ª Springer Science+Business Media New York 2015
Abdominal Imaging
Abdom Imaging (2015) DOI: 10.1007/s00261-015-0419-7
Impact of CT in elderly patients presenting to the emergency department with acute abdominal pain Carly S. Gardner,1 Tracy A. Jaffe,2 Rendon C. Nelson2 1 2
The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Unit 1473, Houston, TX 77030-4009, USA Duke University Medical Center, 2301 Erwin Road, Box 3808, Durham, NC 27710, USA
Abstract Purpose: The purpose of the study was to document the clinical impact of CT in elderly patients presenting to the emergency department (ED) with abdominal pain. Methods: This retrospective IRB-approved study from 2006 to 2013 evaluated 464 patients ‡80 years (mean 89 years, range 80–100: M150, W314), who presented to the ED with acute abdominal symptoms and underwent CT. CTs were divided into those negative and positive for actionable findings, defined as potentially requiring a change in surgical or medical management. Physician diagnosis, treatment plan, and disposition before and after CT were reviewed in the electronic medical record to assess CT influence on management and disposition. CT diagnosis was confirmed with final clinical diagnosis, surgical intervention, pathology, and follow-up. Descriptive statistics were used. Results: CTs were positive in 55%. The most common diagnoses were SBO (18%), diverticulitis (9%), non-ischemic vascular-related emergency (6%), bowel ischemia (4%), appendicitis (3%), and colonic obstruction (2%). These diagnoses were clinically unsuspected prior to CT in 43% (p < 0.05), with significant difficultly in diagnosing SBO (p < 0.05), diverticulitis (p < 0.01), and colonic obstruction (p < 0.01). Positive CT results influenced treatment plans in 65%, surgical in 48%, and medical in 52%. Disposition from the ED was significantly affected by CT (p < 0.001), 65% of admissions with positive CT (p < 0.001) and 63% of discharges with negative CT (p < 0.001). Conclusion: Utilization of abdominopelvic CT in geriatric patients presenting to the ED with acute abdominal symptoms strongly influences clinical management and significantly affects disposition. As the US population ages, the clinical impact of emergent CT in the elderly will intensify.
Correspondence to: Carly S. Gardner; email: [email protected]
Key words: Abdominal pain—CT—Emergency department—Elderly
Acute abdominal pain is one of the most common reasons for older patients to present to the emergency department (ED) in the United States [1]. The ability to accurately and effectively determine the cause of abdominal pain decreases with advancing patient age [2–4]. Because elderly patients require more time and resources, they often have prolonged ED visits, longer wait times before seeing a physician and undergo additional laboratory testing [5–7]. The evaluation of abdominal pain in the elderly may be confounded by late and/or atypical presentation, limitations in history taking and physical exam, unreliable vital signs and laboratory values [8–14]. Furthermore, acute abdominal pain in the elderly is more commonly due to life-threatening vascular and surgical emergencies than in younger populations [4, 15– 18]. Van Geloven et al. [17], moreover, found that mortality was particularly high (17%) in those over age 80 admitted to the hospital through the ED with abdominal pain. As the U.S. population continues to age, it is estimated that 20% of Americans will be over the age of 65 by 2030, with the fastest growing subset over the age of 85 years [12]. There is national concern regarding the appropriateness of imaging in the era of healthcare reform and costeffectiveness. This dramatic growth in the elderly population has put imaging in the spotlight. CT utilization in the ED has increased at an annual rate of 14%, from 2.7 million exams in 1995 to 16.2 million exams in 2007 [19]. In parallel, the Emergency Medicine literature stresses the importance of early, liberal imaging in the elderly population and advocates a low threshold for hospital admission in elderly patients with undifferentiated abdominal pain [20].
C. S. Gardner et al.: Impact of CT in elderly patients
As the subset of elderly patients grows, the utilization of resources such as CT in the ED becomes increasingly important. The incidence of CTs in elderly patients presenting with abdominal symptoms is as high as 59% [21]. In one study, CT of the abdomen and pelvis was shown to alter decision-making in elderly patients, changing diagnosis in 45% of patients and increasing the diagnostic confidence of emergency physicians in 81% of patients [22]. Another study published in 2007 demonstrated no significant change in disposition in the elderly population undergoing CT compared with those undergoing physical examination alone [21]. Physical exam findings, however, have been shown to be unreliable for predicting or excluding clinically significant disease in the elderly with abdominal pain [20], thereby stressing the importance of CT. The purpose of this study is to evaluate and explore the clinical impact of CT imaging of the abdomen and pelvis in patients 80 years or older presenting to the ED with acute abdominal symptoms.
