Table of Contents

HK J Paediatr (New Series)
Vol 1. No. 1, 1996

HK J Paediatr (New Series) 1996;1:82-85

Occasional Survey

Development of Neonatal Intensive Care in Hong Kong

CY Yeung


Attempts to introduce neonatal intensive care in Hong Kong had been fragmented and not supported with enthusiasm until recently. Proper neonatal intensive care was commenced in our Department in 1981 and a combined paediatric and neonatal intensive care programme was established at Queen Mary Hospital in 1982. Our neonatal survival results compare favourably with many reputed North American centres. It is hoped that Hong Kong may develop proper Perinatal Care Programmes and an efficient transport system to further improve our health care deliverance not only to the critically illed neonates but also to the at risk fetuses and their mothers.

Keyword : Intensive care; Neonate

The need for providing special and intensive care for neonates has long been recognised in Hong Kong. However, the priorities of distributing the 1.7% of the GNP to the public hospitals or less than 4% of the GNP for health care,1 were not placed on the development of neonatal medicine nor on other paediatric subspecialties. A proper neonatal intensive care programme was not established in our Department until 19811 after much negotiations and convincing, both with the administration and even within the Department.

The infant mortality rate of Hong Kong has been reduced from 100/1000 live births in the post-Second World War years to only 4.9/1000 in the year ending March 31, 1993 (Figure 1).2 Much of this improvement was due to a most effective child health programme ever introduced to Hong Kong, the "Maternal and Child Health Centres", which are scattered all over the territory. It offers free health check and immunizations for infants and health education to mothers living within the vicinity of the centre. The programme has effectively eradicated many of the potentially fatal and disabling infectious diseases in childhood.3 It is probably one of the main contributors to reducing infant mortality. The decline in neonatal mortality however, has not been as dramatic except in several major hospitals with special neonatal services.

Fig. 1 Infant and neonatal mortality rate of Hong Kong (1946-1992)

The Early Years

The first attempt to reduce the alarming incidence of "kernicterus" occurring in jaundiced term infants was made in late 1961 by Dr S.C. Hu who performed the first exchange transfusion (ET) at Tsan Yuk Maternity Hospital (TYH). It was a big event, drawing a huge crowd of onlookers besides having a number of senior paediatricians performing and assisting in the procedure. This was followed by a rapid increase of exchange transfusions being done, with a record number of seven ETs in one night at the Paediatric B unit of Queen Elizabeth Hospital (QEH) in 1970.4 Universal phenobarbitone prophylaxis was adopted which has effectively reduced the frequency of severe hyperbilirubinaemia and to go with it the number of exchange transfusions.5

In 1969, the first sets of air-conditioners ever installed in the public wards of government hospitals were put into operation at the "new nursery" of Tsan Yuk Hospitals. They were not intended to cold-stress the neonates but to filter off the visible air pollulants emitted from the vicinity, such as the smoke from the restaurants and the log and kerosene-burning homes. A side room was later set aside and installed with local-made phototherapy units for additional treatment of neonates with moderately severe jaundice.

Much emphasis was also put on controlling infections in the neonates especially preventing contamination and infection of the umbilicus and the decaying cord. Problems were often related to the traditional practice of "preventing the umbilicus from exposure" ( 露風 ) by covering with rags and even ashes of burnt incense. Meanwhile, rectal paraldehyde drip was introduced to control seizures occurring among infants with neonatal tetanus from umbilical infection. This form of therapy was found to be highly successful in controlling the seizures and the board-like rigidity in addition to improving survival (unpublished ovservation).4 although there was no artificial ventilation support.

Introduction of Ventilator Care

Early paediatric trainees of Queen Mary Hospital (QMH) traditionally spent a year to train in internal medicine. During this period, most of them would receive some training in cardio-respiratory resuscitation. But the first attempt to ventilate a neonate with severe respiratory failure did not occur until late 1968. It was in a preterm Indian infant boy with severe RDS. The infant was ventilated with the old Bennett respirator, an inappropriate equipment borrowed from the Medical Unit of the hospital. The baby was successfully ventilated and stayed in a private ward for over 100 days, 10 days more than he should, because the parents had received instruction from the high priest of the Sikh Temple that the child "would come out of all dangers if he can stay in the hospital with his doctor for the first 100 days". Subsequently, with the help of Dr Y.C. Tsao, the "purchasing officer" of the paediatric unit, a Loosco (Amsterdam) or "animal" respirator and a set of Hewlett-Packard cardio-respiratory monitor were acquired. I was intrigued to find, on my returned from Canada in late 1980, that the same Hewlett-Packard monitor half broken by then, was still the only set of monitor available in the unit, although apart from the "animal respirator" there was another set of Bournes BP 200 ventilator newly purchased for the department.

