JUM Track the topics, authors and articles important to you
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sato, M.
Right arrow Articles by Yoshii, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sato, M.
Right arrow Articles by Yoshii, H.
© 2004 by the American Institute of Ultrasound in Medicine
J Ultrasound Med 23:1583-1596 • 0278-4297

Reevaluation of Ultrasonography for Solid-Organ Injury in Blunt Abdominal Trauma

Michihiro Sato, MD and Hiroshi Yoshii, MD

Departments of Radiology (M.S.) and Surgery (H.Y.), Saiseikai Kanagawaken Hospital and Kanagawaken Traffic Trauma Center, Yokohama, Japan.

Address correspondence and reprint requests to Michihiro Sato, MD, 6-6 Tomiya-cho, Kanagawa-ku, Yokohama 221-8601, Japan. E-mail: michi-s{at}mub.biglobe.ne.jp.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Objective. To reevaluate the usefulness of ultrasonography for detecting and classifying solid-organ injuries from blunt abdominal trauma by comparing ultrasonography with computed tomography (CT) and laparotomy. Methods. Six hundred four patients with blunt abdominal trauma were examined by both B-mode ultrasonography and CT for a study period of 14 years. The ultrasonographic examiners were divided into 2 groups depending on their experience with ultrasonography. The ultrasonographic results were then compared with CT and surgical findings. This was a retrospective study. Results. In 198 patients, solid-organ injuries were identified on CT, laparotomy, or both. Sensitivity values in group A (experts) were 87.5% for hepatic injuries, 85.4% for splenic injuries, 77.6% for renal injuries, and 44.4% for pancreatic injuries. Sensitivity values in group B were 46.2% for hepatic injuries, 50.0% for splenic injuries, and 44.1% for renal injuries. The detection rates in group A were 80% to 100% for different types of hepatic injuries except superficial injuries (20%) and 70% to 100% for different types of splenic injuries. The detection rates for renal parenchymal and pancreatic duct injuries were 53.3% and 80%, respectively. The detection rates for injuries requiring intervention were 86.1% in group A and 66.7% in group B. Conclusions. The sensitivity of ultrasonography with the use of CT and surgical findings as reference standards decreased compared with our prior study. However, ultrasonography was found to enable experienced examiners to detect and classify parenchymal injuries efficiently, despite disadvantages in detecting superficial and vascular injuries. Ultrasonography should be used to explore not only free fluid but also solid-organ injuries.

Key Words: blunt abdominal trauma • computed tomography • solid-organ injury • ultrasonography

Abbreviations: BAT, blunt abdominal trauma • CT, computed tomography • DPL, diagnostic peritoneal lavage • FAST, focused assessment with sonography for trauma • FN, false-negative • FP, false-positive • TAE, transarterial embolization • TN, true-negative • TP, true-positive


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
A scanning method called focused assessment with sonography for trauma (FAST)1 was devised with a primary objective of developing a procedure that could detect intraperitoneal fluid and could be used easily, after training for brief periods, by surgeons and emergency medicine physicians with limited experience in ultrasonography. Numerous reports have appeared in which the usefulness of FAST has been studied. Generally, such FAST examinations have been performed to detect free fluid in patients with abdominal trauma, and computed tomography (CT) has been performed to detect parenchymal injuries as well as free fluid. Hemodynamically stable patients with positive or indeterminate FAST results undergo CT scanning. Hemodynamically stable patients with negative FAST results are followed by clinical observation and repeated FAST to confirm the absence of injury because organ injuries are not necessarily accompanied by hemoperitoneum. In hemodynamically unstable patients, a positive FAST result leads to an emergency laparotomy, and an indeterminate FAST result leads to a diagnostic peritoneal lavage (DPL) or a CT scan. (Few trauma centers currently perform DPL, either in Japan or in North America.) In this algorithm, ultrasonography is limited to detecting free fluid (ie, FAST). A problem arises as to the performance of an entire abdominal ultrasonographic examination that includes evaluation of organ parenchyma in this algorithm. If the initial ultrasonographic examination is performed according to the FAST procedure, we think that an entire abdominal ultrasonographic examination can be performed during a follow-up examination in a hemodynamically stable patient with a negative FAST result and before a DPL or an emergency laparotomy is performed in an unstable patient.

We reported the usefulness and limitations of ultrasonography in the evaluation of blunt abdominal trauma (BAT) in 1998.2 In that report, ultrasonographic results were compared not only with CT and surgical findings but also with repeated ultrasonographic examinations and the clinical courses of the patients. If the clinical course is used as a reference standard, a minor injury missed on ultrasonography may be defined as a true-negative (TN) finding, even though the injury’s presence would make the result false-negative (FN). It was possible, therefore, that the sensitivity was considered inappropriately high. It seems to be of considerable clinical relevance, therefore, to assess the usefulness of ultrasonography in detail using CT and surgical findings as reference standards and to analyze the detection rates for injuries classified according to their severity. Furthermore, we retrospectively evaluated the diagnostic capability for parenchymal injury of the examiners, who must usually perform FAST in the emergency department, with limited ultrasonographic experience.

The purpose of this study was to reevaluate the usefulness of ultrasonography for assessing solid-organ injuries in BAT by comparing ultrasonography with CT and laparotomy and to evaluate the probability of direct detection of parenchymal injuries.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
During a 14-year period from January 1988 through December 2001, 2127 patients with suspected BAT were examined with ultrasonography, and 824 patients among this group were examined with both ultrasonography and CT at a trauma center affiliated with this hospital. Of these patients, 604 were included in this study. The first criterion for inclusion was that the patient had to have had contrast-enhanced CT performed from the diaphragm to the lower pole of the kidney with 10-mm collimation and at 10-mm intervals. The second criterion was that the intervals between the ultrasonographic and CT examinations had to have been less than 12 hours. This retrospective study was approved by the Institutional Review Board.

