The Association Between On-Scene BP and END in SICH Patients
The Association Between On-Scene BP and END in SICH Patients
This retrospective cohort study was conducted at the emergency department (ED) of a university-affiliated medical centre that has treated 85 500 ± 4520 patients and 203 ± 42 (mean ± SD) SICH patients on average for the past 10 years, as per an annual census. The hospital is also an emergency medical technician (EMT) training centre and provides several educational programmes for personnel working in the local emergency medical system (EMS). Consecutive adult patients with a first-ever SICH referred to our ED by ambulance between 1 January 2007 and 31 December 2012 were potential subjects of this study. As per general practice, patients suspected of having a stroke were treated according to a pre-established protocol. After SICH was confirmed, patient data were recorded in the SICH databank established in January 2007 primarily for quality assurance of stroke care. Most SICH patients are admitted to the neurocritical care unit (NCU) after treatment at the ED. The management of patients with SICH followed the pre-established protocol with regard to assessment frequency, BP control, antipyretic use, airway control and oxygen supplementation, osmotic agent use, anti-epileptic use, procedures for increased intracranial pressure control, and so on. It is the general practice to check the GCS every 15 min during the first 24 h for patients with SICH. Any decline in the GCS and its associated cause must be recorded in the SICH databank. Detailed patient data for this study were retrieved from the databank and cross-checked with medical charts. Patients with ICH, apparently occurring secondary to ischaemic stroke or intracranial disorders (such as aneurysm, moyamoya disease, arteriovenous malformations, brain tumour or brain abscess), with >6 h elapsed time before ED arrival, receiving coumadin, with GCS score of <9 on ED arrival, referred from other hospitals, or with incomplete data were excluded. The study protocol was approved by our hospital's Institutional Review Board (IRB). The informed consent requirement was waived by the IRB.
END was defined as a ≥2-point decrease in GCS noted within 24 h of ED arrival. Baseline neurological status regained within 3 h of an episode of neurological worsening due to seizures or effects of medications for seizure control was not considered as END. END was determined by two emergency physicians (EPs) after they independently reviewed the medical records. The agreement level of END determination was assessed by a κ value. Final judgement of the presence or absence of END was done by a third EP after comprehensive discussion in case of any disagreement in END determination.
To test the exact relationship between on-scene BP and END in SICH patients, we adjusted age, gender, Charlson Index, aspirin use, smoking, elapsed time, consciousness level on ED arrival, haematoma size, intraventricular haemorrhage, midline shift, and haematoma location (infra- or supratentorial) using binary logistic regression analyses. The above-mentioned confounders had been reported to have association with SICH outcome in many studies. The on-scene BP indices analysed in this study were systolic BP (SBP), diastolic BP (DBP), mean arterial pressure (MAP) and pulse pressure (PP). A midline shift was defined as a lateral pineal gland shift of ≥5 mm from the falx at the level of the septum pellucidum, as observed on a CT scan. Because there is a high correlation between BP indices (SBP, DBP, MAP and PP) and BP values measured at different times (on-scene, on ED arrival, on NCU admission), we compared the correlation strength between each BP value using −2 log-likelihood value (model fitness) by adding them into the regression model individually rather than together to avoid multicollinearity. This method has been used previously in similar studies. Statistical analysis was performed using SPSS V.19.0.
Methods
This retrospective cohort study was conducted at the emergency department (ED) of a university-affiliated medical centre that has treated 85 500 ± 4520 patients and 203 ± 42 (mean ± SD) SICH patients on average for the past 10 years, as per an annual census. The hospital is also an emergency medical technician (EMT) training centre and provides several educational programmes for personnel working in the local emergency medical system (EMS). Consecutive adult patients with a first-ever SICH referred to our ED by ambulance between 1 January 2007 and 31 December 2012 were potential subjects of this study. As per general practice, patients suspected of having a stroke were treated according to a pre-established protocol. After SICH was confirmed, patient data were recorded in the SICH databank established in January 2007 primarily for quality assurance of stroke care. Most SICH patients are admitted to the neurocritical care unit (NCU) after treatment at the ED. The management of patients with SICH followed the pre-established protocol with regard to assessment frequency, BP control, antipyretic use, airway control and oxygen supplementation, osmotic agent use, anti-epileptic use, procedures for increased intracranial pressure control, and so on. It is the general practice to check the GCS every 15 min during the first 24 h for patients with SICH. Any decline in the GCS and its associated cause must be recorded in the SICH databank. Detailed patient data for this study were retrieved from the databank and cross-checked with medical charts. Patients with ICH, apparently occurring secondary to ischaemic stroke or intracranial disorders (such as aneurysm, moyamoya disease, arteriovenous malformations, brain tumour or brain abscess), with >6 h elapsed time before ED arrival, receiving coumadin, with GCS score of <9 on ED arrival, referred from other hospitals, or with incomplete data were excluded. The study protocol was approved by our hospital's Institutional Review Board (IRB). The informed consent requirement was waived by the IRB.
Definition of END
END was defined as a ≥2-point decrease in GCS noted within 24 h of ED arrival. Baseline neurological status regained within 3 h of an episode of neurological worsening due to seizures or effects of medications for seizure control was not considered as END. END was determined by two emergency physicians (EPs) after they independently reviewed the medical records. The agreement level of END determination was assessed by a κ value. Final judgement of the presence or absence of END was done by a third EP after comprehensive discussion in case of any disagreement in END determination.
Methods of Measurement
To test the exact relationship between on-scene BP and END in SICH patients, we adjusted age, gender, Charlson Index, aspirin use, smoking, elapsed time, consciousness level on ED arrival, haematoma size, intraventricular haemorrhage, midline shift, and haematoma location (infra- or supratentorial) using binary logistic regression analyses. The above-mentioned confounders had been reported to have association with SICH outcome in many studies. The on-scene BP indices analysed in this study were systolic BP (SBP), diastolic BP (DBP), mean arterial pressure (MAP) and pulse pressure (PP). A midline shift was defined as a lateral pineal gland shift of ≥5 mm from the falx at the level of the septum pellucidum, as observed on a CT scan. Because there is a high correlation between BP indices (SBP, DBP, MAP and PP) and BP values measured at different times (on-scene, on ED arrival, on NCU admission), we compared the correlation strength between each BP value using −2 log-likelihood value (model fitness) by adding them into the regression model individually rather than together to avoid multicollinearity. This method has been used previously in similar studies. Statistical analysis was performed using SPSS V.19.0.
