Surgeon-Performed Ultrasound in Acute Care Surgery

Christopher J. Dente and Grace S. Rozycki

INTRODUCTION

For nearly two decades, acute care surgeons have successfully performed, interpreted, and taught bedside ultrasound examinations of patients who are injured or critically ill.^1–13 Real-time imaging allows the surgeon to receive instantaneous information about the clinical condition of the patient and, therefore, helps to expedite the patient’s management, which is important in patients with time-sensitive diagnoses. In many centers, ultrasound machines are owned by surgeons or surgical departments and are part of the standard equipment in the trauma resuscitation area as well as in the intensive care unit (ICU). While diagnostic peritoneal lavage (DPL) and computed tomography (CT) scanning are still valuable diagnostic tests for the detection of intra-abdominal injury in patients, ultrasound is not only faster but also noninvasive and painless.

As an extension of the physical examination, acute care surgeons routinely use ultrasound in the trauma setting to augment their physical examination in patients with suspected torso and extremity trauma, not only within the standard hospital resuscitation area but also in a variety of other locales. Additionally, ultrasound may be used to supplement the history and physical examination of nontrauma patients presenting with acute abdominal pain and a variety of other time-sensitive diagnoses in the emergency department. Finally, ultrasound may be used in the ICU in a variety of ways to facilitate procedures, detect complications, and augment a surgeon’s physical examination.

As such, this chapter begins with a basic introduction to select principles of ultrasound physics and then covers the components, indications, and pitfalls of the common, focused ultrasound examinations used by acute care surgeons who are evaluating trauma patients, nontrauma patients presenting with acute symptoms, and, finally, critically ill patients in the ICU.

ULTRASOUND PHYSICS

Ultrasonography is operator dependent and, therefore, an understanding of select principles of ultrasound imaging is necessary so that images may be acquired rapidly and interpreted correctly. Knowledge of some of these basic principles enables the acute care surgeon to select the appropriate transducer, optimize resolution of the image, and recognize artifacts. Some basic terms and principles of physics relative to ultrasound imaging in the acute setting are defined in Tables 16-1 to 16-3.

TABLE 16-1 Ultrasound Physics Terminology Relevant to Ultrasound Imaging

TABLE 16-2 Essential Principles of Ultrasound

TABLE 16-3 Terminology Used in Interpretation of Ultrasound Images

In general, an ultrasound system includes the following components: (1) a transmitter that controls electrical signals sent to the transducer; (2) a receiver or image processor that admits the electrical signal; (3) a transducer containing piezoelectric crystals to interconvert electrical and acoustic energy; (4) a monitor to display the ultrasound image; and (5) an image recorder or printer.^14,15 The ultrasound images that are obtained depend on the orientation of the transducer or probe relative to the structure or organ being imaged, with each transducer having an indicator that directs its orientation to the screen. The indicators on most probes are oriented such that placing the indicator cephalad in a coronal or sagittal plane or to the left in the transverse plane will orient the image correctly left to right. As the image may be purposefully reversed by a machine adjustment, the surgeon should confirm the orientation of the probe by adding ultrasound gel to the probe’s footprint and gently rubbing it to visualize where the motion is detected on the screen. This should allow the surgeon to determine how the indicator will orient the image. The orientations or scanning planes are described in Table 16-4 and the projected patient positions on the ultrasound monitor are shown in Fig. 16-1.¹⁶

TABLE 16-4 Scanning Planes Used in Ultrasound Imaging¹⁶

FIGURE 16-1 Scanning planes used in ultrasound imaging. (Adapted with permission from Tempkins BB. Scanning Planes and Methods. Ultrasound Scanning: Principles and Protocols. Philadelphia, PA: WB Saunders Company; 1993, © Elsevier.)

Although diagnostic ultrasound uses transducer frequencies ranging from 1 MHz (megahertz = 1 million cycles/s) to 30 MHz, medical diagnostic imaging most often uses frequencies between 2.5 and 10 MHz (Table 16-5). Accordingly, transducers are chosen on the basis of the depth of the structure or organ to be imaged. High-frequency transducers (≥5 MHz) provide excellent resolution for imaging superficial structures such as an abscess in the soft tissue of an extremity. Lower-frequency transducers emit waves that penetrate deeply into the tissue and, therefore, are preferred for visualizing organs such as the liver or spleen.^14,17,18 Tips to maximize accuracy and quality of ultrasound imaging relative to the above-mentioned physics principles are listed in Table 16-6.

