Misdiagnosis of Small Bowel Obstruction in the Setting of Previous Abdominal Operations

Learning Objectives

At the conclusion of this case, participants should be able to:

• Identify characteristics of “high-risk” abdominal pain, including prior abdominal operations and involuntary guarding.
• Recognize that free intraperitoneal fluid identified on imaging without a history of ascites is abnormal and should be further investigated.
• Discuss indications for cross sectional imaging in patients with high-risk abdominal pain.
• Avoid premature diagnostic closure when the clinical history and the results of imaging studies are contradictory.
• Consider surgical consultation in patients with severe abdominal pain and in patients with peritoneal tenderness.

The Case

A man in his mid-40s with a history of hypertension treated with atenolol and losartan, obesity, and an umbilical hernia repair, presented to the emergency department (ED) with abdominal pain. The ED physician documented 3 days of constant, non-radiating epigastric abdominal pain associated with nausea but no vomiting or diarrhea, fever, or chills. The triage nurse documented pain intensity 8 of 10, blood pressure 150-160 mm Hg systolic, and heart rate 70-100 beats per minute. The physical exam was notable for guarding and a small periumbilical knot, which was not new. Laboratory studies included a hemoglobin of 18.4 (i.e., slightly elevated) and leukocyte count of 8.4 with 75% neutrophils. Liver function tests were normal except for an elevated total bilirubin of 2.4, which was mostly indirect. The metabolic panel was normal except for borderline elevation of serum creatinine, higher than his baseline. Abdominal ultrasound showed fatty liver, mild ascites, and gallbladder sludge without gallstones.

While awaiting test results, the ED physician ordered a “GI cocktail,” which provided no relief. Two small doses of intravenous morphine followed. A nurse documented his pain going from 8 to 10 before the morphine, then back to 8 afterwards. Repeat abdominal exam was unchanged. The patient was discharged home with a diagnosis of “cholelithiasis”, prescriptions for ondansetron and hydrocodone/acetaminophen, and an outpatient referral to general surgery.

The following day, the patient’s wife took him to another ED with similar symptoms, but this time the pain was documented as periumbilical. It was also noted that he had not been passing flatus. Vital signs and physical examination were similar, but a right lower quadrant scar was documented. Laboratory test results were also similar except that the hemoglobin was even higher as was the urine specific gravity at 1.039. Abdominal computed tomography (CT) demonstrated a high-grade small bowel obstruction with a small amount of free fluid or ascites. The patient was taken to the operating room for lysis of adhesions but died 3 days later. An autopsy listed the cause of death as multi-organ failure and sepsis caused by necrotic bowel and peritonitis from small bowel obstruction caused by a ventral abdominal hernia with adhesions.

The Commentary

By Samantha Brown, MD, Garth Utter, MD, MSc, and David K. Barnes, MD

This case describes a relatively healthy, middle-aged man with a history significant for previous abdominal operations who presented with severe midline abdominal pain. It is not clear if the clinician considered an appropriately broad differential diagnosis, as their thought process was incompletely documented in the electronic health record (EHR). Nevertheless, after laboratory testing and a right upper quadrant ultrasound showed gallbladder sludge with some intraperitoneal fluid, the patient’s symptoms were attributed to chronic biliary disease despite important inconsistencies with that diagnosis. A more appropriate diagnosis would have been “symptomatic cholelithiasis” or “biliary colic,” but these terms imply that the pain should have resolved after a few hours. Clearly the patient’s pain did not resolve, so a more plausible diagnosis would have been “acute cholecystitis,” which should be treated by admission and urgent cholecystectomy. Unfortunately, the patient was given an inadequate diagnosis and was discharged while still in pain.

Given a history of previous abdominal operations, severe abdominal pain not responsive to opioid analgesia (and unresolved at the time of discharge), an exam concerning for peritonitis, and an ultrasound demonstrating free fluid, cross-sectional imaging should have been obtained and a surgeon should have been immediately consulted. CT would probably have demonstrated the small bowel obstruction, allowing earlier intervention that may have prevented his clinical decline and death. Simply put, there seems to have been a failure to entertain a differential diagnosis and to recognize the high likelihood of a “surgical abdomen.”

