YF Cheung, MP Leung, KT Chau, KW Hung, MH Cheung

Abstract

This audit reports on the results of two-way interhospital transport of sick infants with cardiac problems from March 1996 to August 1997, using the T1 Globe-Trotter Neonatal Transport System. There were 33 transfers, of which 21 were retrieval transfers of 22 patients, 9 return transfers and 3 transfers to other hospital. Of the 22 patients retrieved, 50% were in significant heart failure requiring ventilatory support, 27% receiving intravenous prostaglandin E1/2 infusions and 23% with anticipated complications due to labile haemodynamic status. Their median age was 19 days (range 0.5 to 193 days) and median weight 2.91 kg (range 1.15 to 5.3 kg). For retrieval transfers by ground transport, the total travel and stabilization time were 88±41 minutes and 57±23 minutes respectively; the average total transport time being 2.4 hours. The single retrieval transfer by air took 2.3 hours in total. There were no equipment failures. Comparison of the pre- and post-transport haemodynamic parameters revealed no significant deterioration. Two-way interhospital transport of paediatric cardiac patients ensures smooth transfer of these high risk patients, and prevents the significant acidosis and hypothermia noted in previous one-way transfer of ventilated patients.

Introduction

Paediatric cardiac patients constitute a high risk group in interhospital transport. The acyanotic group of hospital inpatients to be referred to the tertiary centre for intervention is frequently in significant heart failure and often requiring ventilatory support. The cyanotic neonates may be receiving prostaglandin infusion to maintain ductal patency which predisposes them to apnoea. The previous practice of one-way interhospital transport where the patient was sent via ambulance with the accompanying doctor and nurse from the referring institution were, in several ways, disadvantageous.¹ The hospital-based equipments are unsuitable for transport conditions, the equipments on board the ambulance are inadequate for paediatric patients and the accompanying staff may have received limited training in the rigour of transporting critically ill cardiac patients.

The sending out of a dedicated team from the receiving institution is the central idea in two-way transport. The aforementioned disadvantages may be overcome, although there is inevitable delay of travelling from the base to the referring institution. Since March 1996, the Grantham Hospital paediatric transport team was established to provide territory-wide two-way transport of paediatric cardiac patients. The guidelines suggested by the American Academy of Pediatrics are followed.² There have been other audits of transfer of paediatric patients, though not exclusively focussing on children with congenital heart disease.^3-5 In this audit, we report on the profile of paediatric cardiac patients requiring transport and discuss the logistics of such transport arrangement and to identify problems that need rectification. The effectiveness of two-way interhospital transport in ensuring a stable haemodynamic status was assessed.

Methods

Criteria for mobilization of the transport team

The criteria for mobilization included the following: i) neonates or young infants on ventilatory support, ii) labile or unstable haemodynamic status that warrants close monitoring during transport, iii) apnoea or hypotension as a result of intravenous prostaglandin infusion in ductal-dependent circulation, iv) severe hypoxaemia or heart failure, and v) anticipated complications which might necessitate procedures or therapeutic measures to be performed during transport.

Infants who were too ill despite attempts of stabilization in the referring institution were not suitable candidates for transport. Severe low cardiac output syndrome and significant metabolic abnormalities constitute an even higher risk of transfer. The on-site assessment and advice by a paediatric cardiologist with performance of appropriate bedside procedures (as balloon atrioseptostomy for complete transposition of great arteries) might be more beneficial. Mobilization of the team beyond the evening time was not possible in the study period as staff redeployment was only feasible during daytime (9am-5pm).

Personnel

The transport team consisted of the paediatrician, nurse and health care assistant. The accompanying paediatrician was expected to have experience in neonatology and paediatric cardiology, be able to evaluate the cardiac status of the patient and have the ability to perform necessary resuscitation and stabilization procedures. A fellow with at least 6 months of experience in paediatric cardiology was the minimal requirement. The role of the doctor after arriving to the referring hospital included preliminary assessment of the patient's clinical condition, discussion with the parents about the patient's condition and to obtain consent for transport, to ensure adequate ventilatory support (checking the blood gas and endotracheal tube security) and homeostasis (glucose, electrolytes and intravenous access security) of the baby prior to transfer, and packaging of patient into the transport incubator. The nurse was responsible for making clinical observations every 15 minutes during transit and recording the relevant clinical information. The health care assistant ensured normal functioning of the various equipments, most importantly the batteries and the oxygen supply.

