Inferior clinical outcomes after femur fracture in the obese are potentially preventable.

INTRODUCTION: Obese patients with operative orthopedic trauma have increased risk of adverse outcomes, although the mechanisms accounting for the relationship remain unknown. This study examines the effect of body mass index (BMI) on outcomes after femur fracture fixation, and explores the mediating effects of pathophysiologic factors and clinical management. METHODS: A retrospective chart review was performed of adult patients with femur fractures undergoing surgical fixation at a Level 1 trauma center from 2010 to 2016. Demographics, Injury Severity Score (ISS), Glasgow Coma Scale (GCS) and mechanism of injury (MOI) were collected along with operative data and complications. Primary outcomes were hospital length of stay (HLOS), ICU length of stay (ICU-LOS), mortality, complications, and time to mobility (time first out of bed, TFOB). Bivariate correlations and multiple regression models were used to examine the relationship between BMI and outcomes. Path analysis tested whether the relationship between BMI and clinical outcomes was mediated by differences in 1) clinical management, or 2) physiologic variables. RESULTS: Of 333 patients included, the majority were male (57.4%) with a mean age of 43.4 (22.7) years and ISS of 12.5 (6.8). Predominant MOIs were motor vehicle crashes (42.8%) and falls (34.5%). There was no association between BMI category and age, ISS, or GCS. In univariate analysis, higher BMI was linked to longer HLOS (r = .12), longer ICU-LOS (r = .15), longer TFOB, (r = .18), and higher number of complications (r = .12), p < 0.05. Controlling for age and ISS, obese patients had 6.66 times the odds of respiratory failure (p = 0.021, 95% CI 1.3,33.3) and a 3.88 odds of any complication (p = 0.020, 95% CI 1.24,12.1) compared to their normal weight counterparts. For every one point increase in BMI, time first out of bed was delayed 2.3 h (p < 0.001; 95% CI 1.08, 3.62). The effect BMI on poor outcomes was accounted for by delayed mobility (longer TFOB) in a mediation model. CONCLUSIONS: Higher BMI increases the risk of longer hospital stays and systemic complications. Mediation models indicate that the adverse clinical outcomes associated with obesity are explained by delays in mobility, an intervenable factor. Clinical strategies should be directed at early mobilization to minimize morbidity.

Inhaled epoprostenol improves oxygenation in severe hypoxemia.

BACKGROUND: Epoprostenol (Flolan), an inhalational epoprostenol vasodilator, increases pulmonary arterial flow and decreases pulmonary pressures, thereby improving gas exchange and arterial oxygenation. We evaluated the benefits of inhaled epoprostenol as a less expensive alternative to nitric oxide in ventilated surgical intensive care patients with severe hypoxemia. METHODS: After institutional review board approval was obtained, the records of mechanically ventilated surgical intensive care unit patients who received epoprostenol as a therapy for severe hypoxia (SaO2 < 90%) in a tertiary care referral center were retrospectively reviewed. Initial PaO2/FIO2 (P/F) ratio and oxygen saturation were compared with values at 12 and 48 hours after the administration of epoprostenol. One-way repeated-measures analysis of variance compared improvements in oxygenation. Further subgroup analyses evaluated differences among trauma, nontrauma patient subgroups, time to initiation of epoprostenol, and age. RESULTS: During a 20 month-interval beginning February 2009, 36 patients (23 trauma and 13 nontrauma; age, 15-80 years) were treated. Epoprostenol significantly improved both P/F ratio and oxygen saturation in both trauma and nontrauma patients. Therewas no difference between subgroups. Larger improvements in P/F ratiowere seen when epoprostenolwas started within 7 days. Response between age groups did not differ significantly. Subgroup analysis of mortality (trauma, 60.9% vs. nontrauma, 61.5%) failed to show any differences. CONCLUSION: Treatment with inhaled epoprostenol improved gas exchange in severely hypoxemic surgical patients. Earlier intervention (within 7 days of intubation) was more efficacious at improving oxygenation.

Florida's trauma surgeons: a vanishing breed.