For the ultrasonographic examinations, several ultrasonography systems (SSD-280, SSD-650CL, SSD-1700, SSD-2000, and SSD-5500 [Aloka Co, Ltd, Tokyo, Japan]; EUB-450 [Hitachi Medical, Tokyo, Japan]; and SSA-260 [Toshiba Medical Systems Co, Ltd, Tokyo, Japan]) were used. Mainly, convex array probes with frequencies between 3.25 and 5 MHz were used, although 3.5-MHz linear array probes were used at the initial stage of the study. Images showing any pathologic changes were recorded on hard copies that were then stored with the reports. The standard ultrasonographic procedure focused on detecting intraperitoneal fluid and parenchymal abnormalities in solid organs.

To identify each parenchymal injury, the liver was examined by the right subcostal and intercostal scans and by the transverse and longitudinal scans in the epigastrium, the spleen by the left intercostal scan, the pancreas by the transverse and longitudinal scans in the epigastrium, and the kidneys by the transverse and longitudinal scans in the flank regions. Color Doppler ultrasonography was performed in some patients, but evaluation of that method was not included in this study.

For the CT examinations, a CT scanner (9200; GE Yokogawa Medical Systems, Tokyo, Japan) was used until October 1993; thereafter, a different CT scanner (ProSeed Accell; GE Yokogawa Medical Systems) was used. Conventional CT was performed from the diaphragm to the iliac crest with 10-mm collimation and at 10-mm intervals and from the iliac crest to the symphysis pubis with 10-mm collimation and at 10- or 15-mm intervals. Helical CT was performed from the diaphragm to the symphysis pubis in 161 patients (10-mm collimation, table speed of 10 mm/s, and pitch of 1.0). The intravenous contrast materials were iothalamic acid (Conray; Daiichi Pharmaceutical, Tokyo, Japan), iohexol (Omnipaque 300; Daiichi Pharmaceutical), and iopamidol (Iopamiron 300; Nihon Schering, Osaka, Japan). In adults, 100 mL of contrast material was administered at a rate of 1 to 1.5 mL/s with a power injector, although it was administered to some patients by means of rapid intravenous drip infusion. For children, the contrast material was administered at a dose of 2 mL/kg as a rule. No oral contrast material was used.

The CT images stored on optical disks were reviewed. Each case was then classified according to the classifications of hepatic, splenic, renal, and pancreatic injuries by the Japanese Association for the Surgery of Trauma (Table 1Go).3–6 Table 1Go also shows the correspondence to the organ injury scale by the American Association for the Surgery of Trauma.7–9 Each case was classified by 2 physicians, a radiologist (M.S.) and a surgeon with considerable expertise in trauma surgery (H.Y.). The findings obtained with ultrasonography were also classified according to stored ultrasonographic images and accompanying reports.


View this table:
[in this window]
[in a new window]
  		Table 1. Classification of Hepatic, Splenic, Renal, and Pancreatic Injuries According to the Japanese Association for the Surgery of Trauma and Correspondence to the Organ Injury Scale According to the American Association for the Surgery of Trauma
 
The diagnostic capability of ultrasonography depends on the skill and experience of the examiners. Ultrasonography shows its full diagnostic capability, therefore, only when used by examiners who have acquired substantial skill. Accordingly, we divided the examiners into 2 groups depending on their skill and experience and compared the detection rates of injuries for individual solid organs between the 2 groups. Group A consisted of radiologists, surgeons, and sonographers who participated on a daily basis in ultrasonographic scanning and had sufficient experience (14 members in total). Group B mainly consisted of resident surgeons who had had limited training and experience (51 members in total).

The first ultrasonographic examination was performed by a member of group A in 251 of the 604 patients and by a member of group B in the remaining 353 patients. Forty-eight of the 353 patients were examined by group A as well, to determine their definitive diagnoses, within 12 hours of the first ultrasonographic examinations by group B. The intervals between performance of the ultrasonographic examinations by group A and CT were also less than 12 hours. Including these 48 patients that were examined by both groups, there were a total of 299 patients examined by group A. The interval of less than 12 hours between the first ultrasonographic examinations by group B and the second ultrasonographic examinations by group A was considered acceptable because we had not experienced any cases in which parenchymal injuries became more evident or in which the ultrasonographic images changed markedly within this 12-hour period. The ultrasonographic examinations were performed before or after CT examinations from case to case. When a CT examination was performed first in a case, the subsequent ultrasonographic examiner was uninformed of the details of the CT results. The sonographers of group A often performed ultrasonography in the morning for the patients who arrived at the trauma center the previous night or in the early morning. They could not read the CT images. A case in which an examiner could detect a parenchymal injury based on the CT results was counted as FN.

In each case, we only analyzed the results from the first ultrasonographic examination for each examiner group. During the analyses, the detection rates for different types of injuries were calculated. In addition, the types of injuries identified on ultrasonography by group A were compared with those identified on CT, and their degree of agreement was determined. Moreover, the FN and false-positive (FP) cases were studied. In addition, the utility of ultrasonography was evaluated in cases in which injuries were so severe that an intervention was required.

As the standard for diagnosis, we used CT, which seemed to be most reliable. When an injury was not detected with CT but was found during surgery, the surgical findings were taken as the standard. Because macroscopic observations during surgery could not reveal morphologic changes inside the abdominal organs, however, identification of certain types of injuries depended on CT findings even in laparotomy cases. For statistical analysis, cases with positive findings on ultrasonography and CT or on surgical evaluation were counted as true-positive (TP); those with negative findings on ultrasonography, CT, and surgical evaluation were counted as TN; those with positive findings on ultrasonography and negative findings on CT and surgical evaluation were counted as FP; and those with negative findings on ultrasonography and positive findings on CT or surgical evaluation were counted as FN. For analysis, Fisher’s exact probability test was used. Statistical significance was assumed to be reached at P ≤ .05.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Of the 604 patients studied, 466 were male and 138 were female; mean age ± SD was 32.9 ± 17.5 years (range, 2–94 years). The causes of BAT and the incidence of solid-organ injuries are listed in Table 2Go.