150 mL (Isovue 300, 300 mg L/mL) injected at a rate of 3 mL/s. Coronal reformatted images (section thickness and interval of 3 mm) are reconstructed from 0.6 to 0.625-mm thick axial sections during the portal venous phase acquisition. According to institution protocol over this time period, a positive oral contrast agent, diluted diatrizoate meglumine (Gastrografin 2%; Schering) was administered, 30 mL mixed in 450 mL of water for the following indications: abdominal pain, small bowel obstruction, abscess, pancreatitis, and enteric fistula or leak.
Review of clinical history and imaging Demographic data including age and gender were collected. The electronic medical record was reviewed by one radiologist (CSG) for diagnosis, treatment plan, and disposition before and after imaging. Determining pre-CT diagnosis
Materials and methods Patient selection Institutional Review Board approval was obtained and patient consent waived for this HIPAA-compliant, retrospective study. A retrospective review of electronic medical records from January 2006 to January 2013 identified 464 geriatric patients age 80 years or older who underwent a CT scan of the abdomen and pelvis after presenting to the ED with acute abdominal symptoms. An age of 80 years old was selected given previous studies suggesting that mortality in this patient subset was particularly high [17, 23]. Patients with traumatic indications were excluded from the study. For instance, a patient undergoing CT of the abdomen and pelvis for abdominal pain in the setting of a fall or trauma was excluded. Patients with known malignancy presenting with acute abdominal symptomatology were included. For example, a patient with colon cancer undergoing CT of the abdomen and pelvis because of concern for acute intestinal obstruction or perforation would be included in the study. In the event that a patient underwent more than one abdominal and pelvic CT in the ED, either during the same or different encounters, the scan from the earliest or first visit was selected.
CT imaging technique Acquisition parameters at our institution for abdominal and pelvic CT imaging in these patients includes helical mode, 120 kVp, beam pitch 0.8–1.375, automated tube current modulation (noise index for GE systems 15–22 HU from 0.625 mm images, reference mA for Siemens systems 200), minimum tube current 100–150 mAs, and reconstructed section thickness and interval of 5 mm. Administration of intravenous (IV) contrast material was
The pre-CT diagnosis was identified by review of the emergency physician note in the electronic medical record. The pre-CT diagnosis was defined as the primary diagnosis under the initial assessment and plan for the patient prior to CT imaging. If no diagnosis was listed by the emergency physician before obtaining the CT, the study indication or reason for CT was used as the principal pre-CT diagnosis. For instance, a study indication of ‘‘rule out sbo’’ was classified as a pre-CT diagnosis of small bowel obstruction, and an indication of ‘‘left flank pain, hematuria’’ was classified as a pre-CT diagnosis of renal/ureteral stone. If more than one diagnosis was recorded in the ED note, the first or leading diagnosis listed in the differential was considered to represent the emergency physician’s pre-CT diagnosis. Determining post-CT diagnosis The post-CT diagnosis was based on final interpretation of the CT. CT report interpretations were categorized into those positive and negative for actionable findings. Actionable findings were defined as those potentially requiring a change in either surgical or medical management. Examples of actionable findings included small bowel obstruction, bowel ischemia or perforation, obstructing renal stone, diverticulitis, appendicitis, and vascular emergency such as aortic dissection, aneurysm rupture, and acute arterial or venous thrombosis. Newly diagnosed malignancy was not considered an actionable finding unless considered to cause the patient’s acute symptomatology, such as a colon cancer resulting in large bowel obstruction. The CT results were corroborated with the final diagnosis in the ED note or discharge summary (in those admitted) and any subsequently available pathology, intervention, surgery or clinical follow-up in patients admitted or referred to a
C. S. Gardner et al.: Impact of CT in elderly patients
subspecialty. Positive CT report interpretations concordant with the clinical diagnosis and follow-up were given a post-CT diagnosis. CTs with non-actionable findings, nonspecific findings, or unconvincing findings, that were not supported clinically, were considered negative. For instance, a patient with an indeterminate renal cystic lesion would be categorized as non-actionable or negative. Similarly, a patient with ‘‘possible bowel wall thickening vs. under-distention’’ without any clinical support of colitis or further workup would be considered negative. Patients with actionable positive CT results were subdivided into medical and surgical treatment categories, summarized in Fig. 1. Analysis of ED records was performed to determine whether the CT results influenced clinical management. Changes in medical management were defined as medication or treatment changes and referrals to a subspecialty consultation based on positive CT results. Changes in surgical management consisted of surgical operations and minimally invasive procedures such as endoscopy and image-guided percutaneous drainage or intervention.
