Janice H.H. Yeung

Prevalence of survivor bias in observational studies on fresh frozen plasma: erythrocyte ratios in trauma requiring massive transfusion

Observational studies on transfusion in trauma comparing high versus low plasma:erythrocyte ratio were prone to survivor bias because plasma administration typically started later than erythrocytes. Therefore, early deaths were categorized in the low plasma:erythrocyte group, whereas early survivors had a higher chance of receiving a higher ratio. When early deaths were excluded, however, a bias against higher ratio can be created. Survivor bias could be reduced by performing before-and-after studies or treating the plasma:erythrocyte ratio as a time-dependent covariate. We reviewed 26 studies on blood ratios in trauma. Fifteen of the studies were survivor bias-unlikely or biased against higher ratio; among them, 10 showed an association between higher ratio and improved survival, and five did not. Eleven studies that were judged survivor bias-prone favoring higher ratio suggested that a higher ratio was superior. Without randomized controlled trials controlling for survivor bias, the current available evidence supporting higher plasma:erythrocyte resuscitation is inconclusive.

Early risk stratification of patients with major trauma requiring massive blood transfusion

BACKGROUND There is limited evidence to guide the recognition of patients with massive, uncontrolled hemorrhage who require initiation of a massive transfusion (MT) protocol. OBJECTIVE To risk stratify patients with major trauma and to predict need for MT. DESIGNS Retrospective analysis of an administrative trauma database of major trauma patients. A REGIONAL TRAUMA CENTRE: A regional trauma centres in Hong Kong. PATIENTS Patients with Injury Severity Score ≥ 9 and age ≥ 12 years were included. Burn patients, patients with known severe anemia and renal failure, or died within 24h were excluded. MAIN OUTCOME MEASURES Delivery of ≥ 10 units of packed red blood cells (RBC) within 24h. RESULTS Between 01/01/2001 and 30/06/2009, 1891 patients met the inclusion criteria. 92 patients required ≥ 10 units RBC within 24h. Seven variables which were easy to be measured in the ED and significantly predicted the need for MT are heart rate ≥ 120/min; systolic blood pressure ≤ 90 mm Hg; Glasgow coma scale ≤ 8; displaced pelvic fracture; CT scan or FAST positive for fluid; base deficit >5 mmol/L; hemoglobin ≤ 7 g/dL; and hemoglobin 7.1-10 g/dL. At a cut off of ≥ 6, the overall correct classification for predicting need for MT was 96.9%, with a sensitivity of 31.5% and specificity of 99.7%, and an incidence of MT of 82.9%. The area under the curve was 0.889. CONCLUSION A prediction rule for determining an increased likelihood for the need for massive transfusion has been derived. This needs validation in an independent data set.

Trauma Care Systems

Background:Despite the high incidence of major trauma, few studies have directly compared the performance of trauma systems. This study compared the trauma system performance in Victoria, Australia, (VIC) and Hong Kong, China (HK). Methods:Prospectively collected data over 5 years from January 2001 from the 2 trauma systems were compared using univariate analysis. Variables were then entered into a multivariate logistic regression to assess differences in outcome between the systems and adjusted for effects of clinically important factors. Results:Five thousand five thirty-six cases from VIC and 580 cases from HK were taken for analysis. The HK group was older, but mechanisms of injury were similar in both systems. Thoracic and abdominal trauma was more common in VIC, compared with more head injuries in HK. More patients were admitted to intensive care in VIC and patients stayed in intensive care 1 day longer on average, despite more comorbidity in HK patients. Overall mortality was 20.2% for HK and 11.9% for VIC (X21 = 32.223, P < 0.001). Conclusion:The performance of the HK trauma system was comparable to international standards, but there was a significant difference in the probability of survival of major trauma between the 2 systems. Possible modifiable factors may include criteria for activation of trauma calls and improved ICU utilization.

