Prevalence of Ordering Computed Tomography Scans for Non-Traumatic Abdominal Pain in the Emergency Department at A Tertiary Care Center

Ali Wakil MD¹*, Mujtaba Mohamed MD², Alsadiq Al Hillan MD³, Abbas Al Shami MD⁴, Mohammad A Hossain, MD⁴, Robert Sweeney, DO⁴

1. Department of Gastroenterology and Hepatology, Rutgers New Jersey Medical School, Newark NJ

2. Department of Gastroenterology and Hepatology, Marshal Medical school. Huntington, WV.

3. Department of Gastroenterology and Hepatology, Baoument Health medical center. Royal Oaks, MI.

4. Department of Medicine, Jersey Shore University Medical Center, Hackensack Meridian Health, Neptune, NJ.

Corresponding Author: Ali Wakil, MD, Gastroenterology and Hepatology, Rutgers New Jersey Medical School, Newark, NJ 07103.

Copy Right: © 2022 Ali Wakil, MD, This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Received Date: July 11, 2022

Published Date: August 01, 2022

Abstract

Background & Aims: Abdominal pain is the most common cause for emergency department (ED) referrals. Computed tomography (CT) scans of the abdomen and pelvis are the standard choice for the diagnosis of acute abdominal pain. However, the cost, radiation exposure, and availability of CT scans may make other imaging modalities preferable as first-line tools for non-traumatic abdominal pain. We retrospectively reviewed patient records of those presenting with non-traumatic abdominal pain to the ED and received an abdominal CT scan. Our study's goal was to identify ED prevalence of ordering CT scans for non-traumatic abdominal pain patients compared with other diagnostic imaging modalities such as ultrasound (US) or x-rays.

Methods: We analyzed demographic characteristics, comorbidities, clinical presentation, time of ordering abdominal and pelvic CT, CT scan findings, US orders, and admission status of patients presenting to the ED with non-traumatic abdominal pain from July 2017 to October 2017. Summary statistics of continuous variables were reported as mean ± standard deviation. We created a regression model to identify predictors of positive results of abdominal and pelvic CT scan

Results: A total of 496 patients were included in our study (mean age: 49.7 years; male-to-female ratio: 40:60). US imaging was ordered for 34 patients (7%), and x-ray/obstruction series were ordered for 55 patients (11%) before orders for abdominal and pelvic CT scans. Most patients were diagnosed with non-specific abdominal pain (n=154, 31%). A total of 173 patients (35%) received orders for abdominal CT scan immediately on evaluation in the ED before basic blood work.  Seventy percent of patients were discharged from the ED after a few hours. In 30% of cases, were admitted for further evaluation.

Conclusion: CT scans are commonly ordered for the diagnosis of acute abdominal pain. Fifty percent of patient presented with abdominal pain had CT scan. The most common cause of abdominal pain based on CT scan results is nonspecific abdominal pain in our study. Majority of CT scans were ordered either at same time or after one hour after ordering preliminary blood work. Findings such as absence of diabetes mellitus, history of renal stones, leukocytosis, and acute kidney injury, were correlated enough to predict a positive CT result.

Keywords: Tomography, X-Ray Computed; Abdominal Pain; Emergency Service, Hospital; Medical Overuse; Radiation, Ionizing; Ultrasonography.

Introduction
Abdominal pain accounts for 7% to 10% of all emergency department (ED) visits, and it is the most common cause for ED referrals from outpatient clinics [1]. Most cases of abdominal pain (31%) are diagnosed as non-specific abdominal pain, followed by renal colic (usually in male patients) [2]. Elderly patients are more likely to have a specific etiology of abdominal pain than younger patients [3].

Computed tomography (CT) scanning is an important diagnostic imaging tool used in the evaluation of abdominal pathology, except for patients suspected to have cholecystitis (US is preferred in those cases). A national hospital ambulatory medical care survey in the United States reported an increase in CT scan use in the ED from 2.6 million in 1995 to 16.2 million in 2007 [4,5].