The delivery of trauma and emergency surgical care is in a state of crisis. We hypothesized that this looming crisis was already manifested in Florida. The trauma medical directors of the 20 state designated trauma centers were surveyed for information pertaining to number of available surgeons for trauma call, number of night calls/month, age of the current trauma surgeons, and the estimated number of years each surgeon planned to continue taking call. We also queried trauma medical directors about recruitment of additional trauma surgeons. Fourteen directors responded. Each program had at least four surgeons taking trauma call on average 5.3 nights/month. Sixty-three per cent of surgeons taking call were less than 50-years-old. Thirty surgeons (39.5%) planned to discontinue trauma call within 10 years, leaving 46 surgeons (60.5%) presently committed to longer than 10 years of call. Nine programs were actively recruiting. Five programs (50%) were recruiting for < 1 year, three programs (30%) were recruiting for 1 to 2 years, and two programs (20%) were recruiting > 2 years. Florida's trauma surgeons are a vanishing breed. Given the recruiting difficulties, the diminishing numbers of Florida's general surgeons will have to fill the gaps.

The use of 23.4% hypertonic saline for the management of elevated intracranial pressure in patients with severe traumatic brain injury: a pilot study.

BACKGROUND: Oncotic agents are a therapeutic mainstay for the management of intracranial hypertension. Both mannitol and varied concentrations of hypertonic saline (HTS) have been shown to be effective at reducing elevated intracranial pressure (ICP). We compared the safety and efficacy of 23.4% HTS to mannitol for acute management of elevated ICP after traumatic brain injury (TBI). METHODS: After approval from our institutional review board, the records of patients admitted with severe TBI who received mannitol or HTS were reviewed. Demographic and physiologic data were recorded. ICP, cerebral perfusion pressure, reduction of ICP after dose administration, serum sodium, osmolality, and magnitude of dose response during the subsequent 60 minutes were analyzed. Efficacy was determined by comparison of proportion of patients with any response and mean change in ICP after dosing with either agent. Safety was determined by recording any new postinfusion electrolyte or neurologic anomalies. Data were compared using chi2 test, accepting p < 0.05 as significant. RESULTS: Twenty-two patients with severe TBI received 210 doses of either mannitol or HTS. All patients suffered severe blunt injury (mean Injury Severity Score 28 +/- 11). HTS patients had a significantly higher ICP at the initiation of therapy than that of mannitol group (30.7 +/- 7.94 mm Hg vs. 28.3 +/- 8.07 mm Hg, respectively). There was no difference in initial cerebral perfusion pressure. Mean ICP reduction in the hour after administration of 102 doses of mannitol and 108 doses of HTS was greater for patients receiving HTS (9.3 +/- 7.37 mm Hg vs. 6.4 +/- 6.57 mm Hg, respectively; p = 0.0028, chi2). More patients responded to HTS (92.6% HTS vs. 74% mannitol; p = 0.0002, chi2). There was no significant difference between groups in the duration of ICP reduction after dose administration (4.1 hours vs. 3.8 hours, respectively). No adverse events after administration of either agent were identified. CONCLUSION: Based on this retrospective analysis, 23.4% HTS is more efficacious than mannitol in reducing ICP. If these results are confirmed in a prospective, randomized study, 23.4% HTS may become the agent of choice for the management of elevated ICP after TBI.

Definitive establishment of airway control is critical for optimal outcome in lower cervical spinal cord injury.

BACKGROUND: Respiratory complications can undermine outcome from low cervical spinal cord injury (SCI) (C5-T1). Most devastating of these is catastrophic loss of airway control. This study sought to determine the incidence and effect of catastrophic airway loss (CLA) and to define the need for elective intubation with subsequent tracheostomy to prevent potentially fatal outcomes. METHODS: A database of 54,838 consecutive patients treated in a level I trauma center between January 1988 and December 2004 was queried to identify patients with low cervical SCI, without traumatic brain injury. Patients were then stratified into complete or incomplete SCI groups, based on clinical assessment of their SCI. Mortality, age, injury severity, need for intubation, and tracheostomy were analyzed for each group using Fisher's exact test or Student's t test, as appropriate, accepting p < 0.05 as significant. RESULTS: One hundred eighty-six patients met inclusion criteria. The majority of low cervical spinal cord injuries were complete (58%). Overall, 127 (68%) patients required intubation, 88 (69%) required tracheostomy, and 27 died (15% of study population). Between each group there were significant differences in age and Injury Severity Score, however, within each group there were no significant differences in either. Eleven CSCI patients were not intubated; four of whom were at family request. Six of the remaining seven patients encountered fatal catastrophic airway loss. One patient was discharged to rehabilitation. Patients with incomplete SCI required intubation less frequently (38%); however, 50% of those required tracheostomy for intractable pulmonary failure. CONCLUSIONS: These data indicate that regardless of severity of low cervical SCI, immediate, thorough evaluation for respiratory failure is necessary. Early intubation is mandatory for CSCI patients. For incomplete patients evidence of respiratory failure should prompt immediate airway intervention, half of whom will require tracheostomy.

