Louise Yeung

Timing is Everything: Delayed Intubation is Associated with Increased Mortality in Initially Stable Trauma Patients

BACKGROUND The indications for immediate intubation in trauma are not controversial, but some patients who initially appear stable later deteriorate and require intubation. We postulated that initially stable, moderately injured trauma patients who experienced delayed intubation have higher mortality than those intubated earlier. METHODS Medical records of trauma patients intubated within 3 h of arrival in the emergency department at our university-based trauma center were reviewed. Moderately injured patients were defined as an ISS < 20. Early intubation was defined as patients intubated from 10-24 min of arrival. Delayed intubation was defined as patients intubated ≥25 min after arrival. Patients requiring immediate intubation, within 10 min of arrival, were excluded. RESULTS From February 2006 to December 2007, 279 trauma patients were intubated in the emergency department. In moderately injured patients, mortality was higher with delayed intubation than with early intubation, 11.8% versus 1.8% (P = 0.045). Patients with delayed intubations had greater frequency of rib fractures than their early intubation counterparts, 23.5% versus 3.6% (P = 0.004). Patients in the delayed intubation group had lower rates of cervical gunshot wounds than the early intubation group, 0% versus 10.7% (P = 0.048) and a trend toward fewer of skull fractures 2.9% versus 16.1%, (P = 0.054). CONCLUSIONS These findings suggest that delayed intubation is associated with increased mortality in moderately injured patients who are initially stable but later require intubation and can be predicted by the presence of rib fractures.

Emergency uncrossmatched transfusion effect on blood type alloantibodies.

Background: Trauma patients receive emergency transfusions of unmatched Type O Rh-negative (Rh-) blood until matched blood is available. We hypothesized that patients given uncrossmatched blood may develop alloantibodies, placing them at risk for hemolytic transfusion reactions (HTRs). Methods: Data regarding alloantibody profiles and HTR occurrence were collected from the records of trauma patients at our university-based trauma center who received emergency uncrossmatched blood from July 2008 to August 2010. Results: A total of 132 patients received 1,570 units of packed red blood cells. Mean injury severity score was 28 ± 1.3. Forty-five (34%) patients died: 27 on hospital day 1; the remaining 18 had no evidence of HTR before death. Four Rh- female patients received Rh+ fresh frozen plasma, but none received Rh+ packed red blood cells. Three Rh- male patients received both Rh+ packed red blood cells and fresh frozen plasma, and one received Rh+ fresh frozen plasma. One patient developed anti-Rh D antibodies. None experienced HTR. One female patient had HTR from reactivation of anamnestic JK antibodies. Thirteen (33%) of 39 patients met criteria for HTR based on urinalysis and 29 (40%) of 72 patients tested met criteria for HTR based on hemoglobin and bilirubin values. Only one patient had confirmed HTR. Conclusion: High rates of injury recidivism in trauma patients increase the likelihood of multiple blood transfusions during their lifetime. Rh- patients who receive Rh+ blood are at risk of developing anti-Rh antibodies, putting them at risk for HTR. The conservation of Rh- blood for use in female patients may be detrimental to Rh- male patients. Laboratory diagnostic criteria for HTR are nonspecific in the trauma population and should be used with caution. Level of Evidence: II, therapeutic review.

Effect of Surgery Resident Change of Shift on Trauma Resuscitations and Outcomes

INTRODUCTION The ability of surgery residents to provide continuity of care has come under scrutiny with work hour restrictions. The impact of the surgery resident sign-out period (6-8am and 6-8pm) on trauma outcomes remains unknown. We hypothesize that during shift change, resuscitation times are prolonged with worse outcomes. METHODS Records of patients treated at a university-based urban trauma center during 2008 and 2009 were reviewed. Patients were separated into a shift change group (6-8am and 6-8pm) and a control group of all other time periods and compared using ANOVA, chi square, and unpaired t-tests. RESULTS We reviewed the charts of 4361 consecutive trauma patients. There was no difference in gender, acuity, resuscitation times, Glasgow Coma Scale, revised trauma score, injury severity score (ISS), or probability of survival score between patients arriving during shift change compared to other times (p>0.2). There was no difference in total emergency department time for patients arriving during shift change (p = 0.07), even when stratified by ISS (ISS<15, p = 0.09; ISS>15, p = 0.2). Length of stay was increased for patients arriving during shift change compared to other times (5 vs 4 days, p<0.05). This was more pronounced for those with ISS>15 (16 vs 11 days, p = 0.03); however, there was no impact on intensive care unit length of stay, ventilator days, and mortality (p>0.3) regardless of ISS. CONCLUSIONS Trauma outcomes are generally unaffected by patient arrival during shift change when resident sign-outs occur. Although adaptations are being made to accommodate trauma patient arrival during these times, we need to continue paying close attention, especially to seriously injured patients, to ensure that there are no delays in care that may potentially affect patient outcomes.

