An accurate assessment of patients with potential thoracoabdominal injury includes a safe and reliable method to determine the need for operative intervention. However, the low sensitivity of physical examination alone has encouraged clinician reliance on diagnostic modalities. Both diagnostic peritoneal lavage (DPL) and computed tomography (CT) have been evaluated prospectively and retrospectively to determine their accuracy in detecting hemoperitoneum after injury and their roles in patient evaluation. [1-15] Based upon such evaluations, these modalities have gained general acceptance with widely understood indications and limitations. Ultrasound diagnostic imaging having been used in Germany in the trauma setting for more than 15 years [16-23] has unique qualities that give it distinct advantages over the other tests and is gradually gaining acceptance by surgeons in the United States. [24,25]
We hypothesized that experienced surgeon sonographers could use the focused ultrasound examination to detect hemoperitoneum/pericardial effusion in injured patients and then use the results in the decision matrix for patient care. The objectives of this prospective study were to assess the accuracy and utility of surgeon-performed ultrasound as the primary adjuvant modality for assessment of injured patients and determine if by using ultrasound in this manner, any potential for cost savings would be appreciated.
MATERIALS AND METHODS
Over a 20-month period, ultrasound was prospectively evaluated as the primary adjuvant modality to physical examination of the chest and abdomen at an urban Level I trauma center. To enter the study, patients had to have a suspected torso injury, physical examination, and the need for a diagnostic modality for complete assessment. Patients who were hemodynamically unstable necessitating immediate operative intervention were excluded from the study. The portable Ultramark IV ultrasound system (Advanced Technology Laboratories, Bothwell, Washington) with a 3.0 MHz transducer and photographic capabilities was used for all studies. Ultrasound examinations were done by attending general surgeons, each of whom had more than three years of experience using ultrasound techniques, or by trauma fellows and senior general surgical residents who were trained by the same attendings. Fellows and residents were first instructed by the experienced examiners in didactics including ultrasound physics and techniques as previously reported.  They then participated in practice sessions that focused on the limited ultrasound examination to identify fluid (blood) in three dependent abdominal areas and in the pericardial sac in trauma patients. 
All sonograms were performed with the patient in the supine position during the secondary survey and within 30 minutes of the patient's arrival in the resuscitation center. The ultrasound examination was performed during the resuscitation simultaneously with physical examination and stabilization so as not to interfere with or delay patient care. [24,26] Although minimal patient preparation was required, a standard protocol for the performance of the examination was followed as previously described. 
After each patient's name and medical record number were entered, the focused ultrasound examination was conducted in the following sequence: (1) pericardial area, (2) right upper quadrant, (3) left upper quadrant, and (4) pouch of Douglas (Figure 1).
When a good quality view was obtained of each area, the automatically timed and dated image, was ``frozen'' and photographed. A good quality view was defined as one that demonstrated the correctly imaged sagittal or coronal section of the body region so that the presence or absence of fluid could be determined. The findings, either positive or negative for fluid, were recorded along with the patient's demographic data and the initials of the surgeon sonographer. For patients with blunt mechanisms, the attending surgeon recorded whether DPL ($150) or CT ($650) would have been chosen to evaluate the patient if ultrasound had not been an option.
If the ultrasound examination was negative for fluid, repeat physical and ultrasound examinations were performed by the same examiner within 12 to 24 hours of the patient's admission, and the results of this follow-up were recorded. Patients were followed through discharge and as outpatients in the clinic, one week after discharge. Inpatient observation was a minimum of 20 hours but more often days to weeks. If the ultrasound demonstrated fluid, the results were noted, surgical exploration was performed, and operative findings were recorded. Findings of hemoperitoneum or hemopericardium were considered positive. Hard copies of all ultrasound examinations were reviewed by the principal investigator, who commented on technique, accuracy of reading, and overall quality of the image. Results were categorized as true negative, true positive, false positive, and false negative (Table 1). This study was approved by the Washington Hospital Center's Institutional Review Board; however, informed consent was not required.
Ultrasound examinations were performed on 371 patients. The average age was 33 years (range 15 to 88), and the majority (73%) were males. Of the total, 295 patients (79.5%) suffered blunt injuries and the remaining 76 had penetrating injuries. The average Injury Severity Score (ISS) was 10, and the Revised Trauma Score (RTS) was 7.56. The average examination time was 2.5 minutes.
Ultrasound evaluations were characterized as 305 true negatives, 53 true positives, 12 false negatives, and 1 false positive. The overall sensitivity of ultrasound in detecting fluid (blood) was 81.5%, with 99.7% specificity. Analysis by etiology, however, showed a 78.6% sensitivity with 100% specificity for patients with blunt injuries. For the patients with penetrating injuries, ultrasound showed an 83.8% sensitivity and 97.4% specificity. The positive predictive value relative to the detection of hemoperitoneum/pericardial effusion was 98.1% and the negative predictive value, 96.2%. The accuracy of ultrasound in detecting fluid was 96.5%, however if the repeat ultrasound was eliminated, the accuracy was 95.6%.
Three hundred-five patients, six of whom were pregnant, had true negative ultrasound scans. All were followed through discharge, and 63% returned to clinic for a follow-up examination. These patients had no adverse abdominal or cardiovascular sequelae.
Fifty-three patients had true positive scans. Two of these patients had positive scans on repeat examination, one at 2 hours and the other at 12 hours after evaluation. Of the 53 patients, 13 died, 6 from exsanguination during exploration, 5 from severe intracranial injuries, and 2 from multisystem organ dysfunction.
(Table 2) lists the number of traumatized organs in the blunt-injured patients with true positive scans. Although most of the patients in this group had solid organ injuries, fluid from hollow viscus rupture was identified on ultrasound examination in four patients. Ultrasound also detected fluid from injuries to hollow viscera in nine patients with penetrating injuries. Six of the 31 patients with penetrating injuries had pericardial tamponade. Once the diagnosis was made, preoperative preparations were expedited and the average time from patient admission to arrival in the operating room was only 23 minutes.
Six patients with blunt injuries and six with penetrating injuries had false negative ultrasound scans. All of the patients with penetrating injuries had surgical explorations based on a mechanism of injury (Table 3). No operation was delayed and all operations were therapeutic.
All six patients with blunt injuries underwent exploration due to changes in clinical examination or the results of another diagnostic test deemed necessary (Table 4).
The patient with a false positive examination result had suffered a stab wound to the parasternal area and was hypotensive on admission. The ultrasound examination demonstrated a pericardial effusion confirmed at exploration, but neither a hemopericardium nor an injury was found.
A tally of the surgeons' diagnostic alternatives (if ultrasound had not been available) for the patients with blunt injuries showed that 194 patients (65.8%) would have had DPL and 101 (34.2%) would have had CTs. Although not intended as a cost/benefit analysis, this assessment suggests the potential for cost savings when ultrasound is used in this manner.
Widely used in Europe to evaluate patients with thoracoabdominal injuries, ultrasound is gradually becoming part of U.S. surgeons' diagnostic armamentarium as well. [24,25] Its user friendly technology has encouraged further interest and studies to assess the full value of ultrasound in the emergent setting. [27-30] Having previously mastered the focused ultrasound examination for the detection of hemoperitoneum/pericardial effusion in traumatized patients,  experienced surgeon sonographers successfully used ultrasound as the primary adjuvant and often only diagnostic modality for the evaluations of these patients.
The majority (82.2%) of the patients in the study were true negatives, suggesting that hemoperitoneum is not common in such a trauma patient population. Similar results have been reported in our previous study and by others. [21,24,26,31,32] Accordingly, this information is important in defining the role for ultrasound as a screening tool in injured patients.  Because hemoperitoneum is uncommon in this patient population, the use of a noninvasive, portable screening modality that can rapidly determine its presence is more practical, timesaving, and potentially more cost effective.
The optimal time for the repeat ultrasound examination, if necessary, is currently unknown. Recommendations such as that by Rothlin and co-workers include follow-up examinations every one to two hours for the first six hours after admission and then every 12 hours for two days.  Overall, such labor-intensive procedures are not supported by patient data and the importance of physical examination is inadequately emphasized. In our study, the 12- to 24-hour interval for repeat examination was chosen based on our experience of caring for these patients. Besides the repeat ultrasound, physical examinations were regularly performed and other parameters such as hematocrit were followed. This ensured that decisions regarding patient management were based on the entire clinical picture rather than on ultrasound alone. 
A potential weakness in the study was the limited patient follow-up after discharge. Because only 63% of patients returned to clinic, it is conceivable that an injury was missed. To our knowledge however, this did not occur. Furthermore, achievement of follow-up on all patients would have been unlikely considering the patient population. Another deficiency was the lack of a consecutive patient enrollment into the study. This occurred because not all of our attending surgeons were proficient in ultrasound techniques. Consequently, the data may be biased, but, overall, are typical of our trauma patient population.
Although all patients in the true positive group underwent therapeutic operations, the question arises whether all blunt injured patients with positive ultrasound examinations require surgical intervention. Several authors have investigated this question relative to the quantity of fluid (blood) identified on ultrasound and the need for celiotomy. [23,26,31,35,36] For example, in a study by Bode and colleagues, patients with blunt injuries were managed non-operatively if their ultrasound examinations showed the absence of large or increasing amounts of intra-abdominal fluid.  In their study, ``large'' was neither defined nor quantified, consequently, the reproducibility of this methodology is unpredictable. Because ultrasound does not distinguish between types of body fluids such as, succus entericus or blood, nonoperative management based on ultrasound alone may be unreliable. This was typified in Bode's study by a patient who was observed and later diagnosed with small bowel injury.  Data from our previous work confirmed that ultrasound was sensitive enough to detect fluid from isolated hollow viscus lacerations in patients with blunt injuries and several patients in this study had similar findings. The combination of ultrasound to screen for fluid (blood) and CT to distinguish the injury may allow for nonoperative management of select patients with solid organ injuries.