Simulation of survivorship bias in observational studies on plasma to red blood cell ratios in massive transfusion for trauma

Observational studies on injured patients requiring massive transfusion have found a survival advantage associated with use of equivalent number of units of fresh frozen plasma (FFP) and packed red blood cells (RBCs) compared with use of FFP based on conventional guidelines. However, a survivorship bias might have favoured the higher use of FFP because patients who died early never had the chance to receive sufficient FFP to match the number of RBC units transfused.

High risk trauma in older adults in Hong Kong: A multicentre study

BACKGROUND Trauma is the eighth leading cause of death in Hong Kong. In 2002, 18.5% of the population of Hong Kong was aged 55 years or above, which increased to 22.1% in 2006. The increasing older population in Hong Kong presents a challenge to the health care system yet there is little local data on older trauma patients. The objectives of this study are firstly to describe the epidemiology of high risk trauma in older patients in Hong Kong, and secondly to identify predictors of trauma mortality. METHOD Retrospective analysis of prospectively collected data from a centralised trauma database; data collected from 2002 to 2004 from four trauma centres in Hong Kong. RESULTS Between 2002 and 2004, the four trauma centres had a total of 2,124,175 emergency department attendances of which 376,021 (17.7%) were trauma patients, and 80,827 (3.8%) were aged 55 years or older. 810 injured older patients met the inclusion criteria for this study. 380 (46.9%) patients had co-morbidity at the time of injury. Common causes of injury were falls (50.0%, 405/810) and motor vehicle crashes (33.6%, 272/810) of which (77.2%, 210/272) were pedestrians. Mortality was 24.4% (198/810) and increased with advancing age (p<0.0001). 53.5% (433/810) of patients had major trauma (ISS>15). Head injury contributed to 80.3% (159/198) of deaths. 38.4% (311/810) of patients required operations. Most patients were discharged home (40.5%, 328/810) and one-third (270/810) required rehabilitation. Significant predictors of mortality included co-morbidity, injury severity score, age and decreasing Glasgow Coma Score. CONCLUSION Pedestrians struck by motor vehicles and falls are the principal causes of trauma in older patients in Hong Kong. Mortality increased with advancing age. The independent indicators of trauma mortality in older patients are co-morbidity, age, ISS and GCS.

Protection from traumatic brain injury in hormonally active women vs men of a similar age: a retrospective international study

BACKGROUND It has been suggested that women with traumatic brain injury have more favorable outcomes than do men because of higher levels of circulating estrogen and progesterone that may reduce brain edema. OBJECTIVES To determine whether there is any association between sex and mortality in TBI patients and whether there is any association between sex and brain edema. DESIGN Retrospective cohort study using data from 2001 to 2007 collected from a trauma registry in Hong Kong and the Victorian State Trauma Registry. SETTING Two regional trauma centers in Hong Kong and 2 adult major trauma centers and 1 pediatric trauma center in Victoria, Australia. MAIN OUTCOME MEASURES Mortality and brain edema. PATIENTS Trauma patients with an Abbreviated Injury Scale score (head) of at least 3 who were aged 12 to 45 years were included. Patients with minor head injury and undisplaced closed skull fracture were excluded. RESULTS Both the Hong Kong and Victorian data showed no significant difference in sex-related mortality. Increased mortality was associated with decreased systolic blood pressure and Glasgow Coma Scale score and with increased New Injury Severity Score or Injury Severity Score. In Hong Kong, brain edema was associated with female sex (P = .02), and the odds of brain edema in females were greater than for males. However, this association was not found in Victorian patients. CONCLUSION This study found no significant association between sex and mortality in either Victoria or Hong Kong and does not support the concept that females have better outcomes after traumatic brain injury.

