John A. Savino

Preparation and achievement of American College of Surgeons level I trauma verification raises hospital performance and improves patient outcome.

OBJECTIVE The purpose of this study was to assess the impact on patient outcome and hospital performance of preparing for and achieving American College of Surgeons (ACS) Level I trauma verification. METHODS The center was a previously designated state regional trauma center located adjacent to a major metropolitan area. Preparation for ACS verification began in early 1996 and was completed in early 1998. Final verification took place in April 1999. Data were analyzed before (1994) and after (1998) the process. There was a marked increase in administrative support with trauma named one of the hospitals six centers of excellence. Two full-time board-certified trauma/critical care surgeons were added to the current six trauma surgeons. Their major focus was trauma care. Trauma support staff was also increased with case managers, a trauma nurse practitioner, additional trauma registrars, and administrative support staff. Education and continuous quality improvement were markedly expanded starting in 1996. RESULTS There were 1,098 trauma patients admitted in 1994, and 1,658 in 1998. Overall mortality decreased (1994, 7.38%; 1998, 5.37%; p < 0.05). There was a marked decrease in mortality for severely injured (Injury Severity Score > 30) patients (1994, 44% mortality [38 of 86]; 1998, 27% [22 of 80]; p < 0.04). Average length of stay also decreased (1994, 12.22 days; 1998, 9.87 days; p < 0.02). This yielded an estimated cost savings for 1998 of greater than

An artificial neural network as a model for prediction of survival in trauma patients: validation for a regional trauma area.

4,000 per patient (total saving estimate of

Assessing educational validity of the Morbidity and Mortality Conference: A pilot study

7.4 million). CONCLUSION Trauma system improvement as related to achieving ACS Level I verification appeared to have a positive impact on survival and patient care. There were cost savings realized that helped alleviate the added expense of this system improvement. The process of achieving ACS Level I verification is worthwhile and can be cost effective.

Preoperative Assessment of High-Risk Surgical Patients

BACKGROUND To develop and validate an artificial neural network (ANN) for predicting survival of trauma patients based on standard prehospital variables, emergency room admission variables, and Injury Severity Score (ISS) using data derived from a regional area trauma system, and to compare this model with known trauma scoring systems. PATIENT POPULATION The study was composed of 10,609 patients admitted to 24 hospitals comprising a seven-county suburban/rural trauma region adjacent to a major metropolitan area. The data was generated as part of the New York State trauma registry. Study period was from January 1993 through December 1996 (1993-1994: 5,168 patients; 1995: 2,768 patients; 1996: 2,673 patients). METHODS A standard feed-forward back-propagation neural network was developed using Glasgow Coma Scale, systolic blood pressure, heart rate, respiratory rate, temperature, hematocrit, age, sex, intubation status, ICD-9-CM Injury E-code, and ISS as input variables. The network had a single layer of hidden nodes. Initial network development of the model was performed on the 1993-1994 data. Subsequent models were generated using the 1993, 1994, and 1995 data. The model was tested first on the 1995 and then on the 1996 data. The ANN model was tested against Trauma and Injury Severity Score (TRISS) and ISS using the receiver operator characteristic (ROC) area under the curve [ROC-A(z)], Lemeshow-Hosmer C-statistic, and calibration curves. RESULTS The ANN showed good clustering of the data, with good separation of nonsurvivors and survivors. The ROCA(z) was 0.912 for the ANN, 0.895 for TRISS, and 0.766 for ISS. The ANN exceeded TRISS with respect to calibration (Lemeshow-Hosmer C-statistic: 7.4 for ANN; 17.1 for TRISS). The prediction of survivors was good for both models. The ANN exceeded TRISS in nonsurvivor prediction. CONCLUSION An ANN developed for trauma patients using prehospital, emergency room admission data, and ISS gave good prediction of survival. It was accurate and had excellent calibration. This study expands our previous results developed at a single Level I trauma center and shows that an ANN model for predicting trauma deaths can be applied across hospitals with good results

Pericardiocentesis Followed by Thoracotomy and Repair of Penetrating Cardiac Injury Caused by Nail Gun Injury to the Heart