A review of the 47 true positivies (intra-abdominal fluid only) demonstrated that Morison's pouch was the most common region for fluid accumulation. Only one patient in our study was an exception to this pattern and he had a large hemoperitoneum secondary to a stab wound to the left ninth intercostal space, which lacerated an intercostal artery (Figure 2) The tendency for fluid to settle in Morison's pouch was previously confirmed by others. [16,37] This is a valuable indicator because the documentation of fluid in one region may be adequate indication for surgical intervention, especially in a hypotensive patient. 
Although ultrasound is not necessary to evaluate all patients with penetrating injuries, several reasons for the performance of the examination in this patient population are: (1) rapid screening to identify life-threatening injuries in patients with multiple wounds, (2) triage of multiple simultaneous trauma admissions, and (3) detection of a massive pleural effusion (hemothorax) without the need for a chest radiograph.  Furthermore, in the training environment, practice--using the proper technique--makes perfect. Because ultrasound is operator dependent, [17,35] those learning to do the focused examination have more opportunities to observe positive examinations that generally occur more often in patients with penetrating injuries. However, the best reason to do ultrasound in penetrating injured patients is for the rapid diagnosis and thus expedited surgical intervention for traumatic hemopericardium. Because the outcome of these patients is so dependent on early recognition and swift operative intervention, ultrasound in this setting is the modality of choice for evaluating these patients. [24,40-42]
A review of the six false negative results among patients with penetrating injuries (Table 3) demonstrated that three scans were of poor quality due to improper setting of the power control. The abdominal exploration of patient M.D. showed a small hemoperitoneum, a grade II hepatic laceration, and a perforated diaphragm. For the remaining two patients (D.B., J.D.), a plausible explanation for their negative ultrasound findings may be that the examination was done so soon after injury that hemorrhage was not yet significant enough to give a positive result. This has been cited by others as a possible explanation for some of their false negative results.  Serial ultrasound examinations, as recommended by several authors, [17,26,31,35,36] might have decreased the false negative rate for this subset of patients, but considering the need for operative intervention, this was not applicable to penetrating injured patients.
Of the six false negative results among patients with blunt injuries (Table 4), two patients (J.A., S.B.), had equivocal abdominal examinations despite negative ultrasound studies. CT and DPL were performed, both yielding positive results. The initial ultrasound of another patient (J.S.) was negative, as was the repeat examination performed 35 minutes later for abdominal pain. Due to the patient's symptoms and the diagnosis of thoracic spine fracture, a CT of the abdomen was performed and a splenic injury was diagnosed. One patient (J.T.) with a severe closed head injury had a negative repeat ultrasound examination 12 hours after admission. Due to a subsequent decrease in hematocrit, a CT was performed that demonstrated a splenic injury. Ultrasound examinations in the two remaining patients (C.B., J.M.) were performed improperly. Adverse patient outcomes were avoided in these situations because sound clinical judgement was used for patient management.
Although ultrasound did not detect fluid in all cases, no particular injury resulting in fluid was consistently missed. Instead of attempting to delineate the areas in which ultrasound is not consistently accurate, clinicians should exercise the same clinical judgement when using ultrasound that they use with other diagnostic modalities. 
Critique of False Negative/Positive Results
The 12 false negatives and 1 false positive were reviewed to decide whether they could have been avoided. Issues that surfaced were timing of repeat examinations, and examiner experience.
Although two patients in the true positive group had negative initial ultrasound examinations that later converted to positive (2 hours and 12 hours, respectively), these numbers are too few to draw any conclusions. Whereas some authors recommend frequent repeat examinations, some 30 minutes after admission and then every 2 hours, [26,31,35] this was not practical for our trauma service. Additional prospective studies should be performed to determine if the time for repeat ultrasound examinations could be better defined.
The learning curve undoubtedly contributed to the fact that the sensitivity rate was not higher than that in our previous study.  Although surgical residents quickly became proficient in the performance of the examination, they rotated on the trauma service for only 3 months and then a new group required training. Both experienced surgeon sonographers, each of whom had completed more than 400 examinations, had no false positive results.
Ultrasound: The Initial Modality
In our practice, initiation of surgeon-performed ultrasound examinations had led to a decrease in the performance of DPL and CT examinations. During this study, the experienced surgeon sonographers performed (or supervised) only three DPLs. CT however, was used more frequently. The cost for an ultrasound examination is modest in comparison to the other modalities. When performed by a surgeon in the resuscitation area, technician and radiologist fees are eliminated. Indirectly, revenue is gained because resources are conserved when resuscitation time is reduced (as seen in those patients diagnosed with pericardial tamponade) and when items such as central venous catheters, nasogastric tubes, and urinary catheters are used less frequently. The cost of the machine varies with its level of sophistication, but once this initial investment is made, photographic paper and ultrasound transmission gel are essentially the only other expenditures. After credentialing and continuous quality improvement (CQI) mechanisms are established, billing for the surgeon-performed ultrasound examination is justified.