Neurological outcome in patients with traumatic brain injury and its relationship with computed tomography patterns of traumatic subarachnoid hemorrhage

OBJECT Traumatic subarachnoid hemorrhage (SAH) is a poor prognostic factor for traumatic brain injury. The authors aimed to further investigate neurological outcome among head injury patients by examining the prognostic values of CT patterns of traumatic SAH, in particular, the thickness and distribution. METHODS The study was conducted using a database in a regional trauma center in Hong Kong. Data had been prospectively collected in consecutive trauma patients between January 2006 and December 2008. Patients included in the study had significant head injury (as defined by a head Abbreviated Injury Scale [AIS] score of 2 or more) with traumatic SAH according to admission CT. RESULTS Over the 36-month period, 661 patients with significant head injury were admitted to the Prince of Wales Hospital in Hong Kong. Two hundred fourteen patients (32%) had traumatic SAH on admission CT. The mortality rate was significantly greater and a 6-month unfavorable outcome was significantly more frequent in patients with traumatic SAH. Multivariate analysis showed that the maximum thickness (mm) of traumatic SAH was independently associated with neurological outcome (OR 0.8, 95% CI 0.7-0.9) and death (OR 1.3, 95% CI 1.2-1.5) but not with the extent or location of hemorrhage. CONCLUSIONS Maximum thickness of traumatic SAH was a strong independent prognostic factor for death and clinical outcome. Anatomical distribution per se did not affect clinical outcome.

Timing of tracheal intubation in traumatic cardiac tamponade: A word of caution

Airway, breathing, and circulation are top priorities in any resuscitation. However, in cardiac tamponade, the decision to intubate the trachea and initiate positive pressure ventilation (PPV) should only be taken after consideration of the deleterious haemodynamic effects of positive intrathoracic pressure. We suggest that the threshold for intubation and PPV should be raised in tamponade and that intubation and PPV should, if possible, be timed so that relief of tamponade can immediately follow. In the trauma setting, emergency thoracotomy is the best approach. When intubation is unavoidable because of very low oxygen saturation or cardiac arrest, high ventilatory pressures should be avoided.

Impact on trauma patient management of installing a computed tomography scanner in the emergency department

BACKGROUND Computed tomography (CT) plays a central diagnostic role for trauma patients. A 16-slice multi-detector CT scanner was installed in the emergency department (ED) of Prince of Wales Hospital in December 2004. The aims of this study were to evaluate the impact of the CT scanner within the ED on trauma management and to compare the utilisation patterns of trauma CT before and after the introduction of EDCT. METHODS Analysis of prospectively collected trauma registry data. All consecutive trauma cases admitted through the ED that underwent CT between June 2004 and June 2005 (6 months before and after EDCT installation) were included. A positive CT was defined as the identification (by a specialist radiologist) of a significant finding which was consistent with injury. RESULTS There were 226 and 202 trauma patients in the 6 months before and after EDCT installation, respectively. 111 (49.1%) patients underwent CT scanning before EDCT compared with 110 (54.5%) afterwards. 72 (65%) patients had CT scans performed before admission to definitive care compared with 99 (90%) after EDCT installed (p<0.0001, chi(2) test). Mean time from arrival to first CT was shorter after EDCT (102 min vs. 197 min, p=0.011). Mean trauma room length of stay increased after EDCT was implemented (106 min vs. 80 min; p<0.001). Median time to urgent operation (<6h) was less with EDCT (134 min before vs. 112 min after). No changes in median time to neurosurgical operation (138 min before vs. 148 min after); mean length of stay (12.8 days before vs. 12.5 days after); or mortality (8 patients before vs. 7 patients after). There were 203 scans (1.8/patient) done before EDCT compared with 226 scans (2.5/patient) after. There was no difference in the number of scans done by body region or the proportion of positive scans (32% before vs. 30% after). Logistic regression confirmed that after adjusting for injury severity and admission physiology, time to first CT was shorter (p=0.0307) but ED length of stay was increased (p<0.0001). CONCLUSION After the installation of EDCT, more trauma patients had CT scanning before definitive care, and scans were done sooner, with no significant increase in the number of unnecessary scans.

Fresh-frozen plasma transfusion strategy in trauma with massive and ongoing bleeding. Common (sense) and sensibility

During trauma resuscitation involving massive transfusion, the best fresh-frozen plasma to packed red blood cells ratio is unknown. No randomised controlled trial (RCT) is available on this subject, although there are plenty of observational studies suggesting that the ratio should be about 1:1. This ratio also makes more physiological sense, and we suggest that in patients with massive and ongoing bleeding, it is a sensible strategy with which to start resuscitation.