Acute abdomen is a common complaint in emergency departments (EDs). Plain abdominal radiography (PAR) is one of the essential tools in the diagnosis of this condition in addition to clinical laboratory data and physical examination. There has been a significant debate over the use of PAR as a diagnostic modality in cases of acute abdomen. Much literature states that the sensitivity of PAR is low and can result in unnecessary radiation exposure in the most of the case.1–7 The 2017 Royal College of Radiologists iRefer guidelines for PAR only lists a few indications for PAR, including bowel obstruction and constipation.8 However, due to some reasons, such as financial concerns or difficult access to the advanced instruments, many hospitals still choose PAR for preliminary diagnosis. For patients whose PAR shows an imaging finding (eg, bowel dilatation, free air, abnormal calcifications, or soft tissue density) or who exhibit severe symptoms, an abdominal computed tomography (CT) scan is additionally performed.
PAR can be categorized into two types: (1) supine abdominal radiograph (SAR) and (2) erect abdominal radiograph (EAR). Both these types have pros and cons, resulting in dilemma in doctors’ decision making over the choice of modality for their patients. An incorrect choice can cause difficulties and doubts in making a diagnosis.
This study aimed to explore which one of the SAR and EAR has a higher diagnostic value in diagnosis of overall acute abdomen and their respective advantages over each other for the various underlying diseases. When we encounter a patient with acute abdomen in ED and are considering a PAR, we want to answer the following question: which should I choose to ensure a better result for a more precise diagnosis, SAR or EAR?
2.1. Study subjects
This study retrospectively included 1009 patients who presented with acute abdomen to the ED of a medical center in northern Taiwan and underwent PAR and subsequent abdominal CT within 24 hours between October 2019 and February 2020. Patients with abdominal trauma were excluded. The patients received one of two types of PAR: SAR and EAR. SAR was performed with radiograph perpendicular to the patient’s abdomen in supine position with the lower margin of the image field including the patient’s symphysis pubis. EAR was performed with radiograph perpendicular to the patient’s abdomen in standing position, with the upper margin of image field including the patient’s diaphragm. These patients with acute abdomen were divided into two groups in the study: SAR group and the EAR group.
2.2. Study design
The final clinical diagnoses of patients were based on the CT reports and clinical data such as symptoms and signs, lab data, endoscopy results, and surgical findings. The clinical diagnoses were categorized according to the disease classification into 6 main groups and 20 subgroups (see Table 1). All PAR images of the patients were reviewed by two certified general radiologists. The PAR images were evaluated first to determine whether there were findings listed in Table 2. The readers are blinded to the clinical data when interpretating the PARs. A PAR is defined to have “diagnostic relevance” (or have relevant findings) when the PAR findings are related to the disease and helpful in the clinical diagnosis. The diagnostic relevance of imaging findings of PAR to the clinical diagnosis were determined by consensus of the two certified radiologists.
Table 1 -
Disease classification (6 main groups and 20 subgroups)
|A: GI system
||A1: Hollow organ perforation
A2: Bowel obstruction
A3: Adynamic ileus
A6: Gastritis, peptic ulcer (without perforation)
A7: GI cancer (without obstruction)
A8: Stool impaction
|B: Hepatobiliary system
||B1: Diffuse liver disease (eg, liver cirrhosis, fatty liver, hepatomegaly, hepatitis)
B2: Focal liver lesion (eg, liver tumor, abscess, cyst)
B3: Gallbladder or biliary tract disease
||Pancreatitis, pancreatic tumor
|D: Urinary system
D2: Urinary tract infection (UTI) (eg, pyelonephritis, ureteritis, cystitis)
D3: Ureter diseases (Not urolithiasis nor UTI, eg, tumor, stricture)
D4: Renal diseases (Not urolithiasis nor UTI, eg, tumor, polycystic kidney disease)
D5: Urinary bladder diseases (Not urolithiasis nor UTI, eg, tumor, acute urine retention)
|E: GYN system
||Tubo-ovarian abscess, pyometra, GYN tumors
||Mesenteric adenitis, panniculitis, peritoneal tumors, abscess (unspecified), hernia
GI = gastrointestinal; GYN = gynecological; UTI = urinary tract infection.
Table 2 -
Plain abdominal radiography findings
|1: Negative or nonspecific
2: Free air
3: GI tract dilatation
4: GI tract fold/wall thickening
5: Air-fluid level in bowel
6: Small bowel feces sign
7: Excessive stool in colon
10: Gallbladder stone
13: urinary bladder distention
14: Obliterated psoas muscle margin
15: Abnormal soft tissue density
16: Abnormal air collection
17: Abnormal calcifications unspecified
19: Intra-abdominal foreign body
20: Postoperative appearance
21: Extra-abdominal position of GI tract
22: Bone lesion
2.3. Statistical analysis
The case numbers and the percentages of each group of SAR and EAR and their diagnostic relevance were calculated. The case numbers and percentages of each disease under the disease classification whose PAR (SAR or EAR) presented with relevant findings were also calculated. For patients whose PAR showed relevant findings, the number of occurrences of findings were calculated. The relevance between the imaging findings of PAR and clinical diagnosis in the group with SAR were compared with those in the group with EAR. The difference in their diagnostic relevance for various diseases between the groups with SAR and EAR was additionally analyzed. A chi-squared test was performed with SPSS for Windows version 19.0 (IBM Corp., Armonk, NY, USA) for statistical analysis. The significance level of the chi-square test was set at 0.05 (two-tailed) with a 95% confidence interval. Institutional review board approval was obtained before conducting the study.
This study included 1009 patients (males, 525; females, 484) aged 4–99 years (mean age 56.6; median age 58) who presented acute abdomen in the ED. Among these, 668 (66.2%) underwent SAR and 341 (33.8%) underwent EAR. All patients underwent CT scan within 24 hours after their PAR. Among all the patients, there were 820 cases whose CT scans indicated a definite clinical diagnosis (81.3% of the total), of which 542 (66.1%) underwent SAR and 278 (33.9%) underwent EAR. The other 189 cases showed negative CT findings and inconclusive clinical diagnosis; therefore, they were excluded in the statistical analysis. Overall, there was no statistical difference in PAR diagnostic relevance for acute abdomen between the SAR and EAR groups (sensitivity: 43.5% vs 43.9%, p > 0.05), as shown in Table 3. For respective disease, a statistical difference in the diagnostic relevance between SAR and EAR was only found in bowel obstruction. There was no definite statistical difference in any of the other abdominal diseases. For patients whose PAR showed relevant findings, the most frequently seen findings and their possible diagnoses were listed in Table 4.
Table 3 -
The case numbers, numbers of those whose PAR showed relevant findings, and the difference between SAR and EAR for each disease
||With relevant finding (%)
||With relevant finding (%)
||With relevant finding (%)
||(SAR vs EAR)
|A: GI system
|A1: Hollow organ perforation
|A2: Bowel obstruction
|A3: Adynamic ileus
|A6: Gastritis, peptic ulcer
|A7: GI cancer
|A8: Stool impaction
|B: Hepatobiliary system
|B1: Diffuse liver disease
|B2: Focal liver lesion
|B3: Gallbladder or biliary tract disease
|D: Urinary system
|D2: Urinary tract infection
|D3: Ureter disease
|D4: Renal disease
|D5: Urinary bladder disease
|E: GYN system
*p < 0.05.
***p < 0.001.
EAR = erect abdominal radiograph; GI = gastrointestinal; GYN = gynecological; NS = no significance; PAR = Plain abdominal radiography; SAR = supine abdominal radiograph.