View this table:
[in this window]
[in a new window]
  		Table 2. Causes of BAT in 604 Patients and Incidence of Solid-Organ Injuries Identified by CT, Laparotomy, or Both in 198 Patients
 
There were 198 patients with solid-organ injuries identified on CT, laparotomy, or both. Multiple-organ injuries were detected in 37 patients. Laparotomies were performed in 91 patients. Surgical treatments, such as organ excision, partial resection, suturing, and removal of hematomas, were performed in the 4 upper abdominal organs in 68 patients (liver in 17, spleen in 22, kidney in 20, and pancreas in 9). Transarterial embolization (TAE) was performed in 3 patients with hepatic injuries. The intervals between sustaining injuries and the first ultrasonographic examinations were less than 48 hours in 97% of all patients.

Detection of Solid-Organ Injuries With Ultrasonography
To assess the probability of detecting injuries in individual organs, the numbers of TP, TN, FP, and FN cases were calculated simply on the basis of whether an injury was detected, and the types of injuries were not taken into account (Table 3Go). Because only 1 patient with a pancreatic injury (FN) was examined by group B, this case was not included in the assessment. The sensitivity of ultrasonography for detecting injuries in the liver, spleen, and kidney in group B was significantly lower than the sensitivity in group A.


View this table:
[in this window]
[in a new window]
  		Table 3. Diagnostic Capability of Ultrasonography for Solid-Organ Injuries (Groups A and B)
 
Ultrasonographic Diagnosis of Different Types of Injuries
Because multiple injuries were found in single organs in some patients, the diagnostic capability of ultrasonography was assessed in detail with the use of injuries (rather than patients) as the basis for evaluation and calculation. Ultrasonographic and CT images of some of the injuries are shown in Figures 1Go–4Go.




View larger version (320K):
[in this window]
[in a new window]
  		Figure 1. Type IIIb hepatic injury from a traffic accident in a 19-year-old man. A, Intercostal sonogram of the liver showing a poorly defined hyperechoic area (arrows) including an irregular anechoic area (arrowhead) in the right hepatic lobe. B, Computed tomographic scan showing a low-attenuation area with extravasation of contrast material (arrowhead) representing extensive parenchymal disruption of the liver. The patient underwent a right lobectomy of the liver.

 




View larger version (376K):
[in this window]
[in a new window]
  		Figure 4. Type IIIa pancreatic injury from a fall in a 30-year-old woman. A, Transverse sonogram of the pancreas showing heterogeneous parenchyma of the pancreatic body and enlargement of the body and tail (arrows). B, Sonographic image of a lower position than A showing a hypoechoic area representing a hematoma (arrows). C, Computed tomographic scan showing a pancreatic transection. The patient underwent pancreatic duct repair.

 
Hepatic Injuries
On the basis of the CT and surgical findings, 72 patients examined by group A were classified as follows: 5 type Ia, 38 type Ib, 4 type II, 3 type IIIa, 20 type IIIb, 1 type Ib + II, and 1 type IIIb + IIIb.

Table 4Go shows the numbers of different types of hepatic injuries classified with the use of ultrasonography by group A in comparison with those classified with the use of CT and surgical findings; it also shows the detection rates for the different types of hepatic injuries in groups A and B. The detection rate for type II was significantly lower (P = .001) than for the other (non–type II) injuries in group A. The detection rate for type IIIb was significantly higher (P = .002) than for the other (non–type IIIb) injuries in group B. The detection rate for type I injury in group B was significantly lower than the corresponding rate noted in group A (P = .00001). We also studied whether the types of individual injuries identified by ultrasonography agreed with those identified by CT and laparotomy in the 63 injuries detected by group A (Table 4Go). An agreement rate of 95.2% (60 of 63) was obtained.


View this table:
[in this window]
[in a new window]
  		Table 4. Numbers of Hepatic Injuries of Different Types Classified by Ultrasonography in Comparison With Those by CT and Surgical Findings (Group A) and Detection Rates (Groups A and B)
 
Splenic Injuries
Forty-one patients with splenic injuries were examined by group A. No patients had multiple injuries of the spleen. Table 5Go shows the numbers of different types of splenic injuries classified with the use of ultrasonography by group A in comparison with those classified with the use of CT; the table also shows the detection rates for the different types of splenic injuries in groups A and B. The detection rates for types IIIc and IIId were high in group B.


View this table:
[in this window]
[in a new window]
  		Table 5. Numbers of Splenic Injuries of Different Types Classified by Ultrasonography in Comparison With Those by Classified by CT (Group A) and Detection Rates (Groups A and B)
 
There were no isolated type IV injuries in the patient examined by either group A or group B. Hilar vessel injuries in association with splenic injuries were found during surgery in 5 patients. None of these patients had vascular injuries detected on ultrasonography. The agreement rate between the injury types identified by ultrasonography and those identified by CT was 80% (28 of 35) in group A (Table 5Go).

Renal Injuries
In 8 of 49 (TP and FN) patients examined by group A, parenchymal injuries were not detected with certainty because perirenal hematomas (7 patients) and extravasated urine (1 patient) alone were found on CT. Forty-one patients in which parenchymal injuries were detected with CT were classified as follows: 5 type I, 11 type II, 12 type IIIa, 3 type IIIb, 0 type IIIc, 4 type IVa, 2 type IVb, 2 type II + I, 1 type IVa + II, and 1 type IVb + IIIb. Thirteen (34.2%) of 38 patients (TP) had ultrasonographic diagnoses of only perirenal hematomas or extravasated urine, that is, without parenchymal abnormalities. The sensitivity and accuracy in detecting perirenal hematomas were 83.8% (31 of 37) and 98.0% (293 of 299), respectively, in 299 patients (37 patients accompanied by perirenal hematoma).

Table 6Go shows the numbers of different types of renal injuries classified with the use of ultrasonography by group A in comparison with those classified with the use of CT; it also shows the detection rates for different types of renal injuries in groups A and B. The detection rate for type II was significantly lower (P = .03) than for the other (non–type II) injuries in group A. The detection rate for type IV was significantly lower (P = .01) than for the other (non–type IV) injuries in group A. The only renal injuries detected by group B were type IIIb and IIIc injuries. The agreement rate between the types identified by ultrasonography and those identified by CT was 83.3% (20 of 24) in group A (Table 6Go).


View this table:
[in this window]
[in a new window]
  		Table 6. Numbers of Renal Injuries of Different Types Classified by Ultrasonography in Comparison With Those Classified by CT (Group A) and Detection Rates (Groups A and B)
 
Pancreatic Injuries
Nine patients with pancreatic injuries examined by group A were classified as follows: 5 type IIIa, 3 type I, and 1 type II on the basis of CT and surgical findings. Of these 9 patients, 4 type IIIa injuries were detected on ultrasonography. No type I or II injuries were detected even with CT. Such injuries were found during surgery for treatment of injuries to other organs. The detection rate was as low as 44.4% (4 of 9) as a whole, although that in detecting type IIIa injuries alone was 80% (4 of 5).

False-Negative and -Positive Cases
The FN and FP cases examined by group A were also analyzed. There were 11 FN injuries in the liver, 6 in the spleen, 21 in the kidney, and 5 in the pancreas. Table 7Go shows the numbers of different types of FN injuries to solid organs, their features, and treatments.


View this table:
[in this window]
[in a new window]
  		Table 7. False-Negative Cases of Solid-Organ Injuries (Group A)
 
There were 2 FP injuries in the liver and 1 in the kidney. In the liver, a hemangioma and a focal fatty infiltration were identified erroneously as type Ib injuries. The definitive diagnosis was obtained with CT and repeated ultrasonography. One FP case with a renal injury had a small type Ib injury suspected on the basis of the ultrasonographic findings, but it was not confirmed by the CT and repeated ultrasonography.

Injuries That Required Intervention
Seventy-one patients in which either surgery or TAE was performed on the liver, spleen, pancreas, or kidney included patients who underwent laparotomies to treat injuries in other organs. Sixty-three injuries (17 in the liver, 20 in the spleen, 20 in the kidney, and 6 in the pancreas) were diagnosed as being severe enough to require intervention. Of the 63 injuries, 36 were scanned by group A, and 27 were scanned by group B. The types of undetected injuries are listed in Table 8Go. Renal injuries accounted for 71.4% (10 of 14) of undetected injuries. The detection rates were 86.1% (31 of 36) in group A and 66.7% (18 of 27) in group B.


View this table:
[in this window]
[in a new window]
  		Table 8. Undetected Injuries in 63 Injuries That Required Intervention
 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Around 1990, the question of whether ultrasonography could be an alternative to DPL was discussed simply with respect to detection of intraperitoneal fluid.10–12 The sensitivity, specificity, and accuracy of FAST in detecting free fluid were hitherto reported to be 81% to 94%, 88% to 100%, and 86% to 98%, respectively.10–19 Miller et al20 reported low sensitivity of FAST (42%) from an analysis of 359 patients with BAT and concluded that the FAST examination resulted in underdiagnosis of intra-abdominal injuries in hemodynamically stable patients. Because hemoperitoneum was not found in 26% to 34% of patients with abdominal organ injuries, however, the limitations of FAST were pointed out in some reports.21–23 The FAST examination was also reported to be insufficient for evaluating children with BAT.24 In hemodynamically stable patients with negative FAST results, therefore, observation of at least 6 hours and follow-up FAST have been recommended to exclude abdominal injury.1 Hemodynamically stable patients with positive FAST results can be examined with CT. Conversely, patients who need emergency surgical treatment and have no time for a CT scan should be examined with ultrasonography in a search for injured organs. Moreover, ultrasonographic scanning will be useful for determining a treatment strategy and for following the clinical courses of patients if organ injuries are excluded or identified with ultrasonography at the first visit. Several authors have used ultrasonography to evaluate parenchymal injuries in addition to identifying intraperitoneal fluid. Because their attention was directed mainly to the presence of intraperitoneal fluid, however, they did not describe injuries in individual organs in detail. They reported 86% to 93% sensitivity, 98% to 100% specificity, and 96% to 99% accuracy for detection of injuries with ultrasonography.25–31 Röthlin et al32 reported sensitivity of 43.6% (36.4% by beginners and 45.2% by experienced examiners) for detecting abdominal organ injuries in 312 patients with blunt thoracic or abdominal trauma. The reason why the values reported by them were so low is that they examined not only solid-organ injuries but also small-bowel and mesenteric injuries, for which the detection rate was particularly low.

The diagnostic capability of ultrasonography increases or decreases depending on the skill and experience of the examiners. Because examiners with considerable skill and experience in performing and interpreting ultrasonography are not in-house for 24 hours a day for emergency examinations in our hospital, the physicians (including mostly resident surgeons with limited experience) have to perform ultrasonographic scans before and after normal office hours (namely, at night and on holidays). Accordingly, the diagnostic capability of ultrasonography can vary widely. Moreover, there are individual variations in the abilities of examiners with similar levels of experience. We hypothesized, therefore, that we could assess the average of the diagnostic capability by grouping examiners.