Background

Abdominal pain is one of the most common chief complaints encountered in emergency medicine (EM). In 2018, abdominal pain accounted for 8.5% of all ED visits representing over 11 million patient encounters.¹ Although the differential diagnosis of abdominal pain is lengthy, a time-honored approach to attentive history-taking and physical examination—frequently without imaging—helps clinicians to identify plausible causes and to develop a differential diagnosis that is neither overly broad nor overly narrow.² Pain attributes such as location, quality, radiation, duration, and associated symptoms; relevant history such as previous abdominal operations; related symptoms such as fever, nausea, vomiting, and diarrhea; and physical exam findings such as the location and quality of abdominal tenderness, and the presence of abdominal distention, masses, and scars, drive the workup, narrow the differential, and assist clinicians to refine their selection of laboratory tests and imaging. EM physicians and general surgeons especially should strive to master their skills as diagnosticians of the acute abdomen.

Abdominal pain can be caused by many benign processes such as viral illness and gastroenteritis. Intra-abdominal pathology can also be mimicked by many conditions unrelated to the abdomen such as pneumonia, myocardial infarction, and diabetic ketoacidosis. However, serious causes of intraabdominal pathology should always be considered and reasonably excluded before attributing pain to a benign etiology. Some serious causes of abdominal pain include mesenteric ischemia, small and large bowel obstruction, appendicitis, ruptured abdominal aortic aneurysm, diverticulitis, and acute cholecystitis. Diagnostic imaging is commonly used to evaluate for these conditions.

Whether to obtain imaging, and the choice of imaging procedures, to investigate acute abdominal pain should be guided by the history and physical examination. The location of the pain may be particularly helpful in determining which imaging modality to pursue.³ Ultrasonography is the initial imaging test of choice for right upper quadrant pain, while CT is recommended for evaluation of lower quadrant abdominal pain.⁴ In this case, the first clinician seems to have attributed the patient’s symptoms to right upper quadrant pathology when the pain was epigastric. Although ultrasound is a reasonable first study to investigate hepatobiliary disease, it misses important intra-abdominal pathology beyond the relatively small corridor of the right upper quadrant. There was also a subsequent failure to appreciate that the ultrasound results were non-diagnostic and did not adequately explain the patient’s pain. A prudent physician should ensure that imaging results comport with the history, physical examination, and any abnormal labs, and consider additional testing—including more comprehensive imaging—if those results are incongruous.

The use of diagnostic imaging has increased significantly over the last 20 years, driven largely by improved availability and quality of CT since the 1980s. The accuracy of CT for diagnosing the cause of abdominal pain depends on the location of pain and the specific condition being investigated. For example, the authors of a 2008 meta-analysis found CT to be 91% sensitive and 90% specific for diagnosing acute appendicitis⁵ with an even greater sensitivity of 95% for diagnosing diverticulitis.⁶ This makes it an attractive tool when the cause of abdominal pain is not evident from the history, exam, and laboratory tests. Accordingly, use of CT for evaluating acute abdominal pain in adult ED patients increased from 3.9% to 37.8% between 1997 and 2016.³ There are circumstances, such as those involving pediatric and pregnant patients, when it is advisable to avoid radiation exposure by choosing ultrasound, magnetic resonance imaging (MRI), or serial exams as a substitute for CT. In fact, ultrasound is now the preferred imaging modality in pediatrics for conditions such as appendicitis and intussusception.⁷ There are also many circumstances when imaging is altogether unnecessary, such as when peritonitis, appendicitis, or strangulated hernia with infarcted bowel are apparent by exam and history. In these cases, immediate surgical consultation is indicated, and imaging may lead to a delayed operative management.

In this case, because the diagnosis was not clear from the ultrasound, the next study should have been CT of the abdomen and pelvis with intravenous contrast. Although this study was obtained at the second ED visit, by then the patient’s condition had deteriorated to include necrosis of the bowel.

Case Analysis

This case involved a patient presenting with high-risk abdominal pain. “High-risk” abdominal pain is characterized by clinical features including age greater than 65 years, immunocompromise, alcohol use disorder, cardiovascular disease, prior abdominal operations, or major comorbidities such as cancer and inflammatory bowel disease. These historical risk factors increase the likelihood that a patient’s pain is due to a serious or potentially surgical condition, lowering the threshold for obtaining diagnostic imaging. Similarly, involuntary guarding, which was present in this case, is considered a high-risk examination finding.⁸

This patient was misdiagnosed with symptomatic cholelithiasis during the first ED encounter. The ultrasound showed sludge in the gallbladder (i.e., gallstones), fatty liver disease, and free intraperitoneal fluid, but no mention of a sonographic Murphy’s sign, gallbladder wall thickening, or pericholecystic fluid to suggest acute cholecystitis.⁹ The patient’s persistent pain, even if misattributed to the gallbladder, should have prompted a surgical consultation for acute cholecystitis, which might have led to the correct diagnosis of intestinal obstruction. Although the patient’s bilirubin was mildly elevated, this is a non-specific finding and could have represented hepatic dysfunction from early sepsis (especially in combination with the mild creatinine elevation, consistent with kidney dysfunction or dehydration). Unfortunately, the clinician failed to recognize that the results of the ultrasound did not explain the patient’s clinical presentation. Had they heeded the red flags of severe and persistent pain, and abdominal guarding with a periumbilical “knot,” they likely would have developed a broader differential diagnosis and searched for another explanation for the patient’s pain by obtaining a CT scan.