Equipments

The transport system used was the T1 Globe-Trotter Neonatal Transport System. The mechanical ventilator was Bio-Med Devices' MVP-10k1, which could be adjusted to deliver flow rates up to 20 L/min, inspiratory time between 0.2 to 2 seconds, expiratory time up to 2.5 seconds, peak pressure to a maximum of 70±10 cm H₂O, PEEP to a maximum of 16±4 cm H₂O. The fractional inspired oxygen ranged from 0.21 to 1.0. The ventilator settings were monitored by the Bio-Med Devices' M-10k ventilator monitor. The additional hand ventilation circuit with manometric monitoring allowed simultaneous positive pressure ventilation in the transport of twins and provided backup should there be malfunctioning of the mechanical ventilator. The system could be supplied from the electric mains or the 12 V car battery.

The clinical parameters of the patients were monitored by the Hewlett-Packard portable monitor which was mounted onto the transport trolley. It provided continuous electrocardiographic monitoring, respiratory monitoring, pulse oximetry, core and skin temperature, continuous intra-arterial pressure monitoring and central venous pressure monitoring if necessary. Up to 4 intravenous or intra-arterial infusions could be given sing 50 ml syringe drivers. The vital signs of the patient was documented by the nurse every 15 minutes during the return trip. Adjustments of the ventilator settings or rate of intravenous infusion and any procedures performed were noted.

Data collection

The data was collected prospectively from 1 March 1996 to 31 August 1997. The data collection included: patient demographic data, cardiac diagnoses and other medical problems, referring institutions, travel time, stabilization time, ventilation requirement and maximal settings, equipment utilization, number and types of intravenous infusions, need for intra-arterial monitoring, problems and complications encountered during transit. The travel time referred to the total time spent in the transporting vehicle, while the time of stabilization referred to the time difference between time of arrival at the referring hospital and the time of departure. The sum of the two time intervals was the total transport time. The clinical parameters that were documented just prior to departure from the referring hospital and immediately on arrival to Grantham Hospital included the followings: heart rate, systemic arterial pressure, oxygen saturation, arterial blood gas if intra-arterial line was in-situ and blood sugar by dextrostix. The temperature on admission was documented. For patients with intra-arterial lines, the blood gas was checked in the referring institution after the child had been stabilized within the transport incubator for 15 minutes; similarly, blood gas was repeated immediately upon arrival while the child was still inside the incubator.

Similar data of patients transferred in by one-way transport from the referring hospital was reviewed. Ten ventilated neonates and young infants admitted between September 1994 to September 1996 were randomly selected and their clinical parameters immediately upon arrival were compared with that of the ventilated patients retrieved by two-way transfer. The ventilated patients were chosen as they represent the more critical ones whereby effects related to suboptimal transfer may easily be apparent.

Statistical analyses

The distribution of demographic data and ventilatory settings was expressed as median and range. This offers important information with regard to the extreme possibilities that the equipments and facilities would need to cover. The results of clinical parameters were expressed as mean±standard deviation. Paired Student's t test was used to compare the clinical parameters prior to leaving the referring hospital and immediately on returning to Grantham Hospital. The comparison with the historic data was performed using unpaired Student's t test, Mann Whitney test and Fisher's exact test where appropriate. The significant level was set at 5%.

Results

Demographic data

There were 33 transfers of 34 patients in the 18 month period, with an average of 1.8 transfers per month. The median age was 25.5 days (range 0.5 to 193 days). The age distribution was skewed to the left (fig. 1) and could broadly be divided into three groups of patients: i) neonates with ductal-dependent cyanotic cardiac lesions , ii) one to two months old preterm babies with significant heart failure due to persistent arterial duct, and iii) young infants with large left-to-right shunt corresponding to the drop in pulmonary vascular resistance. The median weight was 2.8 kg (range 1.04 to 5.3 kg). The were 17 boys and 17 girls, including one pair of monozygotic twins retrieved in the same transfer.

