Traditional injury scoring underestimates the relative consequences of orthopedic injury

Traditional injury scoring underestimates the relative consequences of orthopedic injury

Vol. 191, No. 4S, October 2000 mix of non-trauma patients and may undermine the sustainability of large urban TC. To assess the potential burden of t...

182KB Sizes 2 Downloads 29 Views

Vol. 191, No. 4S, October 2000

mix of non-trauma patients and may undermine the sustainability of large urban TC. To assess the potential burden of the trauma patient on the urban Level ITC we evaluated the payer-mix of trauma patients relative to non-trauma patients at different levels of trauma care in a mature state-wide trauma system. Methods: Patients admitted to a hospital in the state over a 3 year period were classified as either trauma (ISS ⱖ 9) or non-trauma and by insurance status as either commercial insurance (CI) (e.g. managed care) or non-commercial insurance (e.g. Medicaid or self-pay). Medicare patients were excluded from analysis. Data were compared using ␹2 analysis.

S83

Surgical Forum Abstracts

23 hour observation solely for identification of missed injuries following trauma is not justified Phillip J Stephan, MD, Clifann McCarley, RN, Grant E O’Keefe, MD, Joseph P Minei, MD. UT-Southwestern Medical Center, Mailing Address: Joseph P Minei, MD; UT-Southwestern, 5323 Harry Hines Blvd.; Dallas, TX 75235-9158, USA; Tel: 214-648-7295. Introduction: 23 hour observation for serial evaluation of trauma patients with unreliable exams, intoxication, or need for further diagnostic work-up is utilized to reduce missed injury. This study was undertaken to assess whether 23 hour observation is an effective adjunct to minimize missed injury after initial emergency room evaluation.

Results: There were 10,386 trauma admissions and 474,944 non-trauma admissions to 87 centers. Trauma patients were less likely to have CI than non-trauma patients (69% vs 74%, p ⬍ 0.001). The proportion of trauma patients with CI treated at the Level ITC was significantly less than at other centers (graph). However, trauma patients treated at the Level I TC were far more likely to have CI than non-trauma patients treated at this same center (graph). By contrast, there was no relationship between payer and trauma status at other levels of care.

Methods: Over a 2-year period at an urban level 1 trauma center, 6749 patients were admitted and 2,458 underwent 23 hour observation. Of the 2,458 observation patients, 303 (12.3%) were converted to full admission and had charts available for review. Of these, 48 were excluded for incorrect classification into an observation status (known multiple injuries, operative procedures, etc.). The remaining 255 converted patients underwent detailed chart review.

Reason for Conversion (N ⴝ 255)

N (%)

% of 2458 observed

Conclusions: Utilization of trauma triage guidelines results in the admission of a disproportionate number of patients with CI relative to non-trauma patients to this urban Level I TC, an effect due to referrals from outside the urban area and across the state. In this environment, designation as a Level ITC may actually improve care for inner city non-trauma patients by ensuring the ready availability of acute care services that follows designation as a TC and by means of cross subsidization of non-trauma care through trauma care reimbursement.

Further evaluation of known injury Pain management Change of plan for known injury Social Psych Missed injury/changed x-ray reading

89 (35) 50 (20) 41 (16) 34 (13) 16 (6) 25 (10)