Table 4 -
Frequently seen PAR findings and their related diseases
|GI tract dilatation (n = 276)
Gastric/small bowel/colon dilatation (n = 219)
Air-fluid level in bowel (n = 57)
|Bowel obstruction, enteritis, ischemic bowel, adynamic ileus, local irritation caused by pancreatitis
|Gastric/small bowel/colon wall thickening (n = 21)
||Ischemia, infectious/inflammatory process, neoplasm
|Excessive stool in colon (n = 40)
|Extraluminal air (n = 29)
Free air (n = 12)
Abnormal air collection (n = 12)
Pneumobilia (n = 5)
|Hollow organ perforation
Cholangitis, postoperative change
|Abnormal calcifications (n = 108)
Gallbladder stone (n = 28)
Urolithiasis (n = 66)
Abnormal calcifications unspecified (n = 14)
Infectious/ inflammatory process, neoplasm
|Abnormal soft tissue density (n = 94)
Urinary bladder distention (n = 11)
Obliterated psoas muscle margin (n = 3)
Abnormal soft tissue density unspecified (n = 56)
Ascites (n = 24)
|Urinary bladder diseases, GYN diseases, neoplasms, abscess, or ascites
Many past studies have indicated that PAR does not help diagnose acute abdomen in EDs.4–6 Others have argued that PAR is only helpful in diagnosing those patients with bowel obstruction or bowel perforation.1–3,7 A meta-analysis in 20207 stated that the sensitivity of PAR for acute abdomen ranged from 30% to 46% with specificity from 75% to 88%, and for suspected bowel obstruction, the range of sensitivity was 48% to 96% with specificity from 50% to 100%. Our study found that among the diseases diagnosed by CT, the overall sensitivity of PAR was 43.7%, while the sensitivity of PAR for bowel obstruction is 92.1%.
In terms of comparing SAR and EAR, two studies9,10 showed that when EAR was added to SAR of acute abdomen, the diagnostic accuracy did not improve significantly. On the other hand, two studies11,12 believe that EAR is very helpful in evaluation of bowel obstruction. There is no clear conclusion about which of EAR or SAR is better. Our study has answered the question: There is no significant difference between SAR and EAR in evaluation of overall acute abdomen.
Regarding gastrointestinal (GI) diseases, PAR was significantly helpful in diagnosing the following diseases (sensitivity of having relevant findings in images ≥50%): hollow organ perforation (HOP), bowel obstruction, adynamic ileus, and stool impaction. In general, in patients having bowel dilatation or free air in their peritoneal cavity, PAR exhibits a satisfactory diagnostic effect. The conditions in which SAR and EAR exhibited significantly different diagnostic relevance were bowel obstruction and stool impaction.
For diagnosing free air, a study13 has shown that the sensitivity of SAR on HOP is 80.4%, which is higher than that of 62.5% in this study. We also unexpectedly noticed that there was no statistical difference in the sensitivities between EAR and SAR in terms of free air detection (Fig. 1).
Regarding bowel obstruction, a meta-analysis in 20207 stated that the sensitivity of PAR for suspected bowel obstruction ranged from 48% to 96% with specificity from 50% to 100%. Our study showed that the sensitivity of PAR was 92.1%. The diagnostic sensitivities of EAR and SAR were 100% and 87.2%, respectively (p = 0.010). By further analyzing, the cases of intestinal obstruction that SAR could not detect, we found that these patients’ bowels were almost completely filled with fluid, which led to a failure in their radiographic visibility. In contrast, in patients who underwent EAR, the fluid in the obstructed bowel had moved to the dependent part, which led to a radiographically notable air-fluid level and thereby identifiable bowel dilation in the abdomen, resulting in a more favorable diagnostic effect (Fig. 2). Two studies11,12 also mentioned the importance of the air-fluid levels in EAR for diagnosing small bowel obstruction and distinguishing its severity.
Regarding stool impaction, SAR showed high diagnostic sensitivity (100% in our study). On the other hand, EAR may miss the stool in rectum because it does not cover the lowermost part of the pelvic cavity. The statistical difference between EAR and SAR may fall short of a clinical significance due to the small sample size.
