| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Emergency Radiology |
1 From the Departments of Radiology (C.B.S., M.A.B., O.A.A., G.C.), Family and Preventive Medicine (R.D.), and Surgery (D.A.F., D.B.H.), University of California, San Diego Medical Center, 200 W Arbor Dr, MC 8756, San Diego, CA 92103-8756. From the 2000 RSNA scientific assembly. Received December 17, 2002; revision requested February 27, 2003; final revision received August 5; accepted August 22. R.D. supported in part by National Institutes of Health grant M01 RR00827. Address correspondence to C.B.S. (e-mail: csirlin@ucsd.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: From a database of 4,000 patients who underwent screening US for suspected blunt abdominal trauma at a level 1 trauma center, the authors retrospectively identified 3,679 patients with negative US findings. In these patients, outcome was determined by means of retrospective review of the trauma registry and all radiologic, surgical, and autopsy reports. In patients with false-negative findings at screening US, all imaging studies and medical charts were also reviewed. Proportions were statistically compared by means of the Pearson 
2 and Fisher exact tests. Monte Carlo estimation was applied when expected frequencies were low.
RESULTS: Among the 3,679 patients with negative findings at screening US, 99.9% (n = 3,641) had no injuries (true-negative findings). Differences in true-negative rates as a function of year (P > .5) or time of day (P > .3) were not significant. Among the 3,641 patients with true-negative findings, 93.6% (n = 3,407) required no additional tests and 6.4% (n = 234) underwent computed tomography or other tests. The percentage of patients who underwent additional tests was significantly higher in the 1st year of the study (19.2%) than in subsequent years (all comparisons, P < .001). Thirty-eight patients had false-negative US findings for abdominal injury. The injuries that were missed in 24 patients were nonsurgical (those that were treated successfully without intervention or were considered minor at autopsy) and those in 14 patients were surgical (required surgical intervention). Cumulatively, 65 injuries were missed. The six most common injuries included retroperitoneal hematoma (n = 13) and injuries in the spleen (n = 10), liver (n = 9), kidney (n = 8), adrenal gland (n = 8), and small bowel (n = 7). Twenty-five of the 38 patients had no or trace hemoperitoneum. Mean diagnostic delay until recognition of missed injury was 16.8 hours ± 4.3 (standard error of the mean). The missed injury was identified within 12 hours in 19 of the 38 patients and within 24 hours in 34.
CONCLUSION: The combination of negative US findings and negative clinical observation virtually excludes abdominal injury in patients who are admitted and observed for at least 12–24 hours.
© RSNA, 2004
Index terms: Abdomen, CT, 70.12112, 70.12115 • Abdomen, injuries, 70.41, 70.43 • Abdomen, US, 70.1298, 70.12984 • Trauma, 70.41, 70.43
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Despite these advantages, the use of screening US for detection of blunt abdominal trauma is controversial. Sensitivity for detection of abdominal injury at US ranges from 63% to 99% in published series (1–8,10–14) and compares unfavorably with that for computed tomography (CT). As a result, there is concern that injuries may be missed in patients with negative US scans, including clinically important injuries that require intervention. Therefore, the clinical value of negative screening US findings is unclear.
Although true- and false-negative US rates are described in previous studies (1–8,10–14), to our knowledge no previous investigation has focused on patients with negative screening US results to comprehensively assess outcome. Thus, the purpose of our study was to assess the clinical and surgical outcome of patients with blunt abdominal trauma and negative screening US findings.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The average age of the 3,679 patients was 40 years (range, 1–99 years): 2,597 were men (mean age, 38 years; age range, 1–95 years) and 1,082 were women (mean age, 42 years; age range, 4–99 years). The difference in mean age between men and women was statistically significant (P < .001, unpaired two-tailed Student t test). This retrospective study was approved by the institutional review board. The board waived informed consent for the retrospective review of patient data.
After screening US, patients were admitted to the trauma surgery service for observation, and some received treatment for other injuries (mean hospital stay, 5.1 days; median, 1.8 days; range, 6 hours to 430 days). Patients who died or who were admitted to the intensive or intermediate care units may have been admitted for less than 72 hours. Patients were given follow-up appointments in the trauma clinic 1 week after discharge from the hospital. CT and other tests—including repeat US, cystography, diagnostic peritoneal lavage, and exploratory surgery—were performed at the discretion of the trauma service. Autopsies were performed in all patients who died.