Best practice determination of timing of spinal fracture fixation as defined by analysis of the National Trauma Data Bank.

BACKGROUND: To examine the efficacy of early versus late spinal fracture fixation, we reviewed National Trauma Data Bank (NTDB) records to identify the breakpoint in reported timing of operative fixation. Using this breakpoint we then analyzed outcome for those treated early versus late, hypothesizing that the early group would experience better outcome as reflected by resource utilization and complications. METHODS: The NTDB was queried for patients with any level spinal fracture that required surgical stabilization. Histogram analysis of the postinjury day of initial operative fixation was used to determine the point at which the majority of operative procedures had been performed, thereby defining early (E) and late (L) groups. Patients in E were matched to a cohort from L with similar age, Injury Severity Score, and Glasgow Coma Scale. Outcome data included hospital length of stay, intensive care unit length of stay, ventilator days, charges, incidence of complications, and mortality. The groups were compared using Student's t test for continuous variables and Fisher's exact test for categorical variables, accepting p < or = 0.05 as significant. RESULTS: Of 16,812 patients who underwent operative fixation, 59% were completed within 3 days of injury and formed E. The 374 L patients whose dataset was complete enough to allow analysis were matched to 497 E patients. There was no significant difference in the presence of spinal cord injury between E and L (51 vs. 48%; p = 0.3735). Complications were significantly higher in L (30% vs. 17.5%; p < 0.0001) yet mortality was similar in both groups (2.0% vs.1.9%; p > 0.05). CONCLUSIONS: NTDB records indicate that the majority of patients with spinal fractures undergo operative fixation within 3 days, and that these patients had less complications and required less resources. Use of a national data bank to compare groups with similar injury severity and presenting physiology can validate best practice and define opportunities for improvement in care.

The effect of early surgical treatment of traumatic spine injuries on patient mortality.

INTRODUCTION: The ideal timing of spinal fixation is controversial. A recent study showed that early spine fixation reduced morbidity and resource utilization. We previously noted a trend toward higher mortality in patients undergoing early spinal fixation. This study was done to analyze whether the timing of spinal fixation had a significant effect on mortality. METHODS: The registry of our Level I trauma program was queried for all patients with at least one spinal vertebral injury. Anatomic and physiologic variables included age, initial Glasgow Coma Scale score, systolic blood pressure, heart rate, and Injury Severity Score. Outcome was evaluated in terms of ventilator days, intensive care unit length of stay, hospital length of stay (HLOS), and mortality. Patients were stratified by day of spinal operative fixation as early when done within 48 hours and late when done after 48 hours. Data were analyzed using chi and an unpaired t test, accepting p < 0.05 as significant. RESULTS: Three hundred sixty-one patients between January 1988 and February 2003 required operative spinal fixation (158 early, within 48 hours vs. 203 late, beyond 48 hours). There was no significant difference between the two groups except mortality, which was significantly higher in the early group (7.6 vs. 2.5%; p = 0.0257), and HLOS, which was significantly shorter in the early group (14.42 vs. 17.64 days; p = 0.025). CONCLUSION: Spinal fixation within 48 hours after vertebral fractures and dislocations appears to increase mortality despite similar anatomic and physiologic parameters in the later operative group. Incomplete resuscitation of patients before surgery may have contributed to this result. The shorter HLOS may have been because of the higher number of early deaths. Prospective studies to identify the optimal timing of spinal fixation and the reason for these outcome differences are warranted.

Urgent airway intervention: does outcome change with personnel performing the procedure?

BACKGROUND: Recent studies indicate that prehospital endotracheal intubation (PHEI) is associated with increased septic morbidity. Because the decision to intubate in the field is considered a life-sustaining mandate we analyzed our experience to validate these reports and to compare field intubation to that done in more controlled circumstances on patient arrival at the trauma center. METHODS: The registry of our Level l trauma center was queried from January 2002 through December 2003 for patients who required emergent EI and had a hospital stay > 2 days. Patients were stratified by site of EI into PHEI and trauma center intubation (TCEI). Demographic data (age, gender, Glasgow Comma Scale, Injury Severity Score) as well as outcome measures (incidence of pneumonia [PNA], Intensive Care Unit length of stay [ICU LOS], hospital length of stay [hospital LOS], and mortality) were compared between groups. Results were subjected to chi2 or unpaired t test, accepting p < 0.05 as significant. RESULTS: The 628 patients requiring EI consisted of 27l in PHEI and 357 in TCEL. When comparing these groups, PHEI were more severely injured (lower Glasgow Comma Scale score and higher Injury Severity Score), but had no other differences in demographics or in measured outcome variables. Within these groups, patients who developed PNA were comparable. They demonstrated similar time of onset of PNA after injury and had similar incidence of resistant organisms (46%). CONCLUSIONS: These data demonstrate no increased risk of PNA for urgent prehospital intubation. Moreover, the onset of PNA and the similar bacteriology is reflective of injury severity and not of additional infectious risk posed by these prehospital lifesaving maneuvers.

