An appraisal of trauma in the elderly

To review the trends of trauma in the elderly experienced at our trauma center compared with other Level I trauma centers. This was a retrospective trauma registry analysis (1996-2003) of 2783 blunt trauma in elderly (BTE) and 4568 adult (BTA) patients in a Level 1 trauma center. Falls and motor vehicular crashes were the most common mechanisms noted in 47 per cent and 31 per cent (84% and 13% in BTE, 25% and 42% in BTA). BTE were sicker, with higher Injury Severity Scores (ISS), lengths of stay, and mortality (5% vs 2%, P value < 0.05). ISS was 5.2-fold higher in nonsurvivors to survivors in BTA and 2.4-fold in BTE. Elevation in ISS resulted in higher linear increase in mortality in BTE (vs BTA) at any ISS level. Mortality in patients with ISS > or = 25 was 43.5 per cent vs 23.8 per cent. ISS > or = 50 had 31 per cent adult survivors but no elderly survivors. Among isolated injuries, head trauma in the elderly carried the highest mortality, at 12 per cent (19% in patients with an Abbreviated Injury Score > or = 3). Abdominal injuries were the most lethal (18.3% and 41.2% in patients with an Abbreviated Injury Score > or = 3) in multiple trauma victims (41% vs 18% in isolated trauma). There was 4.4-fold increased mortality in the presence of thoracic trauma. Combined head, chest, and abdominal trauma carried the worst prognosis. Thirty-four per cent of BTE and 88 per cent of BTA patients were discharged home. Elderly patients need more aggressive therapy, as they are sicker with higher mortality.     

Pattern and severity of injury sustained by pedestrians in road traffic accidents with particular reference to the effect of alcohol

The injuries sustained by 102 pedestrians consecutively admitted to the Accident and Emergency Department of the Royal Infirmary of Edinburgh were prospectively recorded. Injuries were categorized by body regions and the severity assessed by the ISS/AIS system. Pedestrian injuries were found to comprise 0.8 per cent of the department workload and 28 per cent of RTA injuries. The commonest sites of injury were the right leg and head. Injuries to the right side of the body were almost three times more frequent than those to the left. Patients considered to be under the influence of alcohol had more severe and widespread injuries, were more likely to have facial injuries, and were more likely to be detained as inpatients. Males were more commonly injured than females and the predominant age group was 20-29 years. The group affected by alcohol was predominantly male and young, whereas the group unaffected by alcohol included a considerable number of elderly people. The possible reasons for the pattern of injuries are discussed and the contribution of alcohol to the morbidity suffered by pedestrians is stressed.     

Recent trends in the management of combined pancreatoduodenal injuries

In an effort to better characterize the natural history of pancreatoduodenal injuries, we present a review of clinical experiences in the treatment of combined traumatic pancreatoduodenal injuries, focusing on patients in extremis. Records of patients with abdominal trauma admitted to a level 1 trauma center from 1997 to 2001 were reviewed. Of 240 patients who sustained a pancreatic or duodenal injury, 33 had combined pancreatoduodenal injuries. Eighty-two per cent of the patients (27/33) in this series had penetrating injuries, 72 per cent (24) sustained gunshot wounds (GSW). Thirty-one patients were male, and the mean age was 33 years (range, 7-74). These patients presented with an average Injury Severity Score (ISS) of 22 +/- 12 and an average Glasgow Coma Score of 14 +/- 2. Overall length of stay was 39 +/- 59 days (range, 0-351 days). These 33 patients underwent a total of 57 laparotomies with an average of 1.7 operations per patient (range, 1 to 5 operations). Eighty-four per cent of the patients had an associated gastrointestinal injury and 45 per cent had a major vascular injury. Thirteen of the 33 (39%) patients presented in extremis, all 13 underwent an abbreviated laparotomy. The complication rate was 36 per cent, including fistula, abscess, pancreatitis, and organ dysfunction. There were 6 hospital deaths for a mortality rate of 18 per cent. Pancreatoduodenal injuries are associated with a variety of other serious injuries, which add to the overall complexity of these patients. Abbreviated laparotomy may be helpful when managing combined pancreatoduodenal injuries in patients who are in extremis.     