Although sonologists may be willing to provide around-the-clock-in-hospital service, having the surgeon perform and interpret the ultrasound examination has distinct advantages. In doing so, ultrasound becomes part of the surgeon's diagnostic armamentarium, in essence, a part of the physical examination. Therefore, by doing the ultrasound, the surgeon has immediate additional information for patient evaluation without the delay of the technician's performance and radiologist's interpretation of the test.
No single diagnostic modality is universally applicable for the evaluation of trauma patients. Therefore, the decision matrix for which modality to use should incorporate information based on the mechanism of injury, hemodynamic stability of the patient, and availability of/expertise needed for the test. [44,45] Our acquired experience with ultrasound has convinced us of its advantages and has significantly influenced this facet of our approach to the injured patient. Because ultrasound is rapid, portable, noninvasive, repeatable and does not expose the patient to ionizing radiation, we found it to be frequently more suitable for the evaluation of injured patients than DPL or CT.
The value of rapidity was demonstrated in many multisystem-injured patients in the true positive group who presented with several potential sources of blood loss. Ultrasound was used rapidly to assess the patient for the presence or absence of hemoperitoneum/hemopericardium, allowing the surgeon to set priorities for resuscitation and evaluation.
Another valuable role for ultrasound is for the evaluation of the injured pregnant patient. Although no diagnostic modality deemed necessary for maternal evaluation should be withheld for concern of a hazard to the fetus, apprehension about the effects of radiation exposure to the fetus exists. The period of organogenesis, the first 8 weeks of pregnancy, is the crucial time for injury from radiation exposure.  Because patients may be unaware that they are pregnant during this early stage, the use of a nonionizing radiation modality is preferable.
Although not included in this study, the unique qualities of ultrasound make it very suitable for the pediatric patient. [47,48] This painless, noninvasive modality is well accepted by children because, in contrast to CT, it is performed at the bedside and is, therefore, less intimidating.
Credentialing and Continuous Quality Improvement
As with any test used in patient care, credentialing and CQI must be addressed. Ultrasonography relates to these issues because patient management decisions are partially based on the result of the diagnostic test. The surgeon should, therefore, accept responsibility for the evaluation and interpretation of the ultrasound examination.
The American Institute of Ultrasound Medicine (AIUM) has issued a statement regarding ultrasound training guidelines. It recommends that physicians who do diagnostic ultrasound examinations document CME activity (specific for ultrasonography) and evaluation and interpretation of 500 examinations.  This statement, however encompasses all disciplines of ultrasonography. Other sub-specialties represented by The American Board of Cardiology  and the Society of OB/Gyn Ultrasonography  have less stringent recommendations than those proposed by the AIUM. Furthermore the surgeon's use of ultrasound is very specific; i.e., to detect the presence of fluid in the abdomen or pericardial sac. Although an adequate background in physics and instrumentation is necessary, the limited scope of these focused examinations does not require extensive training. The technology, although it originated in the radiology department, is routinely used by cardiologists, obstetricians, ophthalmologists, urologists, and emergency medicine physicians and, therefore, is not domain specific.
Recognizing the value of this technology, a position statement was published by emergency medicine physicians on the importance of training and credentialing in ultrasonography. [52,53] Recently, the American College of Emergency Physicians and the Society for Academic Emergency Medicine organized an Ultrasound Task Force, formulated credentialing criteria, and published a document that defines a curriculum for the training of emergency medicine residents in ultrasonography.  General surgeons with expertise in trauma should pursue a similar course to establish reasonable guidelines for ultrasound credentialing that would include didactics, practice sessions, and monitoring of actual examinations. Additionally, to ensure that it is one of the surgeon's diagnostic tools, we suggest that ultrasound instruction be incorporated into general surgical residency training as it is in Germany. 
An integral part of the credentialing process is the issue of continuous quality improvement. The following are suggestions regarding CQI and ultrasound: (1) Videotape all scans for review by a sonologist or physician qualified to interpret ultrasounds. The videotape is important because the dynamic scan provides more information than the hard copy, therefore increasing the confidence level associated with each interpretation.  From an educational perspective, these videotapes contribute to a working library for resident instruction. (2) Supervise new sonographers. (3) Complete a CQI form (specifically designed for ultrasound) and attach a hard copy of the ultrasound examination. (4) Monitor progress and provide feedback. 
Data from this study underscore the conclusion from our former work that ultrasound is a rapid and accurate diagnostic modality for the detection of a hemoperitoneum and pericardial effusion.  Surgeon sonographers, having previously mastered the performance of the focused ultrasound examination, effectively used ultrasound in this limited diagnostic spectrum as the primary adjuvant modality for assessment of injured patients.
Confident in these results and in those from our previous study, we now routinely use ultrasound initially and almost exclusively for the evaluation of injured patients. Consequently, our indications for DPL and CT have narrowed substantially.
We conclude that ultrasound should be the primary adjuvant instrument for evaluation of injured patients especially those with blunt thoracoabdominal and penetrating thoracic injuries because it is rapid, accurate, and a potentially costeffective addition to the surgeon's diagnostic capabilities.
The authors wish to acknowledge the valuable assistance of Ellen Shair and Nevelle Owens in the compilation of this manuscript.
Dr. Lewis M. Flint (New Orleans, Louisiana): I'd like to congratulate Dr. Rozycki and her colleagues on their continued efforts to define a setting in which ultrasound will be useful in the diagnostic process for evaluating an injured patient.
In 1988, I visited a European trauma unit and learned of their interest in the use of ultrasound to facilitate the diagnostic process for the injured patient. When I examined the data available at that time, and in discussion with several other European trauma surgeons, I concluded that the claims being made on behalf of ultrasound were somewhat exaggerated. I have not since had the opportunity or the reason to change that opinion.
Since that time I have examined each of the subsequent reports and have been consistently skeptical of ultrasound and critical of the results purporting to show enhancement of the diagnostic process by ultrasound. I have been, as far as surgeon ultrasonographers are concerned, the fly in their ointment, or perhaps the fly in their conductive jelly. I imagine that the recorder selected me because of this. I'll try not to disappoint him. I have comments in three areas and questions in some others.
First, as to the structure of this report. Although it is billed as a prospective study, it really is a feasibility assessment.
For example, since the authors state they will not use the ultrasound to determine anything but the presence of blood, the use of the modality in penetrating injury cannot have been for any other reason than to determine the safety of the modality and the prowess of the surgeon ultrasonographer since virtually all the decisions regarding need for operation in patients with penetrating injuries are based on the clinical examination.
The fact that most of the false negative exams clustered in patients with penetrating trauma confirms the lack of value to this modality in this category of patients.
Second, it seems to me that this modality will not become known as the American Stethoscope, as it is known as the German Stethoscope. If we follow Dr. Rozycki's recommendations it may become known as the American Physical Examination.
Thirteen of 53 patients with positive ultrasound tests died, 6 from exsanguination. Is not the physical examination an even less invasive and less costly way to determine that these are really sick patients?
Finally, it seems a stretch to state that the ultrasound, a test which seems to have its major value as a means to make the surgeon feel more secure, has shortened the interval to the operating room and is cost-free. There really is no free lunch in medicine. If nothing else, you will need to be reimbursed to justify using the equipment. Now, the questions.
What's the next step? This is the twenty-third meeting of the AAST that I've attended. I've seen the fact that diagnostic peritoneal lavage averaged three or four papers in the AAST meetings after 1970 and that CT did the same. Do you think that we will be regaled with a similar proportion of ultrasound papers from here on out?
Second, what will be your response when the chief of radiology in your hospital offers to place a trained attending-level ultrasonographer in the hospital 24 hours a day with a guaranteed response time to the emergency department? I can tell you that in the hospitals that I have visited for trauma center verification, many of the radiology departments are capable and willing to do this.
Do you have any experience with ultrasound in the intensive care unit? The Germans use it in this setting as well and, in my opinion, with no better results.
Finally, what do you think the reimbursement should be for the test in your hands?
Dr. Michael L. Hawkins (Augusta, Georgia): I'm not quite sure what Dr. Flint thinks about ultrasound!
My question is about a small subset of patients, blunt trauma, very, very stable, and ultrasound shows a small amount of fluid. Would you then go to CT to confirm what the injuries are, to define liver or spleen injury, et cetera?
Dr. Michael Rhodes (Allentown, Pennsylvania): Grace, another well-presented paper. I have several questions. Where is the machine located and who is responsible for maintenance? Who bills for this procedure? Does a radiologist review the study and is there a video record?
Dr. R. Stephen Smith (Oakland, California): I have three questions and comments. Dr. Rozycki. First of all, I'd like to congratulate you and your co-authors for I think leading the American contribution to ultrasound in trauma. My questions are as follows:
What political problems did you encounter with your radiology department and how did you overcome it?
Secondly, I'd like to point out that there is a certification available for nonradiologist MDs from the Registry of Diagnostic Medical Sonographers if physicians can document adequate experience and pass a written examination. Is this a certification that surgeons should have if they intend to use this in the trauma setting and will it help to overcome the resistance of our radiologist colleagues?
And the third point is, we use a 7.5-MHz probe to delineate wound track to the anterior abdominal wall for tangential gunshot wounds and we feel that we can fairly reliably determine whether or not peritoneal penetration has occurred. Have you had a similar experience and would you comment?
Dr. Gerald O. Strauch (Chicago, Illinois): This was really a nicely presented paper. Could we ask how many people in this room are in hospitals where surgeons are doing bedside ultrasound? It is not surprising how few hands I see.
Even from Lew's comments and others, it's certainly apparent that turf may have the biggest influence on this entire issue. One would think that perhaps the AAST might want to be influential in at least helping this effort along. I think there's little doubt that this is extraordinarily useful technology in the hands of surgeons and that we ought to be supporting it.
