| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Beth Israel Deaconess Medical Center and Harvard
Medical School, 330 Brookline Ave., Boston, MA 02215.
2
Department of Clinical Computing, Beth Israel Deaconess Medical Center and
Harvard Medical School, Boston, MA 02215.
3
Department of Emergency Medicine, Beth Israel Deaconess Medical Center and
Harvard Medical School, Boston, MA 02215.
Received August 23, 1999;
accepted after revision November 1, 1999.
Presented at the annual meeting of the Society of Gastrointestinal
Radiologists, Palm Beach, FL, March 1999.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Fifty-seven patients were enrolled in this prospective study. Using a computer order entry system, emergency department physicians were required to report their most likely diagnosis, level of certainty, and management plan for their patients before ordering abdominal CT. After CT was performed, each physician was required to provide again his or her diagnosis, level of diagnostic certainty, and treatment plan. The outcome of each patient was evaluated by either surgery, other imaging studies, or clinical follow-up.
RESULTS. After the abdominal CT, physicians' mean level of certainty in their diagnoses increased by 1.5 points (on a five-point scale; p < 0.0001). Patient management was changed in 33 (60.0%) of 55 patients. Planned treatment before CT was admission in 42 patients. Actual admissions after CT totaled 32 patients (excluding the two patients in whom preimaging information was not recorded). Thus, the net effect of abdominal CT scanning was to avert 10 (23.8%) of 42 hospital admissions.
CONCLUSION. CT performed in the emergency department increases the physician's level of certainty, reduces hospital admission rates by 23.8%, and leads to more timely surgical intervention.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Abdominal pain unrelated to trauma is one of the three most common symptoms in patients presenting to the emergency department or admitted to general hospitals. The differential diagnosis of an acute abdomen includes a broad spectrum of causes ranging from self-limiting benign causes for which surgery may be not indicated [1, 2] to high-morbidity or high-mortality causes that necessitate prompt surgical intervention. Because many causes may have similar early clinical presentations, imaging studies are frequently used to identify specific causes when possible and to prevent delay in diagnosing urgent surgical conditions. For these reasons, we chose to investigate the way in which CT alters patient management in patients presenting to the emergency department with acute nontraumatic abdominal pain.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CT Ordering and Protocol
CT scans were ordered by physicians using a computerized order entry
system. Twenty-six CT scans were ordered by interns, 22 by residents, and nine
by attending physicians. Before completing the order, the physician was
required to answer the following questions, administered through an online
questionnaire [3]:
All CT was then performed according to the following protocols. The specific protocol was chosen depending on the indication entered by the ordering physician. All CT scans were performed on a HiSpeed Advantage CT scanner (General Electric Medical Systems, Milwaukee, WI) using a large field of view and a 30- to 48-cm display field of view. Helical scanning was performed at 220-240 mA and 120 kVp. The nonionic IV contrast medium ioversol (Optiray 320; Mallinckrodt, St. Louis, MO), was used if the patient had a history of contrast media allergy or other allergies; otherwise iothalamate (Conray 60; Mallinckrodt) was administered. Unless otherwise noted, all scans were of 7-mm-thick slices acquired with a pitch of 1.5 and 7-mm interval spacing.
If the clinical diagnosis was abscess, cholecystitis, or a gynecologic process, the patient was given 900 ml of oral contrast material 1 hr before CT. The following technique was used: 100 ml of IV contrast material was injected at 2 ml/sec. After a 60-sec delay, the abdomen was scanned. An additional 50 ml of IV contrast material was then injected at 2 ml/sec, and after an additional 60-sec delay, helical scanning of the abdomen and pelvis was performed.
If the clinical diagnosis was appendicitis, bowel perforation, diverticulitis, ischemic bowel, small-bowel obstruction, or solid organ or bowel malignancy, the patient was given 1200 ml of oral contrast material at least 1 hr before CT. One hundred milliliters of IV contrast material was administered at 2 ml/sec, and after a 60-sec delay the abdomen and pelvis were scanned using 5-mm-thick slices, a pitch of 1.5-1.8, and 3-mm interval spacing.