Trauma in cirrhotics: survival and hospital sequelae in patients requiring abdominal exploration

Hepatic cirrhosis significantly increases the mortality and morbidity of elective surgery; therefore we hypothesized that cirrhosis would adversely impact outcome after abdominal trauma. We used the trauma registry to identify 17 patients with cirrhosis who sustained trauma injuries requiring emergent exploratory laparotomy. Patients were characterized with respect to age, sex, hospital days, intensive care unit days, and trauma scores. A control group (n = 73) was constructed from the registry by matching age, sex, Injury Severity Score (ISS) and Abbreviated Injury score. Mortality rates were compared by Fisher's exact test and age, ISS, Revised Trauma Score 2, and hospital and intensive care unit days were compared by Student's t test. Despite similar ISS between cirrhotic patients and controls, patients with cirrhosis had a fourfold increase in mortality (mortality odds ratio = 7.2; 95% confidence interval = 2.2-24.0). Cirrhotic trauma patients had a complication rate of 71 per cent and a mortality of 44 per cent. We conclude that cirrhosis is a major independent risk factor for mortality in trauma patients with injuries that require emergent abdominal surgery.     

Pediatric blunt trauma: a retrospective analysis in a Level I trauma center

This study consisted of an 8-year retrospective trauma registry analysis of blunt trauma and comprised of 2458 children (<18 years of age) and 4568 adults (18-64 years of age). Falls and motor vehicular crashes were seen in 30.4 per cent (749) and 23 per cent (566) of children, and 25.4 per cent (1158) and 41.9 per cent (1914) of adults. Children had a higher mean revised trauma score (7.69 vs 7.66) and Glasgow Coma Score (14.5 vs 14.3), and a lower Injury Severity Score (ISS; 6.68 vs 7.83 and hospital length of stay (2.8 vs 3.8 days) with P < 0.05. Overall mortality was 1.3 per cent in children and 1.9 per cent in adults (P = 0.05). Pedestrian accidents resulted in a 3.8 per cent (6/161) mortality rate. Pediatric nonsurvivors had a 6.4-fold higher ISS than survivors compared with a 5.2-fold increase in adults. Mortality progressively increased with higher ISS; 0.09 per cent in <15, 1.3 per cent, in 15 to 24, and 17 per cent in children with > or = 25 ISS. Mortality in multiple chest injuries was 19 per cent. The presence of chest trauma resulted in a 46-fold higher mortality in children. Most lethal injuries were combined head, chest, and abdomen trauma with a 25 per cent mortality in children and 28 per cent in adults. Admission Glasgow Coma Score <9 and systolic blood pressure below 100 mm Hg carried high mortality: 39 and 6 per cent in children vs 31 and 24 per cent in adults. Ninety-seven per cent of children and 89 per cent of adults were discharged home.     

IS PENETRATING INJURY ON THE INCREASE IN SOUTH-WESTERN SYDNEY?

Background: Few Australian studies describe the epidemiology of penetrating trauma. This study describes the incidence and demographic features of penetrating injuries with emphasis on trends and severity analysis. Methods: Case analysis was performed utilizing data from the Liverpool Hospital Trauma Registry (1989–94). NSW Department of Health Hospital Separations (1991–93), and the NSW Bureau of Crime Statistics (1991–93) with reference to the Liverpool and Fairfield Local Government Areas (LGA). Results: The Trauma Registry revealed 251 of penetrating trauma. The median age was 26 years (interquartile range 21–33). Ninety-one per cent of the victims were male. Fourteen per cent of patients had an Injury Severity Score (ISS) > 15. Sixty-five per cent of cases were stab injuries and 20% gunshot injuries with the abdomen being the most commonly injured site. Twenty-one per cent of patients underwent laparotomy, 1.6% thoracotomy and 1.2% thoracotomy and laparatomy. There were 10 (4%) deaths. Trends in incidence of penetrating trauma and violent crime involving weapons were analysed. Static trends were observed for the annual incidence of penetrating trauma from the Liverpool Hospital Trauma Registry. Separations for penetrating trauma from Liverpool and Fairfield hospitals showed a slightly increasing trend. Violent crimes involving weapons in the Liverpool and Fairfield LGA showed a static trend. Nevertheless, separations for penetrating trauma and rates of violent crimes involving weapons were higher in south-western Sydney than metropolitan Sydney or NSW. Eight per cent of the LGA population are Vietnamese but this study identified 16% of victims as being Vietnamese. Conclusions: This study found no significant increase in penetrating trauma or violent crime predisposing to penetrating injury in south-western Sydney.     