3.6 2.0 1.7 1.4 0.7 1.0

Variations in the care of the head injured patient EM Bulger, AB Nathens, FP Rivara, DC Grossman, M Moore, GJ Jurkovich. University of Washington, Harborview Medical Center, Box359796, 325 Ninth Ave., Seattle, WA, 98104, USA Introduction: In spite of the availability of evidenced based guidelines from the Brain Trauma Foundation, the optimal management of the head injured patient remains controversial. Perhaps the most controversial guideline is the recommendation for intracranial pressure (ICP) monitoring for patients with a GCS ⱕ 8 and an abnormal head CT scan. We sought to evaluate the variations in care of head injured patients and to determine the impact of ICP monitoring on outcome. Methods: Data were collected from 34 academic trauma canters of the University HealthSystem Consortium concerning consecutive admissions with a Head AIS (HAIS) score ⱖ2, age ⬎ 18, and at least one long bone fracture from 5/98 to 12/98, n ⫽ 621. Areas where variation in care were assessed included: pre-hospital intubation, ICP monitor placement, use of osmotic agents, hyperventilation, and CT scan utilization. The impact of ICP monitor use on mortality and length of stay (LOS) was evaluated using logistic regression and linear regression, respectively, to control for confounding effects of age, gender, injury severity score (ISS), shock on admission, mechanism of injury, GCS, HAIS, and head CT findings. Results: Considerable variation in care was evident. The use of prehospital intubation ranged from 0 –56% of patients; ICP monitor placement, 0 – 49%; use of osmotic agents, 0 – 63%; and hyperventilation, 0 –38%. The mean number of head CT scans obtained per patient varied from 1– 4, while the median time to first CT scan varied from 0.3–2.4 hours. Overall, ICP monitors were placed in only 103/621 patients (16%). ICP monitoring was performed in 42% of patients with a GCS ⱕ 8 and 34% of patients with an abnormal CT scan. While 57% of patients with both a GCS ⱕ 8 and an abnormal CT result had ICP monitoring, the use of ICP monitors did not influence mortality (adjusted odds ratio for death, 0.89 (95% CI 0.26 –3.05)). Use of ICP monitors was associated with an increase in LOS of 1.9d (95% CI ⫺7.5 to 11). Conclusions: Considerable variability persists in the management of the head injured patient. The use of ICP monitoring in the highest risk patient stratum did not significantly alter mortality or hospital length of stay.

Results: There were 164 (64.3%) males and 91 (35.7%) females ages 15– 83 (mean of 35). Reasons for conversion to full admission from observation status are listed in the table below.

Of the 25 patients that had a missed injury or changed radiographic finding, 10 did not have a clinically significant change in management that prolonged hospital stay (median LOS ⫽ 2 days). The 15 remaining patients had significant missed injuries (0.6% of all observed). All required prolonged hospital admissions (median LOS ⫽ 7 days) and 4 underwent invasive procedures and/or surgery. No patient died from a missed injury. Conclusions: In a cohort of over 2400 observed trauma patients, less than 1% remained hospitalized for significant missed injuries. We conclude that 23 hour observation for the purpose of identifying missed injuries after emergency room evaluation may not be justified.

Traditional injury scoring underestimates the relative consequences of orthopedic injury AJ Michaels, MD, MPH, S Madey, MD, J Krieg, MD, WB Long, MD, FACS. Legacy Emanuel Hospital, Portland, OR and the University of Michigan, Ann Arbor, MI; A Michaels, MD Trauma Services Legacy Emanuel Hospital 2801 Gantenbien Ave Portland, OR 97227, USA (503)413-2100 Introduction: The objective of this study is to illustrate that polytrauma patients with orthopedic injuries (ORTHO) face greater challenges regarding functional outcome than those without. Methods: A convenience sample of adult blunt force trauma patients admitted to a Level I trauma center was evaluated during admission and 12 months after injury. Data were collected from the trauma registry (Trauma One姞), chart review, and interviews. Mailed surveys were completed 12 months after injury. The SF36 general health survey and the Sickness Impact Profile work scale (SIPw) were administered at both time points. Data are presented as mean ⫾ SEM or percent (%). T-tests were conducted to compare means, and ISS was controlled by linear regression prior to the evaluation of the role of ORTHO injury pattern on outcome measures. Significance is noted at the 95% confidence level (p ⬍ .05). Results: The 165 patients studied were 37.2 ⫾ 1.1 years old and 67% male. The mean ISS was 14.4 ⫾ .59 and 61% had ORTHO injury. ORTHO patients were no different from nonORTHO in any measure of baseline status including the SIPw and all domains of the SF36, except that the ISS was greater in the ORTHO group (15.6 ⫾ .96 vs 12.7 ⫾ .73, p ⫽ .017). Baseline SF36 values were similar to national norms. Follow-up at 12 months was 51%. Those lost to follow-up differed only

