METHODS

Data regarding all aspects of transport have been collected prospectively on our PICU from 1993. This dataset was accessed to provide data relevant to this study, which included details regarding the initial airway and vascular access management procedures undertaken in all children for whom a paediatric retrieval team was launched in two separate one year time periods. The first one year period was October 1993 to September 1994; the second was October 2000 to September 2001 (some of the data included in this study were published in 1995).¹⁰ These periods were chosen deliberately separated by many years to allow sufficient changes to have occurred within the model of paediatric retrieval services in the Southeast of England. It was postulated that in 1993 (the year in which the British Paediatric Association report was published), a specialist transport team was not widely used by DGHs; whereas in the latter period, most paediatric retrievals were undertaken by such a specialist team.

We used as outcome measures: the proportion of children in whom the DGH staff performed the initial airway management procedure (endotracheal intubation) and vascular access procedures (arterial and central venous access) in each of the two time periods.

Comparison of proportions was performed by using the χ² test for independence (or Fisher’s exact test when one of the sample sizes was small). Results were expressed in terms of odds ratios. Statistical significance was defined as p < 0.05.

RESULTS

Complete data were available on 340/353 children retrieved in 2000–01, compared to 64/65 children retrieved in 1993–94. Reasons for referral covered a wide range of diagnoses. Table 1 provides a breakdown of the diagnostic groups, and shows that respiratory illness and sepsis were the main diagnostic categories (32.8% and 48.4% respectively in 1993–94; 39.6% and 27.5% respectively in 2000–01).

Table 2 provides a comparison of the two groups of children included in the analysis, with regard to the overall rate of each intervention. The proportion of children retrieved in whom endotracheal intubation was performed did not change significantly in the two time periods (83.9% and 79.1% respectively, p = 0.69). This was in contrast to central venous and arterial access procedures. The proportion of children retrieved in whom a central venous access procedure was performed decreased from 70.3% in 1993–94 to 51.8% in 2000–01 (p = 0.006); for arterial access, these proportions were 83.9% and 69.8% respectively (p = 0.024).

Table 3 shows the proportion of endotracheal intubation performed by DGH staff in the two time periods. This proportion increased from 59.6% in 1993–94 to 84.4% in 2000–01 (p < 0.001, odds ratio 3.66, 95% CI 1.92 to 6.97). From table 4, it is clear that the proportion of central venous access procedures performed by DGH staff also increased from 11.1% to 18.9%. However, this change was not statistically significant (p = value 0.31, odds ratio 1.86, 95% CI 0.68 to 5.07). Table 5 suggests hardly any change in the proportion of arterial access procedures performed by DGH staff.

DISCUSSION

The results of this study indicate that the number of retrievals to our unit increased fivefold from 1993–94 to 2000–01; however, a smaller proportion of children retrieved had interventions performed on them. Of these children in the 2000–01 cohort, a greater proportion were intubated and managed with invasive haemodynamic monitoring by DGH staff in the initial few hours preceding arrival of the specialist retrieval team. This confirms that staff at referring hospitals are not performing fewer resuscitation and stabilisation procedures during the management of critically ill children than in the past. This trend is most obvious with airway management (endotracheal intubation rate), and less marked for circulatory management (central venous and arterial access). The overall decreased rate of intervention for these latter procedures over the years may be explained by a relative increase in the proportion of children who were intubated, but did not need central venous or arterial access.

Although an index of disease severity (such as PRISM III¹⁵ or PIM¹⁶) was not used to formally compare the two groups, the rate of endotracheally intubated patients in the cohorts remained similar, suggesting that there was no large change in severity of illness. The assumption that the rate of endotracheal intubation can be used as a surrogate marker for severity of illness has previously been made in a study assessing changes in the profile of PIC associated with regionalisation.¹⁷

Early stabilisation is a crucial step in the initial management of a sick child presenting to an emergency department.¹⁸ Failure to stabilise the child may result in continuation of physiological deterioration and an inability to reverse the pathological processes that lead to further patient morbidity and mortality.¹⁹ We chose airway and vascular access procedures as markers of stabilisation skills in critically ill children. Endotracheal intubation is a vital skill that is commonly employed in airway management in an emergency setting. The ability to achieve central venous access is considered a necessary stabilisation skill: it allows the measurement of central venous pressure to guide further fluid resuscitation and the delivery of multiple infusions as well as inotropic support centrally (such as adrenaline in children with septicaemia).²⁰ The ability to perform invasive blood pressure monitoring and measure blood gases regularly makes arterial access an important procedure in the management of a sick child. Therefore, familiarity with these procedures is crucial to the emergency stabilisation of a sick child.

A previous study has shown that physicians at local hospitals continue to perform adequate clinical assessment of critically ill children,²¹ and refer to tertiary centres appropriately. We believe that there are several reasons why DGH staff have retained their assessment and stabilisation skills. An increasing number of DGH staff are now trained in Advanced Paediatric Life Support. Better training of middle grade staff, increasing “hands on” consultant involvement in the emergency department, better organisation of training in paediatric anaesthesia,²² and rotation of trained senior specialist registrars to DGHs (previously confined to teaching hospitals) may all play a significant part in accounting for these findings. Active support from the tertiary centre, beginning with telephone advice while the retrieval team is en route, and dissemination of best practice management guidelines to local hospitals, may also play a major part in explaining some of these results.²³ The role of DGH staff in the management of sick children prior to intensive care needs to be recognised and supported by tertiary PICUs within the model of centralised paediatric intensive care. Training of front line staff at these centres will need to focus on the initial resuscitation and stabilisation of critically ill children.

Limitations of the study

The increasing use of resuscitation and stabilisation procedures may not, by themselves, adequately represent proficiency in the management of critically ill children by DGH staff. Management of the sick child also includes other vital clinical skills such as early recognition of illness, adequate seizure control, and the institution of appropriate antibiotic therapy. As such, the use of procedural skills as the only markers of management of the child at the referring hospital is difficult to generalise to the overall quality of initial stabilisation. Moreover, the quality of interventions performed at the referring hospital was not examined in this study. It is possible that the increased number of interventions by DGH staff was associated with a higher complication rate. It is also possible that in the absence of the centralised PIC model, quality of initial management at the DGH may have improved even further due to increasing recognition of the need for early and aggressive stabilisation in this subset of children. A study undertaken from the DGH perspective, which examines the number and quality of initial stabilisation procedures in sick children, may answer many of these questions.

This study does not describe the specific designation and specialty breakdown of staff involved in performing the greater proportion of vital procedures. It is highly likely that this increase could all be attributed to the involvement of anaesthetic staff rather than paediatricians; this information would have been useful to examine issues regarding further medical training and education of medical and nursing staff at the DGH.

Conclusions

The widespread use of a specialist paediatric retrieval service has not resulted in loss of vital stabilisation skills among referring hospital staff. In fact, a greater proportion of initial stabilisation, notably endotracheal intubation, is being performed by staff at the referring hospital. This finding will help to define the future role of DGHs within the evolving model of centralised paediatric care, and provide guidance on issues pertaining to training of front line staff involved in the initial management of the majority of sick children.