Effect on outcome of early intensive management of geriatric trauma patients

BACKGROUND: Despite significant injuries elderly patients (aged 70 years or more) often do not exhibit any of the standard physiological criteria for trauma team activation (TTA), i.e. hypotension, tachycardia or unresponsiveness to pain. As a result of these findings the authors' TTA criteria were modified to include age 70 years or more, and a protocol of early aggressive monitoring and resuscitation was introduced. The aim of the present study was to assess the effect of the new policy on outcome. METHODS: This trauma registry study included patients aged 70 years or more with an Injury Severity Score (ISS) greater than 15 who were admitted over a period of 8 years and 8 months. The patients were divided into two groups: group 1 included patients admitted before age 70 years and above became a TTA criterion and group 2 included patients admitted during the period when age 70 years or more was a TTA criterion and the new management protocol was in place. The two groups were compared with regard to survival, functional status on discharge and hospital charges. RESULTS: There were 336 trauma patients who met the criteria, 260 in group 1 and 76 in group 2. The two groups were similar with respect to mechanism of injury, age, gender, ISS and body area Abbreviated Injury Score. The mortality rate in group 1 was 53.8 per cent and that in group 2 was 34.2 per cent (P = 0.003) (relative risk (RR) 1.57 (95 per cent confidence interval 1.13 to 2.19)). The incidence of permanent disability in the two groups was 16.7 and 12.0 per cent respectively (P = 0.49) (RR 1.39 (0.59 to 3.25)). In subgroups of patients with an ISS of more than 20 the mortality rate was 68.4 and 46.9 per cent in groups 1 and 2 respectively (P = 0.01) (RR 1.46 (1.06 to 2.00)); 12 of 49 survivors in group 1 and two of 26 in group 2 suffered permanent disability (P = 0.12) (RR 3.18 (0.77 to 13.20)). CONCLUSION: Activation of the trauma team and early intensive monitoring, evaluation and resuscitation of geriatric trauma patients improves survival.     

Deep venous thrombosis and pulmonary embolism in trauma patients: an overstatement of the problem?

Deep venous thrombosis (DVT) and pulmonary embolism (PE) affect high-risk trauma patients (HRTP). Accurate incidence and clinical importance of DVT and PE in HRPT may be overstated. We performed a ten-year retrospective analysis of HRTP of the Pennsylvania Trauma Outcome Study. High-risk factors (HRF) included pelvic fracture (PFx), lower extremity fracture (LEFx), severe head injury (CHI) (AIS - head > or =3), and spinal cord injury. HRF alone or in combination, age, Injury Severity Score (ISS), and Glasgow Coma Score (GCS) were examined for association with DVT/PE. A total of 73,419 HRTP were included: 1377 (1.9%) had DVT, 365 (0.5%) had PE. The incidence of DVT in level I trauma centers was 2.2 per cent and was 1.5 per cent in level II centers. The lowest incidence of DVT was 1.3 per cent for isolated LEFx; highest was 5.4% for combined PFx, LEFx, and CHI. Variables associated with DVT included age, ISS, and GCS (all P < 0.001). In logistic regression analysis, only ISS was consistently predictive for DVT and PE. Though increased during the past decade, the overall incidence of DVT in HRTP remains below 3 per cent. Only the combination of multiple injuries or an ISS >30 result in DVT incidence of > or =5 per cent. We believe that current guidelines for screening for DVT may need to be reevaluated.     