Patients were entered prospectively into the trauma registry by the trauma service at the time of discharge from the hospital. The registry contains information about all documented injuries, both abdominal and nonabdominal, and is updated regularly by the trauma service if new injuries are discovered after discharge. To identify any missed injuries after discharge, the trauma service performed systematic monthly audits of all trauma centers in the county.
US Technique
US examinations were performed during resuscitation after trauma by certified sonographers (with 1 year to more than 20 years of experience) with 2.25-, 3.5-, or 5.0-MHz sector transducers or 5.0-MHz curved-array transducers of Doppler US scanners (ATL HDI 3000, Advanced Technologies Laboratories, Bothell, Wash, or model 128-XP, Acuson, Mountain View, Calif). If a patient’s bladder was empty at the start of the examination, it was distended with 200–300 mL of sterile saline administered via a Foley catheter. The number of times that a bladder was filled with saline was not recorded because the bladder is routinely catheterized during resuscitation by the trauma service. Seven bladder regions were examined for fluid, including the upper quadrants, epigastrium, pelvis, paracolic gutters, and retroperitoneum. Visceral organs were also evaluated for parenchymal abnormalities that were suspected to be injuries. Although it was not routinely recorded, scanning time typically ranged from 3 to 5 minutes.
US Interpretation
Until April 2001, US scans were obtained (Image Link; Eastman Kodak, Rochester, NY) and reviewed on an alternator. After April 2001, US scans were archived digitally with a picture archiving and communication system network (Healthcare IMPAX; Agfa, Ridgfield Park, NJ) and reviewed with a 19-inch 1,024-pixel–resolution color monitor (MWD 421; Barco Display Systems, Kortrijk, Belgium).
US scans were interpreted prospectively by the resident and staff radiologist in the US service at the time the scans were obtained. US scans were considered positive if they depicted free fluid or a parenchymal abnormality suspected of being an injury; otherwise, US scans were considered negative. On the basis of findings in a previous study, small quantities of anechoic fluid less than 3 cm in maximum anteroposterior dimension and isolated to the cul-de-sac or paraovarian recesses in women of reproductive age were considered physiologic and not traumatic in the absence of other suspect findings (14).
Data Collection and Statistical Analysis
For each patient in the study, the trauma registry information and all radiologic, surgical, and autopsy reports were reviewed by two authors (C.B.S., M.A.B.) in consensus. Negative screening US scans were labeled as false- or true-negative by the same two authors in consensus. US scans were considered true-negative if findings at follow-up CT, repeat US, intravenous urography, cystography, diagnostic peritoneal lavage, laparotomy, and/or autopsy showed no evidence of abdominal injury. US scans were also considered true-negative in patients who did not undergo additional tests and had a benign course without clinically apparent abdominal injuries during admission, at follow-up trauma clinic appointments, and in audits of other county trauma centers. US scans were considered false-negative if abdominal injuries or unexplained intraabdominal fluid was found at follow-up.
In patients with false-negative US scans, findings in all sonograms, CT scans, and medical charts were reviewed in detail by three abdominal radiologists (G.C., C.B.S., M.A.B., with 20, 5, and 4 years of US experience, respectively) in consensus. US scans were scrutinized for errors of interpretation (free fluid described in the prospective dictated report that was misinterpreted as being nontraumatic) and of observation (obvious free fluid depicted on the US scan but not mentioned in the dictated report). CT scans were scrutinized for free fluid and extraparenchymal hematomas missed at US. Free fluid was considered trace if the maximal dimension was 1 cm or less, andhematomas were considered focal if they were not more than 2 cm in maximal dimension. In addition, missed injuries were categorized as surgical or nonsurgical. Surgical injuries were defined as those that required surgical intervention. Nonsurgical injuries were those that were treated successfully without intervention or were considered minor at autopsy. The specific organ(s) and general location (intraperitoneal, extraperitoneal, or combined intra- and extraperitoneal) of an injury was documented.
True- and false-negative rates were calculated. False-negative rates were subdivided on the basis of whether the missed injuries were surgical or nonsurgical. Comparisons were repeated after stratification for patient sex. To assess whether US performance changed over the course of the study or was affected by the time at which the study was performed, true- and false-negative rates were also calculated separately for each year of the study and for each of four 6-hour time intervals: 12–6 AM, 6 AM to 12 PM, 12–6 PM, and 6 PM to 12 AM.
