الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 153 |  |  |
| | | from | 153 | | |
Abstract
Traumatic brain injury (TBI) constitutes a major health and socioeconomic problem throughout the world. It is a leading cause of death and disability worldwide, and responsible of several millions of ED visits, hospitalizations, and deaths every year with common lifelong disability among survivors. It is prevalent in both low- and high-income countries and affects people of all ages.
Aside from the physical impact, TBI has wide-ranging psychological and socio-economic consequences on survivors, including behavioral, mental and social problems that have an extensive impact on individuals, families and communities. Health effects associated with TBI can be broadly categorized into cognitive, behavioral/emotional, motor, and somatic symptoms.
TBI can be classified in numerous ways, including type, severity, location, mechanism of injury, and physiological response to injury. Its classification into primary impact and secondary cerebral damage (e.g., hypoxia and hypotension) is the commonly used way of classification that affects management and neurological outcome after TBI. Immediately after TBI, cerebral blood flow (CBF) is extremely low and near the ischemic threshold. Prevention and treatment of such secondary injuries are the focus of modern TBI management. In this context, the maintenance of adequate cerebral blood flow (CBF) is critical. However, measuring the real-time CBF changes during therapeutic intervention at the bedside is challenging, and cannot be achieved routinely using imaging techniques such as perfusion magnetic resonance imaging (MRI) and perfusion computed tomography (CT) scan.
Transcranial Doppler (TCD) is a noninvasive study used to measure real-time cerebral blood flow velocity (CBF-V) in the major intracranial arteries. It involves use of low-frequency US waves to insonate the basal cerebral arteries through relatively thin bone windows. TCD allows dynamic monitoring of CBF-V and vessel pulsatility, over extended time periods with high temporal resolution. It is relatively inexpensive, repeatable, and its portability offers increased convenience over other imaging modalities, which is particularly useful in emergency settings.
Over the past 30 years significant efforts have been made to improve the early management of patients with severe traumatic brain injury (GCS score <9). However, the vast majority of TBIs (more than 80%) are mild to moderate TBI, that is, Glasgow Coma Scale (GCS) between 9 and 15. Despite their reassuring presentation, 5 to 20% of these patients will develop secondary neurologic deterioration (SND) within the first post-traumatic week. In spite of that detection of those at risk for early SND after mild to moderate TBI on admission has received relatively little attention. Neurological worsening has several causes including cerebral edema, growth of intracerebral hematoma, seizures, or posttraumatic hydrocephalus.
This is an observational prospective cohort study which was carried out on 120 adult patients of both sexes who fulfilled the inclusion and exclusion criteria and were consecutively admitted to the Emergency Department in Alexandria main university hospital. The aim of the study was to assess the accuracy of early use of Transcranial Doppler in predicting neurologic outcome in patients with mild and moderate traumatic brain injury.
Enrolled patients were assessed by the standard phases of evaluation and management for trauma according to advanced trauma life support guidelines namely; the ABCDE approach of primary survey and resuscitation, secondary survey including; detailed history, head to toe physical examination, laboratory investigation and imaging studies as required, and finally definitive care. GCS score was assessed on admission and monitored throughout the study, ISS was also determined. Computerized Tomography (CT) scans of the head on admission and follow up after 48 hours were done and classified according to Rotterdam and Marshal CT classification models.
TCD ultrasonography was performed in the ED immediately after hemodynamic and respiratory stabilization within the first 8 hours after trauma and repeated after 6 hours. The 2MHz ultrasound transducer was placed over the temporal area. Tracings of right and left middle cerebral arteries were done and by applying pulsed wave on the insonated segment, the peak systolic (PSV), end diastolic (EDV) velocities were measured and mean flow velocity (MFV) was calculated using the formula: (PSV+ (EDVx 2)/3. Pulsatility index (PI), provides information on downstream cerebral vascular resistance, was also calculated using the formula: (PSV-EDV)/ MFV.
The outcome measures were; secondary neurological deterioration (SND) at day 7 after trauma, Glasgow outcome score extended (GOS-E) at 28 days after trauma either by clinical assessment or by structured telephone interview, in-hospital mortality and hospital length of stay.
The demographic data of this study showed higher percentage of males (81 patients representing 67.5% of cases) over females (39 patients representing 32.5%). Age ranged between 18 to 55 years with a mean age of 33.77 ± 11.79 years and median value of 32. Regarding the mode of trauma, road traffic accidents was responsible for TBI in 48 cases (40%), falls in 36 cases (30%), alleged assault in 23 cases (19.2%).