Disposition Anticipated disposition prior to CT imaging and final disposition after CT imaging were recorded. Details of anticipated disposition status prior to CT imaging were obtained from review of the emergency physician note in the electronic medical record and categorized into the following: (1) anticipated admission, (2) anticipated discharge, and (3) unknown or pending CT examination. Under ‘‘disposition’’ in the initial assessment and plan of the ED note, ‘‘likely admit’’ was categorized as an-
Elderly with CT A/P (n=464)
CT Posi ve (n=257, 55%)
Change in Management (n=166, 65%)
Opera ve (n=49, 61%)
CT Nega ve (n=207, 45%)
No Change in Management (n=91, 35%)
Surgical (n=80, 48%)
Medical (n=86, 52%)
Interven onal Radiology (n=20, 25%)
Other Minimally Invasive (n=11, 14%)
Fig. 1. Summary of CT effect on medical and surgical management.
ticipated admission and ‘‘likely discharge’’ as anticipated discharge. Anticipated disposition in the ED note that was either blank, ‘‘unknown’’ or ‘‘pending CT’’ was placed in the third category above. Final dispositions from the ED were as follows: (1) admission, (2) discharge, and (3) death.
Statistical analysis Statistics were performed using Microsoft Excel version 12.1.3 (Microsoft, Redmond, WA) and Statistical Program for the Social Science (SPSS, Chicago, IL) version 11.0, Windows 2000. The incidence of positive CTs, most common indications, most common diagnoses, changes in clinical management, and disposition were reported along with proportions with 95% confidence intervals (CIs). Z test was used to assess proportional differences between pre- and post-CT diagnosis for the most commonly observed diagnoses. The effect of positive vs. negative CT results on disposition, admission, and discharge rates, was compared using z test and Chi square test with Yate’s continuity correction; p < 0.05 indicated statistical significance.
Results Our study identified 464 geriatric patients (mean 89 years, range 80–100 years), including 150 men and 314 women who underwent a CT scan of the abdomen and pelvis in the ED for acute abdominal symptoms. Sixty-nine percent (321/464) of the CTs were performed following the administration of intravenous (IV) contrast material and 31% (141/464) were performed without IV contrast material. The average serum creatinine, in those who received IV contrast material, was 1.1 mg/dL (range 0.5–8.3 mg/dL) and in those who did not 1.4 mg/dL (range 0.4–8.7 mg/dL) (p = 0.01). The most common indications for CT were small bowel obstruction (SBO) (83/464, 18%; 95% CI 15%– 22%), generalized abdominal pain (77/464, 17%; CI 14%–20%), non-ischemic vascular-related indications including abdominal aortic aneurysm rupture or dissection (67/464, 14%; CI 11%–17%), diverticulitis (51/464, 11%, CI 8%–14%), renal or ureteral calculi (32/464, 7%; CI 5%–9%), bowel ischemia (29/464, 6%; CI 4%–8%), appendicitis (12/464, 3%; CI 1%–5%), and pancreatitis (5/464, 1%; CI 0%–2%). CT results were positive for actionable findings in 55% (257/464; CI 50%–60%) and negative in 45% (207/ 464; CI 40%–50%). Table 1 summarizes the most common final diagnoses pre- and post-CT. Significant differences in pre- vs. post-CT diagnosis were seen with SBO, diverticulitis, and large bowel obstruction. No significant difference in pre- vs. post-CT diagnosis was found in vascular emergency, bowel ischemia, or appendicitis. These six most common CT diagnoses
C. S. Gardner et al.: Impact of CT in elderly patients
Table 1. Most common diagnoses pre- and post-computed tomography (CT) Final diagnosis SBO Diverticulitis Vascular emergency Ischemia Appendicitis Large bowel obstruction Totals
Pre-CT diagnosis, n (%, CI 95%)a 34 8 16 5 4 0 67
Post-CT diagnosis, n (%, CI 95%)*
(13, 9–17) (3, 1–5) (6, 3–9) (2, 0–4) (2, 0–4) (0, 0–0) (26, 19–29)
50 25 18 10 8 6 117
(20, 14–24) (10, 6–14) (7, 4–10) (4, 2–6) (3, 1–5) (2, 0–4) (42, 36–48)
p value
Abdominal Imaging
Abdom Imaging (2015) DOI: 10.1007/s00261-015-0419-7
Impact of CT in elderly patients presenting to the emergency department with acute abdominal pain Carly S. Gardner,1 Tracy A. Jaffe,2 Rendon C. Nelson2 1 2
The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Unit 1473, Houston, TX 77030-4009, USA Duke University Medical Center, 2301 Erwin Road, Box 3808, Durham, NC 27710, USA