With true-negative US scans, the number and proportion of patients who required additional tests to confirm the negative US findings were calculated. To determine if the tendency to order additional tests changed over the course of the study, these proportions were also calculated separately for each year of the study. Additional tests that were ordered were recorded.
Finally, we delineated the clinical course of patients with false-negative US scans. We listed all follow-up tests that were performed and determined the test or intervention that first indicated the missed injury; we also recorded the delay (in hours) between the US examination with negative findings and the diagnostic test and the clinical indication(s) that led to the test. The mean delay was then calculated and expressed as the mean plus or minus the standard error of the mean.
Proportions were statistically compared by means of the Pearson 2 and Fisher exact tests. Monte Carlo estimation was applied when expected frequencies were low. A P value of less than .05 was considered to indicate a statistically significant difference. Power to detect pairwise differences in specified effect sizes at the 5% significance level was a post hoc calculation by means of the Fisher exact test. To perform these power calculations, we used the actual sample size and observed frequency of the grouping variable with the highest observed frequency and the sample size of the grouping variable with the smallest remaining sample size. Our decision to use the smallest remaining sample size was a conservative one because comparisons with larger samples would have yielded greater power estimates.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
|
As shown in Figure 4, the percentage of patients who underwent additional tests was significantly higher in 1994 (the 1st year of the study, 19.2% [52 of 271 patients]) than in any of the subsequent years (all comparisons, P < .001). The percentage was fairly constant after 1994, ranging from 4.1% (in 1996 [29 of 699 patients] to 7.2% (in 2001 [24 of 333 patients]); differences in these annual rates were not statistically significant (P = .40). Post hoc power analysis (with assumptions specified in Materials and Methods) showed that our study had greater than 99% power to detect an absolute 10–percentage point difference in additional test rate among true-negative findings in pairwise annual comparisons.
False-Negative Findings
Of the 3,679 patients, 38 (1.0%) with negative US findings had abdominal injuries, and these findings are considered false-negative. Twenty-four (0.7%) patients had nonsurgical injuries, and 14 (0.4%) had surgical injuries (Fig 1). False-negative rates were identical for the 2,597 men (any injury, 1.0% [27 men]; nonsurgical injuries, 0.7% [17 men]; surgical injuries, 0.4% [10 men]) and for the 1,082 women (any injuries, 1.0% [11 women]; nonsurgical injuries, 0.6% [seven women]; surgical injuries, 0.4% [four women]). Since the true-negative rates were uniform during the study period and were unaffected by the time of the examination, it follows that the false-negative rates were also uniform (Fig 2) and unaffected by time (Fig 3). Cumulatively, 65 injuries were missed (Table 1), of which 33 were intraperitoneal and 32 were extraperitoneal. Nineteen patients had isolated intraperitoneal injuries (Fig 5), of whom 14 had isolated extraperitoneal injuries and five had combined intra- and extraperitoneal injuries. Eighteen patients had one missed injury, 14 had two missed injuries, five had three, and one had four. The six most commonly missed injuries included retroperitoneal hematoma (13 injuries) and injuries in the spleen (10 injuries), liver (nine injuries), kidney (eight injuries), adrenal gland (eight injuries), and small bowel (seven injuries).
|
|
|
Including one of the patients with an error of interpretation and both patients with errors of observation, 13 (34%) of the 38 patients with missed injuries had obvious hemoperitoneum at follow-up CT, surgery, or autopsy. Twenty-five (66%) of the patients had no or trace hemoperitoneum. These 25 patients included 17 with no extraparenchymal fluid (Fig 5), six with trace pockets of free fluid no more than 1 cm thick, and three with focal hematomas or sentinel clots up to 2 cm thick at the injury site without free fluid elsewhere (one patient had trace pockets of free fluid and a focal hematoma).