Considering SND at day 7 as primary outcome, studied patients were classified into two primary outcome groups, who were developed SND; 84 cases (70%) and who didn’t developed SND; 36 cases (30%), at day 7 after trauma. There was no significant difference between the two groups as regards age, gender, mode of trauma, time from trauma to TCD study, vital signs and laboratory investigations. While there was significant difference between the two groups as regards time from trauma to admission, airway status, signs of hypoperfusion as well as O2 saturation and pupillary examination on admission. Also patients who developed SND had significantly lower GCS score on admission and after 48 hours, and significantly higher Rotterdam and Marshal CT scores on admission and on follow up CT after 48 hours as well as significantly higher Injury Severity Score (p value of <0.001 for all mentioned scores).
Based on TCD measurements recorded within 8 hours after trauma, patients with SND had significantly lower ED and MF velocities (p value <0.001 and 0.022 respectively), and higher PI with mean of 1.36 ± 0.21 and median of 1.41 in SND group versus mean of 1.07 ± 0.17 and median of 1.06 in patients without SND (p value of <0.001). However PSV showed no significant difference. Of all the TCD variables measured after 6 hours from the 1st study, only pulsatility index showed significant difference between the two outcome groups (p value of < 0.001). EDV, MFV and pulsatility index, when measured early, were independent predictors of SND at day 7 in both univariate and multivariate analyses along with Marshal CT score.
When assessing the AU-ROC using PI it was 0.837 and showed sensitivity of 72.22%, specificity of 91.67%, PPV of 78.8% and NPV of 88.5% with best cut off value of >1.28 in predicting patients with SND at day 7 after trauma, it was interesting to know that it was higher than that of GCS (0.752) and ISS (0.689) and as higher as CT scores agreements (0.826 and 0.885 for Rotterdam and Marshal CT scores respectively) in predicting patients at risk of SND. Similar to PI, EDV was found to be an independent factor for predicting SND with an AU-ROC of 0.759 (with best cut off value of ≤28) which was higher than that of ISS and almost equal to that of GCS and very close to that of PI and CT scores.
As regards GOS-E at day 28 after trauma, patients were dichotomized into favorable 90 cases (75%) and unfavorable 30 cases (25%) outcome groups. There was strong association between EDV and PI with GOS-E, patients who had unfavorable outcomes (GOSE=1-4) had significantly lower EDV and significantly higher PI in comparison with those who had favorable outcomes (GOSE=5-8) at day 28 after injury with p value < 0.001 for both variables. Of all the TCD variables, pulsatility index remained independent predictor of 28 days Glasgow outcome score extended in multivariate regression analysis with p value of 0.023. It showed sensitivity of 70%, specificity of 87.78%, PPV of 65.6% and NPV of 89.8% with best cut off value of >1.32 in predicting patients with unfavorable outcome at day 28 after trauma with AU-ROC of 0.805.
In correlation to hospital length of stay, EDV showed moderate negative correlation (r: -0.226, p value 0.013) while PI showed strong positive correlation (r: 0.430, p value: <0.001). Among factors appeared to be significant in predicting hospital LOS in univariate analysis, only PI out of TCD measurements appeared to be independent predictor in multivariate one with p value of 0.007, along with GCS on admission (p value <0.001) and Rotterdam CT score (p value of 0.034).
Again, PI showed moderate negative correlation with GCS score on admission, moderate positive correlation with ISS and Rotterdam CT score, and strong positive correlation with Marshal CT score, that means PI might be correlated indirectly to severity and prognosis of TBI.
As regards in-hospital mortality only 6 cases died (5%), of all TCD measurements only PI showed significant difference between died and survived groups which was higher in the dead cases with p value < 0.001. Also it was significant predictor of in-hospital mortality in univariate analysis. However this effect disappeared in multivariate analysis. AU-ROC for In-hospital mortality prediction using PI was 0.934 (p value < 0.001) with sensitivity of 83.33%, specificity of 80.7%, PPV of 18.5% and NPV of 98.9% with best cut off value of 1.36 in predicting in-hospital mortality.
Accordingly TCD measurements obtained early upon admission could be useful predictors for neurological outcomes in mild and moderate TBI and might be of value if added in the multimodality monitoring of TBI not only for severe but also in mild and moderate cases.