S84

Surgical Forum Abstracts

J Am Coll Surg

in that they were more likely to be male. 64% had returned to work 12 months after injury. After controlling for ISS with linear regression, the ORTHO patients had greater hospital lengths of stay (p ⬍ .001) than nonORTHO patients. They also had relatively worse scores 12 months after injury on the SIP work score (p ⫽ .016) and 6 of 8 SF36 domains (bodily pain, physical function, role physical, mental health, role emotional, and social function - all p ⬍ .05). Conclusions: Injury severity affects both mortality and the potentially more consequential issues of long-term morbidity. The ISS was designed to quantify mortality risk across the anatomic boundaries of acute injury. Patients with ORTHO injury have significantly worse functional recovery and greater long-term costs than those without ORTHO injury. As trauma systems approach the limits of achievable survival, new advances in trauma care can be directed more towards the quality of recovery for our patients. Patients with ORTHO injury require additional resources for optimal recovery. We must focus our efforts on the development of more sensitive scoring systems designed to predict functional outcome across injury boundaries for those who survive.

Predictors of mortality in adult trauma patients: the physiologic trauma score (PTS) is equivalent to TRISS DA Kuhls MD, DL Malone MD, LM Napolitano MD, R McCarter ScD, TM Scalea MD. R. Adams Cowley Shock Trauma Center and the University of Maryland, 22 South Greene St., Room N4E27, Baltimore, MD, USA Phone (410)328-5830 Introduction: Several statistical models (TRISS, ICISS, NISS) have been developed during recent decades in an attempt to accurately predict outcome in trauma patients. However, the anatomic portion of these models makes them difficult to use when performing a rapid initial trauma assessment. We sought to determine if a physiologic trauma score (PTS) could accurately predict mortality in trauma. Methods: Prospective data were analyzed in 9539 trauma patients at a Level I Trauma Center over 30 months (1/97–7/99). A SIRS score (1– 4) was calculated on admission (1 point for each present: Temp. ⬎ 38° C or ⬍ 37° C, Pulse ⬎ 90, RR ⬎ 20, WBC ⬎ 12 or ⬍ 4). SIRS score, ISS, RTS, TRISS, age, sex and race were used in logistic regression models to predict trauma patients’ risk of death. The area under ROC curves of sensitivity versus 1-specificity was used to assess the predictive ability of the models. Results: The study cohort had a mean ISS of 9 ⫾ 9; mean age of 37 ⫾ 17. SIRS (SIRS score ⱖ 2) was present in 2165 patients (28.5%). In single-variable models, TRISS and ISS were most predictive of outcome. A multiple-variable model (PTS) combining SIRS score with RTS and age (variables easily calculated at the patient bedside) was equivalent to TRISS and superior to ISS in predicting mortality. Single Indices Area under ROC TRISS ISS RTS SIRS

0.97 0.92 0.88 0.75

Multiple Indices SIRS, SIRS, SIRS, SIRS,

Age, Gender ISS, Age, Race RTS, Age ISS, RTS, Age

Area under ROC 0.82 0.95 0.95 0.97

Conclusions: The PTS, including only physiologic variables, accurately predicts mortality in trauma. Furthermore, the predictive ability of this model is equivalent to other complex models that utilize both anatomic and physiologic data (TRISS, ISS, ICISS). A multicenter prospective assessment and validation of the physiologic trauma score (PTS) is warranted.