Dr. A. Brent Eastman (La Jolla, California): Grace, I congratulate you on a beautiful presentation and the courage of choosing Dr. Flint as your discussant.
I have one question. In looking at your algorithm, if your ultrasonography was positive you went to laparotomy. My question is, have you missed the opportunity for conservative management of solid viscous injuries which is based on CT grading systems? Second, do you think we will develop a grading system as we move to this ultrasound modality? That is do you think we can develop a grading system of solid organ injury based on ultrasound?
Dr. Sheldon Brotman (Danville, Pennsylvania): You said you were doing two examinations, one at entry and if negative a second exam 12 to 24 hours later. My question is, how many of your positives were identified the first time around and how many the second time around?
I think that's important, because Flint noted the fact that there was a relatively high rate of mortality in your series. The question must be asked, did this delay account for the high rate of mortality? Thank you.
Dr. Grace S. Rozycki (closing): I'd like to express my appreciation to Dr. Flint, and other members of the Association for their comments and questions.
Although I am a proponent of ultrasound, I believe that there is no substitute for sound clinical judgment and physical examination. As we proceed through the prospective trials for the evaluation of ultrasonography as a diagnostic tool, we should apply the same common sense approach to ultrasound as we do with any other diagnostic modality.
Cost, which was addressed by several of the discussants, is a timely and important issue. I believe a cost or charge to the patient is valid only with a credentialing and continuous quality improvement process in place. These are currently in progress. Just as other subspecialties, such as cardiology, urology, and ophthalmology have established their own guidelines regarding ultrasonography, so should general surgeons do similarly.
In response to Dr. Flint's question about the placement of an ultrasound technician in-house, I believe that this would not replace the surgeon. The surgeon is present in the resuscitation center prior to patient arrival. Making use of ultrasound's most valuable quality, rapidity, the surgeon's ultrasound assessment is completed prior to the arrival of the technician.
The questions asked by Drs. Hawkins and Eastman regarding the amount of fluid identified on ultrasound conservative management and the need for operative management is part of our next study. An ultrasound followed by the performance of a CAT scan in hemodynamically stable patients may provide some valuable information on quantity of blood and indications for operation. In this study, however, all true positives had therapeutic operations.
The machine is placed in the resuscitation unit and the surgeon is responsible for it. However, routine biannual mechanical inspections by the Biomedical Engineering Department are performed as part of an internal quality control process.
In this study, video recording was not available. However, its value for resident education is well recognized.
As expected, the turf issues were addressed by several of the discussants. I would like to share with you some thoughts on what I have learned over the past few years:
Make a friend.
Consult an ultrasound technician at another institution. They are frequently eager to share information with you and they are free of the political issues. The obstetricians are also valuable resources. No one questions their practice of ultrasonography. You may want to consider purchasing an ultrasound machine that is similar to the one they use so that their staff is comfortable using it to evaluate an injured pregnant patient.
I suspect that a 7.5-MHz probe would provide some detailed information for wound evaluation, however, in this study we were limited to the 3-MHz probe.
Dr. Strauch, thank you so much for your favorable comments. Such support from the Association would provide an incentive for other institutions to use this diagnostic modality.
Almost all patients in the true positive group were positive on initial examination. Only two patients converted to positive, one at 2 hours and the other at 12 hours respectively. These numbers are too small to draw any conclusions regarding show often repeat examinations should be performed. Additionally, other parameters such as the hematocrit were followed which ensured that patient management decisions were based on the entire clinical picture and not just the results of the ultrasound.
I'd like to thank the Association for the privilege of presenting our data.
The paper by Rozycki and colleagues is an important contribution to the exploding literature on the use of ultrasound in the diagnostic evaluation of the trauma victim. The work comes from a group that is intensely interested in the technique and is actively encouraging its use, especially by surgeons. This conviction in the technique underlies both the strengths and weaknesses of the article.
The strengths of this article are in the number of patients studied and the careful attention of the surgeon-ultrasonographers in the performance of the procedure. The section on quality assurance and credentialing will undoubtedly serve as a model for those centers wishing to embark on having trauma surgeons provide this diagnostic evaluation.
The overriding conviction in ultrasound also underlies the weakness of the paper and it is possible that the authors beliefs have interfered with their evaluating the technique with an appropriately jaundiced eye. The article fails to critically evaluate which patient groups would really benefit from ultrasound in place of another test. An example of this is an analysis of the true positive group where the authors state that all patients with scans positive for intrabdominal fluid required surgical intervention. Since a significant number of these patients had liver and splenic injuries it is hard to understand why the author's thought that no patient could have been managed non-operatively. According to the authors definitions in Table 1 there does not appear to be a place for nonoperative management of organ injuries based on the presence or absence of fluid nor could patients' operations be considered nontherapeutic. Although it is almost irresistable not to put a stitch or place some hemostatic agent when you are in the operating room staring at an injured liver or spleen, it is hard to believe that all of these celiotomies were truly therapeutic.