Abstract Purpose: The purpose of the study was to document the clinical impact of CT in elderly patients presenting to the emergency department (ED) with abdominal pain. Methods: This retrospective IRB-approved study from 2006 to 2013 evaluated 464 patients ‡80 years (mean 89 years, range 80–100: M150, W314), who presented to the ED with acute abdominal symptoms and underwent CT. CTs were divided into those negative and positive for actionable findings, defined as potentially requiring a change in surgical or medical management. Physician diagnosis, treatment plan, and disposition before and after CT were reviewed in the electronic medical record to assess CT influence on management and disposition. CT diagnosis was confirmed with final clinical diagnosis, surgical intervention, pathology, and follow-up. Descriptive statistics were used. Results: CTs were positive in 55%. The most common diagnoses were SBO (18%), diverticulitis (9%), non-ischemic vascular-related emergency (6%), bowel ischemia (4%), appendicitis (3%), and colonic obstruction (2%). These diagnoses were clinically unsuspected prior to CT in 43% (p < 0.05), with significant difficultly in diagnosing SBO (p < 0.05), diverticulitis (p < 0.01), and colonic obstruction (p < 0.01). Positive CT results influenced treatment plans in 65%, surgical in 48%, and medical in 52%. Disposition from the ED was significantly affected by CT (p < 0.001), 65% of admissions with positive CT (p < 0.001) and 63% of discharges with negative CT (p < 0.001). Conclusion: Utilization of abdominopelvic CT in geriatric patients presenting to the ED with acute abdominal symptoms strongly influences clinical management and significantly affects disposition. As the US population ages, the clinical impact of emergent CT in the elderly will intensify.
Correspondence to: Carly S. Gardner; email: [email protected]
Key words: Abdominal pain—CT—Emergency department—Elderly
Acute abdominal pain is one of the most common reasons for older patients to present to the emergency department (ED) in the United States [1]. The ability to accurately and effectively determine the cause of abdominal pain decreases with advancing patient age [2–4]. Because elderly patients require more time and resources, they often have prolonged ED visits, longer wait times before seeing a physician and undergo additional laboratory testing [5–7]. The evaluation of abdominal pain in the elderly may be confounded by late and/or atypical presentation, limitations in history taking and physical exam, unreliable vital signs and laboratory values [8–14]. Furthermore, acute abdominal pain in the elderly is more commonly due to life-threatening vascular and surgical emergencies than in younger populations [4, 15– 18]. Van Geloven et al. [17], moreover, found that mortality was particularly high (17%) in those over age 80 admitted to the hospital through the ED with abdominal pain. As the U.S. population continues to age, it is estimated that 20% of Americans will be over the age of 65 by 2030, with the fastest growing subset over the age of 85 years [12]. There is national concern regarding the appropriateness of imaging in the era of healthcare reform and costeffectiveness. This dramatic growth in the elderly population has put imaging in the spotlight. CT utilization in the ED has increased at an annual rate of 14%, from 2.7 million exams in 1995 to 16.2 million exams in 2007 [19]. In parallel, the Emergency Medicine literature stresses the importance of early, liberal imaging in the elderly population and advocates a low threshold for hospital admission in elderly patients with undifferentiated abdominal pain [20].
C. S. Gardner et al.: Impact of CT in elderly patients
As the subset of elderly patients grows, the utilization of resources such as CT in the ED becomes increasingly important. The incidence of CTs in elderly patients presenting with abdominal symptoms is as high as 59% [21]. In one study, CT of the abdomen and pelvis was shown to alter decision-making in elderly patients, changing diagnosis in 45% of patients and increasing the diagnostic confidence of emergency physicians in 81% of patients [22]. Another study published in 2007 demonstrated no significant change in disposition in the elderly population undergoing CT compared with those undergoing physical examination alone [21]. Physical exam findings, however, have been shown to be unreliable for predicting or excluding clinically significant disease in the elderly with abdominal pain [20], thereby stressing the importance of CT. The purpose of this study is to evaluate and explore the clinical impact of CT imaging of the abdomen and pelvis in patients 80 years or older presenting to the ED with acute abdominal symptoms.