The use of CT scanning results in financial consequences for health care organizations and exposure to radiation for patients [6]. One study found that low-dose CT scanning to minimize exposure to excessive radiation was non-inferior to high-dose CT scanning [4]. In our institution, CT scan of the abdomen and pelvis charges range from $2000 to $4000 depending on the type of study. CT scan of the abdomen and pelvis without intravenous contrast charge is $2655; with contrast, the cost is $3419. The price for a combination of a CT scan without contrast and CT scan with contrast is $3900. CT has the highest sensitivity and specificity in patients with acute abdominal pain. However, the positive predictive value of ultrasound (US) is comparable with CT in certain clinical conditions [4,5].  Therefore, considering the cost, radiation exposure, and availability, abdominal US may be offered for patients as an alternative first-line imaging study to evaluate non-traumatic abdominal pain who don’t display an explicit peritoneal sign, renal colic pain without acute renal failure, gastroenteritis patients, or when high suspicion for cholecystitis. This can be followed by an abdominal CT if US findings are negative or inconclusive in the presence of persistent symptoms [4]. Yet, it is important to realize that Ultrasound is operator dependent and affected by body habitus.

 In this retrospective medical records review study, our goal was to assess prevalence of ordering CT scans in patients who presented with non-traumatic abdominal pain to the ED in a tertiary care academic hospital. Our secondary goals were to identify the most common CT diagnoses based on scan findings for this patient population, assess the correlation with clinical presentation, determine whether US was considered before abdominal CT scan was ordered (for appropriate cases), and evaluate the number of patients admitted or discharged.

Materials and Methods

Study duration

We conducted a search for via Information Technology department for patients who presented to Jersey Shore University Medical Center who presented with abdominal pain from 07/01/2017 to 06/30/2018. The search yielded a total of 5,623 patients presented with abdominal pain (excluding patients with traumatic abdominal injury). Total number of 1,539 presented to ER with abdominal pain between 07/01/2017 to 10/01/2017 (see figure 1).  Demographic characteristics including age, sex, ethnicity, comorbidities, clinical presentations, time of ordering CT abdomen, CT scan findings, US orders, and admission status were further analyzed for patients presented from 07/01/2017 to 10/01/2017. In that group, we excluded any patients who presented with abdominal pain yet did not receive orders for abdominal CT, pregnant patients, those with a history of cancer, presented with traumatic abdominal pain, and any patient younger than 18 years. Total number of 819 patients received CT scan in ER. 323 Patients were excluded. Of this group, 231 patients were cancer patients (We didn’t track the type of cancer), 73 patients were below 18 (between 16-17), 19 patients had abdominal pain related to trauma.  Medical records were retrieved from the ED registry database (Medhost). Positive findings on CT scan were defined as imaging suggestive of any pathological changes including inflammatory conditions e.g, appendicitis, diverticulitis, colitis, enteritis), bleeding (eg, retroperitoneal bleed, abdominal wall hematoma, intra-abdominal hemorrhage), ischemia (eg, bowel ischemia), perforation (eg, bowel perforation, gall bladder perforation, urinary bladder perforation), and obstruction (eg, small bowel obstruction, urinary stone causing obstruction). that explains the patient’s abdominal pain.

The timing of ordering a CT scan was also estimated upon arrival to the ED, along with timing of ordering complete blood count, comprehensive metabolic panel, or urinalysis. We sought to determine the association between clinical presentation and positive imaging studies.

Settings

The ED at Jersey Shore University Medical Center consists of 60 beds (8 pediatric, 12 fast track, 5 behavioral health,5 trauma, 30 adult treatment beds). It is a suburban, tertiary care medical center that has a level 2 trauma center, cardiac catheterization lab, comprehensive stroke center and multiple residency programs (there is not an Emergency Medicine residency). There is 24/7/365 access to all radiology modalities (CT, US, MRI) with CT being the most rapid and easily accessible.

There are no specific written protocols for imaging of abdominal pain. Testing is ordered based on the history and physical and personal preference of the physician. At the time of the study ER physicians had access to imaging history for tests that were performed in part of our healthcare system, but not surrounding hospitals. The ED is staffed 24 hours a day by predominately board certified or residency trained Emergency Medicine physicians. There are approximately 90 thousand total visits per year. 20% of which are pediatric (72 thousand adult visits).

