Indications for abdominal imaging: When and what to choose?

US is the most widely used imaging technique in abdominal emergencies⁽⁴⁾. It can be used in the setting of both acute and chronic pain. US excludes important pathology, and is sometimes the only imaging technique required to make a full diagnosis (e.g. in biliary lithiasis or cholecystitis). It can also guide decisions about further investigations. In the case of acute abdominal pain, different studies have shown that US adds 40% more information than clinical examination alone and changes the management in 20% of cases⁽⁵⁾. Using US in patients with acute abdominal pain can decrease the number of emergency abdominal CT examinations by a half. The combined use of US and CT in patients with inconclusive US examinations in cases of acute abdominal pain will reduce the percentage of missed urgent diagnoses to 6%⁽⁶⁾.

Ultrasound is the imaging technique of choice in patients with jaundice. It can demonstrate obstruction by showing dilated biliary ducts, and sometimes it can identify the cause of obstruction. Sensitivity for the detection of choledocholithiasis varies considerably across different centers, with values between 25% and 100% being reported⁽⁷⁾. Endoscopic ultrasound is the method of choice to rule out microlithiasis⁽⁸⁾.

Ultrasound should be used in patients with symptoms related to the urinary tract, such as hematuria (Fig. 1). The most important indications for ultrasound in urinary tract infection are to check for complications, such as nephritic foci in the kidneys or renal abscesses, or to exclude an obstructive cause of pyelonephritis.

Fig. 1.

Sensitivity of conventional B-mode ultrasound in detecting renal tumors depends on the size and location of the tumor. In one study the detection rate was 65% for tumors <2.5 cm and 80% for tumors >2.5 cm⁽⁹⁾. Contrast-enhanced ultrasound (CEUS) results are comparable to contrast-enhanced CT for characterization of focal renal lesions. US has a sensitivity of 87.1% and a specificity of 98.1% in diagnosing bladder tumors⁽¹⁰⁾.

Both US and CT are excellent techniques for confirming or excluding the presence of an abdominal mass, with sensitivity and specificity higher than 95%⁽¹¹⁾. The accuracy of US in determining the organ of origin is estimated at between 88% and 91%⁽¹¹⁾.

For patients with elevated liver enzymes, ultrasound should be the first imaging technique used. US performs well in the diagnosis of diffuse liver disease, with a sensitivity >90% for the diagnosis of advanced liver cirrhosis with complications⁽¹²⁾.

US is recommended as a follow-up technique for ectatic abdominal aorta (diameter between 2.6 and 2.9 cm). Aneurysms with a diameter >3 cm should be examined by CT. In pancreatitis, the first evaluation is often done by means of CT. Ultrasound is useful in follow-up since frequent CT scans lead to excessive irradiation exposure.

Where there is good visualization of the liver, contrast-enhanced ultrasound has a comparable sensitivity to contrast-enhanced CT or MRI for the diagnosis of liver metastasis. CEUS of the liver has been shown to detect more metastases in the follow-up of colorectal metastases than conventional B-mode US⁽¹³⁾.

Congenital abnormalities can be diagnosed with both antenatal and postnatal US evaluations. Ultrasound is typically the first imaging method used in urinary tract congenital abnormalities because of its easy availability, non-invasiveness and the fact that it is free from ionizing radiation. CT and MRI are indicated in complex urinary malformations to evaluate the collecting system and vascular anatomy.

CEUS has significantly improved the diagnostic performance of US in the diagnosis of parenchymal organ injuries, with sensitivity and specificity of >90% and up to 99% under certain circumstances; the performance is then similiar to that of CT. CEUS can prevent overutilization of CT⁽¹⁴⁾. Ultrasound is generally used as the diagnostic tool of choice in low-energy trauma limited to the abdomen. In a series of 57 patients with blunt abdominal trauma, the diagnostic accuracy of ultrasound in evaluating the existence of peritoneal fluid was 91%, although it was only 56% for the evaluation of parenchymal injury. However, another study reported considerably better accuracy, with a value of 94.7% for the combined presence of parenchymal injury and free peritoneal fluid⁽¹⁵⁾. The different results reported may be due to differences in experience between operators and centers. Focused Assessment with Sonography for Trauma (FAST) shows high sensitivity up to 99% for the detection of free fluid, but the sensitivity in the diagnosis of parenchymal injuries is poor.

Ultrasound is an additional method to CT for pre- and post-transplantation evaluations. CT is considered the technique of choice due to its better suitability for assessing vascular structures. US can be used in diagnosis and follow-up of non-vascular complications of transplants, such as abscesses or the presence of free fluid.

US can be used for guiding both diagnostic and interventional procedures in abdominal pathology. Interventional US benefits from the development of new techniques, such as fusion imaging (a technique that uses data from two different imaging modalities to improve the quality of information for increased diagnostic accuracy) and CEUS. The complication rate in interventional ultrasound is low, ranging from 0.51% to 0.81% in US-guided fine needle biopsy, although the rate ranges from 0.4% to 2.5% when a needle with a diameter over 1 mm is used⁽¹⁶⁾.

Ultrasound is the method of choice for the detection, quantification and localization of peritoneal fluid, with results superior to CT. US can detect even small amounts of fluid in the peritoneum⁽¹⁷⁾.

Ultrasound is the recommended imaging method for the diagnosis of abdominal emergencies in newborns and young children, such as hypertrophic pyloric stenosis or intussusception. For suspected hypertrophic pyloric stenosis, ultrasound, in experienced hands, has sensitivity and accuracy close to 100% and a diagnostic accuracy of 97%–100% for the diagnosis of intussusception⁽¹⁸⁾. In children, ultrasound is also the technique of choice to identify treatable abnormalities that favor upper urinary tract infections, other than vesico-ureteral reflux. Contrast-enhanced voiding urosonography is replacing radiographic voiding cystourethrography as the technique of choice in diagnosing vesico-ureteral reflux.

CEUS is indicated in both hepatic and non-hepatic pathology. The role of CEUS in characterizing focal liver lesions is well-established. When US is technically satisfactory, it offers comparable results to those of CT and MRI. A large multicenter study showed the value of CEUS in the characterization of focal liver lesions⁽¹⁹⁾. CEUS has also proved its value in renal, pancreatic and small bowel pathology. In most of the cases evaluated by CEUS, ultrasound contrast agents (UCA) are administered intravenously. Intracavitary administration of UCA can also offer useful diagnostic information. UCAs can be injected in physiologic cavities, such as the bladder for vesico-ureteral reflux or the uterine cavity for assessing tubal patency or in pathologic cavities for the characterization of fistulae. CEUS can be used intraoperatively especially in gastrointestinal surgery, but also in neurosurgery and interventional procedures, such as biopsies and interstitial ablation therapies which can be guided using CEUS.

CT has both very good sensitivity and specificity in diagnosing urinary lithiasis, even if low-dose protocols are used. The sensitivity of low-dose abdominal CT in the diagnosis of urolithiasis ranges from 90 to 98%, with only very small calculi (smaller than 3 mm) being missed. Specificity of low-dose CT in the diagnosis of urinary calculi ranges from 88 to 100%⁽²¹⁾. Bowel obstruction, suspicion of gastric or bowel perforation or lower gastrointestinal tract hemorrhage are also evaluated by CT. The accuracy of CT in diagnosing pneumoperitoneum is as high as 99%⁽²²⁾. A meta-analysis found that CT angiography has a sensitivity of 82.5% and a specificity of 92.1% in detecting gastrointestinal bleeding⁽²³⁾. The sensitivity of CT angiography in determining the underlying cause of gastrointestinal bleeding has been reported to be above 90% in multiple studies. Sensitivity and specificity of CT in the diagnosis of small bowel obstruction range from 81 to 100% and from 68 to 100%, respectively⁽²⁴⁾. CT has a lower sensitivity, estimated at 82%, for the diagnosis of enteromesenteric ischemia, a potentially letal condition⁽²⁵⁾.

The use of CT imaging reduces morbidity and mortality in hemodynamically stable patients with high-energy and severe multi-trauma and is, therefore, the standard technique. Clinical examination is notoriously unreliable in abdominal trauma, particularly in bowel and pancreatic injuries, and the misdiagnosis of these lesions is a wellknown cause of increased morbidity and mortality in patients who survive the initial phases of multiple trauma. The addition of CT to diagnostic protocols of patients with abdominal trauma will lead to a substantial decrease, of approximately 20%, of missed injuries by means of clinical examination and abdominal ultrasound.

CT has some disadvantages:

The latter is an important limiting factor, especially when there is a low-risk mechanism of injury and the patient’s condition would not necessarily warrant a CT examination, even though an imaging investigation is required. If possible, minor abdominal trauma should be managed by means of ultrasound and clinical surveillance, while CT is mandatory for major abdominal trauma or polytrauma. CT has good accuracy in the evaluation of solid abdominal visceral trauma and bowel/mesenteric trauma. Different papers report a sensitivity of CT ranging between 92 and 98% in depicting liver and spleen injuries⁽²⁶⁾. For small bowel and mesenteric injuries, CT is reported to have a sensitivity that varies between 70 and 95% and specificity between 92 and 100%⁽²⁷⁾. The sensitivity of CT in pancreatic injuries has been reported by different authors to range from 70 to 95%⁽²⁸⁾. CT can visualize the existence of active hemorrhage by showing contrast media extravasation.

CT has both better sensitivity and specificity than ultrasound in the diagnosis of renal tumors. The sensitivity of CT in detection of small renal masses is higher than 90% and approaches 100% in tumors larger than 2 cm⁽²⁹⁾.

CT is a useful imaging tool when there is a suspicion of tumor in the abdomen or pelvis and also for follow-up of patients with known oncologic pathology (Fig. 2). Local staging of pelvic tumors, such as uterine, prostatic or rectal tumors, is an indication for MRI and not CT. The presence of gastric or colonic tumors needs to be confirmed using endoscopy. Virtual CT-colonoscopy can be used as a substitute for conventional colonoscopy for the diagnosis of colonic polyps and tumors. Follow-up of oncologic disease should in most cases be performed using CT.

Fig. 2.

Patients with abdominal pain, fever or developing biologic changes, such as leukocytosis or elevated CRP, after surgical procedures should be evaluated by CT to look for the presence of complications, e.g. an abscess or fistula.

Patients with liver cirrhosis and nodules discovered on ultrasound screening should be evaluated by means of contrast-enhanced CT (CE-CT) or, preferably, contrast-enhanced MRI (CE-MRI). CE-MRI has, when compared to CT, higher sensitivity (0.82 vs 0.66) and lower negative likelihood (0.20 vs 0.37) for the diagnosis of nodules in a cirrhotic liver⁽³⁰⁾.

CE-CT/CT angiography (CTA) is a very good imaging technique for assessing congenital or acquired abnormalities of the abdominal vessels, such as evaluation of abdominal aneurysms or atherosclerotic disease leading to occlusion or critical stenosis of the abdominal vessels. Aortic aneurysms should be periodically evaluated by means of CT-angiography.

Among the available imaging techniques, including positron emission tomography-computed tomography (PET-CT), MRI has the highest accuracy for the characterization of focal liver lesions in both cirrhotic and non-cirrhotic patients. With a combination of T1- and T2-weighted images, diffusion-weighted imaging (DWI), and injection of hepatobiliary agents, such as Gd-EOB-DTPA, the majority of focal liver lesions can be accurately characterized and a differential diagnosis between benign and malignant focal liver lesions can be made⁽³²⁾. Overall, sensitivity and specificity of a complex MRI protocol, including DWI and Gd-EOB-DTPA injection, for detection and further classification of a focal liver lesion are high, ranging between 90 and 95%⁽³²⁾.

MRI can also be used to evaluate diffuse liver diseases. In liver cirrhosis, the main indication for MRI is detection of hepatocellular carcinoma. MRI is also very useful for predicting the grade of liver fibrosis, as there is a strong correlation between gadolinium accumulation in tissues in the late phases after injection and hepatic fibrosis. The diagnostic performance of MRI increases with the stage of liver fibrosis and is as high as 0.92 for fibrosis stage 4⁽³³⁾. MRI can also be used for the detection and quantification of other diffuse liver diseases, such as hemochromatosis, hemosiderosis, and steatosis.

MRI is an appropriate technique for evaluating liver infections and detecting liver abscesses in inconclusive findings. Its sensitivity to identify small differences in tissue composition leads to a very good specificity for certain hepatic infections, including hydatid cyst and candidiasis. MR imaging shows 100% sensitivity and 96% specificity for the diagnosis of hepatosplenic fungal disease⁽³⁴⁾.

In the evaluation of liver donors, liver transplants, and postoperative complications, MRI has advantages over other imaging techniques, particularly in the evaluation of post-transplant biliary pathology. Also, it can offer additional information to sonography-based techniques in the evaluation of posttransplant liver fibrosis. Magnetic resonance cholangiography provides a panoramic and detailed representation of the bile ducts, which is not achievable by other techniques. Accuracy of MRI in detecting posttransplant biliary strictures is as high as 92.3%⁽³⁵⁾.

MRI is an excellent technique for the assessment of biliary tract pathology. MRI may be preferred over CT for biliary tract pathology, because it has superior stone conspicuity and does not utilize ionizing radiation. It is preferred over diagnostic endoscopic retrograde cholangiopancreatography (ERCP), as the latter has high complication (3%–9%) and mortality rates (0.2%–0.5%)⁽³⁶⁾.

MRI is also used in pancreatic disease, with the indications for pancreatic MRI including characterization of suspected parenchymal abnormalities found on computed tomography (CT) or ultrasound (US), detection and staging of pancreatic neoplasms as an adjunct to CT, characterization of cystic pancreatic lesions, detection of small non-organ-deforming pancreatic ductal adenocarcinomas, detection of neuroendocrine tumors, evaluation of acute and chronic pancreatitis when CT fails to be diagnostic, detection of choledocholithiasis as a cause of acute pancreatitis, detection of intraluminal pancreatic calculi and staging of chronic pancreatitis.

MRI of the abdomen with positive digestive contrast media (usually water with an osmotic agent) should be used for the assessment of the extent and complications of inflammatory bowel disease. MRI has comparable results to CT in the detection of complications of inflammatory bowel disease, such as stenosis and fistula formation. The lack of ionizing radiation makes MRI a better technique than CT, particularly in young patients, as patients with inflammatory bowel disease often require numerous imaging studies during their lifetime.

MRI is an indispensable imaging tool in the preoperative local staging of rectal tumors. An initial local staging is performed to determine which patients require preoperative radiochemotherapy, or to plan surgery in those who do not require radiochemotherapy. MRI has an excellent soft tissue contrast, which facilitates assessment of tumor spreading into the perirectal fat, involvement of the mesorectal fascia (which is an indication for preoperative radiotherapy) and involvement of the sphincteric complex. MRI is indicated before and after radiochemotherapy to assess the response to therapy (Fig. 3).

Fig. 3.

MRI is the preferred imaging tool for the characterization of perianal fistulas. Imaging of perianal fistulas facilitates the detection of secondary fistulous tracts and the formation of abscesses. Information about the relationship between fistulous tracts and the sphincteric complex is also required by the surgeon prior to intervention⁽³⁷⁾.