The probability of detecting injuries with ultrasonography was first evaluated on the basis of the number of patients in which the presence or absence of an injury could be correctly distinguished by ultrasonography. The types of injuries were not considered in this evaluation. Because there were a large number of TN cases, the specificity, accuracy, and negative predictive value were expected to be high even when the examinations were insufficient. Therefore, the parameters important for the evaluation were the sensitivity and positive predictive value. In 1998, we made a study of 1239 patients who had BAT within a 15-year period from October 1982 through December 1996.2 We reported sensitivity of 92.4% for the liver, 90.0% for the spleen, 92.2% for the kidney, and 71.4% for the pancreas. Compared with those values, the data for group A obtained in this study were considerably lower. This can be explained by the difference of the reference standards adopted for comparison with the ultrasonographic findings. In addition to CT and laparotomy, the repeated ultrasonographic examination within 24 hours and the clinical course were taken as reference standards in the prior study; therefore, it was possible that cases with minor injuries were considered negative, and cases that would have been counted as FNs were counted as TNs. As a consequence, the sensitivity became unduly high in the prior study. Several authors have pointed out this shortcoming and the fact that use of the clinical course as a reference standard unduly increases the sensitivity.33–36 Because contrast-enhanced CT and surgical findings were taken as reference standards in all patients in this study, patients with minor injuries who might have been counted as having TN results in the prior study were counted as having FN results in the present study. This 5% to 15% decrease in sensitivity may reveal the possibility of missing minor injuries in the liver, spleen, and kidney with ultrasonography. However, most minor injuries missed by ultrasonography have no clinical importance. We think that detection of a minor injury with ultrasonography is an indicator of the accuracy of the ultrasonographic examination.

Goletti et al37 attempted to detect intraperitoneal fluid and solid-organ injuries in 250 patients and reported sensitivities of 80% (8 of 10), 93% (29 of 31), 100% (3 of 3), and 66% (2 of 3) for the liver, spleen, kidney, and pancreas, respectively. Their reference standards included the clinical course. McGahan et al34 analyzed findings in 500 patients with abdominal trauma, using CT and surgical findings as reference standards, and reported detection rates of 14% (1 of 7), 69% (9 of 13), and 25% (1 of 4) for the liver, spleen, and kidney, respectively. They concluded that splenic injuries were most easily detected. Kuligowska et al38 compared the findings from ultrasonographic examinations with those from helical CT examinations from 268 patients with BAT. They detected 21 of 26 injuries in 22 patients and reported detection rates of 90% (9 of 10), 90% (9 of 10), and 50% (3 of 6) for the liver, spleen, and kidney, respectively. The results of the studies cited above cannot be compared directly with those of our study because the numbers of the injuries were small in those studies, and the degrees of experience of the examiners were unspecified.

The sensitivity of ultrasonography in detecting injuries in the liver, spleen, and kidney in group B was significantly lower than the sensitivity in group A, although there was no significant difference between the clinical outcomes for the patients in group A and those in group B. This difference in the sensitivity can be accounted for mainly by the variable levels of skill and experience among examiners in ultrasonography as a matter of course. The differences in the settings and conditions of the ultrasonographic examinations and the differences in the ultrasonographic equipment might affect the diagnostic capability between groups in some cases. The finding that the detection rates for solid-organ injuries were close to 50%, even though the examiners had only limited training and experience, indicates that the attempt to detect not only intraperitoneal fluid but also solid-organ injuries by ultrasonography is clinically important.

The proportions of type Ib and IIIb injuries were high among those with hepatic injuries. The detection rate for type II injuries (20%) was significantly lower than the rates for other types (80%–100%). This is accounted for by the difficult nature of imaging type II injuries, which are shallow (<3 cm in depth) and small in extent. The detection rate for type IIIb injuries was significantly higher than the rates for other types in group B. The detection rate for type I injuries in group B was significantly lower than that in group A. These facts indicate that, although more severe injuries are detected more easily by experienced as well as by inexperienced examiners, the severity of injuries (except type II) has no marked influence on imaging for experienced examiners. According to Richards et al,39 who studied 146 patients with blunt hepatic injuries, the sensitivity was 98% in detecting grade III or higher injuries (according to the liver injury scale).9 Such a high value may be attributable to the high detection rate for intraperitoneal fluid, given that the number of solid-organ injuries detected directly was only 18 (12%).39

The detection rates for splenic injuries of types I and III in group A were as high as 70% to 100%. Because the depth of parenchymal lesions is 2 to 3 mm in type II injuries, they cannot be imaged with ultrasonography and CT. Splenic FN injuries of type IIIa included isoechoic lesions and a small lesion located in the peripheral portion of the spleen. We think that detection of these injuries is beyond the capability of ultrasonography. The detection rate in group B was 50.0% but increased in severe injuries such as type IIIc and IIId injuries. Siniluoto et al40 reported a rate of 62.5% (35 of 56 patients) for detecting splenic parenchymal injuries and subcapsular hematomas by the primary ultrasonographic examination. They reported that the rate could be raised by repeated ultrasonography.

In contrast to injuries to the other solid organs, renal injuries have a characteristic feature: the diagnosis of renal injuries is made merely by finding a perirenal hematoma or extravasated urine even when the presence of parenchymal injuries has not been shown directly. This seems to be due to the morphologic complexity of the kidney, which makes detection of injuries difficult. The sensitivity in detecting perirenal hematomas was 83.8%, but, conversely, the direct detection rate of renal parenchymal injuries was as low as 53.3% in group A. It was difficult to detect type II injuries (superficial lacerations) because they were small. In addition, it was also possible that the examination of the renal parenchyma was insufficient after a perirenal hematoma was detected. The type IV (vascular) injuries were detected less than the other injury types. Detection of vascular injuries seems to be possible with color Doppler ultrasonography. The direct detection rate of renal parenchymal injuries was only 17.1% in group B, but the detection rate for type IIIb and IIIc injuries was not low (75%).