Ethical considerations

Our Institutional Review Board approved the study protocol. All study procedures were conducted under the Declaration of Helsinki regarding research involving human subjects.

Statistical analyses

Continuous variables were reported as mean ± standard deviation. Categorical variables were reported as frequencies. Correlations between categorical variables were obtained using Chi-Square test. A Logistic regression (enter) model that included demographic variables (age, sex, and race) and variables that were statistically correlated to the outcome (positive CT result) was used to predict the outcome. One variable (loss of appetite), though didn’t statistically correlate to the outcome, was enforced in the model due to its clinical importance [6]. Hosmer-Lemeshow test was used to ascertain the goodness of fit of the model. An alpha value (p) less than 0.05 was considered statistically significant. Statistical analyses were done using IBM SPSS Statistics for Windows Version 25.0 (IBM Corporation, Artmonk, NY).

Results

During the study duration, (07/01/2017 to 10/01/2017) total number of patients with abdominal pain was 1,539, and 819 (53.2%) received a CT scan of abdomen and pelvis. After applying the exclusion criteria on these 819 patients, a total of 496 medical records were manually reviewed.

Patients: The average age of our patients was 50±18.6 years, and the male to female ratio was 40:60. Most patients were white (73%). The most common comorbidities found in our patients were a history of prior abdominal surgery and hypertension (Table 1). 170 patients had no past medical history. The most common anatomically specific location of abdominal pain on presentation to the ED was left lower quadrant pain (18%) (Table 1) and most common findings on positive CT scans for pain on that site was diverticulitis.

Imaging findings: Only 6% of the patients (n=34) had sonographic studies before CT scans. Of these, 6 were positive for cholecystitis, leiomyoma, and ovarian cyst, 5 were inconclusive, and 23 were negative. Abdominal x-rays (including obstructive series) was ordered in 11% of patients (n=55), and the majority were negative (n=53). Sixty percent of abdominal and pelvic CT scans did not show remarkable findings (n=300), and 40% (n=196) of scans revealed pathologies that explain the patients’ abdominal pains, or “positive” findings. Twenty-six patients with positive CT findings had abdominal x-ray and/or US study prior to CT but were unremarkable. Three patients had negative abdominal x-ray and negative US study, yet CT scan came back positive.

Outcomes: The most common diagnosis observed in these patients was nonspecific abdominal pain (n=154, 31%), followed by appendicitis (n=42, 8%), renal colic (n=41, 8%), diverticulitis (n=40, 8%), gastroenteritis (n=39, 7%), gastritis (n=30, 6%), colitis (n=27, 5%), pyelonephritis (n=28, 5%), bowel obstruction (n=15, 3%), and perforation (n=11, 2%). Other miscellaneous diagnoses included mesenteric adenitis, pancreatitis, uterine leiomyoma, ovarian mass, ovarian abscess, ovarian cyst, mesenteric ischemia, constipation, pelvic inflammatory disease, abdominal wall hematoma, cholecystitis, and cholelithiasis, accounting for 13% of patients (n=69). At the end of the ED visit, 151 patients (30%) were admitted and 345 patients were discharged (70%).

Result of CT scan based on timing: When CT abdomen and pelvis ordered before basic blood tests that resulted in 85 positive CT scans.  97 CT scans were ordered within an hour from ordering Complete blood count and Comprehensive metabolic panel and 14 within 2 hours after ordering preliminary blood work (see Figure 2).

Discussion

Multiple pathological processes can cause abdominal pain, which makes imaging essential for identifying specific pathology. Common etiologies include inflammatory processes such as diverticulitis, pancreatitis, obstructive process (eg, bowel obstruction, obstructive renal stones), bleeding (eg, retroperitoneal bleed, abdominal wall hematoma), perforation (eg, bowel perforation), and ischemia (eg, bowel ischemia) [7].

We found that most patients (60%) who received CT orders had negative findings on CT (table 2). Similarly, Bhatt et al reported that 50% of patients had negative findings on abdominal CT scans of the abdomen [8]. Mindelzun et al reported that nearly one-third of patients with abdominal pain who present to the ED never had an established diagnosis [9]; this aligns with our findings where approximately 31% of our patients' non-specific abdominal pain with no abnormalities in the CT imaging. 

