Abstract Diagnostic peritoneal lavage (DPL) and computed tomography (CT) are the primary diagnostic modalities used in the evaluation of patients with suspected blunt abdominal trauma (BAT). DPL is fast and accurate but is associated with complications. CT is also accurate, yet requires stability and transportability of the patients. Ultrasound (US) has been suggested as an aid in evaluating BAT. We evaluated US in the initial assessment of BAT in 1000 patients. Patients were eligible for the study if they met specified trauma criteria and had suspected BAT. We then followed the outcome of the patients and their further work-up. US showed a sensitivity of 88%, a specificity of 99%, and an accuracy of 97% for detecting intraabdominal injuries. We conclude that emergency ultrasound may be used as the initial diagnostic modality for suspected blunt abdominal trauma. Evaluating the abdomen in blunt trauma remains a clinical challenge. Physical examination may be misleading when there are proximity injuries such as lower rib fractures or pelvic fracture. Injuries may be missed by physical examination when there is altered mental status from drugs, alcohol, or associated head trauma. The physical examination is unreliable when there is associated spinal cord injury. Using the vital signs as a guide to intraabdominal injury is also unreliable, since hemorrhage may be from other sites and stable vital signs can be associated with an intraabdominal injury. [1] For these reasons diagnostic peritoneal lavage (DPL) and computed tomography (CT) are used as diagnostic modalities in the evaluation of patients with suspected blunt abdominal trauma (BAT). Since described by Root [2] in 1965, DPL has been proved to be fast and accurate, but it is associated with complications. [2-6] Major complications occur in about 1% of cases [5,6] and include laceration of iliac or mesenteric vessels and perforation of intestines or bladder. CT is a useful adjunct in evaluating BAT but requires a stable patient who can be transported to the scanner. [7-12] Ultrasound (US) has been shown to be accurate and useful in the evaluation of BAT. [13-21] US combines the advantages of DPL (fast and accurate) with those of CT (non- invasive and accurate). In June of 1993 we completed a prospective study comparing US to DPL and CT examination. [22] With the results of this study we concluded that US is reliable in the detection of free intraperitoneal fluid and could be utilized in suspected BAT. We designed a prospective study to determine whether US could be used as the initial diagnostic modality in patients who would otherwise have a DPL or CT examination. Patients with suspected BAT were evaluated with US. The result was then compared to observation, CT, or operation. PATIENTS AND METHODS From July 1993 until November 1994, 1000 patients with suspected BAT were evaluated with US. Patients were evaluated at the Ryder Trauma Center of the University of Miami/Jackson Memorial Hospital, the only Level Trauma Center for all of Dade County, Florida. Patients were brought to the Trauma Center if they met trauma criteria (Table 1). US was performed if the attending surgeon felt physical examination was unreliable for ruling out intraabdominal injury. Exclusion criteria were hemodynamic stability with gross hematuria, suspected retroperitoneal hemorrhage, severe pelvic fracture, or attending surgeon preference. Stable patients with gross hematuria or suspected retroperitoneal hemorrhage normally have a CT examination at our institution to evaluate the retroperitoneum. Furthermore, patients with severe pelvic fracture are given a CT examination to evaluate the fracture and the retroperitoneum. To avoid duplication of tests these patients were excluded from our study. After US was performed, a data collection sheet was completed and patients were followed for their hospital stay. US was performed on patients during resuscitation by a technologist using an Accuson 128X P/10 (Mountain View, California) with a 3.5-MHz sector or curvilinear transducer. All USs were immediately interpreted by an attending radiologist or senior radiology resident, from real-time images or hard copies. US was utilized to evaluate the presence of free fluid in seven areas: the subphrenic space bilaterally, subhepatic space, paracolic gutters bilaterally, splenic tip, and the pelvis. The liver and spleen were investigated for parenchymal injury. US was deemed positive if free intraperitoneal fluid or parenchymal injury was clearly seen. A negative result signified that no free intraperitoneal fluid was identified or visceral injury noted on an adequately performed examination. An indeterminate result was recorded if there was questionable fluid or visceral injury or if the examination was technically limited. A positive CT was defined as the presence of free fluid or visceral injury identified. Criteria for a positive DPL were (1) initial aspirate of 5 cc of blood, (2) RBCs > 100,000 RBCs/mm 3 , or (3) WBCs > 500 WBCs/mm 3 . After a negative US examination patients were admitted for observation. If subsequent gross hematuria developed, persistent abdominal pain occurred, or brief hypotension ensued, then a follow-up CT was obtained (see Figure 1, Algorithm 1). The work-up and treatment of a patient with a positive US were based on vital signs. If the US was positive and the patient had stable vital signs, then a CT was obtained to further delineate the injury. A patient with a positive US and hypotension was either taken directly to the operating room or was given a subsequent DPL, based on the attending surgeon's preference (see Figure 2, Algorithm 2). If an indeterminate result was obtained, the patient was either given a DPL or examined by CT scan, based on the vital signs (see Figure 3, Algorithm 3). For statistical analysis, true negative (TN) was defined as a negative US and no subsequent injury found. True positive (TP) was defined as a positive US and confirmation. False negative (FN) was defined as a negative US and subsequent identification of intraabdominal 1,000 Consecutive Ultrasounds for Blunt Abdominal Trauma ISSN: 0022-5282 Accession: 00005373-199604000-00015 Author(s): McKenney, Mark G. MD; Martin, Larry MD; Lentz, Kimberley MD; Lopez, Cristina MD; Sleeman, Danny MD; Aristide, George BS; Kirton, Orlando MD; Nunez, Diego MD; Najjar, Rony MD; Namias, Nicholas MD; Sosa, Jorge MD Issue: Volume 40(4), April 1996, pp 607-612 Publication Type: [Article] Publisher: © Williams & Wilkins 1996. All Rights Reserved. Institution(s): From the University of Miami School of Medicine (M.G.M., L.M., K.L., D.S., G.A., O.K., D.N., J.S.) and Jackson Memorial Hospital (C.L., R.N., N.N.), Miami, Florida. Presented at the Eighth Annual Scientific Session of the Eastern Association for the Surgery of Trauma, January 11-14, 1995, Sanibel, Florida. Address for reprints: Mark McKenney, MD, Assistant Professor of Clinical Surgery (D-40), University of Miami School of Medicine, P.O. Box 016960, Miami, FL 33101. Page 1 of 7 Ovid: 1,000 Consecutive Ultrasounds for Blunt Abdominal Trauma. 21-02-2009 http://ovidsp.tx.ovid.com/spa/ovidweb.cgi