The burden of noncompliance with seat belt use on a trauma center.

INTRODUCTION: It is well-known that noncompliance with seat belt use results in worse injury. The impact of noncompliance on hospital resource consumption and hospital charges is less well known. This study was carried out to examine the economic burden of noncompliance with seat belt use. METHODS: Trauma registry data were reviewed for patients involved in motor vehicle crashes in 2003 and 2004. Routine demographic data were analyzed. Outcome data included hospital length of stay, intensive care unit length of stay, number of ventilator days, and mortality. Hospital charges, rate of collection, hospital use (measured by need for admission), operating room use, and intensive care unit use were calculated to determine the burden of noncompliance with seat belt use. RESULTS: There were 3,426 patients identified for analysis. Of these patients, 1,744 (51%) were compliant with seat belt use (SEAT) while 1,682 were not compliant (NO SEAT). Patients in the NO SEAT group were significantly younger (31.2 versus 37.4 years old) and significantly more severely injured (Injury Severity Score of 11 versus 7) than those in the SEAT group. Patients in the NO SEAT group had a significantly longer hospital length of stay (4.4 versus 2.2 days) and intensive care unit length of stay (1.4 versus 0.3 days), as well as significantly more ventilator days (1.2 versus 0.2 days) than those in the SEAT group. Mortality was more than doubled in the NO SEAT group (2.2 versus 0.9%) as compared with the SEAT group. Resource consumption was significantly greater in the NO SEAT group, as evidenced by increased hospital use (64.9 versus 39%), increased critical care unit use (22.9 versus 10.3%) and increased operating room use (9.2 versus 4.9%) when compared with the SEAT group. Subsequently, hospital charges were significantly higher in the NO SEAT group ($32,138 versus $16,547) than in the SEAT group. Charge collection rate was lower in the NO SEAT group (30.5 versus 42.5%) than in the SEAT group. CONCLUSIONS: These data quantify the burden placed on a trauma center by noncompliance with seat belt use. This information should drive more focused education and injury prevention programs. It should also be clearly articulated to legislators to stimulate more support for more stringent legislative policy and improved trauma center funding.

The effect of early spine fixation on non-neurologic outcome.

INTRODUCTION: It has been shown that spinal fracture fixation within 3 days can reduce the incidence of pneumonia, length of stay, number of ventilator days, and hospital charges. Our institutional protocol calls for surgical stabilization of spinal fractures within 3 days of admission. We hypothesized that compliance with an early spinal fracture fixation protocol (within 3 days of admission) would improve non-neurologic outcome in patients with spinal fractures. METHODS: The trauma registry was queried for the period January 1988 through October 2001 to identify patients with spinal fractures requiring surgical stabilization. Patients were analyzed to determine the compliance with our protocol and to determine whether early spinal fixation can reduce the incidence of pneumonia, reduce length of stay, and reduce mortality. RESULTS: 1,741 patients with spinal fractures were identified. 299 (17.2%) required surgical stabilization. 174 (58.2%) had surgical stabilization within 3 days while 125 (41.8%) had surgical stabilization greater than 3 days from admission. There were no significant differences between the two groups with regards to age (37.9 versus 42.5), admission GCS (14.1 versus 13.9), or ISS (22 versus 20.8). The incidence of pneumonia was similar in both groups (21.8 versus 25.6%). The mortality was higher in the early group as compared with the late group (6.9 versus 2.5%), although it did not reach statistical significance. The hospital length of stay was significantly shorter (14.3 versus 21.1) for patients who had early spine fixation, however there was no statistically significant difference between the two groups with regards to intensive care unit length of stay (7.2 versus 7.9) or number of ventilator days (5.02 versus 1.9). Patients who were severely injured (ISS > 25) also had a significantly shorter hospital length of stay (19.6 versus 29.1) if they underwent early spinal fixation. Patients with thoracic spine injury and associated spinal cord injury had a significantly shorter HLOS (10.1 versus 30.5), ICULOS (2.3 versus 13.1), and lower incidence of pneumonia (6.5 versus 33.3%). CONCLUSIONS: Reasonable compliance with an early spinal fracture fixation protocol produced some outcome improvements in non-neurologic outcome. Early spine stabilization reduced hospital length of stay in all patients. Patients with thoracic spine trauma and a spinal cord injury had the greatest benefit in reduction of morbidity, HLOS and ICULOS from early stabilization. There was a trend toward poorer outcome in some groups with early spine stabilization. A rigid protocol requiring early surgical spine stabilization in all patients does not appear justified. Although early spine stabilization should be performed whenever possible to reduce hospital length of stay, the timing of this procedure should be individualized to allow patients with the most severe physiologic derangements to be optimized preoperatively.