Trauma in nonagenarians and centenarians: review of 137 consecutive patients

The purpose of this study was to examine the hospital course and outcomes of elderly trauma patients. We accomplished a retrospective review of all consecutive trauma patients admitted to a level II trauma center from January 2000 to April 2002. Gender, Injury Severity Score (ISS), length of stay (LOS), operative procedure, morbidity, and mortality of patients > or = 90 years of age were compared with younger patients. Of 2645 trauma admissions, 137 patients (5%) were > or = 90 years (range, 90 to 108 years; mean, 93.1 years); 5 patients were > or = 100 years. One hundred eleven (81%) patients were female; 26 (19%) male. Average ISS for patients > or = 90 was 8.75 and was 7.78 for younger patients. One hundred sixteen elderly patients (85%) had ISS < 15. Falls were the most common mechanism of injury (93%), usually ground-level falls (64%). Two hundred ninety-two injuries included 133 fractures and 102 soft tissue injuries. Thirty-four elderly patients (25%) and 733 younger patients (29%) required surgery. Complications developed in 8 per cent of older and 6 per cent of younger patients. Hospital LOS averaged 4.36 days for older and 3.51 days for younger patients. Six older (4.4%) and 63 younger (2.5%) patients died. ISS scores and LOS were slightly higher in elderly patients, but morbidity and mortality were comparable in both groups.     

Preinjury psychopathology in trauma patients

One hundred patients admitted to a Level I trauma center were interviewed using the Langner 22-item index of psychiatric symptoms to evaluate pre-existing psychological pathology. The medical chart was also examined for any type of note or evidence of psychiatric consultation efforts by the surgical attending staff. There were 74 males and 26 females in the sample, with a mean age of 33 years. Types of injuries included blunt trauma in 71% and penetrating trauma in 29% of the group, respectively. Eight per cent of these injuries were self inflicted. The mean Injury Severity Score was 17. Alcoholic intoxication was documented in 49%. We found that severe psychopathology as defined by the Langner Index was present in 88% of those admitted with penetrating trauma, in 47% of those admitted with blunt trauma, and in 75% of the intoxicated group. In only 14% of the sample was psychopathology documented in the chart by the attending staff. Preinjury psychopathology in trauma patients was commonly present and seemed to be most highly associated with penetrating trauma and alcohol use.     

In the wake of Hurricane Isabel: a prospective study of postevent trauma and injury control strategies

Preventing hurricane-related injuries (HRI) has historically centered on the pre-event and event phases of the disaster. To date, no study has focused on injuries occurring during the postevent phase. We examined HRI that occurred after Hurricane Isabel struck a U.S. urban city. HRI presenting 1 week prior to the hurricane were collected from emergency department electronic records. HRI that presented to our level 1 trauma center were prospectively collected for 1 week after the hurricane. Nine hundred seventy-eight patients with possible HRI were identified. Fifty-one patients with trauma directly attributed to the hurricane were used for analysis. The number of HRI occurring before, during, and after the hurricane were 7 (14%), 3 (6%), and 41 (80%), respectively. The majority of HRI (37%) occurred on posthurricane day 1. Head, chest, upper and lower extremities accounted for 9 (18%), 8 (16%), 13 (26%), and 14 (28%) of HRI. More than one third of HRI patients were admitted to the hospital, and 12 (24%) underwent an operation. The average hospital length of stay was 4.7 days. Of our trauma alerts, 75 per cent had an Injury Severity Score (ISS) >8, and 20 per cent had an ISS >15. Tree-related injuries (TRI) accounted for 59 per cent of HRI. Males, ages 50-60, had the highest incidence of injury (63%). Significant injuries occur in the wake of a hurricane. Optimization of disaster preparation must include prevention strategies targeted to the postevent recovery phase of disasters.     

Epidemiological study on high grade trauma. Friuli VG Major Trauma Study Group.