In this study, eight percent (n=42) of patients diagnosed with acute appendicitis. Aranda-Narváez et al reported that Sensitivity of CT scan is significantly superior to ultrasound scan (97% vs. 86%) in diagnosing acute appendicitis, but positive predictive value is similar in both tests (92% vs. 94%) [10]. Johansson et al reported that diagnostic accuracy is high for US as well as for CT. US is better for diagnosing positive findings (Guarding, fever, nausea), while CT was better for excluding diagnosis of appendicitis [11]. With this comparable sensitivity and specificity Ultrasound could have been used as an initial diagnostic test for patients with low clinical suspicion of other concomitant or alternative pathologies. In conclusion, CT and ultrasound scan are excellent diagnostic tools for acute appendicitis and have contributed to a significant increase in surgical explorations with correct diagnosis [10].

Abdominal abscess is a feared complication of diverticulitis, appendicitis (most common diagnosis in our study), if the treatment is not administered in a timely manner. Abdominal US is less sensitive than CT in detecting abdominal abscess pathology. Dobrin et al conducted a retrospective review of 92 patients with abdominal US and CT scans of the abdomen and pelvis to evaluate an intra-abdominal abscess [12]. Dobrin et al reported a sensitivity and specificity for US of 75% and 91%, respectively, compared to 88% and 93%, respectively, for CT scan [12]. No data were available regarding the yield of abdominal US in the evaluation of abdominal pain in postoperative patients and neutropenic patients.

In addition, the radiation dose of the abdominal x-ray is comparable to that of a low-dose CT scan, however, with significantly less sensitivity (75% vs 46%) [13]. In our study, 11% of the patients (n=55) received abdominal x-ray, and 96% of them (n=53) had negative results.

Therefore, it’s safe to conclude that abdominal x-ray should be avoided in the evaluation of patients with abdominal pain. Seven percent of patients (n=39) in our study presented with gastroenteritis and diarrhea, yet they received a CT scan of the abdomen and pelvis, a procedure that could possibly have been avoided. Aisenberg et al reported that CT scanning of the abdomen and pelvis bore no impact on management in most cases [14]. Instead, a thorough history and examination of those patients in addition to preliminary laboratory workup can prevent potential radiation toxicity and the high costs associated with CT use [14]. Should a CT scan be not obtained in these 39 patients, $117,000 of roughly estimated total charges could have been saved in our institution.

The most common site of pain in our patients was the left lower quadrant. The most common cause of left lower quadrant pain in adults is acute sigmoid or descending colonic diverticulitis [15]. Only 8% of our patients were diagnosed radiologically with diverticulitis. Patients with acute diverticulitis with typical symptoms of diverticulitis, previous history of diverticulitis, no guarding, and no lactic acidosis may not require any imaging [16]. Those patients can be treated medically without the need for radiologic examinations.

CT scanning is the investigation of choice for evaluating patients with suspected descending or sigmoid colon diverticulitis due to its high sensitivity and specificity and its ability to diagnose complications such as small bowel perforation. Therefore, CT helps determine whether surgical or medical treatment is warranted [17]. Also, CT scanning can rule out other causes of left lower quadrant pain that mimic diverticulitis [18] with a reported overall accuracy of 98% [19].

When comparing CT and US in diagnosing diverticulitis, graded-compression sonography and CT are both effective initial diagnostics according to a meta-analysis [20]. However, CT is more sensitive (50% to 100%) for revealing alternative diagnoses for left lower quadrant pain than US (33% to 78%) [18]. None of our patients was diagnosed with diverticulitis using the US. Abdominal radiography is extremely limited in the evaluation of suspected complications of diverticulitis. Abdominal radiography can demonstrate large amounts of retroperitoneal or intraperitoneal air but is significantly less sensitive than CT for small amounts of air. All diverticulitis complicated by perforation in our patients were diagnosed using CT scan.