When acute pancreatitis was the clinical diagnosis, CT of the abdomen was performed 1 hr after administration of 900 ml of oral (not IV) contrast material. The abdomen was then scanned with 5-mm-thick slices through the pancreas. One hundred fifty milliliters of IV contrast material was then administered, and the abdomen and pelvis were rescanned.
If renal colic was the most likely diagnosis, CT urography was performed according to the following protocol: unenhanced CT of the abdomen was performed at 5-mm slice thickness, a pitch of 1.5-1.8, and 5-mm interval spacing. If a renal or ureteral stone was identified, no further imaging was performed. If a stone was not clearly identified, 100 ml of IV contrast material was administered at 2 ml/sec, and after a 60-sec delay the abdomen was scanned. After a 3-min delay, the patient was placed prone. An additional 50 ml of IV contrast material was administered at 2 ml/sec, and after a 60-sec delay the abdomen and pelvis were scanned to include the kidneys and bladder. Five-millimeter-thick slices were acquired at a pitch of 1.5-1.8 with 3-mm interval spacing.
If an abdominal aortic aneurysm was considered, unenhanced scanning of the abdomen was performed. One hundred fifty milliliters of IV contrast material was then administered at 3.5 ml/sec, and after a 20-sec delay, scanning of the abdomen and pelvis was performed at 3-mm slice thickness, a pitch of 2.0, and 2-mm interval spacing. If retroperitoneal bleeding was considered the most likely diagnosis, the abdomen and pelvis were scanned without IV or oral contrast material.
Between 8:00 A.M. and 10:00 P.M., all CT scans were interpreted by an attending radiologist. All scans obtained between 10:00 P.M. and 8:00 A.M. were initially interpreted by a radiology resident and subsequently reviewed by an attending radiologist at 8:00 A.M. No CT scans included in this study were obtained on the weekend, when attending radiologist coverage is less available. Forty-five (78.9%) of 57 CT scans were performed between 8:00 A.M. and 10:00 P.M. If CT was performed between 10:00 P.M. and 8:00 A.M., the emergency department physician made his or her decision based on the resident's interpretation. None of the final interpretations provided by the attending staff would have altered patient management.
Within 24 hr of the CT, the referring clinician was sent an e-mail with the following questions: (1) What was your diagnosis after the CT scan? (The list of the most likely diagnoses that had been presented at the time the CT scan was ordered was again presented to the referring doctor). (2) On a scale of 0-100%, what was the certainty of your diagnosis after the CT scan: very low (<20%), low (20-39%), moderate (40-59%), high (60-79%), very high (>80%)? The patient's final diagnosis was then determined by a review of the patient's medical record using either pathology, surgery, additional imaging studies, or clinical follow-up. ("Clinical follow-up" was defined as a second examination by a physician for the chief complaint that initiated the CT.) The change in the physician's level of diagnostic certainty before and after CT was calculated, as was the impact of CT on patient management. Differences in the change in the physician's level of certainty were calculated using a two-sided, paired Wilcoxon's signed rank test. The sensitivity and specificity of the CT scan were then calculated using pathology, surgery, clinical follow-up, or other imaging studies. The sensitivity and specificity of CT were also calculated adjusting for verification bias [4].
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Complete data on the planned treatment before CT were available in 55 (96.5%) of 57 patients. In two cases, physicians chose "other" from the set menu but failed to enter a plan using the free text option. Data for the 55 patients in whom complete data were available are presented in Table 2. Patient management was changed in 33 (60.0%) of 55 patients. In 12 patients (21.8%), the initial treatment plan included hospital admission but was changed to a treatment plan that allowed the patients to be sent home. In two (3.6%) of 55 patients, the initial treatment plan would have sent the patient home, but after CT these patients were admitted. Forty-two patients would have been admitted on the basis of the management plan before CT. