compared with infants in the 1000-1099 g weight group (Table 7), the odds of survival to hospital discharge were much lower among infants weighing less than 1000 g and increased with increasing BW almost by a factor of 2 to 3 for each 100 g increment except in the BW 1300-1399 G and there was no significant difference in odds of survival for infants in the categories 900-999 g and 1000-1099 g.


This three year review study reported the survival rates of VLBW infants admitted to a regional hospital, the only referring center serving a catchment population area of 2.5 million, essentially rural. Poverty rate and unemployment rates stand at 63.5% and 53.7% respectively. The proportion of households with access to safe water (32%) and sanitation (24%) are far below the national figures (79% and 62%, respectively). During the 3 year period, the low birth weight birth (LBW) rate was 20.9%. Saving babies report 2003 documented rates of 17.6% and 13.3% in Metropole, rural areas, respectively. High average LBWR of 24.4% was reported in the Boland/Overberg rural areas in the Western Cape Province between 2001 and 2005 (17). Although the VLBW infants contributed only 5.7% to total live births, they accounted for 18.2% and 60% of neonatal admissions and neonatal deaths respectively. The combination of high LBWR, high number of deliveries and the overall low capacity of neonatal services in remote areas exacerbate the uneven balance between demand and offer for neonatal care. The general infrastructure and logistic problems in the rural settings compound the health challenges and create recipe for low survival rates among this vulnerable group more likely in need of advanced neonatal services, unfortunately so unevenly distributed.

The current overall VLBW survival rate of 72.5% (2011-2013) is in keeping with developing setting figures and compare favourably with published rates from non- rural settings in South Africa ( CMJAH 2006/2007 : 70.5% (8); CH Baragwanath (7) 2000-2002 : 72%). This also represents an improvement if compared to 69.9% survival rate reported in the Unit during the period 2006 to 2010 (18), prior to the introduction of Unit policy which expanded cpap and surfactant therapy to the BW group 800- 999 g. This could explain in part our better survival rate of 40.5% compared to 32% (CH Baragwanath) and 35% (CMJAH). From Turkey (high middle income country like South Africa), D. Atalay et al. (19) reported in 2010 survival rates of 40% and 86.2% for infants weighing < 1000 g and 1000 g to 1499 g , respectively. Our current rates of 40.5% and 85.1% compare favorably but these remain low if compared to developed world rates (4, 5, 6).

In this review, extreme prematurity was the single most significant cause of death in agreement with reported PPIP national data (2, 3). Infants with BW < 800 g were the predominant group (36.8%) among the VLBW infants’ deaths. Their low mean hospital stay (3.2 days) and the fact that 25.5%, 59.2% and 76.3% of them died before 24 hours, 48 hour and 72 hours, respectively are indications of non- active intervention in this BW category. The expansion of nasal CPAP & InSurE to the BW group 800-999 g, was introduced selectively in 2008 and it became routine and official policy in January 2011. The survival rate in this BW group is 57.7%. This is a significant improvement compared to 45.5% rate before the policy shift (OR =1.68, 95% CI: 1.18-2.31; P< 0.0028). This review joins the national debate as to whether it is appropriate to continue using a cut-off point of 1000 g , and whether BW is an appropriate variable to use in deciding neonatal intensive care provision. The big concern and argument has been that because of limited resources, attempt to expand NICU access to the ELBW infants may compromise the care and therefore the outcome not only for ELBW infants but for all neonates sharing these facilities. Table 9 summarized survival data prior and after the expansion of ventilation support to 800-999 g group at LURWMH. Despite the increased burden of VLBW infant admissions (31.3% in annual mean admissions) and despite the deterioration of nurse to patient ratio, all BW groups showed increased survival rates, probably due to the general improvement in neonatal care. The most significant improvement in survival rates was in the 800-999 g group, most likely as a positive impact of nasal CPAP and surfactant therapy. As this positive impact of expansion of NICU programs to lower categories of VLBW infants is duplicated in Public sector around the country, section 27 (2) provision of South Africa constitution may be used to challenge the arbitrary cut-off of 1000 g to provide intensive cares services. This policy, more than 2 decades old, may not stand the legal challenge using the concept of “progressive realisation” of the right of people of South Africa to have access to health care services. It is reasonable that the balance between resources and justice need to be considered at all times but the question is how much time we still need to achieve this “progressive realisation” evoked in section 27(2) of the constitution.

The increase in odds of survival by 100 g increments was also reported by Velaphi et al. in Soweto, South Africa. In this review, there were 12.5 fold increases in survival odds from 600 g to 1000 g and 7 fold increases from 1000 to 1499 g reflecting both severe organ immaturity at the limits of viability and to some degree lack of active intervention among lower range BW categories. The lack of significant difference in survival odds between the 900-999 g and 1000-1099 g BW groups suggests the impact of providing mechanical ventilation and surfactant therapy in both groups. Deliveries by C/S and female gender were advantageous for survival. This has been reported elsewhere and in the Public sector in South Africa (7, 8, 20,). Indications of C/S were not recorded and information about the presenting part is unknown. Further studies are warranted to determine whether this may be a causal association or C/S is just a marker of quality care (good antenatal attendance, antenatal steroid provision, excellent Intrapartum monitoring, skilled paediatrician attending to the infant). Malloy MH (21) analysed the impact of C/S on VLBW infants’ mortality and concluded that the positive impact on mortality was independent from maternal risk factors for Cesarean section.

At this stage of LURWMH neonatal Unit performance, almost achieving a survival rate close to 60% in the BW group 800-999 g, strategically, the Unit should aim at improving this figure to a minimum of 75 to 80% before to consider any expansion of intensive care services below the 800 g mark. Collaborative quality assurance projects with Obstetricians and community services aimed at addressing the burden of VLBW infants (31% increases in annual mean admissions, 2006-2010 vs 2011- 2013. See Table 9). As previously reported (22), pregnancy induced hypertension accounted for a third of NICU admissions at LURWMH in 2011. Improved management of this obstetric complication may reduce the burden of VLBW infants. The recent launch in South Africa of subcutaneous implant for extended period of contraception could be a mid- term to long term leverage in controlling VLBW output as consequence of unwanted or teenage pregnancies. Scaling up low cost obstetric interventions such as antenatal steroid , antibiotic for preterm rupture of membranes, detection and treatment of bactriuria, clean child birth. Neonatal quality projects on top of on-going efforts of scaling up basic neonatal care training and implementation, should include avoidance of nursery overcrowding, strict infection control measures, improving nurses to infant ratios and special NICU skills training for nurses, 100% breast milk provision and aggressive nutrition with early provision (12-24 hours) of parenteral nutrition combined to breast milk as trophic feed. The human factor is crucial to such enterprise, motivating staff with simple communication strategy based on whatever we do let us do it well and responsibly.


Survival of ELBW infants is low. This can be significantly improved. BW and GA were the strongest predictors of survival. Female gender and Caesarean section seem to provide survival advantages. Despite the rural status challenges, survival rates compare favourably with South African published data from non-rural settings.


  1. Who/Preterm birth.www.who.int/entity/mediacentre/factsheet363/en/-32k

  2. RC Pattison for PPIP users. Saving babies 2008-2009: seventh report on perinatal care in South Africa. www.ppip.co.za

  3. D Greenfield, N Rhoda, RC Pattison. A decade of perinatal care survey. www.ppip.co.za

  4. B A Darlow, A. E Cust, D A Donoghue. Improved outcomes for very low birth weight infants: evidence from New Zealand national population based data. Arch Dis Child Fetal Neonatal Ed 2003; 88: F23-F28.

  5. Maureen Hack, Jeffrey D. Horbar, Michael H. Malloy, Linda Wright, Jon E. Tyson, Elizabeth Wright. Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Network. www.pediatric.aappublications.org

  6. Anthony S, Ouden LD, Brand R, Verloove-Vanhorick P, Gravenhorst JB. Changes in perinatal care and survival in very preterm and extremely preterm infants in the Netherlands between 1983 and 1995. Eur J Obstet Gynecol Reprod Biol 2004, 112: 170-7. PubMed Abstract/ www.ncbi.nlm.nih.gov/pubmed/14746953

  1. S C Velaphi, M Mokhachane, R M Mphahlele, E Beckh-Arnold, ML Kuwanda, PA Cooper. Survival of very-low-birth weight infants according to birth weight and gestational age in a public hospital. S Afr Med J 2005; 95: 504-504.

  2. Daynia E ballot, Tobia F Chirwa, and Peter A cooper. Determinants of survival in very low birth weight neonates in public sector hospital in Johannesburg. BMC Pediatrics 2010; 10 -30. www.biomedcentral.com/1471-2431/10/30

  3. Manoel de Carvalho, Maria Auxilliadora S. M. Gomes. Mortality of very low birth weight preterm infants in Brazil: reality and challenges. J Pediatr (Rio J). 2005; 81(1 Suppl): S111- S118.

  4. Grupo Colaborativo Neocosur. Very low birth weight infants outcomes in 11 South American NICUs. Journal of perinatology 2002; 2-7.

  5. P Garg and S Gogia. Reducing neonatal mortality in developing countries: low-cost interventions are the key determinants. www.nature.com/jp/journal/v29/nl

  6. Gary L. Darmstadt, Neff WALKER, Joy E Lawn, Zulfiqar A Bhutta, Rachel A Haws and Simon Cousens. Saving newborn lives in Asia and Africa: cost and impact of phased scale-up of interventions within the continuum of care. Health Policy and Planning 2008; 23: 101-117.

  7. Carlo WA, Goudar SS, Jehan I et al. Newborn care training and perinatal mortality in developing countries. N Engl J Med. 2010; 362(7): 614-623.

  8. Carlo WA et al. High mortality rates for very low birth weight infants in Developing Countries despite training. Pediatrics 2010; 126:e1072-e1080.

  9. Robert Pattisson, David Woods, David Greenfield and Sithembiso Velaphi. Improving survival rates of newborn infants in South Africa. Reproductive Health 2005,2:4:1-8

  10. G F Kirsten, C L Kirsten, P A Henning, J Smith et al. The outcome of ELBW treated with NCPAP and InSurE in a resource –limited Institution.Pediatrics2012; 129: e952-e959.

  11. Western Cape Maternal, Child and Women Health Directorate. Appendix 2 Low Birth Weight. www.westerncape.gov.za/text/2007/cd/volume-7-appendix 2

  12. Kapongo NC et al. Pattern and outcome of neonatal admissions at a regional Hospital, Northern Kwazulu Natal: January 2006 to December 2010. Perinatal priorities conference 2011. www.perinatalpriorities.co.za/proceedingdb

  13. D. Atalay, O Salihoglu, E Can, A .Beskardes, S Atipoglu. Short-Term outcome of Very Low birth weight infants born at a tertiary care Hospital, Istanbul, Turkey. Iran J Pediatr. April 2013; 23(2): 205-2011.

  14. HC. Lee; JB. Gould. Survival advantage associated with cesarean delivery in very low birth weight vertex neonates. Obstet Gynecol, 2006, 107(1):97-105.

  15. Michael H. Malloy. Impact of Cesarean Section on Neonatal Mortality Rates among Very Preterm Infants in the United States, 2000-2003. Pediatrics 2008; 122; 285-292.

  16. Kapongo NC et al. Pattern and outcome of neonatal intensive care admissions at a regional Hospital, Northern Kwazulu Natal: Neonatal database review. Perinatal priorities conference 2012. www.perinatalpriorities.co.za/proceedingdb

Tables and Figures

DH Greenfield

Neonatal Medicine, University of Cape Town, and Limpopo Initiative for Newborn Care

Yüklə 1,08 Mb.

Dostları ilə paylaş:
1   2   3   4   5   6   7   8   9   ...   15

Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©muhaz.org 2022
rəhbərliyinə müraciət

    Ana səhifə