PURPOSE To assess inter-rater agreement in perceptions of cases presented during Morbidity & Mortality conference (M&M) and changes associated with initiation of a modified M&M. METHODS Faculty, residents, fellows, and students at weekly M&M between June 2001 and March 2002 voluntarily completed an anonymous questionnaire after each M&M case presentation, which asked: if the complication was avoidable (yes/no/not sure), if consensus was reached among participants (yes/no/not sure), the primary cause of the complication (diagnostic error, technical error, judgment error, nature of disease, other), when the primary cause occurred (preoperatively, intraoperatively, postoperatively), and which, if any, of 16 listed actions could prevent similar future problems. On September 24, 2001, the conference was lengthened and modified. Data collected before (n = 30 cases) and after M&M modification (n = 46 cases) were compared. RESULTS A total of 76 cases were evaluated for a total of 860 completed forms. In 57 cases (75%), majority opinion (ie, > or =50% of participants) indicated that the complication resulted from either nature of disease (n = 32, 41%), or error in diagnosis (n = 5, 7%), technique (n = 8, 11%) or judgment (n = 12, 17%). There was no clear majority for the remainder of the cases. Relative to cases presented prior to M&M modification, for those presented post-modification, the majority perceived that consensus was reached more often (96% of cases vs. 70% of cases, p<.01), and that complications were more often avoidable (54% of cases vs. 23% of cases, p <.05), more likely caused during the preoperative period (26% of cases vs. 7% of cases, p <.01) and less likely caused during the postoperative period (28% of cases vs. 67% of cases, p <.01). CONCLUSIONS The variability in questionnaire responses suggests that an evaluation instrument such as that reported here can be useful in assessing educational needs, quantifying the efficacy of case presentations, and assessing the effects of modifications to conference content and structure. Modifying M&M in accordance with published recommendations appears to improve case analysis and consensus among participants.

Preoperative Assessment of the High-Risk Elderly

From the foregoing accounts of preoperative assessment of myocardial performance, as well as preload and afterload status it is clear that the proper anesthesia techniques and agents can be selected. Physiologically optimal adjustments of preload, afterload, and myocardial function can be attained by the appropriate, harmonious selection of anesthesia technique and vasoactive drugs made on the basis of close hemodynamic monitoring preoperatively, intraoperatively, and in the immediate postoperative period.

An Indwelling Oncometer-Catheter

Highlights • Nail gun injuries to the heart are potentially fatal and difficult to recognize.• Timely recognition and definitive repair of the injury is essential for patient survival.• Pericardiocentesis using a catheter capable of repeated and/or sustained aspiration/drainage can be a temporizing life-saving measure allowing for definitive repair in a controlled setting.• Pericardiocentesis must be considered as a bridge to definitive therapy in select cases of penetrating cardiac injury, particularly in rural settings with limited surgeon availability.

The Automated Renal Profile

The ability of patients to meet the stresses of surgery and anesthesia should be assessed in the preoperative period in order to quantitate the cardiovascular, respiratory, and metabolic functional deficits. The high-risk patient who must undergo surgery presents a judgment decision for the internist, surgeon, and patient. Unfortunately, preoperative risk assessment without optimization of the patient does little to diminish intraoperative and postoperative morbidity and mortality.

Hemodynamic and pulmonary changes after drainage of significant pleural effusions in critically ill, mechanically ventilated surgical patients.

Measurement of colloid osmotic pressure (COP) and hydrostatic pressure within the intravascular space is an important factor in the evaluation of body fluid shifts. This paper describes the development of an oncometercatheter, which allows the in vivo measurement of COP. The device utilizes multiple lumen catheter technology and standard intensive care equipment for monitoring. The physiologic basis for measurement is described and the catheter function explained. Sensitivity and lack of complications make this catheter invaluable in the laboratory determination of COP. The oncometer may be incorporated in a balloon flotation catheter, which may allow for its eventual use in the monitoring of the critically-ill human.

Multi-Institutional Validation of a Web-Based Core Competency Assessment System

An automated renal profile has been developed based on 30 input and derived parameters. This profile allows a rapid assessment of the glomerular and tubular functions, the excretion factor of the kidney, along with the hemodynamic data of critically ill patients. The diagnosis of prerenal azotemia or acute tubular necrosis can be easily recognized on the profile, allowing immediate therapeutic intervention (fluids or diuretics) prior to permanently established renal failure.