On my unexpected transfer to head the second paediatric unit of Queen Elizabeth Hospital in early 1970, there were no equipment support for the neonatal service apart from exchange transfusion sets. Dr R. Johnson Lee and I were instrumental to introduce HIV (Human Intermittent Ventilation) to the paediatric service of the hospital. We utilized an under-water-glass-tube (Figure 2) as a manometer to ensure appropriate pressure on ventilating the infant with the anaesthetic bag. Usually, a student nurse or the most junior of the medical or nursing staff was taught to inspire by squeezing the blown up anaesthetic bag, ensuring a constant stream of air bubbling through the under-water seal of 15-20 cm water. Releasing of the small open end of the bag would permit expiration. This labour-intensive HIV system had helped quite a few critically illed neonates.

Fig. 2 'HIV' - Human intermittent ventilation

The year 1972 saw the influx of "modern" infant incubators (the Air-Shield) with servo-temperature-control-system to meet the need of the sick neonates on both sides of the harbour in addition to replacing the few broken "Armstrong incubators", which necessitated hot water bottles placed inside to keep warm.

Re-organisation of Neonatal Service

Before 1980, sick neonates at Tsan Yuk Hospital were looked after by an obstetric medical officer in consultation with a paediatric doctor, and those at Queen Mary Hospital by the on-call paediatric M.O. Re-organisation was achieved by setting up a small team of paediatric doctors to station at TYH and another team to provide services to both the inborn and outborn sick neonates at QMH. Paediatric doctors were trained locally in the first instance and then sent overseas for additional experience. With the support of the then DMHS, accommodations were re-arranged and necessary medical equipments ordered; and a new laboratory facility was also planned. By the latter part of 1982, several monitors, ventilators and a proper neonatal laboratory were in place. In 1982, our Department also organised the first paediatric intensive care course and graduated 34 paediatric nurses for the first time in Hong Kong. Three more batches of paediatric nurses have gone through such course since.

We have planned a combined intensive care programme to embrace both general paediatric and neonatal ICU as one service at QMH.6 We have included paediatricians with training in general ICU and neonatology in the service and also respirologist, nephrologist and cardiologists to form a core-group of intensivists (Table I). Residents do not only learn intensive care neonatal medicine, they also learn various intensive care procedures which are needed but seldom available in most neonatal centres. For example, central venous pressure monitoring, peritoneal dialysis, doppler ultrasonogram for identifying ductus or septal defects etc. are some of the techniques that are now regularly performed by our ICU residents under supervision of our core-intensivists.

Table I Development of Neonatal Intensive Care in Hong Kong
Date Event Hospital
1961 First Exchange Transfusion TY
1968 Ventilatory therapy for neonates on borrowed equipment QM
1969 Ventilator + CR Monitor acquired QM
  Air-conditioning to filter off air pollutants TY
1970 HIV-Human Intermittent Ventilation QE
  Rectal paraldehyde for neonatal tetanus QE
1972 Bulk supply of Incubators QM, QE
1982 Proper NICU Beds QM, TY, GH
  Neonatal Laboratory TY
1990 New Intensive Care Unit (combined) QM
1991 New Intensive Care Unit (neonatal) TY

Our arrangement may not be most conducive to the development of neonatology per se, nevertheless it encourages more academic and professional crosspollinations among various intensivists of the paediatric discipline. For a small and relatively insulated place like Hong Kong, it could be a useful model and a more cost-effective intensive care programme in a regional hospital. Our combined Paediatric and Neonatal Intensive Care Programme has recently been chosen by British Paediatric Resource Committee for discussion as a possible model to adopt for service provision in U.K. also (Chantler C, personal communication).

The range of our current activities can be seen in Table III. Our core intensivists have received both local and overseas training and experience in many neonatal and paediatric intensive care centres, with at least 5 years of post-MRCP experience in the field (Table II). A little over half of the patients referred to our ICU programme in the past 5 years were neonates (Table III). The spectrum of the major neonatal illnesses referred to us are immaturity and related respiratory problems followed by other system disorders as indicated in Figure 3. The same core group of intensivists have been providing the service coverage of Tsan Yuk Maternity Hospital also since 1985.

Table II Medical Staff of the Combined Paediatric ICU at QMH & TYH
Staff No. Training / Experience Background
Paediatricians with 5+ years post-basic experience or training in ICU 5 Hong Kong, London, McMaster, Cambridge, Melbourne, Pittsburg, Edmonton, Chapelhill, Stanford
Others with MRCP (or equivalent) HK, UK, USA
Residents HK


Table III Admissions to QMH Combined ICU (1985-1992)
Age Distribution Major Problems
Age Proportion Condition Proportion
< 1 month 50.4% Immaturity 35.9%
1-12 mo 23.7% Respiratory 35.7%
1-5 yr 15.1% Gastrointestinal 34.6%
1-10 yr 6.7% Infection 42.8%
10-15 yr 4.1% Neurologic 36.8%
    Cardiac 28.1%

Effect of Our Neonatal ICU Programme

The introduction of our neonatal ICU programme in late 1980 has resulted in a significant reduction of neonatal mortality in both Tsan Yuk and Queen Mary Hospitals. At QMH, where the programme was a combined paediatric and neonatal one, the survival results of the low birth weight infants born in our hospital compared favourably with that of McMaster University (Table IV).7,8 In fact, infants of 1.0 to 1.5 kg birth weight have demonstrated much more dramatic improved survival (86.5% in Hong Kong Vs 40.9% at McMaster) in the post-ICU years compared with McMaster. Although infants of less than 1 kg showed results similar to McMaster (20.6% Vs 21.2%), it has however not taken into account of many infants less than 1000 gm born before 1980 who were probably recorded as abortions. While in the post ICU era all live-births with birth weight over 500 grams were included in the calculation. The actual improvement in the survival of the Extreme Low Birth Weight Infants therefore must have been more significant.

