Predicting Acute Compartment Syndrome (PACS)

Specific Aim 1: Prospectively enroll and follow for 6 months a sample of 200 patients. Patients will receive continuous tissue perfusion monitoring using NIRS in all 4 leg compartments and intramuscular pressure (IMP) via indwelling catheters placed in the anterior and deep posterior compartments. These measures will be blinded and not provided in real time to treating physicians. All clinical care, including diagnosis of ACS, will be according to current standard-of-care practiced at each institution. Specific Aim 2: Convene expert panels of 5 orthopaedic surgeons experienced in the diagnosis and treatment of ACS to retrospectively assess the likelihood that each patient had ACS. This retrospective assessment will be based on a 'patient profile' summarizing data collected as part of this study. Specific Aim 3: Determine the extent to which clinicians agree in retrospective assessments of the likelihood of ACS. Hypothesis: On the basis of known clinical and functional outcome at 6 months and monitoring information, clinicians will agree on the likelihood of ACS in \< 90% of cases. Specific Aim 4: Model the panel's assessment of the likelihood of ACS as a function of data available to the clinician within the first 48 hours of injury using a training set of the data. This model can then be used to compute a point estimate of the risk of ACS (and associated 95% confidence interval) for any given patient. Specific Aim 5: Assess, for patients in a test/validation data set, the performance of the model in predicting the panel's assessment of the likelihood of ACS. Hypothesis: In \< 95% of the cases, the panel's assessment of the likelihood of ACS will fall within the 95% interval of uncertainty predicted by the model.