I am also troubled by the need for two ultrasound examinations to achieve the accuracy reported in the paper which is similar to one abdominal CT scan or one DPL. The authors reiterate in the paper that they did not ultrasound everyone because it is a busy trauma center, yet people who are evaluated by ultrasound needed two examinations. This would make their protocol more labor intensive.
In summary, this paper begins to evaluate a potentially useful diagnostic modality and like all good studies, the paper poses more questions than it answers. Whether the results reported here can be duplicated by other trauma centers less committed to the technique will ultimately influence how ultrasound will be used in the evaluation of the trauma patient. Until the data presented here are confirmed or contradicted it is unlikely that the benefits of the procedure would warrant a blanket elimination of our standard diagnostic modalities without a more careful and controlled evaluation.
David H. Livingston, MD
Department of Surgery
Newark, New Jersey
1. Buntain WL, Gould HR, Maull KI: Predictability of splenic salvage by computed tomography. J Trauma 28:24, 1988
2. Davis RA, Shayne JP, Max MH, et al: The use of computerized axial tomography versus peritoneal lavage in the evaluation of blunt abdominal trauma: A prospective study. Surgery 98:845, 1985
3. Engrav LH, Benjamin CI, Strate RG, et al: Diagnostic peritoneal lavage in blunt abdominal trauma. J Trauma 15:854, 1975
4. Fabian TC, Mangiante EC, White TJ, et al: A prospective study of 91 patients undergoing both computed tomography and peritoneal lavage following blunt abdominal trauma. J Trauma 26:602, 1986
5. Fischer RP, Beverlin BC, Engrav LH, et al: Diagnostic peritoneal lavage: Fourteen years and 2,586 patients later. Am J Surg 136:701, 1978
6. Freeman T, Fischer RP: The inadequacy of peritoneal lavage in diagnosing acute diaphragmatic rupture. J Trauma 16:538, 1976
7. Goldstein AS, Sclafani SJA, Kupferstein NH, et al: The diagnostic superiority of computerized tomography. J Trauma 25:938, 1985
8. Kearney PA Jr, Vahey T, Burney RE, et al: Computed tomography and diagnostic peritoneal lavage in blunt abdominal trauma: Their combined role. Arch Surg 124:344, 1989
9. Meyer DM, Thal ER, Weigelt JA, et al: The role of abdominal CT in the evaluation of stab wounds to the back. J Trauma 29:1226, 1989
10. Olsen WR, Hildreth DH: Abdominal paracentesis and peritoneal lavage in blunt abdominal trauma. J Trauma 11:824, 1971
11. Peitzman AB, Makaroun MS, Slasky BS, et al: Prospective study of computed tomography in initial management of blunt abdominal trauma. J Trauma 26:585, 1986
12. Phillips T, Sclafani SJA, Goldstein A, et al: Use of the contrast-enhanced CT enema in the management of penetrating trauma to the flank and back. J Trauma 26:593, 1986
13. Powell DC, Bivins BA, Bell RM: Diagnostic peritoneal lavage. Surg Gynecol Obstet 155:256, 1982
14. Root HD, Keizer PJ, Perry JF Jr: Peritoneal trauma: Experimental and clinical studies. Surgery 62:679, 1967
15. Thal ER, May RA, Beesinger D: Peritoneal lavage: Its unreliability in gunshot wounds of the lower chest and abdomen. Arch Surg 115:430, 1980
16. Chambers JA, Pilbrow WJ: Ultrasound in abdominal trauma: An alternative to peritoneal lavage. Arch Emerg Med 5:26, 1988
17. Forster R, Pillasch J, Zielke A, et al: Ultrasonography in blunt abdominal trauma: Influence of the investigators' experience. J Trauma 34:264, 1992
18. Goldberg BB, Clerfield HR, Goodman GA, et al: Ultrasonic determination of ascites. Arch Intern Med 131:217, 1973
19. Gruessner R, Mentges B, Duber C, et al: Sonography versus peritoneal lavage in blunt abdominal trauma. J Trauma 29:242, 1989
20. Gruessner R, Ruckert K, Klotter JH, et al: Ultrasound and lavage in polytrauma patients with blunt abdominal trauma. Dtsch Med Wochenschr 110:1521, 1985
21. Hoffman R, Pohlemann T, Wipperman B, et al: Management der Sonographie bei stumpfem Bauchtrauma. Unfallchirurg 92:471, 1989
22. Ruf W, Friedl W, Weber G, et al: Stellt der sonographische Nachweis von Blut im Abdomen nach stumpfem Bauchtrauma in jedem Fall eine Operationsindikation dar? Unfallchirurg 93:132, 1990
23. Tiling T, Bouillon B, Schmid A, et al: Ultrasound in blunt abdominothoracic trauma. In Border JR, Allgower M, Hansen ST, Ruedi TP (eds): Blunt Multiple Trauma. New York, Marcel Dekker, 1990
24. Rozycki GS, Ochsner MG, Jaffin JH, et al: Prospective evaluation of surgeons' use of ultrasound in the evaluation of trauma patients. J Trauma 34:516, 1993