A better understanding of trauma epidemiology may allow to enhance the organisation of trauma systems with a potentially relevant impact on the level of trauma care. A one year epidemiology study (1st March 1998-28th February 1999) was planned in Friuli Venezia Giulia with the aim to collect all prehospital, hospital and outcome data of patients who sustained a major trauma (ISS > 15) within the regional border. In 12 months 15,429 traumatized patients (14,108 residents) were admitted to any one of the Regional hospitals. Over 1% of the whole population sustained injuries severe enough to cause hospital admission. 630 people (77.3% male, 27.7% female average age 42 ys) had a major trauma. The incidence of major trauma is 525 per million people per year. RTA was by far the most important cause of major injuries (78.6%) followed by work accidents (6.8%), domestic (5.9%) and sport accidents (1.9%). Only 1.2% of all the major injuries was the consequence of interpersonal violence. One hundred-sixty-six trauma victims died on the spot (149) or before hospital arrival (17). 464 patients with major injuries reached the hospital alive. More than two third of the patients with ISS > 15, suffered from a multiple trauma. 70% had a severe injury to the head (AIS > or = 3). Head trauma occurred as an isolated injury in only 35.3%. Hospital mortality within 30 days from admission (trauma death) was 25.1%. The results of the follow-up at 6 months are still incomplete. However the preliminary data clearly show that a high percentage of the patients who were discharged alive from the ICU had a good neurologic recovery.     

Clinical and autopsy analysis of delayed diagnosis and missed injuries in trauma patients

Delayed diagnosis of injury (DDI) during hospitalization and missed injuries (MI) on autopsy in trauma deaths result in untoward outcomes. Autopsy is an effective educational tool for health care providers to evaluate trauma care. A retrospective study of trauma registry patients and coroner's records was categorized into groups 1 (alive patients) and 2 (trauma deaths) and analyzed. DDI incidence was similar in group 1 (1.8%) and group 2 (1.9%). Autopsy analysis (163 patients) yielded 139 MI in 94 patients (57.6%), <3 per cent of MI had negative impact on survival. Bony injuries comprised 68 per cent of DDI and 19 per cent of MI. Group 1 DDI patients were sicker with higher injury severity score (ISS: 16.07) than their cohorts (ISS 7.13, P value <0.05). These patients had higher Glasgow Coma Scale (14.41) and lower ISS (16.07) as compared with group 2 MI patients (ISS: 33.49, GCS: 6.45, P value < 0.05). Autopsy rate was 99.5 per cent in trauma deaths, 57 per cent for nontrauma deaths, and 79 per cent for all deaths. Less than 3 per cent of MI had negative impact on survival. Routine ongoing patient assessment with pertinent diagnostic workup is essential in reducing DDI. Trauma autopsies reveal MI, which aid performance improvement (PI).     

Teaching Hospital Perspective of the Quality of Trauma Care in Lagos, Nigeria

Information including respiratory rate, blood pressure, Glasgow Coma Scale, and anatomic injuries were obtained for all injured patients consecutively admitted to the Lagos University Teaching Hospital (LUTH) during the 9-month period from September 1995 to May 1996. The data were analyzed using the UK Major Trauma Outcome Study (MTOS) methodology. The Revised Trauma Score (RTS), the Injury Severity Score (ISS), and the Probability of Survival (Ps) were calculated for each patient. In addition, the “M,”“z,” and “w” scores were calculated for blunt injuries. Of the 253 patients (median age 28 years; male/female ratio 4:1) studied, 85% suffered blunt trauma, 75% of which were due to RTA. About 60% of RTA victims were pedestrians. Only 8% of nonpedestrian victims wore protective devices. The mean ISS was 10, the mean RTS 6.98, and the mean Ps 0.96. The “m” statistic for blunt trauma was 0.91, and there were 122 survivors compared to the 163 predicted. The “z” and “w” scores were −18.49 and −24.55, respectively. Altogether, 42.7% of patients with ISS > 15 and 40.7% with RTS < 7.5 died. According to these results, the severity of injuries encountered in Lagos is comparable to those reported from the United Kingdom and the United States, but the death rate is higher.     