Fig. 3 Disease systems of neonatal admissions (85-89) Queen Mary Hospital


Table IV Improvement of Neonatal Survival of Inborn Infants after ICU at QMH
(BW in Kg)
Before ICU After ICU Percentage Improvement
Live Births Deaths /1000 Live Births Deaths /1000
< 1 Kg HK 23 826 61 656 20.6
  Mc 160 881 98 694 21.2
1-1.5 Kg HK 64 797 83 108 86.5
  Mc 213 366 167 216 40.9
Overall HK 87 805 144 340 57.8
  Mc 373 587 265 392 33.2
Mc = A North American Centre (NEJM 1983)
HK = Hong Kong; before ICU = 1975-80;
after ICU = 1981-86

Improvement in neonatal survival results at Tsan Yuk Maternity Hospital upon introduction of ICU programme was equally impressive.9 More recently the quality of the survivors also appears comparable to most of the published series.10

Future Development

To ensure the most successful outcome, a neonatal programme must be run in close collaboration with an obstetric component i.e. a really perinatal programme. The early Wisconsin experience11 has demonstrated that by combining the separate working routine of the neonatal paediatricians and obstetricians into one functional programme, they were able to dramatically improve both the perinatal and neonatal survival statistics. Hong Kong might do even better by a really combined perinatal programme.

I sincerely hope that the day will come soon that neonatal paediatricians in Hong Kong will get involved in looking after the fetuses of complicated pregnancies, meeting and counselling the mothers together with their obstetric colleagues. They should not present themselves to the fear-stricken mothers as a strange breed of physicians who are taken by surprise with severe neonatal illnesses which usually demand their impromptu attention under the current traditional practice pattern. Most of these illnesses can be managed more effectively if a therapeutic programme can be planned more prospectively with the obstetricians and participation of the parents before the babies are born. We also need to develop an effective in-utero and neonatal transport system so that high risk pregnancies and critically illed neonates may be transferred to tertiary centres for further expert and intensive care.


In most industrialised communities, like USA, where 11-12% of their GNP are spent on health care, even private hospitals are equipped to care for critically-illed neonates. Hong Kong however is very different, most tertiary clinical services are mainly available in public hospitals resourced by only 1.7% of the GNP. Neonatal intensive care is not yet among the most established tertiary services in our public hospital system. With increasing interest in neonatology and more human and other resources channelled to the perinatal service, I am confident that a much improved neonatal service will be forthcoming in most regional hospitals in the near future. It is also my wish that a much closer liaison between obstetric and neonatal services can be achieved to implement a real perinatal service to our community.


I am grateful to the Hong Kong Paediatric Society for giving me the honour and inviting me to deliver this special lecture to trace the development of neonatal intensive care in Hong Kong.


1. Hong Kong Annual Reports, 1980-1992. Hong Kong Government Printers.

2. Yeung CY Tradition and child health. Supplement to the Gazette of the University of Hong Kong. Jan 1990.

3. Department of Health, Hong Kong Government; Annual Report, 1992.

4. Annual Report of Paediatric B Unit, Queen Elizabeth Hospital, 1970.

5. Yeung CY, Lee KH, Chan A, Tam LS. Phenobarbitone prophylaxis for neonatal hyperbilirubinemia. Paediatrics1971;372-6.

6. Wong SN, Tsoi NS, Yeung CY. Renal replacement therapy in a combined paediatric and neonatal intensive care program A Hong Kong experience. Paediatric Nephrology, 1994; (In press).

7. Cheung PT, Yeung CY, Tam YC. Impact of a neonatal intensive care program on neonatal mortality of VLBW babies. Special Silver Jubilee Publication of Department of Paediatrics, University of Hong Kong, 1987;p140.

8. Boyle MH, Torrance GW, Sinclair JC, Horwood SP. Economic evaluation of neonatal intensive care of very-low-birth-weight infants. N Engl J Med 1983;308(22):1330-7.

9. Lau SP, Fok TF. Impact of neonatal intensive care on neonatal survival at Tsan Yuk Maternity Hospital. HK J Paediatr 1985;2:131-7.

10. Tang TS, Yeung CY Outcome among surviving very low birth weight infants. HK J Paediatr 1992;9:144-9.

11. Gravins SN, Howe G, Callon HF. Perinatal Care in Wisconsin 1964-1970. In Stetson JB and Swyer P editors : Neonatal Intensive Care, St. Louis 1975, Warren H Green Inc.


This web site is sponsored by Johnson & Johnson (HK) Ltd.
©2024 Hong Kong Journal of Paediatrics. All rights reserved. Developed and maintained by Medcom Ltd.