TABLE 16-5 Clinical Applications of Selected Transducer Frequencies

TABLE 16-6 Maximizing Accuracy in Ultrasound Imaging

SURGEON-PERFORMED ULTRASOUND IN TRAUMA

FAST

Developed for the evaluation of injured patients, the Focused Assessment for the Sonographic Examination of the Trauma Patient (FAST) is a rapid diagnostic examination to assess patients with potential injuries to the torso. The test sequentially surveys for the presence or absence of fluid in the pericardial sac and in the dependent abdominal regions, including Morison’s pouch region in the right upper quadrant (RUQ), the left upper quadrant (LUQ) behind the spleen and between the spleen and kidney, and the pelvis posterior to the bladder. Surgeons can perform the FAST during the primary or secondary survey of the American College of Surgeons Advanced Trauma Life Support¹⁹ algorithm and, although minimal patient preparation is needed, a full urinary bladder is ideal to provide an acoustic window for visualization of blood in the pelvis.

Blood, as any fluid, will accumulate in dependent regions of the abdomen.²⁰ In the supine position, this corresponds to Morison’s pouch, the splenorenal recess, and above the spleen as well as in the pelvis posterior to the bladder. All these regions may be visualized rapidly and dependably with the FAST. Furthermore, ultrasound is an excellent modality for the detection of intra-abdominal fluid, having been shown to detect ascites in small amounts.^21,22 Although the exact minimum amount of intraperitoneal fluid that can be detected by ultrasound is not known,²³ most authors agree that it is a sensitive modality.

The FAST is performed in a specific sequence for several reasons. The pericardial area is visualized first so that blood within the heart can be used as a standard to set the gain (Table 16-1). Most modern ultrasound machines have presets so that the gain does not need to be reset each time the machine is turned on. Occasionally, if multiple types of examinations are performed with different transducers, the gain should be checked to ensure that intracardiac blood appears anechoic. This maneuver ensures that a hemoperitoneum will also appear anechoic and will be readily detected on the ultrasound image. The abdominal part of the FAST begins with a survey of the RUQ that is the location within the peritoneal cavity where blood most often accumulates and is most readily detected with the FAST. Indeed, investigators from four Level I trauma centers examined true-positive ultrasound images of 275 patients who sustained either blunt (#220) or penetrating (#55) injuries.²⁴ They found that regardless of the injured organ (with the exception of those patients who had an isolated perforated viscus), blood was most often identified on the RUQ image of the FAST. This can be a time-saving measure because when hemoperitoneum is identified on the FAST examination of a hemodynamically unstable patient, that image alone, in combination with the patient’s clinical picture, is sufficient to justify an immediate abdominal operation.²⁴ In a stable patient, following the exam of the RUQ, the LUQ and pelvis are visualized as discussed below.

Technique

Ultrasound transmission gel is applied on four areas of the thoracoabdomen, and the examination is conducted in the following sequence: the pericardial area, RUQ, LUQ, and the pelvis (Fig. 16-2).

FIGURE 16-2 Schematic diagram of transducer positions for FAST: pericardial, right upper quadrant, left upper quadrant, and pelvis.

A 3.5-MHz convex transducer is oriented for sagittal or longitudinal views and positioned in the subxiphoid region to identify the heart and to examine for blood in the pericardial sac. The normal and abnormal views of the pericardial area are shown in Fig. 16-3. The subxiphoid image is usually not difficult to obtain, but a severe injury to the chest wall, a very narrow subcostal area, subcutaneous emphysema, or morbid obesity can prevent a satisfactory examination.²⁵ Both of the latter conditions are associated with poor imaging because air and fat reflect the wave too strongly and prevent penetration into the target organ.¹⁴ If the subcostal pericardial image cannot be obtained or is suboptimal, a parasternal ultrasound view of the heart should be performed (Figs. 16-4 and 16-5).

FIGURE 16-3 (Left) Sagittal view of pericardial area showing pericardium as single echogenic line (normal). (Right) Sagittal view of pericardial area showing separation of visceral and parietal areas of pericardium with blood (arrow) that appears anechoic.