Abdominal injury missed at screening US was first identified at follow-up CT in 29 patients, repeat US in four, cystography in one, diagnostic peritoneal lavage in two, fetal heart rate monitoring in one, and emergent exploratory laparotomy in one (Table 2). The most common clinical reasons for the ordering of additional tests cited in these patients’ charts were pain, hematuria, decreasing hematocrit, abdominal wall ecchymosis, hypotension, fractures (of the ribs, lumbar spine, or pelvis), and sepsis (Table 3).
|
|
|
Four patients died. Two patients had nonsurgical abdominal injuries on the basis of imaging and autopsy findings, and they died of nonabdominal causes (fat emboli in one case, and closed head injury in the other). The remaining two patients had surgical abdominal injuries that likely contributed to their deaths. One of these patients had massive retroperitoneal bleeding. Laparotomy was performed, but the patient died on the table of cardiac arrest before repairs could be made. The missed retroperitoneal hematoma was diagnosed at cystography performed 15 minutes after screening US and was then confirmed at CT performed immediately afterward. The other patient had a colonic perforation that necessitated multiple surgical procedures, and the patient eventually died of sepsis. The colonic perforation was identified at diagnostic peritoneal lavage performed 9 hours after screening US.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In our study, the majority of negative US findings were true-negative: 99.0% of patients had no injury, and 99.6% had no surgical injury. Although the men were significantly younger than the women, results were identical for men and women analyzed separately. The true-negative rate was relatively constant over the course of the study, ranging from 98.2% to 99.7% on an annual basis.
Of patients with true-negative results, 6.4% underwent additional tests to evaluate abdominal injury. The remaining 93.6% did not undergo procedures such as CT and diagnostic peritoneal lavage, which minimized radiation exposure and potential complications. The tendency to order additional tests was highest (19%) in the 1st year of the study; it then remained fairly constant over the next several years (range, 4%–7%). One explanation for the reduction in additional tests ordered between the 1st and subsequent years was that, as experience accrued, confidence in screening US increased among both trauma surgeons and radiologists. The institutional learning curve was relatively rapid: The ordering of additional tests stabilized within 1 year.
False-negative screening US scans were rare; they occurred in 1.0% of patients. When false-negative screening US findings occurred, injuries were recognized within 12 hours in 50% of patients and within 24 hours in 89%. False-negative screening US findings were usually identified at follow-up CT and less commonly with other tests. The most common indications for the ordering of additional tests in patients with missed injuries were pain, hematuria, decreasing hematocrit, abdominal wall ecchymosis, hypotension, fracture (of the ribs, lumbar spine, or pelvis), and sepsis.
After assessment of the effect of delayed diagnoses on patient outcome, a few observations can be made. Most (63%, 24 of 38) missed injuries were nonsurgical and therefore probably had little effect on patient outcome. Two false-negative findings were trivial injuries depicted at CT that were officially listed as true-negative findings in the trauma registry. Moreover, some patients with surgical injuries likely would have undergone similar treatment even if the injuries had been detected sooner. For example, the woman with retromembranous bleeding probably would have been treated with fetal monitoring rather than immediate surgery if the placental abnormality had been detected at initial US screening. Similarly, it is not clear that the clinical course of the patient with delayed splenic rupture would have been altered by earlier diagnosis because most splenic lacerations are treated conservatively, and delayed rupture is not necessarily preventable.
The cases of the four patients with diagnosis delayed longer than 24 hours and of the four patients who died merit extra scrutiny. Although three of the former patients required surgery, all four patients eventually had a good outcome without complications despite the delay. Among the latter patients, delayed diagnosis may have contributed to the death of one woman (with colonic perforation identified 9 hours after screening US that necessitated several surgeries to repair the colon), who eventually died of sepsis. We doubt that the delay in diagnosis contributed to the deaths of the other three patients. Two of these patients died of nonabdominal causes, and the third died of massive retroperitoneal bleeding identified 15 minutes after screening US.
Identification of factors that led to false-negative US results is important. False-negative findings at screening US were seen in two patients early in our experience who had minimal intraperitoneal fluid that was dismissed as unimportant (error of interpretation). The finding of minimal intraperitoneal fluid is now considered positive for injury (17), and these injuries would no longer be missed. Other false-negative findings included two gross errors of observation (one surgical injury and one nonsurgical injury). False-negative findings in one patient (the pregnant woman with retromembranous bleeding seen at cesarean section in whom US images of the placenta and uterus were not acquired) represented a technical error. Since then, our standard protocol in pregnant women who experience blunt abdominal trauma mandates assessment of the placenta and a limited evaluation of the fetus (16). Most (66%, 25 of 38) of the patients with missed injuries had no clinically important free fluid on follow-up images. The lack of fluid is clearly important in the explanation of missed injuries because depiction of free fluid is the hallmark of screening US, while depiction of parenchymal abnormalities is difficult even when scanning is performed by experienced sonographers (19). On the other hand, 13 of the 38 patients had obvious hemoperitoneum. In three of these 13 patients, the diagnosis of hemoperitoneum can be explained as an error of interpretation or observation, as delineated earlier. In one other patient, shadowing from bowel gas may have obscured paracolic gutter fluid identified at CT. In another patient, who had delayed splenic rupture, hemoperitoneum may not have been present at screening US. In the remaining eight patients, hemoperitoneum may have developed over time; the mean diagnostic delay in these eight patients was 18 hours ± 8.