Eventually, this patient was diagnosed with a small bowel obstruction related to adhesions from a previous abdominal operation (ventral herniorrhaphy). Small bowel obstruction secondary to adhesions occurs in approximately 15% of patients with previous abdominal surgery.¹⁰ Adhesion disease is the cause in the majority (>70%) of small bowel obstruction cases.¹¹ A prospective study in the United Kingdom found 68% of patients who underwent operative management for small bowel obstruction had previous abdominopelvic surgery.¹² Less common causes of obstruction include abdominal wall or internal hernia, volvulus, malrotation, and foreign bodies.¹¹ Therefore, a history of prior abdominal operations should always raise suspicion for adhesions and small bowel obstruction in patients presenting with acute abdominal pain.

Although there were several red flags that were ignored or minimized in this case, including the history of abdominal surgery, severe abdominal pain, and guarding on exam, the first clinician’s failure to appreciate the significance of free intraperitoneal fluid was particularly notable. The presence of ascites or free fluid in the absence of a known condition that causes ascites (e.g., cirrhosis) should always be considered pathological in men, and in women of non-childbearing age, and should be investigated accordingly. In retrospect, the appearance of free fluid was likely “the earliest indicator of an acute abdomen needing surgery.”¹³

Approaches to Improving Patient Safety

Imaging in High-Risk Abdominal Pain

Patients presenting to the ED with abdominal pain of uncertain etiology and high-risk features should generally undergo diagnostic imaging. The American College of Radiology (ACR) Appropriateness Criteria (AC) are annually updated “evidence-based guidelines to assist referring physicians and other providers in making the most appropriate imaging or treatment decision for a specific clinical condition.”¹⁴ AC are available for localized abdominal pain (e.g., right lower quadrant with suspected acute appendicitis) as well as non-localizing abdominal pain.¹⁵

Plain radiography has historically been used for diagnosing small bowel obstruction. If the pathognomonic findings of dilated small bowel loops with air-fluid levels are present, this can provide rapid confirmation of the diagnosis without higher levels of radiation and without the risk of traveling from the ED to the radiology suite. However, diagnostic radiographs are only 69% sensitive and 57% specific.⁷ CT is 92% sensitive and provides ancillary information about the presence of a closed-loop obstruction, intestinal ischemia, or intestinal perforation, and has therefore become the preferred imaging study when small bowel obstruction is considered.¹⁶However, not every patient with abdominal pain should undergo CT imaging. Depending on the location of the pain, ultrasound may be a more useful test.

The Role of Surgical Consultation

Any imaging study should be used in conjunction with the history, physical exam, and any laboratory results to confirm or refute the diagnoses under consideration. For certain patients, surgical consultation should be obtained regardless of imaging findings. While consultation with a surgeon is mandatory for a patient who is hemodynamically unstable and cannot safely undergo diagnostic imaging, consultation should also be strongly considered in any patient with signs of peritoneal tenderness (i.e., with finger percussion, guarding, or “rebound tenderness”), especially if they have a history of abdominal surgery. Patients who have undergone bariatric or other complex foregut operations are particularly prone to diagnostic challenges and complications, and benefit from the involvement of a surgeon familiar with the sequelae of those operations.

The Importance of Free Fluid

The initial clinician in this case failed to recognize that the ultrasound was not diagnostic and did not explain the patient’s condition. The presence of new ascites or free fluid has significant implications for mortality and morbidity. In patients with cirrhosis, the presence of ascites is associated with a significantly increased risk of death with a hazard ratio of 27.6.¹⁷ This fact alone should have justified additional workup. Instead, the clinician either dismissed the presence of free fluid as irrelevant or somehow associated it with symptomatic cholelithiasis. Because this patient had no history of ascites or liver disease, the clinician should have pursued additional imaging to further investigate the etiology of the fluid.