FIGURE 16-4 Transducer position for left parasternal view of heart.

FIGURE 16-5 Normal (left) and abnormal (right) heart, parasternal view.

Next, the transducer is placed in the right anterior or midaxillary line between the 11th and 12th ribs to obtain sagittal images of the liver, kidney, and diaphragm (Fig. 16-6) and determine the presence or absence of blood in Morison’s pouch and in the right subphrenic space. Next, attention is turned to the LUQ. With the transducer positioned in the left posterior axillary line between the 10th and 11th ribs, the spleen and left kidney are visualized and the presence or absence of blood between the two organs and in the left subphrenic space is determined (Fig. 16-7). The splenic window is often the most difficult window to adequately visualize and the probe should be placed significantly more posterior (posterior axillary line) and superior (one to two rib spaces higher) than with the RUQ window.

FIGURE 16-6 (A) Normal sagittal view of liver, kidney, and diaphragm. Note Gerota’s fascia is hyperechoic. (B) Abnormal sagittal view of liver, kidney, and diaphragm. Note fluid (blood) in between liver and kidney (arrows).

FIGURE 16-7 (Left) Normal sagittal view of spleen, kidney, and diaphragm. (Right) Abnormal sagittal view of spleen, kidney, and diaphragm with fluid (blood) in between spleen and kidney and above the spleen in the subphrenic space.

Finally, the transducer is directed for a transverse view and placed about 4 cm superior to the symphysis pubis. It is slowly swept inferiorly to obtain a coronal view of the full bladder and the pelvis examining for the presence or absence of blood (Fig. 16-8).

FIGURE 16-8 (Left) Normal coronal view of full urinary bladder. (Right) Abnormal coronal view of full bladder with fluid in pelvis. (Note the bowel floating in fluid.)

Accuracy of the FAST

Improper technique, inexperience of the examiner, and inappropriate use of ultrasound have long been known to adversely impact the accuracy of ultrasound imaging. More recently, the etiology of injury, the presence of hypotension on admission, and select associated injuries have also been shown to influence the accuracy of this modality.^2,3,8 Failure to consider these factors has led to inaccurate assessments of the accuracy of the FAST by comparing it inappropriately to a CT scan and not recognizing its role in the evaluation of patients with penetrating torso trauma.^26,27 Both false-positive and -negative pericardial ultrasound examinations have been reported to occur in the presence of a massive hemothorax or mediastinal blood.^4,8,10,28 Repeating the FAST after the insertion of a tube thoracostomy improves the visualization of the pericardial area and decreases the number of false-positive and -negative studies. While false studies may occur, a rapid focused ultrasound survey of the subcostal pericardial area is a very accurate method to detect hemopericardium in most patients with penetrating wounds in the “cardiac box.”^4,10 In a large study of patients who sustained either blunt or penetrating injuries, the FAST was 100% sensitive and 99.3% specific for detecting hemopericardium in patients with precordial or transthoracic wounds. Furthermore, the use of pericardial ultrasound has been shown to be especially helpful in the evaluation of patients who have no overt signs of pericardial tamponade. This was highlighted in a study in which 10 of 22 patients with precordial wounds and a hemopericardium on an ultrasound examination had admission systolic blood pressures >110 mm Hg and were relatively asymptomatic. Based on these signs and the lack of symptoms, it is unlikely that the presence of cardiac wounds would have been strongly suspected in these patients and, therefore, this rapid ultrasound examination provided an early diagnosis of hemopericardium before the patients underwent physiologic deterioration.

The FAST is also very accurate when it is used to evaluate hypotensive patients who present with blunt abdominal trauma. In this scenario, ultrasound is so accurate that when the FAST is positive, an immediate operation is justified.^4,8,10,29

Because the FAST is a focused examination for the detection of blood in dependent areas of the abdomen, its results should not be compared to those of a CT scan because the FAST does not readily identify intraparenchymal or retroperitoneal injuries. Therefore, select hemodynamically stable patients considered to be at high risk for occult intra-abdominal injury should undergo a CT scan of the abdomen regardless of the results of the FAST examination. These patients include those with fractures of the pelvis or thoracolumbar spine, major thoracic trauma (pulmonary contusion, lower rib fractures), and hematuria. These recommendations were based on the results of two studies by Chiu et al. in 1997³⁰ and Ballard et al. in 1999.³¹ Chiu et al. reviewed their data on 772 patients who underwent FAST examinations after sustaining blunt torso injury. Of the 772 patients, 52 had intra-abdominal injury but 15 of them had no hemoperitoneum on the admitting FAST examination or on the CT scan of the abdomen. In other work conducted by Ballard et al. at Grady Memorial Hospital, an algorithm was developed and tested over a 3.5-year period to identify patients who were at high risk for occult intra-abdominal injuries after sustaining blunt thoracoabdominal trauma.³¹ Of the 1,490 patients admitted with severe blunt trauma, there were 102 (70 with pelvic fractures, 32 with spine injuries) who were considered to be at high risk for occult intra-abdominal injuries. Although there was only 1 false-negative FAST examination in the 32 patients who had spine injuries, there were 13 false negatives in those with pelvic fractures. Based on these data, the authors concluded that patients with pelvic fractures should have a CT scan of the abdomen regardless of the result of the FAST examination. The lower accuracy of the FAST in patients with pelvic fractures was again noted in a recent series published by Friese et al., in which an initial FAST examination had an 85% positive predictive value but only a 63% negative predictive value in 146 patients with pelvic fractures.³² These studies have helped provide guidelines to decrease the number of false-negative FAST studies, but, as with the use of any diagnostic modality, it is important to correlate the results of the test with the patient’s clinical picture. Suggested algorithms for the use of FAST are depicted in Fig. 16-9A and B. Indeed, the FAST exam has been included in the most recently published evidence-based guidelines for the evaluation of patients with blunt abdominal trauma from the Eastern Association for the Surgery of Trauma (EAST) with reported accuracy rates of 96–98%.³³

FIGURE 16-9 (A) Algorithm for the use of ultrasound in patients with penetrating chest wounds. (B) Algorithm for the use of ultrasound in patients with blunt abdominal trauma.

Quantification of Blood

The amount of blood detected on the abdominal CT scan³⁴ or in the DPL aspirate (or effluent) has been shown to predict the need for operative intervention.³⁵ Similarly, the quantity of blood that is detected with ultrasound may be predictive of a therapeutic operation.^36,37 Huang et al. developed a scoring system based on the identification of hemoperitoneum in specific areas such as Morison’s pouch or the perisplenic space.³⁶ Each abdominal area was assigned a number from 0 to 3, and the authors found that a total score of ≥3 corresponded to more than 1 L of hemoperitoneum. This scoring system had a sensitivity of 84% for determining the need for an immediate abdominal operation. Another scoring system developed and prospectively validated by McKenney et al. examined the patient’s admission blood pressure, base deficit, and the amount of hemoperitoneum present on the ultrasound examinations of 100 patients.³⁷ The hemoperitoneum was categorized by its measurement and its distribution in the peritoneal cavity, so that a score of 1 was considered a minimal amount of hemoperitoneum but a score of >3 signified a large hemoperitoneum. Forty-six of the 100 patients had a score >3, and 40 (87%) of them underwent a therapeutic abdominal operation. This scoring system had a sensitivity, specificity, and accuracy of 83%, 87%, and 85%, respectively. The authors concluded that an ultrasound score of >3 is statistically more accurate than a combination of the initial systolic blood pressure and base deficit for determining which patients will undergo a therapeutic abdominal operation. Although the quantification of hemoperitoneum is not exact, it can provide valuable information about the need for an abdominal operation as well as its potential to be therapeutic.

Recent Advances and Organ Specificity

As surgeons have become more facile with ultrasound exams and as technology has improved, extensions of the FAST exam have been described. Again, it is noted that the standard FAST exam is designed to accurately answer two simple questions: Is there fluid in the peritoneal cavity and is their fluid in the pericardial sac? The use of ultrasound for more complex diagnostic interventions is described below, but these areas are less well studied and beyond the purview of the traditional FAST exam.