Radiographic clearance of blunt cervical spine injury: plain radiograph or computed tomography scan?

BACKGROUND: The purpose of this study was to evaluate the roles of cervical spine radiographs (CSR) and computed tomography of the cervical spine (CTC) in the exclusion of cervical spine injury for adult blunt trauma patients. METHODS: At the authors' institution, all adult blunt trauma patients with physical findings of posterior midline neck tenderness, altered mental status, or neurologic deficit are considered at risk of cervical spine injury and undergo both CSR and CTC for evaluation of the cervical spine. The TRACS database at level 1 of the trauma center at this institution was queried for all blunt trauma patients from November 2000 to October 2001. Patient injury severity score (ISS), Glascow Coma Score (GCS), age, gender, CSR results, CTC results, and treatment data were analyzed. RESULTS: The review included 3,018 blunt trauma patients with appropriate data. For 1,199 of these patients (779 men and 420 women) (40%) at risk for cervical spine injury, both CSR and CTC were performed for cervical spine evaluation. The average age of these patients was 39.4 years (range, 18-89 years). The average GCS was 13 and the average ISS was 8.4 in this study population. In 116 (9.5%) of these patients, a cervical spine injury (fracture or subluxation) was detected. The injury was identified on both CSR and CTC in 75 of these patients. In the remaining 41 patients (3.2%), the CSR results were negative, but injury was detected by CTC. All these injuries missed by CSR required treatment. For this group with false-negative CSR, the average GCS was 12 and the average ISS was 14.6. There were no missed cervical spine injuries among the patients with negative CTC results. CONCLUSION: No identifiable factors predicted false-negative CSR. There does not appear to be any role for CSR screening in this setting. The data from this study add to the growing body of evidence that CTC should replace CSR for the evaluation of the cervical spine in blunt trauma.

Pediatric trauma is very much a surgical disease.

OBJECTIVE: The evolution of nonoperative management of certain solid visceral injuries has stimulated speculation that management of the severely injured child is no longer a surgical exercise. The authors hypothesized that the incidence of injuries that require surgical evaluation is disproportionately high in children at risk of death or disability from significant injury. METHODS: National Pediatric Trauma Registry data were queried for all patients with ICDA-9-CM diagnoses requiring at least surgical evaluation. Selected diagnoses included CNS: 800 to 804, 850 to 854; thoracoabdominal: 860 to 870; pelvic fracture: 808; and acute vascular disruption: 900 to 904. Operative intervention was identified by ICDA-9-CM operative codes less than 60 and selected operative orthopedic codes between 79.8 and 84.4. At-risk patients were identified as those with at least one of the following: Glasgow Coma Scale score less than 15, Glasgow Coma Scale motor score less than 6, initial systolic blood pressure less than 90, or Injury Severity Scale score more than 10. The incidence of a surgical diagnosis in at-risk children was compared to the incidence in the population with no identifiable risk. Within the population undergoing surgical evaluation, resource utilization, as reflected by operative intervention and ICU days, and outcome, as reflected by mortality, were compared between the at-risk group and the group with no identifiable risk. RESULTS: From 1987 to 2000, 87,424 records were complete enough for analysis. Of those, 48,687 (55.6%) patients sustained at least one injury requiring a surgical evaluation and 28,645 (32.7%) children were determined to be at risk. Mortality for at-risk children was 5.8% versus 0.02% for those with no identifiable risk. Of the children at risk, 24,706 (86.2%) had at least one injury requiring a surgical evaluation. Of the 58,779 children with no risk, 23,981 (40.8%) also had at least one injury requiring a surgical evaluation. Operative intervention for surgical injuries was required in 20.5% of cases (n = 10,015). Of these, 5,562 (56%) were at-risk children, and they had a mortality rate of 11.5%. Of the children not at risk, 4,453 required operative care, and they had a mortality of 0.1%. At-risk children undergoing surgery required an average of 5.02 days of ICU care compared to 1.2 for cases performed on children without risk. CONCLUSIONS: These data clearly demonstrate the primacy of surgical pathology as the major determinant of outcome in pediatric injury. Operative intervention and the option of timely operative care remain major components of clinical management of children with injuries that pose a significant risk of morbidity or mortality.

