Материал: Advanced Imaging of the Abdomen - Jovitas Skucas

obstruction. A voiding cystourethrogram reveals a hypertrophied bladder wall. The presence of a urachal remnant is also common. The prostatic urethra is dilated, and a dilated prostatic utricle is often seen. Vesicoureteric reflux is common into dilated, aperistaltic ureters. Renal dysplasia is identified in some patients.

Diaphragmatic Abnormalities

Diaphragmatic development is complex, with the ventral portion evolving from the septum transversum and the dorsal portion originating from the pleuroperitoneal membrane. A Morgagni hernia anteriorly or a Bochdalek hernia posteriorly is a result of incomplete formation of these two diaphragmatic segments. A Bochdalek hernia is more common, usually is on the left, and can be massive to the point that respiratory distress is evident.

Diaphragmatic agenesis is occasionally detected in an asymptomatic adult. Imaging reveals associated herniation of colon, small bowel, and even kidney into the chest, together with a hypoplastic underlying lung.

Inadequate striated muscle development leads to diaphragmatic eventration. Eventration can be complete or partial. When extensive, imaging findings mimic a diaphragmatic hernia. Minor diaphragmatic eventration is of little significance and tends to resolve with age. A partial eventration is difficult to detect in neonates.

Conventional radiography detects most diaphragmatic hernias, although US is useful not only for diagnosis but also for follow-up. A peroral contrast study is generally diagnostic.

Noonan’s Syndrome

Noonan’s syndrome is a mostly autosomaldominant condition with facial dysmorphism, a number of congenital cardiac defects, and short stature. It is linked to the cardiofaciocutaneous syndrome, and both probably represent a variable expression of the same genetic defect. An association also exists between neurofibromatosis type 1 and Noonan’s syndrome. The reason for including this syndrome in a book on abdominal disorders is that some of these indi-

ADVANCED IMAGING OF THE ABDOMEN

viduals have abdominal lymphangiectasia, at times involving the gastrointestinal tract. About two thirds of children with Noonan’s syndrome have poor feeding and gastrointestinal dysfunc- tion—findings suggesting delayed gastrointestinal motor development—and require tube feedings. Also, a number of these patients suffer from various bleeding disorders. In some, lymphangiectasia is associated with pleural effusions, lymphedema, or a protein-losing enteropathy and a resultant hypoproteinemia. Lymphangiectasia can be identified in these individuals by lymphangiography, or, currently more often by lymphoscintigraphy, which reveals dilated and tortuous abdominal and pelvic lymphatic channels and abnormal lymphatic flow.

Computed tomography after bipedal lymphangiography in a 21-year-old man with Noonan’s syndrome and protein-losing enteropathy confirmed intestinal lymphangiectasia (5). After cardiac catheterization, a 15- year-old girl with this syndrome developed cutaneous lymphatic fluid oozing from a groin site (6).

Trauma

Unstable Patient

A hemodynamically unstable trauma patient requires immediate resuscitation. Exploration is considered in a patient with clinically evident abdominal trauma who is unresponsive to resuscitation. Imaging simply delays therapy in such a setting. A possible exception is a limited US study for intraperitoneal fluid while the patient is being resuscitated, but keep in mind the limitations of such a study.

Computed tomography of children in shock reveals dilated, fluid-filled bowel; intense enhancement of bowel wall, mesentery, pancreas, kidneys, and adrenal glands; and enhancement of a smaller than normal aorta and inferior vena cava (7). Similar but less pronounced changes are found in adults.

Stable Patient

A pneumoperitoneum in a trauma patient, regardless of whether detected with conven-

PERITONEUM, MESENTERY, AND EXTRAPERITONEAL SOFT TISSUES

tional radiography or CT, generally is an indication for exploration. Other findings, such as organ damage, peritoneal fluid, or a hematoma, in a stable patient are more judgmental. In some,whether to proceed with exploration in an otherwise stable patient with imaging-evident peritoneal fluid or organ damage is not clear, and a policy of observation is adopted; a repeat of appropriate imaging studies is often in order for these patients.

Currently most minor liver and splenic trauma is managed conservatively, although a trend is evident toward nonoperative management of hemodynamically stable patients even with more severe injury. Thus stable patients even with a moderate amount of hemoperitoneum have been managed conservatively, with no differences found between nonoperative and operative groups in resultant abdominal complications and hospital length of stay

(8).

Penetrating Injury

The vast majority of gunshot wounds involving the peritoneal cavity require surgical repair. The diagnostic dilemma is determining which of these injuries penetrate the peritoneum. Several diagnostic peritoneal lavage studies in gunshot wound patients achieved a sensitivity of and specificity of over 95% in determining peritoneal penetration. To a large degree CT has supplanted lavage in patients with penetrating trauma, also achieving sensitivities and specificities of over 95% (9).

Most traumatic visceral artery aneurysms (pseudoaneurysm) are due to penetrating injury. A not uncommon scenario is a patient who has surgery shortly after trauma, undergoes arterial ligation, and then presents with a gastrointestinal bleed several weeks later from an aneurysm.

A reasonable approach in stable patients with abdominal stab wounds is to obtain initial CT or US, and in the absence of evidence for immediate surgery to follow them with serial imaging.

Diagnostic Peritoneal Lavage

In the 1980s diagnostic peritoneal lavage was generally considered superior to CT, although

its use has decreased markedly over the last decade, having been supplanted by CT and US. Nevertheless, an occasional clinician still recommends that lavage be performed first in a setting of blunt trauma if no contraindications exist.

Diagnostic peritoneal lavage relies on detecting blood in the peritoneal cavity. Generally an arbitrary threshold for a positive test, such as 10,000 red blood cells per cubic millimeter, is assumed. A higher threshold increases the missed injury rate and a lower one increases the false positive rate. The advantages of diagnostic peritoneal lavage include its simplicity and its relatively high sensitivity in detecting intraperitoneal blood. It does not evaluate the severity of injury, and thus is limited in predicting a need for surgery. It is insensitive for retroperitoneal injuries. Even with intraperitoneal injuries, it may miss blood in patients with previous abdominal surgery and extensive adhesions.

A comparison of diagnostic peritoneal lavage and CT in patients with blunt trauma is difficult because each study evaluates different findings.

Peritoneal Fluid

Although a number of investigators believe that US readily detects intraperitoneal fluid, less often discussed is how much fluid is necessary for detection with US. In a blinded prospective study of 100 patients undergoing diagnostic peritoneal lavage, continuous US scanning of Morison’s pouch revealed that the mean volume of infused fluid first detected was 619mL and that detection sensitivity after infusing 1L was 97% (10). Even keeping in mind that intraperitoneal fluid appears to be twice as common in the pouch of Douglas than in Morison’s pouch, statements in the literature about small, moderate, and large amounts of fluid detected with US should be viewed with a jaundiced eye.

Multiple US scans are necessary to detect abnormal fluid; a single view, such as only of Morison’s pouch, misses intraperitoneal fluid in a number of patients. In general,in patients with acute trauma evaluated with US, the sensitivity for detecting free fluid is about 65% to 80% and the specificity about 95%, with free fluid in the pelvis being the most common reason for a false-negative finding. Most peritoneal fluid

detected after trauma represents blood; less common is urine, bile, or intestinal content. Pus and chyle develop if presentation is delayed.

The presence of intraperitoneal fluid correlates with injury but does not predict whether surgery will be necessary. Surgical teaching often mandates laparotomy after blunt trauma if isolated intraperitoneal fluid is detected by imaging, yet this is a complex and controversial topic. In general, blunt trauma patients eventually requiring laparotomy have more intraperitoneal fluid than those managed conservatively; an association also exists between the amount of mesenteric fluid and mesenteric laceration.

General Imaging Considerations

Multiorgan damage is common in major trauma; thus a finding of an abnormal collection of intraor extraperitoneal fluid is not necessarily due to a visible liver or splenic injury, but could also represent a synchronous mesenteric or bowel injury. A US study of over 1000 women of reproductive age with blunt trauma concluded that fluid isolated to the cul-de-sac is likely physiologic, but those with free intraperitoneal fluid usually have clinically important abdominal injuries (11).

If one has the luxury of time, chest and abdominal radiography are reasonable studies, although if CT is available, a strong argument can be made for using it initially. The ready availability of diagnostic CT and its ability to detect other conditions continue to expand the indications for CT in patients with abdominal trauma, and in many centers CT is the first imaging examination performed in a hemodynamically stable patient with a suspected intraabdominal injury. Exceptions include the hemodynamically unstable patient, one with an immediate life-threatening condition, or the patient who is to undergo emergent surgery for nonabdominal trauma.

In spite of an occasional admonition by emergency physicians, radiologists in the United States administer both IV and oral contrast prior to CT to most trauma patients. Very few complications due to contrast are reported. An extensive literature exists on IV contrast reactions, and this topic is beyond the scope of this book. Most radiologists believe that oral contrast aids in study interpretation, and that the advantages of contrast use outweigh any possi-

ADVANCED IMAGING OF THE ABDOMEN

ble disadvantages. Gastroesophageal reflux and aspiration is uncommon even in obtunded or uncooperative patients, and oral contrast is often administered through a nasogastric tube. Lung contrast results if contrast is instilled through a tube placed into a bronchus. The oral iodinated contrast agents used in CT are hypoosmolar and should not be compared to hyperosmolar full-strength contrast agents employed for other examinations.

A pneumoperitoneum can develop after chest trauma. Some of these patients also have an associated pneumothorax, pneumomediastinum, or a retropneumoperitoneum. The pneumoperitoneum can generally be identified with conventional radiography, although occasionally it is detected only with CT.

The vital signs of trauma patients are generally being monitored, and an imaging study should not be relied on to detect hypotension. Nevertheless, on a contrast-CT study the presence of a prolonged nephrogram without excretion into collecting systems should suggest hypotension. A collapsed inferior vena cava should suggest hypovolemia, with or without hypotension. Likewise, the spleen may become smaller than normal. Small bowel ischemia, manifesting as diffuse bowel wall thickening, may develop.

Computed tomography evaluates both the presence of fluid and organ injury. At times a CT study is equivocal. If surgical exploration is not contemplated, repeat CT is often helpful in monitoring the progression of any abnormalities.

In some institutions, especially outside the United States, US rather than CT has replaced diagnostic peritoneal lavage and is often used as a screening modality for suspected abdominal trauma. Use of US in trauma patients has generated strong opinions. Statements such as CT “. . . is costly, time-consuming, requires sedation, and may be associated with complications in young children . . .” while US “. . . is quick, noninvasive, repeatable, and costeffective . . .” have appeared in the trauma literature (12). Numerous studies extol the virtues of US in trauma patients, yet operator experience is difficult to place in perspective. Pediatric surgeons in particular advocate US as a triage tool in pediatric trauma patients and believe that it alone is sufficient to evaluate children after blunt abdominal trauma. Some believe that only

PERITONEUM, MESENTERY, AND EXTRAPERITONEAL SOFT TISSUES

those with abnormal US should be further studied with CT, a conclusion of dubious validity.

Advocates of US in a setting of trauma rely primarily on detecting intraperitoneal fluid, yet such reliance as a sole indicator of visceral injury does not appear warranted. Although a minority, some patients with later proven visceral injuries develop little or no hemoperitoneum. Nevertheless, screening US studies in patients with blunt abdominal trauma have published sensitivities of 85% to 95% for detecting injuries severe enough to require laparotomy. Abdominal US is considered positive if either intraor retroperitoneal fluid is detected. False negative studies include retroperitoneal injury, bowel injury and intraperitoneal solid organ injury without presence of a hemoperitoneum. Yet critical analyses of US point to a study of limited value. Major organ damage is missed and active hemorrhage is not identified. The bowel and pancreas are poorly visualized. In a trauma setting, the superiority of CT over US has been established by a number of studies. In one study, CT revealed fluid, organ injury, or both in 33% of consecutive children with blunt abdominal trauma (13); the sensitivity and specificity of US for fluid detection only was 47% to 59% and 79%, respectively (for two observers), and the sensitivity and specificity of US when fluid and organ injury were considered was 65% to 71% and 71% to 79%, respectively; the authors concluded that the low US sensitivity suggests that “a normal screening sonography alone in the setting of blunt abdominal trauma fails to confidently exclude . . . intraabdominal injury” (13). Ultrasonography can probably be justified, however, in those institutions lacking the ready availability of CT.

At times CT is used in patients with minimal trauma to decide whether to discharge a patient or not. Ultrasonography is generally considered not adequate to answer this question and many trauma US studies rely on keeping a patient under observation for some time.

Currently MR is not considered appropriate for screening trauma patients.

Diagnostic/therapeutic laparoscopy has been performed in patients with suspected abdominal trauma, with conversion to open exploration as needed. The role for such laparoscopy is yet to be established.

Bowel Injury/Perforation

In a trauma setting, CT detection of peritoneal fluid, in the absence of any visible solid organ injury, suggests bowel injury.About half of these patients have small bowel or diaphragmatic injury, although isolated intraperitoneal fluid can be associated with unsuspected injury from bowel and mesenteric injuries, to solid organ trauma.

Complicating the issue is that some patients with subsequently detected major bowel injury have no hemoperitoneum on admission CT and US, but bowel and mesenteric injury is detected only hours later; even then, bowel and mesenteric injury can be difficult to diagnose. Currently such injury is probably best studied with CT. Both IV and oral contrast are helpful. A prospective CT study achieved a sensitivity of only 64% but a specificity of 97% in detecting bowel injury in patients with blunt abdominal trauma (14); findings used to detect bowel injury included mesenteric infiltration, bowel wall thickening, extravasation either of vascular or enteric contrast, and the presence of pneumoperitoneum. Bowel wall thickening, in particular, is difficult to put in proper perspective as a finding of major bowel injury. If associated with a mesenteric hematoma, sufficiently severe mesenteric or bowel injury is generally presumed to warrant considering surgery. On the other hand, a focal mesenteric hematoma without adjacent bowel wall thickening occurs both in those patients requiring surgery and those who do not. Computed tomography has a high specificity in detecting a mesenteric hematoma. Nevertheless, the true accuracy of CT in establishing major bowel or mesenteric injury is difficult to judge, and published conclusions vary.

With a perforation, imaging rarely identifies bowel wall discontinuity. Intraperitoneal spill of oral or rectal contrast identified by CT is usually assumed to represent a bowel perforation, but although diagnostic, it is rarely detected. Spill of instilled contrast from a urinary tract perforation is in the differential diagnosis.

In pediatrics the role of CT in detecting bowel perforation appears even more limited than in adults, and CT identifies small bowel injury only in a minority. Clinicians should be aware of this CT limitation and not be lulled into a false sense of security, leading to a delay in surgery.