A more recent prospective, multicenter trial conducted by the Western Trauma Association reported on the use of ultrasound to serially evaluate patients with documented solid organ injuries (SOI) after trauma.³⁸ The so-called BOAST exam, or the bedside organ assessment with sonography after trauma, was performed by a limited number of experienced surgeon sonographers in 126 patients with 135 SOI in 4 American trauma centers. This study, performed over nearly 2 years, was designed to be a more thorough abdominal ultrasound examination with multiple views obtained of each solid organ (kidneys, liver, and spleen). Criteria for enrollment included normal hemodynamics, absence of peritonitis or other need for urgent laparotomy, and lack of excessive blood transfusion in the attending physician’s judgment. All patients were victims of blunt trauma with a mean Injury Severity Score (ISS) of nearly 15.

Overall, only 34% of injuries to solid organs were seen with BOAST yielding an error rate of 66%. None of the 34 grade I injuries were identified and only 13 (31%) of the grade II injuries were identified. Sensitivities for grade III and IV injuries ranged from 25% to 75% and only one grade V injury (to the liver) was examined and positively identified. Eleven patients developed 16 intra-abdominal complications (8 pseudoaneurysms, 4 bilomas, 3 abscesses, and 1 necrotic organ), and 13 (81%) were identified by the sonographers. This study emphasizes that ultrasound, in most surgeons’ hands, should not be considered a reliable modality for diagnosis and grading of SOI although it may be acceptably accurate in the diagnosis of post-traumatic abdominal complications in patients with SOI managed nonoperatively.

In Europe, preliminary work using Power Doppler ultrasonography to identify specific organ injuries has been published in recent years.^39,40 Many of these exams include the use of a sonographic contrast agent injected peripherally during the scan. In one study, the authors were able to document extravasation of contrast in 20 of 153 patients (13%). Extravasation was seen not only from the spleen, liver, and kidney after trauma but also in postoperative patients (aortic aneurysm repair, postsplenectomy) and in a patient with a ruptured aortic aneurysm. In 9 of 20 patients, CT scan was performed and all 9 confirmed extravasation of contrast. In the 133 patients without extravasation, the absence of active bleeding was inferred by a subsequent CT scan in 82 patients, surgical data in 13 patients, and clinical follow-up in 38 patients, with no cases of active bleeding missed by ultrasound. Thus, the addition of an ultrasonic contrast agent and Power Doppler may be of some benefit in the diagnosis of specific injuries. It should be emphasized, however, that the FAST exam in most American trauma centers is used simply as a screening tool to identify the presence or absence of hemoperitoneum or hemopericardium in a trauma patient.

Traumatic Hemothorax

A focused thoracic ultrasound examination was developed by surgeons to rapidly detect the presence or absence of a traumatic hemothorax in patients during the ATLS secondary survey.⁹ This focused thoracic ultrasound examination employs the ultrasound physics principles of the mirror image artifact and tissue acoustic impedance as presented in Table 16-1. A test that promptly detects a traumatic effusion or hemothorax is worthwhile because it dramatically shortens the interval from the admission of the patient with hemothorax to the insertion of a thoracostomy tube.

Technique

The technique for this examination is similar to that used to interrogate the upper quadrants of the abdomen in the FAST and also uses the same type and frequency transducer. In point of fact, it is performed one to two rib spaces higher than the RUQ and LUQ FAST views using the same probe. Ultrasound transmission gel is applied to the right and left lower thoracic areas in the midaxillary to posterior axillary line between the 9th and 10th intercostal spaces (Fig. 16-10). The transducer is slowly advanced cephalad to identify the hyperechoic diaphragm and to interrogate the supradiaphragmatic space for the presence or absence of fluid (Fig. 16-11A and B) that appears anechoic. In the positive thoracic ultrasound examination, the hypoechoic lung can be seen “floating” amidst the fluid. The same technique can be used to evaluate a critically ill patient for a pleural effusion as discussed earlier.

FIGURE 16-10 Transducer positions for thoracic ultrasound examination (detection of hemothorax).

FIGURE 16-11 (A) Sagittal view of liver, kidney, and diaphragm. Note supradiaphragmatic (lung) area but absence of pleural effusion. (B) Sagittal view of right supradiaphragmatic space. The right hemithorax contains fluid (blood) that appears anechoic.

Accuracy

One of the earliest reports on the use of ultrasound for the evaluation of fluid collections in the pleural space was described by Joyner et al. in 1967.⁴¹ Later, Gryminski et al. documented the superiority of ultrasound over standard radiography for the detection of pleural fluid.⁴² In that study, they reported that ultrasound detected even small amounts of pleural fluid in 74 (93%) of 80 patients, whereas plain radiography detected pleural fluid in only 66 (83%) of these patients.