Of the 33 transfers, 21 were retrieval transfers of 22 patients, 9 return transfers to the referring hospital after intervention, and 3 transfers to other hospitals for other aspects of care. All the 22 requests assessed on-site were judged to be suitable for transport. The median age and weight were 19 days (range 0.5 to 193 days) and 2.91 kg (range 1.15 to 5.3 kg), respectively. The average retrieval transfer was 1.2 per month.

Fig Age distribuuuu us use patients at the time of transfer.

Indications for and mode of transport

The indications for which the transport team was mobilized were shown in table I. Of the 22 patients retrieved, 50% were in significant heart failure requiring ventilatory support, 27% (6/22) were given continuous intravenous prostaglandin E1 / E2 infusions for ductal-dependent cyanotic heart conditions. The remaining 23% (5/22) of patients had labile haemodynamic status with anticipated complications during transport. Twelve patients were either sent back to the referring hospital after interventions in Grantham Hospital or transferred out for other aspects of care.

Ground transport by ambulance was employed in 31 transfers (20 retrieval, 11 return or to other hospitals), while helicopter was used in 2 transfers. Air transport was employed in one return and one retrieval transfer to and from Tuen Mun Hospital, respectively. The baby girl retrieved was 4 month old and weighing 3.3 kg, with a cardiac diagnosis of truncus arteriosus and ventricular septal defect. She was complicated by severe heart failure, for which ventilatory support was instituted. To avoid delay due to traffic jam on the day of transfer, air transport was selected. The hospitals involved in this transport network were listed in table II.

Transport times

For retrieval transfers by ambulance, the travel time to-and-fro was 80 ± 41 minutes per transfer. For the retrieval flight from Tuen Mun Hospital, the flying time was 10 minutes in contrast to the 60 minute-journey of sending the incubator to Tuen Mun Hospital by ambulance.

The stabilization time in all the retrieval transfers was 57 ± 23 minutes. The stabilization time, as defined by the difference between the time of arrival and time of departure, actually included the time of mobilizing the central supporting staff and moving the incubator to the ward. The activities performed during this period included gathering of clinical information, assessment of the patient and performing necessary procedures, obtaining consent and transferring the patient to the incubator.

The average total transport time per retrieval transfer by ambulance was 145±55 minutes, the single retrieval by helicopter took a total of 140 minutes. For back transfer or transfer to other hospitals, the total transport time was 129±49 minutes.

Equipment use and medications

All patients had continuous electrocardiographic and respiratory monitoring. One patient with prolonged QT syndrome and episodes of ventricular tachycardia had a defibrillator standby. Two patients had the chest drains left in-situ after ductal ligation. Intra-arterial lines were inserted in 50% of the patients, which allows continuous blood pressure monitoring during transit. Continuous intravenous infusions were required in 82% (28/34) of patients. Fifteen patients required only one intravenous infusion, while 12 required 2 or more. The maximum needed was four. The types of infusions were listed in Table III. Intravenous boluses of medications were required during transit in 3 patients, midazolam in two and pancuronium in one. There were no equipment failures or dislodgement of lines being documented, irrespective of the mode of transport. Battery change, however, was required for the Hewlett-Packard transport monitor if it was switched on for more than 1 hour. Transient intermittent loss of signal from the pulse oximeter was the only common problem.

Respiratory support

Positive ventilatory support was required in 65% of the patients (22/34). Twelve of 22 patients retrieved were ventilated, the median ventilatory settings were: peak inspiratory pressure 22 cm H₂O (range 15-25), PEEP 4 cm H₂O (range 4-6), rate 38/minute (range 26-50), inspired oxygen concentration 50% (range 25-78). No serious complications of ventilation occurred. There were no documented dislodgement of endotracheal tube during transit. After the initial adjustment of the ventilator settings based on the blood gas result after stabilizing the baby within the transport incubator, no further adjustments were required throughout the return journey.