More than one third of CT scans ordered in 0 hours (CT scan ordered before or / at the same time Comprehensive Metabolic panel, Complete blood count, urinalysis). De Burlet el al reported the results of her study suggests that a significant proportion of CT scans in patients with acute abdominal pain are not clinically indicated or are being performed prior to adequate clinical workup [21].

However, we must acknowledge the fact that Some diagnoses are especially time sensitive and need immediate investigation (aortic dissection, perforated viscus, hemodynamically unstable patients) and therefore, the study may be ordered immediately without preliminary work up.

As well, a healthy appearing CT scan has clinical utility because it allows for physicians to eliminate a diagnosis rapidly and reduce length of stay.  Most of our patients (n=345, 70%) were discharged from the ED; only 151 (30%) patients were admitted for further evaluation.  When we stratified discharges by CT scan results, forty nine percent of patients with positive CT scan were admitted (n=97), and 50.9 % were discharged (n=99). Patients with negative CT scan who were discharged from ED was 82% (n=246) and 18 %were admitted (n=54). The CT scan result was an important factor for whether to admit or discharge these patients. Barksdale et al reported in their prospective trial of 547 patients presenting to the ED with abdominal pain [22]. CT scanning altered the diagnosis in 54% of patients and frequently changed disposition patterns, with a greater proportion of patients discharged instead of admitted for observation [22]. In our study it was hard to determine whether CT scan alone was the primary factor to determine patient disposition as it is very difficult to find that clearly on patients’ charts.

Limitations

Our study was limited in that it was conducted at a single institution, and the study was retrospective. None of the ordering physicians were aware of the study being conducted so it was a mirror of common practice. The other limitation is the clinical reason for doing the scan was not always easy to find in the medical records. If physicians are looking for a kidney stone or an abdominal aortic aneurysm, they may not wait for laboratory results. Some diagnoses are especially time sensitive and need immediate investigation, and therefore, the study may be ordered immediately. Also, a healthy appearing CT scan has clinical utility because it allows for physicians to eliminate a diagnosis rapidly and reduce length of stay. Length of stay prior to the diagnosis also has a cost and having appropriate information early in the assessment allows the care team to proceed with a diagnostic pursuit.

Conclusion

CT scans are commonly ordered for the diagnosis of acute abdominal pain. Fifty percent of patient presented with abdominal pain had CT scan. The most common cause of abdominal pain based on CT scan results is nonspecific abdominal pain in our study. Majority of CT scans were ordered either at same time or after one hour after ordering preliminary blood work. Findings such as absence of diabetes mellitus, history of renal stones, leukocytosis, and acute kidney injury, were correlated enough to predict a positive CT result.

References

1. Caporale N, Morselli-Labate AM, Nardi E, Cogliandro R, Cavazza M, Stanghellini V. Acute abdominal pain in the emergency department of a university hospital in Italy. United European Gastroenterol J. 2016;4(2):297?304. doi:10.1177/2050640615606012

2. Cervellin G, Mora R, Ticinesi A, Meschi T, Comelli I, Catena F, Lippi G. Epidemiology and outcomes of acute abdominal pain in a large urban Emergency Department: retrospective analysis of 5,340 cases. Ann Transl Med. 2016;4(19):362. doi:10.21037/atm.2016.09.10

3. Laurell H, Hansson LE, Gunnarsson U. Acute abdominal pain among elderly patients. Gerontology. 2006;52(6):339?344. doi:10.1159/000094982

4. Stoker J, van Randen A, Laméris W, Boermeester MA. Imaging patients with acute abdominal pain. Radiology. 2009;253(1):31?46. doi:10.1148/radiol.2531090302

5. Larson DB, Johnson LW, Schnell BM, Salisbury SR, Forman HP. National trends in CT use in the emergency department: 1995-2007. Radiology. 2011;258(1):164?173. doi:10.1148/radiol.10100640

6. Ohle R, O'Reilly F, O'Brien KK, Fahey T, Dimitrov BD. The Alvarado score for predicting acute appendicitis: a systematic review. BMC Med. 2011;9:139. Published 2011 Dec 28. doi:10.1186/1741-7015-9-139

7. Matz K, Britt T, LaBond V. CT ordering patterns for abdominal pain by physician in triage. Am J Emerg Med. 2017;35(7):974?977. doi:10.1016/j.ajem.2017.02.003

8. Bhatt A, Yang X, Karnik N, Sill A, Kowdley G. Use of Computerized Tomography in Abdominal Pain. Am Surg. 2018;84(6):1091?1096.