It is likely that at least some of the 38 false-negative findings would have been missed if CT had been performed as the initial screening test instead of US. Six false-negative findings were isolated enteric injuries, which are diagnostically challenging with both US (20) and CT (21). Although all six findings were diagnosed eventually at CT, four of the six CT examinations were performed at least 5 hours after screening US. Delayed CT demonstrated the injury, but it is uncertain whether immediate CT would have. Speed and portability are qualities that make US appealing in the setting of trauma, where rapid diagnosis is critical. Because of the development of hemoperitoneum over time, however, the same qualities can make detection of injuries with slower bleeding difficult. Also, the false-negative injuries in solid organs in at least three patients (the patient with posttraumatic pancreatitis [the pancreas itself was intact at laparotomy without evidence of laceration], the pregnant patient with retromembranous bleeding, and the patient with delayed splenic rupture) might have been missed at initial CT.
Several limitations of our study merit discussion. The most important limitation is that a definitive reference standard— CT, laparotomy, or autopsy—was available in a small minority of patients. As a result, a clinical reference standard was used in the majority (90.9%, 3,343 of 3,679) of patients. To optimize clinical follow-up, we limited our study to patients who met the entry criteria for the Mutiple Trauma Outcomes Study. Thus, all surviving patients were either admitted to the surgical ward for a minimum of 72 hours or were admitted to the intensive or intermediate care units, where more continuous monitoring is possible, for at least some time. Patients initially admitted to the intensive or intermediate care units were typically transferred to the surgical ward before discharge from the hospital; for these patients, duration of stay in the surgical ward was variable and was not subject to the 72-hour inclusion criterion. Thus, patients admitted to the intensive or intermediate care units may have been hospitalized for less than 72 hours.
In the 3,343 patients without a definitive reference standard, mean duration of hospitalization was 4.4 days (median, 1.7 days; range, 6 hours to 367 days); 97.5% (3,261 patients) were admitted for at least 12 hours, and 81.5% (2,723 patients) were admitted for at least 24 hours. All patients were given follow-up appointments after discharge from the trauma clinic, and monthly audits of all trauma centers in the county were performed. Although minor injuries may have been missed in patients who remained asymptomatic, no clinically important injuries were missed.
Another limitation of the current study is that cases were not consecutive because US examination was not available 24 hours a day for the entire study period and because we chose to limit patients to those who met the entry criteria for the Mutiple Trauma Outcomes Study. As a result in the 3,679 patients, 11.7% (n = 431) underwent screening US during the 6-hour interval between 12 and 6 AM, and the remaining 88.3% (n = 3,248) underwent screening US during the 18-hour interval between 6 and 12 AM. The false-negative rate among patients evaluated between 12 and 6 AM (0.9%, four of 431 patients) was essentially identical to that among patients evaluated at other times (1.0%, 34 of 3,248 patients). Therefore, any bias caused by the fluctuating hours of US availability was minor.
A third limitation in the current study is its retrospective nature, which required us to extract information from medical charts and dictated radiology, surgery, and autopsy reports. In particular, the true indications for performance of additional tests may not have been accurately reflected in these records.
We believe that a negative screening US result has high clinical value for two related reasons. First, the majority of patients with negative US scans have no abdominal injuries; an even larger percentage have no injuries that require surgical intervention. Therefore, negative findings at screening US assure the trauma surgeon that abdominal injury, particularly surgical injury, is highly unlikely, and attention can be appropriately directed toward management of other injuries. Because US is rapid, portable, and easily integrated into the resuscitation effort, this information can be made available within minutes of the patient’s arrival in the resuscitation suite.
Second, our study findings suggest that routine confirmation of negative findings at US by performing other tests is unnecessary. In our series, 0.5% (19 of 3,679) of patients with negative screening US findings had injuries that were not detected for more than 12 hours, and 0.1% (four of 3,679) of patients had injuries that were not detected for more than 24 hours. At our institution, all patients treated by the trauma resuscitation team, even those with negative CT findings, are admitted and observed, usually for at least 12–24 hours. Thus, the combination of negative findings at screening US and a negative clinical course during routine inpatient observation virtually excludes abdominal injury, and confirmation with other tests is unnecessary.
Despite the favorable results in our study and those of others, we advocate US as a screening modality and not as a definitive test for blunt abdominal trauma. Because US occasionally does not depict injuries, particularly those associated with little or no hemoperitoneum, US findings must be interpreted in the context of clinical and other imaging findings, with a low threshold for additional testing. In our opinion, certain clinical findings mandate additional testing despite negative screening US findings, including persistent pain, decreasing hematocrit levels, abdominal wall ecchymosis, hemodynamic instability, and sepsis. Hematuria and fractures of the lower ribs, lumbar spine, and pelvis may also warrant additional tests. Because many of these findings evolve over time, patients with negative US scans should be admitted and observed. At institutions where a period of observation is not routine practice, a negative US scan by itself may not be adequate to exclude occult injury, and patients should undergo CT as a primary study.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. C. Lee, E. L. Ormsby, J. P. McGahan, G. M. Melendres, and J. R. Richards The Utility of Sonography for the Triage of Blunt Abdominal Trauma Patients to Exploratory Laparotomy Am. J. Roentgenol., February 1, 2007; 188(2): 415 - 421. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. McGahan, S. Horton, E. O. Gerscovich, M. Gillen, J. R. Richards, M. S. Cronan, J. M. Brock, F. Battistella, D. H. Wisner, and J. F. Holmes Appearance of solid organ injury with contrast-enhanced sonography in blunt abdominal trauma: preliminary experience. Am. J. Roentgenol., September 1, 2006; 187(3): 658 - 666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. G. ODLE Blunt Pelvic Trauma Radiol. Technol., January 1, 2006; 77(3): 200 - 219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Sosna, T. Sella, D. Shaham, S. C. Shapira, A. Rivkind, A. I. Bloom, and E. Libson Facing the New Threats of Terrorism: Radiologists' Perspectives Based on Experience in Israel Radiology, October 1, 2005; 237(1): 28 - 36. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Stengel, K. Bauwens, G. Rademacher, S. Mutze, and A. Ekkernkamp Association between Compliance with Methodological Standards of Diagnostic Research and Reported Test Accuracy: Meta-Analysis of Focused Assessment of US for Trauma Radiology, July 1, 2005; 236(1): 102 - 111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Farahmand, C. B. Sirlin, M. A. Brown, G. P. Shragg, D. Fortlage, D. B. Hoyt, and G. Casola Hypotensive Patients with Blunt Abdominal Trauma: Performance of Screening US Radiology, May 1, 2005; 235(2): 436 - 443. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Catalano, R. Lobianco, M. M. Raso, and A. Siani Blunt Hepatic Trauma: Evaluation With Contrast-Enhanced Sonography: Sonographic Findings and Clinical Application J. Ultrasound Med., March 1, 2005; 24(3): 299 - 310. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Brown, C. B. Sirlin, N. Farahmand, D. B. Hoyt, and G. Casola Screening Sonography in Pregnant Patients With Blunt Abdominal Trauma J. Ultrasound Med., February 1, 2005; 24(2): 175 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Megremis, N. Segkos, A. Andrianaki, G. Gavridakis, K. Psillakis, L. Triantafyllou, N. Katsougris, M. Michalakis, and E. Sfakianaki Sonographic Diagnosis and Monitoring of an Obstructing Duodenal Hematoma After Blunt Trauma: Correlation With Computed Tomographic and Surgical Findings J. Ultrasound Med., December 1, 2004; 23(12): 1679 - 1683. [Full Text] [PDF]
|
|
|
|
|
|
G. T. Shaw, C. B. Sirlin, M. A. Brown, R. Deutsch, O. A. Andrade-Barreto, D. Fortlage, D. B. Hoyt, and G. Casola The Golden Hour: How to Spend Your Time and Money in Trauma Radiology * Dr Sirlin and colleagues respond: Radiology, August 1, 2004; 232(2): 622 - 624. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