150 mL (Isovue 300, 300 mg L/mL) injected at a rate of 3 mL/s. Coronal reformatted images (section thickness and interval of 3 mm) are reconstructed from 0.6 to 0.625-mm thick axial sections during the portal venous phase acquisition. According to institution protocol over this time period, a positive oral contrast agent, diluted diatrizoate meglumine (Gastrografin 2%; Schering) was administered, 30 mL mixed in 450 mL of water for the following indications: abdominal pain, small bowel obstruction, abscess, pancreatitis, and enteric fistula or leak.
Review of clinical history and imaging Demographic data including age and gender were collected. The electronic medical record was reviewed by one radiologist (CSG) for diagnosis, treatment plan, and disposition before and after imaging. Determining pre-CT diagnosis
Materials and methods Patient selection Institutional Review Board approval was obtained and patient consent waived for this HIPAA-compliant, retrospective study. A retrospective review of electronic medical records from January 2006 to January 2013 identified 464 geriatric patients age 80 years or older who underwent a CT scan of the abdomen and pelvis after presenting to the ED with acute abdominal symptoms. An age of 80 years old was selected given previous studies suggesting that mortality in this patient subset was particularly high [17, 23]. Patients with traumatic indications were excluded from the study. For instance, a patient undergoing CT of the abdomen and pelvis for abdominal pain in the setting of a fall or trauma was excluded. Patients with known malignancy presenting with acute abdominal symptomatology were included. For example, a patient with colon cancer undergoing CT of the abdomen and pelvis because of concern for acute intestinal obstruction or perforation would be included in the study. In the event that a patient underwent more than one abdominal and pelvic CT in the ED, either during the same or different encounters, the scan from the earliest or first visit was selected.
CT imaging technique Acquisition parameters at our institution for abdominal and pelvic CT imaging in these patients includes helical mode, 120 kVp, beam pitch 0.8–1.375, automated tube current modulation (noise index for GE systems 15–22 HU from 0.625 mm images, reference mA for Siemens systems 200), minimum tube current 100–150 mAs, and reconstructed section thickness and interval of 5 mm. Administration of intravenous (IV) contrast material was
The pre-CT diagnosis was identified by review of the emergency physician note in the electronic medical record. The pre-CT diagnosis was defined as the primary diagnosis under the initial assessment and plan for the patient prior to CT imaging. If no diagnosis was listed by the emergency physician before obtaining the CT, the study indication or reason for CT was used as the principal pre-CT diagnosis. For instance, a study indication of ‘‘rule out sbo’’ was classified as a pre-CT diagnosis of small bowel obstruction, and an indication of ‘‘left flank pain, hematuria’’ was classified as a pre-CT diagnosis of renal/ureteral stone. If more than one diagnosis was recorded in the ED note, the first or leading diagnosis listed in the differential was considered to represent the emergency physician’s pre-CT diagnosis. Determining post-CT diagnosis The post-CT diagnosis was based on final interpretation of the CT. CT report interpretations were categorized into those positive and negative for actionable findings. Actionable findings were defined as those potentially requiring a change in either surgical or medical management. Examples of actionable findings included small bowel obstruction, bowel ischemia or perforation, obstructing renal stone, diverticulitis, appendicitis, and vascular emergency such as aortic dissection, aneurysm rupture, and acute arterial or venous thrombosis. Newly diagnosed malignancy was not considered an actionable finding unless considered to cause the patient’s acute symptomatology, such as a colon cancer resulting in large bowel obstruction. The CT results were corroborated with the final diagnosis in the ED note or discharge summary (in those admitted) and any subsequently available pathology, intervention, surgery or clinical follow-up in patients admitted or referred to a
C. S. Gardner et al.: Impact of CT in elderly patients
subspecialty. Positive CT report interpretations concordant with the clinical diagnosis and follow-up were given a post-CT diagnosis. CTs with non-actionable findings, nonspecific findings, or unconvincing findings, that were not supported clinically, were considered negative. For instance, a patient with an indeterminate renal cystic lesion would be categorized as non-actionable or negative. Similarly, a patient with ‘‘possible bowel wall thickening vs. under-distention’’ without any clinical support of colitis or further workup would be considered negative. Patients with actionable positive CT results were subdivided into medical and surgical treatment categories, summarized in Fig. 1. Analysis of ED records was performed to determine whether the CT results influenced clinical management. Changes in medical management were defined as medication or treatment changes and referrals to a subspecialty consultation based on positive CT results. Changes in surgical management consisted of surgical operations and minimally invasive procedures such as endoscopy and image-guided percutaneous drainage or intervention.
Disposition Anticipated disposition prior to CT imaging and final disposition after CT imaging were recorded. Details of anticipated disposition status prior to CT imaging were obtained from review of the emergency physician note in the electronic medical record and categorized into the following: (1) anticipated admission, (2) anticipated discharge, and (3) unknown or pending CT examination. Under ‘‘disposition’’ in the initial assessment and plan of the ED note, ‘‘likely admit’’ was categorized as an-
Elderly with CT A/P (n=464)
CT Posi ve (n=257, 55%)
Change in Management (n=166, 65%)
Opera ve (n=49, 61%)
CT Nega ve (n=207, 45%)
No Change in Management (n=91, 35%)
Surgical (n=80, 48%)
Medical (n=86, 52%)
Interven onal Radiology (n=20, 25%)
Other Minimally Invasive (n=11, 14%)
Fig. 1. Summary of CT effect on medical and surgical management.
ticipated admission and ‘‘likely discharge’’ as anticipated discharge. Anticipated disposition in the ED note that was either blank, ‘‘unknown’’ or ‘‘pending CT’’ was placed in the third category above. Final dispositions from the ED were as follows: (1) admission, (2) discharge, and (3) death.
Statistical analysis Statistics were performed using Microsoft Excel version 12.1.3 (Microsoft, Redmond, WA) and Statistical Program for the Social Science (SPSS, Chicago, IL) version 11.0, Windows 2000. The incidence of positive CTs, most common indications, most common diagnoses, changes in clinical management, and disposition were reported along with proportions with 95% confidence intervals (CIs). Z test was used to assess proportional differences between pre- and post-CT diagnosis for the most commonly observed diagnoses. The effect of positive vs. negative CT results on disposition, admission, and discharge rates, was compared using z test and Chi square test with Yate’s continuity correction; p < 0.05 indicated statistical significance.
Results Our study identified 464 geriatric patients (mean 89 years, range 80–100 years), including 150 men and 314 women who underwent a CT scan of the abdomen and pelvis in the ED for acute abdominal symptoms. Sixty-nine percent (321/464) of the CTs were performed following the administration of intravenous (IV) contrast material and 31% (141/464) were performed without IV contrast material. The average serum creatinine, in those who received IV contrast material, was 1.1 mg/dL (range 0.5–8.3 mg/dL) and in those who did not 1.4 mg/dL (range 0.4–8.7 mg/dL) (p = 0.01). The most common indications for CT were small bowel obstruction (SBO) (83/464, 18%; 95% CI 15%– 22%), generalized abdominal pain (77/464, 17%; CI 14%–20%), non-ischemic vascular-related indications including abdominal aortic aneurysm rupture or dissection (67/464, 14%; CI 11%–17%), diverticulitis (51/464, 11%, CI 8%–14%), renal or ureteral calculi (32/464, 7%; CI 5%–9%), bowel ischemia (29/464, 6%; CI 4%–8%), appendicitis (12/464, 3%; CI 1%–5%), and pancreatitis (5/464, 1%; CI 0%–2%). CT results were positive for actionable findings in 55% (257/464; CI 50%–60%) and negative in 45% (207/ 464; CI 40%–50%). Table 1 summarizes the most common final diagnoses pre- and post-CT. Significant differences in pre- vs. post-CT diagnosis were seen with SBO, diverticulitis, and large bowel obstruction. No significant difference in pre- vs. post-CT diagnosis was found in vascular emergency, bowel ischemia, or appendicitis. These six most common CT diagnoses
C. S. Gardner et al.: Impact of CT in elderly patients
Table 1. Most common diagnoses pre- and post-computed tomography (CT) Final diagnosis SBO Diverticulitis Vascular emergency Ischemia Appendicitis Large bowel obstruction Totals
Pre-CT diagnosis, n (%, CI 95%)a 34 8 16 5 4 0 67
Post-CT diagnosis, n (%, CI 95%)*
(13, 9–17) (3, 1–5) (6, 3–9) (2, 0–4) (2, 0–4) (0, 0–0) (26, 19–29)
50 25 18 10 8 6 117
(20, 14–24) (10, 6–14) (7, 4–10) (4, 2–6) (3, 1–5) (2, 0–4) (42, 36–48)
p value