Clinical parameters

Comparisons of the clinical parameters to identify possible deterioration as a result of transport showed no significant difference (Table IV). The comparisons were made only for retrieval transfers as the data collection is more complete to allow meaningful interpretation. There was no significant change in oxygen saturation in children with cyanotic heart disease after transport and the haemodynamic status remained relatively stable. The rectal temperature on admission was 37.2 ± 0.4°C (range from 36.5 to 38.1°C). The comparison with the historic data was illustrated in Table V. The previous practice of one-way transport was associated with patients being admitted in a more acidotic, hypercarbic and hypothermic status.

Outcomes

No patient died or deteriorated during transit. The arrhythmia was well controlled with no breakthroughs throughout the journey. There were no serious complications or equipment failure. All the 22 patients retrieved underwent medical or surgical interventions: 17 patients received corrective or palliative cardiac surgery, two received interventional cardiac catheterizations, two for control of arrhythmia with one requiring pacemaker insertion, one for continuous renal replacement therapy after cardiac surgery. The median duration of stay for these patients was 17 days (range from 2-57 days). Fifty % of the patients could be discharged directly from Grantham Hospital, 41% (9/22) being transferred back to the referring hospital or other hospital for continual care, while only 9% (2/22) died eventually after surgical interventions.

Discussion

The provision of two-way transport of sick infants with cardiac problems aims to minimize the possible deterioration in transit in view of the labile haemodynamic status. Historical data suggests common findings of hypothermia and mixed respiratory and metabolic acidosis with the previous practice of one-way transport. The major difference between the planned two-way transfers and the previous ad-hoc transfers appears to be the provision of optimal pre-transfer stabilization and means of tackling with anticipated problems during transfer. A poor clinical status upon arrival might require a longer period of stabilization is required before any surgical or transcatheter interventions could be performed. Furthermore, significant deterioration during transit with low cardiac output or severe hypoxaemia would worsen the eventual outcome despite subsequent rectification of the cardiac lesions.

Pre-transfer stabilization in the referring hospital is important for a smooth transport. Early and correct diagnosis of the underlying heart condition dictates appropriate management and allows ample time for stabilization before referral. The transport team should optimize the patients's condition prior to transport.^6-8 Four aspects that need to be focussed on in paediatric cardiac patients are: i) the need for ventilatory support, ii) institution of inotropic support, iii) control of arrhythmia and iv) maintenance of ductal patency by prostaglandin E1/2 infusion. Intubation during transport should be avoided if possible. The decision to intubate should be made while in the referring hospital after initial assessment. Patients with heart failure and significant respiratory distress, those with recurrent prolonged apnoea after starting intravenous prostaglandin E1/2 infusions are candidates for elective intubation before transfer. The position and security of the endotracheal tube should be ascertained. Infusions of inotropes and antiarrhythmic medications should be continued. For patients with refractory, life-threatening arrhythmia, a defibrillator with the paediatric pads should be available on board the ambulance should emergency defibrillation be required in transit. Continuous prostaglandin E1/2 infusion is lifesaving for neonates with ductal-dependent cardiac lesions as pulmonary atresia, transposition of great vessels and severe left ventricular outflow obstruction. This drug should be stocked in the transport incubator as it is not universally available in all the private hospitals.

The mean stabilization time of 57 minutes compares favourably with the another reported audit of neonatal transport.³ The calculation of the stabilization time in this study included the time required to wait for the central supporting team to help transferring the incubator from the ambulance to the ward. There is usually a delay of 10 to 15 minutes, which in essence is wasted. The notification of the central support team of the referring hospital through the mobile phone when the ambulance is arriving would probably minimize such delay. The trip to Tuen Mun Hospital by helicopter is merely only one-sixth the time that of ground transport. Air transport to-and-fro Tuen Mun Hospital avoids the potential problems relating to unexpected traffic congestion (such as inadequate oxygen or battery), reduces the vulnerable period that patients with unstable haemodynamic status are exposed to and reduces the total transfer time. The use of helicopter as a routine in such circumstances should be considered. Unfortunately, the current policy did not allow air transport of the vacant transport incubator to Tuen Mun Hospital.