McGahan et al41 reported 37 injuries in 32 patients with blunt renal trauma; all these ultrasonographic findings were compared with CT or surgical findings. The direct detection rate for renal parenchymal injuries was as low as 22% (8 of 37). The rates for grade I, II, and III injuries in the CT classification of Mirvis42 were 15% (3 of 20), 17% (2 of 12), and 60% (3 of 5), respectively. Grade I in the Mirvis classification corresponds fairly well to types I and II in our study, and grade II in the Mirvis classification corresponds fairly well to types IIIa and IIIb in our study, whereas grade III corresponds fairly well to types IIIc and IV. Disagreement between the types of renal injuries identified by ultrasonography and by CT was found with respect to differentiation of type II injuries from type IIIa injuries. The difference between type II and IIIa injuries depends on whether an injury is superficial or deep, namely, whether a laceration reaches the collecting system. This determination cannot be made with ultrasonography, which is unable to visualize the collecting system directly unless the system is distended.

Detection of pancreatic injuries of types I and II by diagnostic imaging is difficult. Type III injuries are so severe that surgical interventions are always required. We should satisfy at present that type III injuries can be detected with ultrasonography even if types I and II injuries cannot be detected. The pancreas was examined via ultrasonography in only a few reports. This fact probably results from the idea that the pancreas should be examined via CT, because imaging of the pancreas with ultrasonography is affected by bowel gas and the patient’s body habitus. Although imaging of the pancreatic tail is difficult, severe injuries occur most frequently in the pancreatic body, which is imaged relatively clearly. The probability of detecting pancreatic injuries with ultrasonography seems to be high, therefore. Active attempts should be made to examine the pancreas with ultrasonography.

The number of undetected injuries that required intervention such as laparotomy or TAE was the largest for injuries of the kidney. This was because vascular injuries (type IV) accounted for most undetected injuries. Except for 8 type IV renal injuries, the detection rates of intervention-requiring injuries in the 4 organs were 89.1% (49 of 55) when the results of groups A and B were combined and 96.9% (31 of 32) when calculated from the results of group A alone. It seems unlikely, therefore, that an experienced examiner fails to detect severe injuries requiring laparotomy by using ultrasonography, including the color Doppler method.

An international consensus conference1 stated the potential advantages of ultrasonography for parenchymal lesions in the following respects: ultrasonography can help determine the need for laparotomy or can help guide the approach to surgery in unstable patients who cannot undergo CT. In stable patients, CT may be avoided with an accurate and reliable organ-specific ultrasonographic study. The conference further stated the clinical importance of attempts to detect solid-organ injuries directly with ultrasonography because hemoperitoneum was undetectable in some patients with abdominal injuries. The conference concluded, however, that because most intraparenchymal lesions do not require intervention and may never become clinically serious, it is not necessary to perform ultrasonographic scanning for organ-specific injury, which depends on the skill and experience of examiners, and it may be impractical for surgeons to acquire and maintain sufficient skill to identify lesions by ultrasonography when CT is readily available. Accordingly, a dominant opinion at the conference was that the need for laparotomy should be determined in unstable patients with the FAST procedure, whereas stable patients should be examined in detail by CT.1

Recently, multi-detector-row CTs have allowed faster scanning than conventional CT; therefore, the diagnostic capability of CT for injuries has increased further. Some physicians may deny the necessity of ultrasonographic scanning for patients with trauma. However, the situation can arise in which ultrasonography is necessary to investigate the cause of a patient’s illness during emergency treatment of an unstable patient who cannot undergo CT. In addition, an artifact caused by movement of the body or by the upper arm that cannot be raised or by metallic foreign bodies in the abdomen may obscure CT imaging at times. Ultrasonographic examiners are required to have skills to cope with such situations. Scanning for abdominal trauma with the FAST procedure takes only several minutes. If an examiner has sufficient daily experience in scanning the abdomen (including detection of pleural and pericardial fluids), the examiner can complete an ultrasonographic scan within 10 minutes even if the extent of scanning is expanded to cover the solid organs (liver, spleen, kidney, and pancreas). Detection of free air is also possible in a certain percentage of cases. Searching for parenchymal abnormalities as well as free fluid improved the sensitivity of ultrasonography in some reports,23,43 but, conversely, ultrasonography was limited mainly by its low sensitivity for identifying organ injuries in hemodynamically stable patients in other reports.44 As shown in this study, experienced examiners do not fail to detect any serious injuries to the solid abdominal organs, and identification of the types of individual injuries on ultrasonography by experienced examiners agrees with CT-based identification in most detected injuries. This study has also shown that examiners with limited experience can detect hepatic, splenic, and renal injuries at rates of 44% to 50% (sensitivity) and can detect many severe injuries. Therefore, we concede that training not only in the FAST procedure but also in scanning organs is effective and necessary.

In conclusion, ultrasonography enables experienced examiners to detect and classify parenchymal injuries efficiently, despite disadvantages in detecting superficial and vascular injuries, and should be used to explore not only free fluid but also solid-organ injuries. Examiners with limited experience (such as residents) should also try to perform not only the FAST procedure but also scanning of solid organs to improve their skills. We think that ultrasonography can play an important role in screening for solid-organ injuries in patients with mild trauma who do not need to undergo CT immediately and in determination of organ-specific injury in unstable patients who cannot undergo CT.




View larger version (289K):
[in this window]
[in a new window]
  		Figure 2. Splenic injury from an assault in a 15-year-old boy. A, Intercostal sonogram of the spleen showing heterogeneous parenchyma including small hypoechoic areas. B, Computed tomographic scan showing splenic lacerations (arrows) with hemoperitoneum (arrowheads). The patient was treated conservatively.