Regarding liver and gallbladder diseases, PAR was found to be only significantly helpful in diagnosing diffuse liver disease (sensitivity ≥50%).
For diagnostic relevance in diffuse and focal liver disease, there was a statistical difference between EAR and SAR; however, given the small sample size of this study, this finding may fall short of a clinical significance. On further analysis of each cases, we found that PAR could diagnose liver cirrhosis mainly by the detection of ascites (Fig. 3). For focal liver lesion, PAR is only helpful in air-containing liver abscesses. For other diseases, such as metastasis, hepatocellular carcinoma, and liver abscess without air, PAR has no diagnostic value.
The diagnostic relevance of PAR on pancreatic diseases was below 50%, and there was no difference in terms of diagnostic relevance between SAR and EAR.
For the diseases of the urinary system, PAR was found to be helpful in diagnosing urolithiasis and urinary bladder problems (sensitivity ≥50%), but the diagnosis of bladder problem can only be made when the patients had a distended urinary bladder. In diagnosis of all urinary diseases, no statistical difference was found between SAR and EAR.
The diagnostic relevance of PAR with regards to gynecological (GYN)–related diseases was below 50%, and there was no statistical difference in diagnostic relevance between SAR and EAR, which was unexpected. In our expectation, SAR should have a higher diagnostic effect than EAR because SAR includes the whole pelvic cavity; however, we did not note this difference in diagnosis by our study. If a GYN lesion can be detected by PAR (which means the lesion is dense or large enough), usually the field covering by the EAR is sufficient for lesion detection in patients with ordinary body size (Fig. 4).
PAR is not adequately sensitive for diagnosing many diseases, but the presence of certain findings can help in detecting abnormalities. Some frequently encountered findings are listed in Table 4. Paying particular attention to these findings is essential for radiologists and emergent physicians in decision making at ED:
- • GI tract dilatation: This often suggest bowel obstruction, but other causes include enteritis, adynamic ileus, and local irritation caused by pancreatitis.
- • GI tract wall thickening: This may relate to ischemia, infectious/ inflammatory diseases of GI tract or even the GI tract neoplasms.
- • Extraluminal air: This may indicate HOP (free air), the existence of abscess (localized mottled air), or pneumobilia (branching air in liver).
- • Abnormal calcification: This includes gallstones, urolithiasis, other calcifications in the inflammatory diseases (Fig. 5), chronic granuloma, organized hematoma, foreign body (eg, fish bone) or calcified neoplasm.
- • Abnormal soft tissue density: This may indicate urinary bladder distention, GYN diseases (Fig. 6), neoplasms (GI, renal, biliary) (Fig. 7), abscess, or ascites.
In comparison of the frequently used image modality for acute abdomen (PAR, ultrasound [US], CT), several studies had discussed the pros and cons of each modality.8,14,15 CT has the most accurate and reproducible diagnosis to acute abdomen, but it has higher cost, lower availability, and high radiation exposure. On the contrary, US and PAR have lower diagnostic sensitivity but lower cost and higher accessibility. PAR has low radiation dose and US has no radiation dose. Because PAR has the lowest cost and highest availability in Taiwan Health Care System, PAR is often the first line imaging evaluation of acute abdomen at ED, which could help the emergency physicians for further decision making. PAR is most valuable in diagnosing bowel obstruction, HOP, urolithiasis, and foreign body, but low sensitivity and specificity for other causes of acute abdomen.
This study has some limitations. The first limitation is the selection bias. All the patients in this study underwent a CT scan following their PAR (which could possibly indicate a more serious condition), and the choice of SAR or EAR depended on the doctors’ preference at the ED. Second, the sample size was small for several diseases. Finally, there may be difference in interpretation between radiologists and emergency physicians.
In conclusion, among the cases with definite clinical diagnosis by CT and clinical data, 43.7% showed relevant findings in PAR. PAR is more helpful in diagnosing the following diseases/conditions: HOP, bowel obstruction, adynamic ileus, stool impaction, liver cirrhosis with ascites, urolithiasis, and urinary bladder distention. PAR is not adequately sensitive for diagnosing many diseases, but the following findings are helpful and should be especially monitored: GI tract dilatation and wall thickening, extraluminal air, abnormal calcifications, and abnormal soft tissue density. There is no significant difference between SAR and EAR in evaluation of overall acute abdomen. However, EAR does have a diagnostic advantage over SAR for the patients with suspected bowel obstruction. Thus, if bowel obstruction is clinically suspected in ED, EAR is strongly advised for the emergency physicians for further imaging diagnosis.
We wish to acknowledge the help of doctor Rivero in English editing.
1. Ahn SH, Mayo-Smith WW, Murphy BL, Reinert SE, Cronan JJ. Acute nontraumatic abdominal pain in adult patients: abdominal radiography compared with CT evaluation. Radiology. 2002;225:159–64.
2. Prasannan S, Zhueng TJ, Gul YA. Diagnostic value of plain abdominal radiographs in patients with acute abdominal pain. Asian J Surg. 2005;28:246–51.
3. Ashindoitiang JA, Atoyebi AO, Arogundade RA. The value of plain abdominal radiographs in management of abdominal emergencies in Luth. Nig Q J Hosp Med. 2008;18:170–4.
4. Gans SL, Stoker J, Boermeester MA. Plain abdominal radiography in acute abdominal pain; past, present, and future. Int J Gen Med. 2012;5:525–33.
5. Marasco G, Verardi FM, Eusebi LH, Guarino S, Romiti A, Vestito A, et al. Diagnostic imaging for acute abdominal pain in an Emergency Department in Italy. Intern Emerg Med. 2019;14:1147–53.
6. Fernandez M, Craig S. Appropriateness of adult plain abdominal radiograph
requesting in a regional Emergency Department. J Med Imaging Radiat Oncol. 2019;63:175–82.
7. Denham G, Smith T, James D, McKiernan S, Evans TJ. Exploring the evidence-practice gap in the use of plain radiography for acute abdominal pain and intestinal obstruction: a systematic review and meta-analysis. Int J Evid Based Healthc. 2020;18:159–69.
8. Susanne S, Catriona R, Akkib R. Indications for plain abdominal films from the Emergency Department. The Royal College of Radiologists. 2018. Available at https://www.rcr.ac.uk/audit/indications-plain-abdominal-films-emergency-department
. Accessed August 26, 2021.
9. Field S, Guy PJ, Upsdell SM, Scourfield AE. The erect abdominal radiograph in the acute abdomen: should its routine use be abandoned? Br Med J (Clin Res Ed). 1985;290:1934–6.
10. Geng WZM, Fuller M, Osborne B, Thoirs K. The value of the erect abdominal radiograph for the diagnosis of mechanical bowel obstruction and paralytic ileus in adults presenting with acute abdominal pain. J Med Radiat Sci. 2018;65:259–66.
11. Lappas JC, Reyes BL, Maglinte DD. Abdominal radiography findings in small-bowel obstruction: relevance to triage for additional diagnostic imaging. AJR Am J Roentgenol. 2001;176:167–74.
12. Thompson WM, Kilani RK, Smith BB, Thomas J, Jaffe TA, Delong DM, et al. Accuracy of abdominal radiography in acute small-bowel obstruction: does reviewer experience matter? AJR Am J Roentgenol. 2007;188:W233–8.
13. Chiu YH, Chen JD, Tiu CM, Chou YH, Yen DH, Huang CI, et al. Reappraisal of radiographic signs of pneumoperitoneum at emergency department. Am J Emerg Med. 2009;27:320–7.
14. Stoker J, van Randen A, Laméris W, Boermeester MA. Imaging patients with acute abdominal pain. Radiology. 2009;253:31–46.
15. Cartwright SL, Knudson MP. Diagnostic imaging of acute abdominal pain in adults. Am Fam Physician. 2015;91:452–9.