Impaired residents: identification and intervention.

PURPOSE: To describe the effect of a psychologist on faculty staff supporting impaired residents to successful program completion in general surgery. METHODS: Since 1996, the Department of Surgery has had a staff psychologist on faculty who works closely with the Trauma service. Duties include provision of patient and family therapies (representing a revenue-generating activity for the department), in-services for staff stress management, research, and community education activities. As resident performance issues have arisen, the psychologist has been instrumental in early identification of and referral for underlying issues affecting resident behavior. RESULTS: From 1996 to October 2001, 12 residents (2 with multiple referrals) have been identified with significant psychological issues impairing performance. The various psychological problems include depression, anger control issues, and addictions; required interventions have included referrals for therapeutic counseling, recommendations for career change, and therapeutic counseling combined with pharmacotherapy. Fellow residents referred 4 of the 12 residents to the psychologist for possible evaluation, and the remaining 8 had referral initiated by performance concerns. CONCLUSIONS: Significant psychological problems can impair resident performance and possible training program completion. Early identification and intervention by a professional psychologist on faculty can provide support for increased opportunity to complete residency training.

Beyond the golden hour: avoiding the pitfalls from resuscitation to critical care.

The transition from resuscitation bay to intensive care unit is wrought with potential problems that can significantly affect patient outcome. Among these are hypothermia, injudicious fluid management, hypocarbia, incorrect drug dosing, and incomplete evaluation. Based on a comprehension of the ongoing pathophysiology associated with injury, steps can be taken to ensure that this process positions the patient on the beginning of the path to recovery and not further into the spiral of worsening organ system dysfunction.

Two careers in one: an analysis of the earning power of certification in surgical critical care.

BACKGROUND: The core of general surgery supports multiple disciplines, each of which entails similar operative care for different diseases. The purpose of this study was to compare variations of practice patterns of four general surgeons to define the general surgical core that each shared in common, and to determine the effect of subspecialization in surgical critical care on the scope of practice and efficiency of revenue production. METHODS: The charges and collections of four members of the same surgical faculty were analyzed for the 6 months beginning July 1, 2000. Three members practiced general surgery with additional specialization in surgical oncology, surgical endoscopy, and trauma/critical care. The fourth covered all aspects of general surgery, including in-house trauma call, but not surgical critical care. Data were stratified by Current Procedural Terminology code and categorized as operative, bedside care (which included minor procedures), and evaluation/consultation care. Scope of practice was defined as the proportion of operative cases represented by the 10 most frequently performed procedures. General surgical core was defined as those cases that were preformed by all four surgeons at the same frequency. Efficiency of revenue generation was defined as collection rate for these procedures divided by the established, budgeted collection rate for each practitioner. All results were compared using chi(2) with significance accepted at p < 0.05. RESULTS: Fifteen operative procedures were performed with equal frequency by each surgeon and represented a broad spectrum of surgical disease. These procedures constituted a similar proportion of operative practice for all specialists (mean, 45.2%; 90% confidence limit, 3.5%), yet occupied 70% of the trauma surgeon's 10 most frequent surgical procedures versus 36% for the surgical oncology and surgical endoscopy. Charges generated by the provision of surgical critical care, especially in bedside procedures commonly performed in the intensive care unit, exceeded all of the other three surgeons and equaled the revenue generated by operative care. Although overall revenue-generating efficiency was less for the trauma surgeons (57% of eventual collections vs. 67%, chi(2) p = 0.1), immediate reimbursement for critical care was higher than for any other clinical services. CONCLUSION: These data demonstrate that subspecialization in surgical critical care provides valid additional earning capacity to surgical practitioners. Reimbursement is at least as good as for traditional operative care, and fees generated can actually exceed revenue from operative care. With impending decreases in global reimbursement, and attempts to unbundle operative fees, this additional capability becomes an important consideration in potential career choice, as well as a major component in the fiscal stability of trauma programs.