All the transfers had been relatively smooth. To minimize the risk of complications during transit, the child should be well sedated before starting the journey. It is common to give a single intravenous bolus of midazolam just before leaving the hospital, and it usually serves the purpose. Some of the patients were already started on continuous morphine, fentanyl, or midazolam infusions. The infrequent use of muscle relaxant may reflect the reluctance of the team to use such medication for fear of accidental extubation. The checking of line patency and endotracheal tube security is especially important. Should reinsertion of the lines or the tube proved necessary during transit, it would be advantageous to request the ambulance be pulled over, rather than struggling with the vehicle moving. This, of course, would not be feasible in a helicopter which therefore emphasize on the importance of pre-transfer stabilization. It is extremely noisy inside the helicopter which makes verbal communications difficult. Furthermore, one is firmly strapped in during the flight which renders performance of procedures difficult. The equipments and the ventilator had been reliable and the readings were quite consistent during transit, whether by ambulance or helicopter transport. Apart from intermittent interference of pulse oximetry signals, the electrocardiographic signals and quality of pressure tracings had been satisfactory.

The establishment of a dedicated transport team is not a minor undertaking. The cost includes a team consisting of at least a third year medical officer, third year nurse and a first year health care assistant, the equipment depreciation cost, cost of medications, ambulance or helicopter transport cost, other public transport cost (for example taxi fare) after return transfer and mobile phone maintenance expenditure; opportunity cost includes the preparation time, waiting time, redeployment with reduction of staff for the established activities. Under the current circumstance, the team cannot be 'dedicated' as such as it is definitely not cost-effective to have a team standby only for the sake of transport in view of an average of 1.8 transfers per month, with two-third being retrieval transfers. Flexible redeployment of staff from the intensive care unit to form the team is necessary for each transfer, so that early notification is important and the team to be mobilized on a semi-elective basis.

The provision of the transport team is currently limited to daytime service. The time taken from receiving the consultation call to actual mobilization of the transport team reflects not only the time necessary for the stabilization of the patient in the referral hospital, the urgency of the situation and the mobilization efficiency, but also the time and personnel constraints in the actual clinical setting. This time slot data was not collected in the present study in view of the semi-elective transfers for some of the patients due to the presently formidable constraints, thus confounding the final interpretation.

Accurate documentation of the timing and any reasons for the inevitable delay does, however, allow for more objective appraisal of the urgency of the transfer and the preparatory phase before the actual transfer. Furthermore, data on the clinical status of patients regarded as stable enough for one-way instead of two-way transfer after initial telephone consultation was not available. The auditing of the outcomes of patients selected for one-way transfer after initial screening would provide suggestions for necessary revision of the guidelines for two-way interhospital transport in the future.

In conclusion, two-way interhospital transport of paediatric cardiac patients by the transport team ensures smooth transfer of these high risk patients. The maintenance of a stable haemodynamic status minimizes the delay for subsequent surgical or transcatheter interventions.

References

1. Pon S, Notterman DA. The organization of a pediatric critical care transport program. Ped Clinics of N Am 1993;40:241-61.

2. American Academy of Pediatrics Committee on Hospital Care: Guidelines for air and ground transportation of pediatric patients, Pediatrics 1996;78:943-50.

3. Leslie AJ, Stephenson TJ. Audit of neonatal intensive care transport. Arch Ds Child 1994;71:F61-6.

4. Robb HM, Hallworth D, Skeoch CH, Levy C. An audit of a paediatric intensive care transfer unit. Examining the workload and practice of the medical team. British Journal of Intensive Care 1992;Nov-Dec:371-9.

5. Black RE, Mayer T, Walker ML, et al. Air transport of pediatric emergency cases. N Engl J Med 1982;307:1465-8.

6. Qslon CM, Jastremski, Vilogi JP, et al. Stabilization of patients prior to inter-hospital transfer. Am J Emerg Med 5:33,1986.

7. Hackel A. A medical transport system for the neonate. Anaesthesiology 43:258,1975.

8. Hicks TC, Danzl DF, Thomas DM, et al. Resuscitation and transfer of trauma patients. A prospective study. Ann Emerg Med 11:296,1982.