Surgeons at Emory University have also examined the accuracy of this examination in 360 patients with blunt and penetrating torso injuries.⁹ They compared the time and accuracy of ultrasound with that of the supine portable chest x-ray and found both to be very similar with 97.4% sensitivity and 99.7% specificity observed for thoracic ultrasound versus 92.5% sensitivity and 99.7% specificity for the portable chest x-ray. Performance times, however, for the thoracic ultrasound examinations were statistically much faster than those for the portable chest x-ray. Although it is not recommended that the thoracic ultrasound examination replace the chest x-ray, its use can expedite treatment in many patients and decrease the number of chest radiographs obtained.

Pneumothorax

The use of ultrasound for the detection of a traumatic pneumothorax is not a new diagnostic test, having been reported by numerous acute care surgeons dating back to the early 1990s.^43–47 This examination is useful to the surgeon to evaluate a patient for a potential pneumothorax in the following circumstances: (1) bulky radiology equipment is not readily available; (2) inordinate delays for obtaining a chest x-ray are anticipated; or (3) numerous injured patients (mass casualty situation) must be rapidly assessed and triaged.^48,49 Although useful in the trauma resuscitation area, surgeons may also find this examination helpful to detect a pneumothorax in a critically ill patient who is on a ventilator, after a thoracentesis procedure, or, potentially, after discontinuing the suction on an underwater seal device.

Technique

A 5.0- to 7.5-MHz linear array transducer is used to evaluate a patient for the presence of a pneumothorax. The examination may be performed while the patient is in the erect or the supine position. Ultrasound transmission gel is applied to the right and left upper thoracic areas at about the third to fourth intercostal space in the midclavicular line and the presumed unaffected thoracic cavity is examined first. The transducer is oriented for longitudinal imaging, is placed perpendicular to the ribs, and is slowly advanced medially toward the sternum and then laterally toward the anterior axillary line. The normal examination of the thoracic cavity identifies the rib (seen as black on the ultrasound image because it is a refraction artifact), pleural sliding, and a comet tail artifact (Table 16-1). Pleural sliding is the identification of the visceral and parietal layers of the lung seen as hyperechoic superimposed pleural lines. When a pneumothorax is present, air becomes trapped between the visceral and parietal pleura and does not allow for the transmission of the ultrasound waves. Therefore, the visceral pleura is not imaged and pleural sliding is not observed. The comet tail artifact is generated because of the interaction of two highly reflective opposing interfaces, that is, air and pleura (Fig. 16-12). When air separates the visceral and parietal pleurae, the comet tail artifact is not visualized. If desired, the examination may be repeated with the transducer oriented for transverse views, with images obtained with the probe parallel to the ribs.

FIGURE 16-12 Comet tail artifact (arrow).

Accuracy

Several studies have documented excellent sensitivity and specificity of ultrasound for the detection of a pneumothorax in the resuscitation area.^43,45,46,50 Dulchavsky et al. from Detroit Receiving Hospital, Wayne State University, showed that ultrasound can be successfully used by surgeons to detect a pneumothorax in injured patients.⁵¹ Of the 382 patients (364 trauma; 18 spontaneous) evaluated with ultrasound, 39 had pneumothoraces and ultrasound successfully detected 37 of them, yielding a 95% sensitivity. Not unexpectedly, pneumothoraces in two patients could not be detected because of the presence of significant subcutaneous emphysema. The authors recommended that when a portable chest x-ray cannot be readily obtained, the use of this bedside ultrasound examination for the identification of a pneumothorax can expedite the patient’s management.

One study, published in 2006, however, documents significant loss of accuracy of an ultrasound examination starting about 24 hours after chest tube insertion.⁵² This study documents the hospital course of 14 patients with tube thoracostomies undergoing 126 ultrasound examinations over their hospital course. While ultrasound detection of pleural sliding in uninstrumented pleural cavities remained 100% accurate over time, the accuracy of ultrasound examination after chest instrumentation fell to 65% after 24 hours.⁵²

Stay updated, free articles. Join our Telegram channel

Join