 



View larger version (256K):
[in this window]
[in a new window]
  		Figure 3. Type IIIa renal injury from a traffic accident in a 6-year-old boy. A, Longitudinal sonogram of the right kidney showing a poorly defined echogenic area with an irregular anechoic area in the midpole of the kidney (arrows). B, Computed tomographic scan showing a low-attenuation area in the renal parenchyma (arrow) with perirenal hematoma (arrowhead), representing a deep laceration. The patient was treated conservatively.

 

   Footnotes
 
Received April 12, 2004, from the Departments of Radiology (M.S.) and Surgery (H.Y.), Saiseikai Kanagawaken Hospital and Kanagawaken Traffic Trauma Center, Yokohama, Japan. Revision requested June 22, 2004. Revised manuscript accepted for publication August 25, 2004.

We thank the ultrasound group of the Department of Clinical Laboratory, Saiseikai Kanagawaken Hospital, for data acquisition.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. FAST Consensus Conference Committee. Focused assessment with sonography for trauma (FAST): results from an international consensus conference. J Trauma 1999; 46:466–472.[Medline]
  2. Yoshii H, Sato M, Yamamoto S, et al. Usefulness and limitations of ultrasonography in the initial evaluation of blunt abdominal trauma. J Trauma 1998; 45:45–51.[Medline]
  3. Yamamoto S, Kasai T, Maekawa K, Yoshii H, Hepatic Injury Classification Committee of the Japanese Association for the Surgery of Trauma. Classification of hepatic injuries according to the Japanese Association for the Surgery of Trauma [in Japanese]. J Jpn Assoc Surg Trauma 1997; 11:29.
  4. Maekawa K, Suzuki T, Yasuda K, Yoshii H, Splenic Injury Classification Committee of the Japanese Association for the Surgery of Trauma. Classification of splenic injuries according to the Japanese Association for the Surgery of Trauma [in Japanese]. J Jpn Assoc Surg Trauma 1997; 11:30.
  5. Tabuse H, Tohma Y, Okamoto K, et al, Renal Injury Classification Committee of the Japanese Association for the Surgery of Trauma. Classification of renal injuries according to the Japanese Association for the Surgery of Trauma [in Japanese]. J Jpn Assoc Surg Trauma 1997; 11:32–33.
  6. Henmi H, Maekawa K, Motegi M, Yasuda K, Kasai T, Sakamoto T, Pancreatic Injury Classification Committee of the Japanese Association for the Surgery of Trauma. Classification of pancreatic injuries according to the Japanese Association for the Surgery of Trauma [in Japanese]. J Jpn Assoc Surg Trauma 1997; 11:31.
  7. Moore EE, Shackford SR, Pachter HL, et al. Organ injury scaling: spleen, liver, and kidney. J Trauma 1989; 29:1664–1666.[Medline]
  8. Moore EE, Cogbill TH, Malangoni MA, et al. Organ injury scaling, II: pancreas, duodenum, small bowel, colon, and rectum. J Trauma 1990; 30:1427–1429.[Medline]
  9. Moore EE, Cogbill TH, Jurkovich GJ, Shackford SR, Malangoni MA, Champion HR. Organ injury scaling: spleen and liver (1994 revision). J Trauma 1995; 38:323–324.[Medline]
  10. Grüessner R, Mentges B, Düber Ch, Rückert K, Rothmund M. Sonography versus peritoneal lavage in blunt abdominal trauma. J Trauma 1989; 29:242–244.[Medline]
  11. Kimura A, Otsuka T. Emergency center ultrasonography in the evaluation of hemoperitoneum: a prospective study. J Trauma 1991; 31:20–23.[Medline]
  12. Hoffmann R, Nerlich M, Muggia-Sullam M, et al. Blunt abdominal trauma in cases of multiple trauma evaluated by ultrasonography: a prospective analysis of 291 patients. J Trauma 1992; 32:452–458.[Medline]
  13. Tso P, Rodriguez A, Cooper C, et al. Sonography in blunt abdominal trauma: a preliminary progress report. J Trauma 1992; 33:39–44.[Medline]
  14. Liu M, Lee CH, P’eng FK. Prospective comparison of diagnostic peritoneal lavage, computed tomographic scanning, and ultrasonography for the diagnosis of blunt abdominal trauma. J Trauma 1993; 35: 267–270.[Medline]
  15. McKenney M, Lentz K, Nunez D, et al. Can ultrasound replace diagnostic peritoneal lavage in the assessment of blunt trauma? J Trauma 1994; 37: 439–441.[Medline]
  16. Ma OJ, Mateer JR, Ogata M, Kefer MP, Wittmann D, Aprahamian C. Prospective analysis of a rapid trauma ultrasound examination performed by emergency physicians. J Trauma 1995; 38:879–885.[Medline]
  17. Boulanger BR, Brenneman FD, McLellan BA, Rizoli SB, Culhane J, Hamilton P. A prospective study of emergent abdominal sonography after blunt trauma. J Trauma 1995; 39:325–330.[Medline]
  18. Rozycki GS, Ochsner MG, Schmidt JA, et al. A prospective study of surgeon-performed ultrasound as the primary adjuvant modality for injured patient assessment. J Trauma 1995; 39:492–500.[Medline]
  19. Lingawi SS, Buckley AR. Focused abdominal US in patients with trauma. Radiology 2000; 217:426–429.[Abstract/Free Full Text]
  20. Miller MT, Pasquale MD, Bromberg WJ, Wasser TE, Cox J. Not so FAST. J Trauma 2003; 54:52–60.[Medline]
