Diagnostic Delays in Pediatric Stroke
Diagnostic Delays in Pediatric Stroke
A population-based cohort of children (aged 29 days to 15.99 years) with onset of AIS or HS between 1 July 2008 and 30 June 2009 and residing in the South of England was identified prospectively. Multiple sources were used to ascertain cases as described in a previous report. The case definition was a child with acute neurological symptoms that were shown on brain imaging to be secondary to acute focal cerebral infarction or haemorrhage in a vascular distribution. HS included non-traumatic intracerebral haemorrhage and subarachnoid haemorrhage but not subdural or extradural haemorrhage. Children with recurrent stroke were excluded from this analysis of diagnostic delay.
Medical records including case notes, electronic admission and discharge databases, and radiological logs were reviewed to confirm eligibility and to establish the timing of onset of symptoms, presentation to hospital, first neuroimaging, and first neuroimaging diagnostic of stroke. Four time intervals were calculated for each case (see figure 1):
(Enlarge Image)
Figure 1.
Diagrammatic representation of time intervals calculated.
It was not possible to ascertain accurately all four time intervals for all cases. Cases were excluded from analysis of diagnostic delays if none of the time intervals could be ascertained. Children with stroke that occurred while already in hospital were excluded from analysis of pre-hospital delays. For children with in-hospital strokes, the time of initial recognition of the symptoms of stroke was considered to be the time of 'arrival in hospital' for the purposes of analysing in-hospital delays. Time of arrival in hospital from 08:00 to 17:59 on a weekday (excluding bank holidays) was regarded as within fully staffed working hours. All children were treated in the UK's National Health Service (NHS). The NHS is publicly funded with universal coverage that is free at the point of use. All acute NHS hospitals within the study area participated in the study. The time of arrival in hospital was recorded at the first hospital to assess the children, although many children were subsequently transferred to other hospitals for access to specialist neuroimaging or paediatric neurology care.
The presenting features were categorised according to a scheme previously published by the International Paediatric Stroke Study (IPSS), that is, focal signs (hemiparesis, facial weakness, aphasia, dysarthria, gaze palsy, visual field defect, pupillary defects, ataxia and other cerebellar signs, and sensory deficits), diffuse signs (reduced level of consciousness, irritability, headache, vomiting and papilloedema) and seizures (both focal and generalised).
Quantitative variables are reported as medians with IQRs if they were not normally distributed. Times were compared between groups using the Wilcoxon rank-sum test or Kruskal-Wallis one-way analysis of variance where appropriate. The χ test was used to compare categorical data. p Values <0.05 were considered statistically significant. Statistical analyses were performed using Stata IC V.11.2 (Statacorp, College Station, Texas, USA).
Methods
A population-based cohort of children (aged 29 days to 15.99 years) with onset of AIS or HS between 1 July 2008 and 30 June 2009 and residing in the South of England was identified prospectively. Multiple sources were used to ascertain cases as described in a previous report. The case definition was a child with acute neurological symptoms that were shown on brain imaging to be secondary to acute focal cerebral infarction or haemorrhage in a vascular distribution. HS included non-traumatic intracerebral haemorrhage and subarachnoid haemorrhage but not subdural or extradural haemorrhage. Children with recurrent stroke were excluded from this analysis of diagnostic delay.
Medical records including case notes, electronic admission and discharge databases, and radiological logs were reviewed to confirm eligibility and to establish the timing of onset of symptoms, presentation to hospital, first neuroimaging, and first neuroimaging diagnostic of stroke. Four time intervals were calculated for each case (see figure 1):
(Enlarge Image)
Figure 1.
Diagrammatic representation of time intervals calculated.
Time from onset of symptoms to arrival at hospital.
Time from arrival at hospital to first neuroimaging.
Time from arrival at hospital to first neuroimaging diagnostic of stroke.
Time from onset of symptoms to first neuroimaging diagnostic of stroke.
It was not possible to ascertain accurately all four time intervals for all cases. Cases were excluded from analysis of diagnostic delays if none of the time intervals could be ascertained. Children with stroke that occurred while already in hospital were excluded from analysis of pre-hospital delays. For children with in-hospital strokes, the time of initial recognition of the symptoms of stroke was considered to be the time of 'arrival in hospital' for the purposes of analysing in-hospital delays. Time of arrival in hospital from 08:00 to 17:59 on a weekday (excluding bank holidays) was regarded as within fully staffed working hours. All children were treated in the UK's National Health Service (NHS). The NHS is publicly funded with universal coverage that is free at the point of use. All acute NHS hospitals within the study area participated in the study. The time of arrival in hospital was recorded at the first hospital to assess the children, although many children were subsequently transferred to other hospitals for access to specialist neuroimaging or paediatric neurology care.
The presenting features were categorised according to a scheme previously published by the International Paediatric Stroke Study (IPSS), that is, focal signs (hemiparesis, facial weakness, aphasia, dysarthria, gaze palsy, visual field defect, pupillary defects, ataxia and other cerebellar signs, and sensory deficits), diffuse signs (reduced level of consciousness, irritability, headache, vomiting and papilloedema) and seizures (both focal and generalised).
Quantitative variables are reported as medians with IQRs if they were not normally distributed. Times were compared between groups using the Wilcoxon rank-sum test or Kruskal-Wallis one-way analysis of variance where appropriate. The χ test was used to compare categorical data. p Values <0.05 were considered statistically significant. Statistical analyses were performed using Stata IC V.11.2 (Statacorp, College Station, Texas, USA).