Bedside thoracic ultrasonography of the fourth intercostal space reliably determines safe removal of tube thoracostomy after traumatic injury.

BACKGROUND Thoracic ultrasonography is more sensitive than chest radiography (CXR) in detecting pneumothorax; however, the role of ultrasonography to determine resolution of pneumothorax after thoracostomy tube placement for traumatic injury remains unclear. We hypothesized that ultrasonography can be used to determine pneumothorax resolution and facilitate efficient thoracostomy tube removal. We sought to compare the ability of thoracic ultrasonography at the second through fifth intercostal space (ICS) to detect pneumothorax with that of CXR and determine which ICS maximizes the positive and negative predictive value of thoracic ultrasonography for detecting clinically relevant pneumothorax resolution. METHODS A prospective, blinded clinical study of trauma patients requiring tube thoracostomy placement was performed at a university-based urban trauma center. A surgeon performed daily thoracic ultrasonographies consisting of midclavicular lung evaluation for pleural sliding in ICS 2 through 5. Ultrasonography findings were compared with findings on concurrently obtained portable CXR. RESULTS Of the patients, 33 underwent 119 ultrasonographies, 109 of which had concomitant portable CXR results for comparison. Ultrasonography of ICS 4 or 5 was better than ICS 2 and 3 at detecting a pneumothorax, with a positive predictive value of 100% and a negative predictive value of 92%. The positive and negative predictive values for ICS 2 were 46% and 93% and for ICS 3 were 63% and 92%, respectively. CONCLUSION Bedside, surgeon-performed, thoracic ultrasonography of ICS 4 for pneumothorax can safely and efficiently determine clinical resolution of traumatic pneumothorax and aid in the timely removal of thoracostomy tubes. LEVEL OF EVIDENCE Diagnostic study, level II.

Does gastric volume in trauma patients identify a population at risk for developing pneumonia and poor outcomes

BACKGROUND Trauma patients may have full stomachs or impaired airway reflexes that place them at risk for aspiration and pneumonia. Our hypothesis was that trauma patients with larger gastric volumes as measured by abdominal computed tomography (CT) at admission have higher rates of pneumonia and worse outcomes. METHODS We matched an initial cohort of 81 trauma patients with an admission CT of the abdomen and a diagnosis of pneumonia by Injury Severity Score and Abbreviated Injury Score of the head and chest with a control group of 81 trauma patients without pneumonia. We estimated gastric volumes on CT and compared variables using chi-square, t-tests, receiver operating curve analysis, and regression analysis. RESULTS Patients with pneumonia had larger gastric volumes than those without pneumonia (879 cm(3)versus 704 cm(3); P = 0.04). Receiver operating curve analysis gave a gastric volume threshold value of 700 cm(3) as a predictor of pneumonia. Patients with a gastric volume ≥ 700 cm(3) had more pneumonia (61% versus 41%; P = 0.01), stayed longer in the hospital (27.6 versus 19.7 d; P < 0.05) and the intensive care unit (18.4 versus 12.5 d; P = 0.01), required more days on the ventilator (18.1 versus 12.0 d; P = 0.02), and had a trend toward increased mortality (17% versus 11%; P = 0.2). On multivariate analysis, nasogastric or orogastric tube (odds ratio 3.0; P = 0.004) and gastric volume >700 cm(3) (odds ratio 2.7; P = 0.004) were independent predictors of pneumonia. CONCLUSIONS Trauma patients who developed pneumonia had larger initial gastric volumes. A straightforward estimate of gastric volume on admission abdominal CT may predict patients at risk for developing pneumonia and poor outcomes. Clinicians should be especially vigilant in taking precautions against pneumonia and have a lower threshold for suspecting pneumonia in patients with abdominal CT gastric volumes ≥ 700 cm(3).