Pilot study of injured patients seen in the University of Port Harcourt Teaching Hospital, Nigeria

A 12-month pilot study of injured patients seen in the Emergency Department of the University of Port Harcourt Teaching Hospital was carried out. Trauma (28.8 per cent) was the main reason for emergency visits; 82.1 per cent of the patients were under 31 years of age. Domestic accidents were the commonest cause of trauma (42.5 per cent), followed by criminally motivated injuries (30.4 per cent), road traffic accidents (26.0 per cent), industrial (0.5 per cent) and other accidents (0.6 per cent). The overall mortality rate due to trauma was 5.4 per cent and within the period of study, trauma was the most common cause of deaths in hospital (10.1 per cent). Road traffic accidents were responsible for 67.9 per cent of these deaths, followed by criminally motivated injuries (16.1 per cent). Domestic, industrial and boat accidents caused 7.6 per cent, 5.4 per cent and 3.1 per cent deaths, respectively. Injury Severity Scores (ISS) among 419 patients showed a 100 per cent mortality for those with scores above 35. There were 48 prehospital, 19 emergency room and 14 in-hospital deaths among patients with an ISS below 35. The deaths resulted largely from delayed transportation of victims to the hospital and partly from inadequate emergency medical services. To reduce the current high rate of preventable deaths from injury, we recommend (i) ambulance services for early transportation of victims to the hospital and (ii) improved emergency medical care.     

Severe trauma caused by stabbing and firearms in metropolitan Sydney, New South Wales, Australia

BACKGROUND: Stabbing and firearm trauma causing severe injuries (injury severity score (ISS) >15) and death is uncommon in Australia. The present study describes the experience with stabbings and firearm trauma causing severe injuries at a major Australian urban trauma centre. METHODS: Data from a prospectively generated trauma registry regarding all patients presenting to Royal Prince Alfred Hospital (RPAH), Sydney, Australia with penetrating trauma causing severe injuries from July 1991 to June 2001 was retrospectively analysed. RESULTS: Of all patients presenting to RPAH with stabbing and firearms wounds over the 11 year study period, 28% received an ISS >15. One hundred and forty patients were identified. 94% were male. The mean age was 34 years (15-82 years). The number of cases/year has not shown an increasing trend. Thirty per cent of patients sustained firearm related injuries, with the remainder mainly caused by knives or machetes. Fifteen per cent of injuries were self inflicted. The most common location of injury was on a public street. Fifty-two per cent of patients were injured in more than one anatomical region, with the abdomen being the most common site of injury (53%). On hundred and seventy-four operations were performed - laparotomies (43%), thoracotomies (26%), craniotomies (5%) and orthopaedic, vascular, wound explorations and other procedures (26%). Twenty-eight per cent of patients suffered at least one complication during their admission, with coagulopathy being the most common complication (20%). Mean length of stay was 10.4 days (1-107 days). The total mortality rate for the severely injured patients was 21%, with gun-related injuries having a higher mortality rate than stabbing injury (36%vs 15%). Sixty per cent of deaths were related to exsanguination. CONCLUSIONS: Stabbings and firearm trauma are associated with significant morbidity, mortality and utilization of hospital resources in metropolitan Sydney. Overall mortality rates are similar to institutions with higher volumes of penetrating trauma.     

Popliteal artery trauma in a rural level I trauma center

Popliteal vascular trauma has historically been an urban phenomenon. We hypothesized that rural popliteal artery injury would result more often from blunt mechanisms of injury (MOI), have a longer time to operation, and result in a higher amputation rate. We retrospectively reviewed all cases of popliteal artery injury from December 1994 to May 2001 at our rural trauma center. Age, gender, Injury Severity Score (ISS), MOI, scene transport versus transfer from a referring hospital, time to operation, and operative times were studied. Significance was determined by Student's t test with a P value < or = 0.05. Thirty-two popliteal artery injuries were found. Blunt trauma accounted for 50 per cent of the injuries. Eighty-eight per cent of the patients were transferred from a referring hospital. Patients transported directly from the scene had a higher ISS. Longer operative times translated into an increased need for fasciotomy. The amputation rate was 19 per cent. This is the first attempt to delineate the specific nature of rural popliteal artery trauma. The amputation rate was not different between the two different MOI and was independent of the time to operation. Of those patients receiving an amputation 83 per cent were transferred from another hospital and despite a statistically lower ISS still required an amputation.     

Adult splenic injuries: treatment patterns and predictive indicators.