  21. Chiu WC, Cushing BM, Rodriguez A, et al. Abdominal injuries without hemoperitoneum: a potential limitation of focused abdominal sonography for trauma (FAST). J Trauma 1997; 42:617–625.[Medline]
  22. Shanmuganathan K, Mirvis SE, Sherbourne CD, Chiu WC, Rodriguez A. Hemoperitoneum as the sole indicator of abdominal visceral injuries: a potential limitation of screening abdominal US for trauma. Radiology 1999; 212:423–430.[Abstract/Free Full Text]
  23. Brown MA, Casola G, Sirlin CB, Hoyt DB. Importance of evaluating organ parenchyma during screening abdominal ultrasonography after blunt trauma. J Ultrasound Med 2001; 20:577–583.[Abstract/Free Full Text]
  24. Coley BD, Mutabagani KH, Martin LC, et al. Focused abdominal sonography for trauma (FAST) in children with blunt abdominal trauma. J Trauma 2000; 48:902–906.[Medline]
  25. Bode PJ, Niezen RA, van Vugt AB, Schipper J. Abdominal ultrasound as a reliable indicator for conclusive laparotomy in blunt abdominal trauma. J Trauma 1993; 34:27–31.[Medline]
  26. McKenney MG, Martin L, Lentz K, et al. 1,000 consecutive ultrasounds for blunt abdominal trauma. J Trauma 1996; 40:607–612.[Medline]
  27. Healey MA, Simons RK, Winchell RJ, et al. A prospective evaluation of abdominal ultrasound in blunt trauma: is it useful? J Trauma 1996; 40: 875–885.[Medline]
  28. McElveen TS, Collin GR. The role of ultrasonography in blunt abdominal trauma: a prospective study. Am Surg 1997; 63:184–188.[Medline]
  29. McKenney KL, Nuñez DB Jr, McKenney MG, Asher J, Zelnick K, Shipshak D. Sonography as the primary screening technique for blunt abdominal trauma: experience with 899 patients. AJR Am J Roentgenol 1998; 170:979–985.[Abstract/Free Full Text]
  30. Bode PJ, Edwards MJR, Kruit MC, van Vugt AB. Sonography in a clinical algorithm for early evaluation of 1671 patients with blunt abdominal trauma. AJR Am J Roentgenol 1999; 172:905–911.[Abstract/Free Full Text]
  31. Dolich MO, McKenney MG, Varela JE, Compton RP, McKenney KL, Cohn SM. 2,576 ultrasounds for blunt abdominal trauma. J Trauma 2001; 50:108–112.[Medline]
  32. Röthlin MA, Näf R, Amgwerd M, Candinas D, Frick T, Trentz O. Ultrasound in blunt abdominal and thoracic trauma. J Trauma 1993; 34:488–495.[Medline]
  33. Pearl WS, Todd KH. Ultrasonography for the initial evaluation of blunt abdominal trauma: a review of prospective trials. Ann Emerg Med 1996; 27:353–361.[Medline]
  34. McGahan JP, Rose J, Coates TL, Wisner DH, Newberry P. Use of ultrasonography in the patient with acute abdominal trauma. J Ultrasound Med 1997; 16:653–662.[Abstract]
  35. McGahan JP, Richards JR. Blunt abdominal trauma: the role of emergent sonography and a review of the literature. AJR Am J Roentgenol 1999; 172:897–903.[Free Full Text]
  36. Benya EC, Lim-Dunham JE, Landrum O, Statter M. Abdominal sonography in examination of children with blunt abdominal trauma. AJR Am J Roentgenol 2000; 174:1613–1616.[Abstract/Free Full Text]
  37. Goletti O, Ghiselli G, Lippolis PV, et al. The role of ultrasonography in blunt abdominal trauma: results in 250 consecutive cases. J Trauma 1994; 36:178–181.[Medline]
  38. Kuligowska E, Barish MA, Soto JA, et al. Real-time ultrasound versus spiral computed tomography in the evaluation of blunt abdominal trauma [abstract]. J Ultrasound Med 1996; 15(suppl):S:33.[Abstract]
  39. Richards JR, McGahan JP, Pali MJ, Bohnen PA. Sonographic detection of blunt hepatic trauma: hemoperitoneum and parenchymal patterns of injury. J Trauma 1999; 47:1092–1097.[Medline]
  40. Siniluoto TMJ, Päivänsalo MJ, Lanning FP, Typpö ABT, Lohela PK, Kotaniemi AE. Ultrasonography in traumatic splenic rupture. Clin Radiol 1992; 46:391–396.[Medline]
  41. McGahan JP, Richards JR, Jones CD, Gerscovich EO. Use of ultrasonography in the patient with acute renal trauma. J Ultrasound Med 1999; 18:207–213.[Abstract]
  42. Mirvis SE. Trauma. Radiol Clin North Am 1996; 34:1225–1257.[Medline]
  43. Richards JR, Schleper NH, Woo BD, Bohnen PA, McGahan JP. Sonographic assessment of blunt abdominal trauma: a 4-year prospective study. J Clin Ultrasound 2002; 30:59–67.[Medline]
  44. Poletti PA, Kinkel K, Vermeulen B, Irmay F, Unger PF, Terrier F. Blunt abdominal trauma: should US be used to detect both free fluid and organ injuries? Radiology 2003; 227:95–103.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 RadioGraphicsHome page
R. C. Alonso, S. B. Nacenta, P. D. Martinez, A. S. Guerrero, and C. G. Fuentes
Kidney in Danger: CT Findings of Blunt and Penetrating Renal Trauma
RadioGraphics, November 1, 2009; 29(7): 2033 - 2053.
[Abstract] [Full Text] [PDF]

Home page
 RadioGraphicsHome page
M. Korner, M. M. Krotz, C. Degenhart, K.-J. Pfeifer, M. F. Reiser, and U. Linsenmaier
Current Role of Emergency US in Patients with Major Trauma
RadioGraphics, January 1, 2008; 28(1): 225 - 242.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sato, M.
Right arrow Articles by Yoshii, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sato, M.
Right arrow Articles by Yoshii, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS