Medical records were retrospectively analyzed for 76 patients who were subjected to abdominal US as the first-line diagnostic imaging modality at the National Hospital Organization Shimoshizu Hospital (Yotsukaido, Japan) from April 2010 to April 2015. Abdominal US was performed at the time of consultation or during off-hours (weeknights, weekends and holidays). Recruited patients were restricted to those in which abdominal US was performed as a first diagnostic approach in order to evaluate the diagnostic performance solely from abdominal US without any potentially confounding information from other diagnostic imaging procedures. Thus, patients were excluded when abdominal US was performed following another diagnostic imaging procedures, such as radiography, CT or MRI because the sonographer may have been informed of the findings obtained. The analyzed patients included 39 males with an average (mean ± standard deviation) age of 65.8±18.8 years and 37 females with an average age of 53.7±19.3 years.

Patients were hospitalized or referred to a different hospital based on the diagnosis obtained by abdominal US, the results of blood examinations, clinical findings or diagnostic imaging following abdominal US. The National Hospital Organization Shimoshizu Hospital does not have a department of gynecology. Therefore, patients were referred to another hospital for gynecological consultation. The referred hospitals were National Hospital Organization Chiba Medical Center (Chiba, Japan) and Seirei Sakura Citizen Hospital (Sakura, Japan). During off-hours, the management of patients was determined on the basis of abdominal US and clinical symptoms. The present study was approved by the Ethics Committee of the National Hospital Organization Shimoshizu Hospital. It was not considered a clinical trial since abdominal US was performed as a part of routine clinical practice. Written informed consent for inclusion in the study was waived. Patient records/information was anonymized and de-identified prior to analysis.

Abdominal US was performed by Senior Fellows of the Japan Society of Ultrasonics in Medicine (Tokyo, Japan; http://www.jsum.or.jp/jsum-e/index.html) using a SSA-700A US system (Toshiba Medical Systems Corporation, Ohtawara, Japan) with a 3.75-MHz curved-array probe (PVT-375BT; Toshiba Medical Systems Corporation) or an 8.0-MHz linear-array probe (PLT-805AT; Toshiba Medical Systems Corporation).

Acute cholangitis is defined as inflammation due to the obstruction of a bile duct (6). Findings detectable by abdominal US include bile duct dilatation and evidence of its etiology, such as stricture, stones or stent (6). In the present study, in the absence of these findings, acute cholangitis was not diagnosed by abdominal US, but was based on a combination of symptoms including systemic inflammation and cholestasis (6). Acute cholecystitis was diagnosed by a combination of local and systemic signs of inflammation (7). Abdominal US supported the diagnosis with findings of distension, wall thickening and sludge in the gallbladder (7). More specifically, the finding of a sonographic Murphy sign was considered the most reliable because it is considered to suggest inflammation of the gallbladder (17).

Acute pancreatitis is typically diagnosed as abdominal pain, elevated serum amylase and swelling of the pancreas by diagnostic imaging (18). In the current study, acute pancreatitis was diagnosed when abdominal US revealed a swollen pancreas, fluid collection and inflammation of adjacent organs. Acute diverticulitis was diagnosed as diverticulum with thickened wall and high echo from the surrounding tissue (Fig. 1A) (12). Acute appendicitis was diagnosed as swollen appendix with a diameter >10 mm, and thickened wall (Fig. 1B) (13,14).

Initial diagnosis with abdominal US was in agreement with the final diagnosis in 66 patients. For the remaining 10 patients, the diagnosis obtained from abdominal US differed from the final diagnosis. To investigate the performance and limitations of abdominal US, patients were divided into two groups. The first included patients in whom abdominal US diagnosis agreed with the final diagnosis (Table I) and the other included patients whose initial abdominal US diagnosis differed from the final diagnosis (Table II).

Solid organ diseases, including acute cholangitis, acute cholecystitis and acute pancreatitis, were correctly diagnosed by abdominal US. Intestinal diseases, including acute appendicitis and colonic diverticulitis, were also correctly diagnosed. Abdominal US was useful for the diagnosis of various types of cancer, including hepatocellular carcinoma, pancreatic cancer, and colorectal cancer (Fig. 2A). Colorectal cancer findings included irregular shaped wall thickening and loss of stratification, also referred to as the ‘pseudokidney’ sign (15). Critical conditions, such as rupture of the spleen, were also successfully diagnosed with abdominal US (Fig. 2B). Fluid in the abdominal cavity and high echo in the spleen suggested bleeding and damaging lesions, respectively (19). A total of 30 patients were admitted outside of regular working h and were subjected to abdominal US. They were correctly diagnosed without being subjected to CT or other diagnostic imaging procedures. These results clearly indicate that abdominal US is useful for the diagnosis of patients presenting with abdominal symptoms specifically during off-hours.

Table II summarizes data from those patients who were misdiagnosed by abdominal US and the potential reasons for misdiagnosis. These patients were subjected to additional diagnostic imaging procedures and were hospitalized for treatment because their conditions suggested the necessity of further management. Two patients were referred to another hospital for gynecological consultation. All patients described in Table II were appropriately managed, according to their condition.

Gas over the site of the disease made the examination difficult to execute. Gas over the bile duct (patient 2) hindered the detection of bile duct cancer. Gas in the sigmoid colon made the examination difficult for patient 3.

A diagnosis of acute appendicitis was hampered when it was not possible to detect the appendix (patient 4; Fig. 3A) (20).

A sonographic Murphy sign markedly suggests acute cholecystitis when abdominal US findings include distension and wall thickening of the gallbladder (17). Since the duodenum is close to the gallbladder, the tenderness perceived in duodenal ulcer may occasionally resemble a sonographic Murphy sign (patient 6; Fig. 3B).

A major difficulty in successful diagnosis with abdominal US was the presence of gynecological disorders. It was not difficult to detect a mass in the pelvis (patient 8) or a cystic lesion in the ovary (patient 9; Fig. 4A), however it was difficult to diagnose the site or organ of the original lesion, or relative torsion. The abdominal US clearly illustrated ascites and weakened peristalsis of the sigmoid colon (patient 10; Fig. 4B), but the correct diagnosis required subsequent referral to a gynecologist. Thus, these results suggested that patients should be referred for gynecological consultation when the abdominal US indicates abnormal findings in the pelvis, particularly in the ovary and uterus.