BACKGROUND: With the trend towards conservation in splenic trauma, the ability to identify a group of patients for whom we can safely offer conservative treatment becomes an important factor. METHODS: Data were reviewed from the trauma register at the Auckland Hospital, Auckland, New Zealand, in an attempt to isolate any predictive factors that may allow more appropriate allocation of treatment modalities in the future. Methods of treatment were determined and the success or failure of conservative management noted. Differences in the demographics, Injury Severity Score (ISS) and computed tomographic (CT) findings were particularly sought. RESULTS: Over a period of 111 weeks 48 patients were admitted with splenic injuries. Fifteen (31.2%) had immediate splenectomy, 27 (56.2%) were initially treated non-operatively and six (10.1%) died pre-operatively. Of the non-operative group eight (29.6%) failed this management at an average of 4.125 days into their hospital stay. No differences were found in age, mechanism, gender or ISS between the failed and successfully treated group. Using the Buntain classification of CT-graded splenic injury, 13 (87%) who had successful non-operative treatment had a grade II or III compared with six (86%) who failed this management being grade IV. CONCLUSION: Although these results did not reach statistical significance, by coupling the trends seen together with other work, CT grading of splenic injury is a predictive indicator and does appear to have a role in the early allocation of patients to appropriate treatment plans.     

ADULT SPLENIC INJURIES: TREATMENT PATTERNS AND PREDICTIVE INDICATORS

Background : With the trend towards conservation in splenic trauma, the ability to identify a group of patients for whom we can safely offer conservative treatment becomes an important factor. Methods : Data were reviewed from the trauma register at the Auckland Hospital, Auckland, New Zealand, in an attempt to isolate any predictive factors that may allow more appropriate allocation of treatment modalities in the future. Methods of treatment were determined and the success or failure of conservative management noted. Differences in the demographics, Injury Severity Score (ISS) and computed tomographic (CT) findings were particularly sought. Results : Over a period of 111 weeks 48 patients were admitted with splenic injuries. Fifteen (31.2%) had immediate splenectomy, 27 (56.2%) were initially treated non-operatively and six (10.1%) died pre-operatively. Of the non-operative group eight (29.6%) failed this management at an average of 4.125 days into their hospital stay. No differences were found in age, mechanism, gender or ISS between the failed and successfully treated group. Using the Buntain classification of CT-graded splenic injury, 13 (87%) who had successful non-operative treatment had a grade II or III compared with six (86%) who failed this management being grade IV. Conclusion : Although these results did not reach statistical significance, by coupling the trends seen together with other work, CT grading of splenic injury is a predictive indicator and does appear to have a role in the early allocation of patients to appropriate treatment plans.     

The Injury Severity Score or the New Injury Severity Score for predicting intensive care unit admission and hospital length of stay?

OBJECTIVES: To compare the New Injury Severity Score (NISS) and the Injury Severity Score (ISS) as predictors of intensive care unit (ICU) admission and hospital length of stay (LOS) in an urban North American trauma population and in a subset of patients with head injuries. METHODS: The study population consisted of 23,909 patients from three urban level I trauma centres in the province of Quebec, Canada. The predictive accuracies of the NISS and the ISS were compared using Receiver Operator Characteristic (ROC) curves and Hosmer-Lemeshow (H-L) statistics for the logistic regression model of ICU admission and using r2 for the linear regression model of LOS. RESULTS: A total of 7660 (32%) patients were admitted to the ICU. Mean LOS was 8.2+/-2.5 days. In the whole sample, the NISS presented equivalent discrimination (area under ROC curve: NISS = 0.839 versus ISS = 0.843, p = 0.08) but better calibration (H-L statistic: 309 versus 611) for predicting ICU admission. In the subgroup patients with moderate to serious head injuries, the NISS was a better predictor of ICU admission in terms of both discrimination (area under ROC curve: NISS = 0.771 versus ISS = 0.747, p < 0.00001) and calibration (H-L statistic: 12 versus 21). The NISS explained more variation in LOS than the ISS for the whole sample (r2 = 0.254 versus 0.249, p = 0.0008) and in the sub-population with moderate to severe head injuries (r2 = 0.281 versus 0.263, p = 0.0002). CONCLUSIONS: The NISS is a better choice for case mix control in trauma research than the ISS for predicting ICU admission and LOS, particularly among patients with moderate to severe head injuries.