25. Tso P, Rodriguez A, Cooper C, et al: Sonography in blunt abdominal trauma: A preliminary progress report. J Trauma 33:39, 1992
26. Bode PJ, Niezen RA, VanVugt AB, et al: Abdominal ultrasound as a reliable indicator for conclusive laporotomy in blunt abdominal trauma. J Trauma 34:27, 1993
27. Jarowenko DG, Hess RM, Herr MS, et al: Use of ultrasonography in the evaluation of blunt abdominal trauma. Abstract presented at 1989 meeting of the American Association for the Surgery of Trauma, Chicago, Illinois
28. Plummer D, Dick C, Ruiz E, et al: Emergency department two-dimensional echocardiography in the diagnosis of nontraumatic cardiac rupture. Ann Emerg Med 23:1333, 1994
29. Schiavone WA, Ghumrawi BK, Catalano DR, et al: The use of echocardiography in the emergency management of nonpenetrating traumatic rupture. Ann Emerg Med 20:1248, 1991
30. Shackford SR: Focused ultrasound examinations by surgeons: The time is now (Editorial). J Trauma 35:181, 1993
31. Goletti O, Ghiselli G, Lippolis PV, et al: The role of ultrasonography in blunt abdominal trauma: Results in 250 consecutive cases. J Trauma 36:178, 1994
32. Sarkisian AE, Khondkarian RA, Amirbekian NM, et al: Sonographic screening of mass casualties for abdominal and renal injuries following the 1988 Armenian earthquake. J Trauma 31:247, 1991
33. Rozycki GS, Kraut EJ: Isolated blunt rupture of the infrarenal vena cava: The role of ultrasound and computed tomography in an occult injury. J Trauma 383:402, 1995
34. Rothlin MA, Naf R, Amgwerd M, et al: Ultrasound in blunt abdominal and thoracic trauma. J Trauma 34:488, 1993
35. Wening JV: Evaluation of ultrasound, lavage, and computed tomography in blunt abdominal trauma. Surg Endosc 31:152, 1989
36. Huang MS, Liu M, Wu JK, et al: Ultrasonography for the evaluation of hemoperitoneum during resuscitation: A simple scoring system. J Trauma 36:173, 1994
37. Kuligowska E, Mueller PR, Simeone JF, et al: Ultrasound in upper abdominal trauma. Semin Roentgenol 14:281, 1984
38. Kimura A, Otsuka T: Emergency center ultrasonography in the evaluation of hemoperitoneum: A prospective study. J Trauma 31:120, 1991
39. Kohan JM, Poe RH, Israel RH: Value of chest ultrasonography versus decubitus roentgenography for thoracentesis. Am Rev Respir Dis 133:1124, 1986
40. Hiatt JR, Yeatman LA, Child JS: The value of echocardiography in blunt chest trauma. J Trauma 28:914, 1988
41. Plummer D, Brunette D, Asinger R, et al: Emergency department echocardiography improves outcome in penetrating cardiac injury. Ann Emerg Med 21:709, 1992
42. Hauser AM: The emerging role of echocardiography in the emergency department. Ann Emerg Med 18:1298, 1989
43. 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 35:267, 1993
44. Feliciano DV: Diagnostic modalities in abdominal trauma: Peritoneal lavage, ultrasonography, computed tomography scanning, and arteriography. Surg Clin North Am 71:241, 1991
45. Mueller PR, Simeone JF: Intraabdominal abscesses: Diagnosis by sonography and computed tomography. Radiol Clin North Am 21:425, 1983
46. Brent RL: The effect of embryonic and fetal exposure to X-ray, microwaves, and ultrasound: Counseling the pregnant and nonpregnant patient about these risks. Semin Oncol 16:347, 1989
47. Filiatrault D, Longpre D, Patriquin H, et al: Investigation of childhood blunt abdominal trauma: A practical approach using ultrasound as the initial diagnostic modality. Pediatr Radiol 17:373, 1987
48. Hoelzer DJ, Brian MB, Balsara VJ: Selection and nonoperative management of pediatric blunt trauma patients: The role of quantitative crystalloid resuscitation and abdominal ultrasonography. J Trauma 26:57, 1986
49. American Institute of Ultrasound Medicine: Training Guidelines for Physicians Who Evaluate and Interpret Diagnostic Ultrasound Examinations. Rockville, Md., American Institute of Ultrasound Medicine, 1993
50. Prop RL, Winters: Clinical competence in adult echocardiography. Circulation 81:2032, 1990
51. Jehle D, Davis E, Evans T, et al: Emergency department sonography by emergency physicians. Am J Emerg Med 7:605, 1989
52. Jehel D: Ultrasound in emergency medicine. SAEM Newsletter, 1990
53. Mateer J, Plummer D, Heller M, et al: Model curriculum for physician training in emergency ultrasonography. Ann Emerg Med 23:95, 1994
54. Sanders C (ed): Clinical Sonography: A Practical Guide. 2nd ed., Boston, Little Brown, 1991