9. Mindelzun RE, Jeffrey RB. Unenhanced helical CT for evaluating acute abdominal pain: a little more cost, a lot more information. Radiology. 1997;205(1):43?45. doi:10.1148/radiology.205.1.9314959

10. Aranda-Narváez JM, Montiel-Casado MC, González-Sánchez AJ, et al. Empleo, eficacia y repercusión clínica del apoyo radiológico al diagnóstico de la apendicitis aguda [Radiological support for diagnosis of acute appendicitis: use, effectiveness and clinical repercussions]. Cir Esp. 2013;91(9):574?578. doi:10.1016/j.ciresp.2013.01.009

11. Johansson EP, Rydh A, Riklund KA. Ultrasound, computed tomography, and laboratory findings in the diagnosis of appendicitis. Acta Radiol. 2007;48(3):267?273. doi:10.1080/02841850601182162

12. Dobrin PB, Gully PH, Greenlee HB, et al. Radiologic diagnosis of an intra-abdominal abscess. Do multiple tests help?. Arch Surg. 1986;121(1):41?46. doi:10.1001/archsurg.1986.01400010047005

13. Alshamari M, Norrman E, Geijer M, Jansson K, Geijer H. Diagnostic accuracy of low-dose CT compared with abdominal radiography in non-traumatic acute abdominal pain: prospective study and systematic review. Eur Radiol. 2016;26(6):1766-1774. doi:10.1007/s00330-015-3984-914.Aisenberg GM, Grimes RM. Computed tomography in patients with abdominal pain and diarrhoea: does the benefit outweigh the drawbacks?. Intern Med J. 2013;43(10):1141?1144. doi:10.1111/imj.12262

15.       Parks TG. Natural history of diverticular disease of the colon. Clin Gastroenterol. 1975;4(1):53?69.

16. Caputo P, Rovagnati M, Carzaniga PL. Is it possible to limit the use of CT scanning in acute diverticular disease without compromising outcomes? A preliminary experience. Ann Ital Chir. 2015;86(1):51?55.

17. Bates DDB, Fernandez MB, Ponchiardi C, von Plato M, Teich JP, Narsule C, Anderson SW, et al. Surgical management in acute diverticulitis and its association with multi-detector CT, modified Hinchey classification, and clinical parameters. Abdom Radiol (NY). 2018;43(8):2060?2065. doi:10.1007/s00261-017-1422-y.

18. Werner A, Diehl SJ, Farag-Soliman M, Düber C. Multi-slice spiral CT in routine diagnosis of suspected acute left-sided colonic diverticulitis: a prospective study of 120 patients. Eur Radiol. 2003;13(12):2596?2603. doi:10.1007/s00330-003-1887-7.

19. Kircher MF, Rhea JT, Kihiczak D, Novelline RA. Frequency, sensitivity, and specificity of individual signs of diverticulitis on thin-section helical CT with colonic contrast material: experience with 312 cases. AJR Am J Roentgenol. 2002;178(6):1313?1318. doi:10.2214/ajr.178.6.1781313.

20. Laméris W, van Randen A, Bipat S, Bossuyt PM, Boermeester MA, Stoker J. Graded compression ultrasonography and computed tomography in acute colonic diverticulitis: meta-analysis of test accuracy. Eur Radiol. 2008;18(11):2498?2511. doi:10.1007/s00330-008-1018-6.

21. De Burlet KJ, MacKay M, Larsen P, Dennett ER. Appropriateness of CT scans for patients with non-traumatic acute abdominal pain. Br J Radiol. 2018;91(1088):20180158. doi:10.1259/bjr.20180158

22.  Barksdale AN, Hackman JL, Gaddis M, Gratton MC. Diagnosis and disposition are changed when board certified emergency physicians use CT for non-traumatic abdominal pain. Am J Emerg Med 2015;33:1646-50.