Methods: 165 of 673 trauma deaths occurred in the ICU at our Level I trauma center during 1997 and 1998. All but 4 underwent autopsy and the records of 139 were available for review. Missed injuries were defined as Class I-major, pre-mortem diagnoses that if recognized and treated appropriately, would have changed patient outcome; Class II-major diagnoses that if recognized and treated appropriately, would not have changed outcome; Class III-minor diagnoses. All preexisting medical conditions identified at autopsy were also recorded. Results: Mean age was 50 (range 9 –104yrs) and 72% were males. Mean ICU stay was 10 ⫾ 15 days, median 4.5days. 15 of 2135 patients with penetrating injuries, 100 of 4255 patients with blunt injuries and 28 of 308 patients with burns admitted to our facility died in the ICU. Significant pre-existing medical conditions found at autopsy included significant cardiac disease (47), emphysema (5), and cirrhosis/hepatic fibrosis (6). There were 5 (3.6%) Class I discrepancies: meningitis; lacerated tricuspid papillary muscle; retroperitoneal abscess; myocardial infarction; bowel infraction. There were 28 (20.1%) Class II discrepancies including pneumonias (11), pulmonary emboli (3) and small lung abscesses (3). 109 (78.4%) of 139 patient autopsies revealed no major missed findings (21 had Class III discrepancies). Conclusions: Major missed findings as a cause of death are unusual in the trauma ICU. While potentially valuable for identifying preventable trauma deaths, autopsy infrequently provides unexpected information in the ICU population.

Changing the face of eldertrauma: one center’s experience Robert D Barraco, MD, MPH, Thomas M Scalea, MD. R Adams Cowley Shock Trauma Center, Baltimore MD Mailing Address Dr. Barraco: Department of Surgery, Division of Trauma/Critical Care SUNY-Stony Brook Health Sciences Center, T19, 060 Stony Brook, NY 11794-8191, USA Phone (631) 444-1045 Introduction: The population ages 65 and over will almost double by the year 2025. By the year 2050, Medicare will comprise 35% of the federal budget. Injury and poisonings are second only to heart disease as hospital discharge diagnosis in the elderly. Many improvements in trauma systems, resuscitation and critical care have been made over the last 10 years. As a result, we hypothesize that the outcomes of eldertrauma have improved. To date, this has not been proven in the literature. Methods: A retrospective database review was conducted on 1107 patients age 65 and over during four non-consecutive years at our Level I trauma center. Data gathered included age, sex, mechanism of injury, Injury Severity Score (ISS), ICU and total length of stay (ICULOS & TLOS), mortality and disposition. Subgroup analyses were performed for ISS and age. Statistical tests included 1- and 2-way ANOVA and linear and logistic regression analyses. Results: The mean ages across years were between 74 and 76 years old. Mechanism did not change appreciably over the years studied. There were statistically significant increases in proportion of female patients and mean ISS (p ⫽ 0.0018 and p ⬍ 0.0001, respectively). Statistically significant decreases were found for the outcome variables ICULOS, TLOS and mortality over the years studied aside from the effect of the decreasing ISS (p ⬍ 0.0001 for each outcome). Each year, disposition to home was the most frequent, followed by rehabilitation centers.

Unexpected findings in trauma patients dying in the ICU: results of a two-year consecutive autopsy series

Conclusions: Outcomes after eldertrauma are improving even after controlling for patient characteristics. Other factors obviously play a role in this improvement, such as the improved care of the injured elderly. The demographics may also be changing, with more females and older patients being seen at our center.

Adrian W Ong MD, Stephen M Cohn MD, Kelly A Cohn RN, MS, Mary Murtha MSN, David Jaramillo BA, Rakesh Parbhu BS, Michael Bell MD. Departments of Surgery and Pathology. University of Miami School of Medicine 1800 NW 10th Ave. Miami, FL 33136, USA, 305-585-1178

A statewide population-based study of gender differences in trauma: validation of a prior singleinstitution study

Introduction: The true incidence of missed injuries in traumatic deaths is unknown; only about 50% undergo autopsies. We determined the rate of missed injuries in our trauma ICU population with a 100% autopsy policy.

L Napolitano M.D., M Pasquale M.D., R McCarter Sc.D. R. Adams Cowley Shock Trauma Center, Baltimore MD and Lehigh Valley Trauma Center, Allentown, PA., 22 South Greene St, Room N4E27, Dept. of Surgery, Baltimore, MD, USA, Phone (410)328-5830.