Anchoring and Premature Closure

Diagnostic anchoring is the tendency to focus on a limited set of clinical facts despite the availability of new, changed, or evolving information. This common cognitive bias leads to premature closure, or the acceptance of a diagnosis before it is fully verified.¹⁸ This case exemplifies closure on biliary disease as the source of the patient’s abdominal pain despite multiple findings to the contrary, such as the lack of gallbladder wall changes on ultrasound. Cognitive errors contribute to roughly one third of all complications.¹⁹ Anchoring and premature closure are two of the most common cognitive biases that contribute to medical errors, with an incidence of 16.8% and 9.5%, respectively, in one surgical series.¹⁹ When clinical information and diagnostic testing results are contradictory, as in this case, the clinician should reassess all relevant data, reconsider the differential diagnoses, and consider involving additional clinicians, if available, to help mitigate cognitive errors and their impact on patient care. Diagnoses are like driving conditions: When they’re unclear, it’s wise to slow down.

The Role of Lactate Levels in Evaluation of Abdominal Pain

Some clinicians advocate checking serum lactate (or lactic acid) levels in patients with abdominal pain. Briefly, lactate is a normal byproduct of mammalian metabolism. Lactate levels rise when aerobic processes are overwhelmed and during pathologic processes such as shock and bowel ischemia.^16,20 Lactate has been shown to be a marker for tissue hypoperfusion and a reliable test for early diagnosis of abdominal catastrophe, including vascular causes of abdominal pain,²⁰ although serum lactate is only obtained in roughly half of ED patients with abdominal pain.²¹ Evidence for the utility of serum lactate as a marker for at-risk bowel and acute intra-abdominal pathology is mixed.²⁰ Obtaining peritoneal fluid for lactate is not practical under most circumstances. Moreover, some authors advise caution when using lactate in the workup of abdominal pain because its low specificity may lead to overuse of diagnostic resources.²² Importantly, lactate levels are not highly sensitive for intestinal ischemia because a strangulated segment of bowel may have no venous outflow (by which the lactate becomes measurable in circulating blood). Based on certain clinical and testing parameters, the use of lactate levels can be considered in the evaluation of abdominal pain. In the present case, it is not clear if lactate would have been elevated at either of the patient’s ED visits. Indeed, a negative result at those times may have been falsely reassuring to clinicians.

The Role of Observation

One option that was not utilized in this case was a period of extended observation. Over the last decade, observation units have been increasingly utilized to determine a patient’s need for inpatient care. Observation units have been shown to be beneficial to patients, allowing further workup and monitoring (e.g., serial exams and labs, treatment with fluids, etc.) while also improving resource utilization and decreasing costs to the health system.²³ In this case, rather than discharging the patient with a questionable diagnosis and unresolved pain, the clinician could have placed the patient in an observation unit for further treatment, especially given his continued guarding on reassessment despite opioid analgesia. Had his examination worsened or lab results changed, additional imaging or surgical consultation could have been obtained at that time.

Take Home Points

• Evaluation of acute abdominal pain should always start with a focused but thoughtful history and physical exam.
• Always have a low threshold to obtain comprehensive abdominal imaging, usually by CT, in patients with high-risk abdominal pain.
• Avoiding cognitive biases, such as premature closure and anchoring, requires constant vigilance. If in doubt, slow down the diagnostic process and gather more information and/or consultations to foster a broader differential diagnosis.
• Consider surgical consultation in patients with high-risk abdominal pain when imaging is non-diagnostic or when pain persists despite aggressive analgesia.
• Utilize imaging studies appropriately. ACR guidelines are available to guide physician choice of appropriate imaging.
• Ascites, or free intraperitoneal fluid, is abnormal in patients without a history of liver disease.
• Don’t ignore abnormal clinical findings. Even mildly abnormal physical exam findings, laboratory results, or imaging findings, taken together, may represent serious pathology.

Samantha Brown, MD
Assistant Professor
Department of Emergency Medicine
UC Davis Health
skbro@ucdavis.edu

Garth Utter, MD, MSc
Consulting Editor, AHRQ’s Patient Safety Network (PSNet)
Professor
Department of Surgery, Division of Trauma, Acute Care Surgery, and Surgical Critical Care
UC Davis Health
ghutter@ucdavis.edu

David K. Barnes, MD, FACEP
Consulting Editor, AHRQ’s Patient Safety Network (PSNet)
Health Sciences Clinical Professor
Director of Faculty Development
Director of ED Sustainability
Department of Emergency Medicine
Physician Advisor
UC Davis Health
dbarnes@ucdavis.edu

References

1. Cairns C, Kang K, Santo L. National Hospital Ambulatory Medical Care Survey: 2018 emergency department summary tables. Accessed December 19, 2024. [Available at]
2. Silen W. Cope’s Early Diagnosis of the Acute Abdomen. 22nd ed. New York, NY: Oxford University Press; 2010.
3. Wang RC, Kornblith AE, Grupp-Phelan J, et al. Trends in use of diagnostic imaging for abdominal pain in U.S. emergency departments. AJR Am J Roentgenol. 2021;216(1):200-208. [Free full text]
4. Cartwright SL, Knudson MP. Diagnostic imaging of acute abdominal pain in adults. Am Fam Physician. 2015;91(7):452-459. [Free full text]
5. van Randen A, Bipat S, Zwinderman AH, et al. Acute appendicitis: meta-analysis of diagnostic performance of CT and graded compression US related to prevalence of disease. Radiology. 2008;249(1):97-106. [Available at]
6. Sai VF, Velayos F, Neuhaus J, et al. Colonoscopy after CT diagnosis of diverticulitis to exclude colon cancer: a systematic literature review [published correction appears in Radiology. 2012 Jul;264(1):306]. Radiology. 2012;263(2):383-390. [Free full text]
7. Wolfe C, Halsey-Nichols M, Ritter K, et al. Abdominal pain in the emergency department: how to select the correct imaging for diagnosis. Open Access Emerg Med. 2022;14:335-345. [Free full text]
8. Kendall JL, Moreira ME. Evaluation of the adult with nontraumatic abdominal or flank pain in the emergency department. UpToDate. 2024. Accessed November 3^rd, 2024. [Available at]
9. Wang HH, Portincasa P, Liu M, et al. Similarities and differences between biliary sludge and microlithiasis: their clinical and pathophysiological significances. Liver Res. 2018;2(4):186-199. [Free full text]
10. Duron JJ, Silva NJ, du Montcel ST, et al. Adhesive postoperative small bowel obstruction: incidence and risk factors of recurrence after surgical treatment: a multicenter prospective study. Ann Surg. 2006;244(5):750-757. [Free full text]
11. Bower KL, Lollar DI, Williams SL, et al. Small bowel obstruction. Surg Clin North Am. 2018;98(5):945-971. [Available at]
12. van Goor H. Consequences and complications of peritoneal adhesions. Colorectal Dis. 2007;9 Suppl 2:25-34. [Available at]
13. Rudralingam V, Footitt C, Layton B. Ascites matters. Ultrasound. 2017;25(2):69-79. [Free full text]
14. American College of Radiology Appropriateness Criteria. Accessed November 3^rd, 2024. [Available at]
15. Expert Panel on Gastrointestinal Imaging:, Scheirey CD, Fowler KJ, et al. ACR Appropriateness Criteria^® Acute Nonlocalized Abdominal Pain. J Am Coll Radiol. 2018;15(11S):S217-S231. [Available at]
16. Li Z, Zhang L, Liu X, Yuan F, Song B. Diagnostic utility of CT for small bowel obstruction: systematic review and meta-analysis. PLOS One. 2019;14(12):e0226740 [Free full text]
17. Tapper EB, Zhao Z, Mazumder N, et al. Incidence of, risk factors for, and outcomes after ascites in a population-based cohort of older americans. Dig Dis Sci. 2022;67(11):5327-5335. [Free full text]
18. Croskerry P. The importance of cognitive errors in diagnosis and strategies to minimize them. Acad Med. 2003;78(8):775-780. [Available at]
19. Antonacci AC, Dechario SP, Antonacci C, et al. Cognitive bias impact on management of postoperative complications, medical error, and standard of care. J Surg Res. 2021;258:47-53. [Available at]
20. Verma I, Kaur S, Goyal S, et al. Diagnostic value of lactate levels in acute abdomen disorders. Indian J Clin Biochem. 2014;29(3):382-385. [Free full text]
21. Khan K, Faisal A, Sridhar S, et al. Underutilization of lactate levels: a missed opportunity for evaluating abdominal pain in the ED? CHEST. 2023;164(4):A1718. [Free full text]
22. Meyer ZC, Schreinemakers JM, van der Laan L. The value of C-reactive protein and lactate in the acute abdomen in the emergency department. World J Emerg Surg. 2012;7(1):22. [Free full text]
23. Perry M, Franks N, Pitts SR, et al. The impact of emergency department observation units on a health system. Am J Emerg Med. 2021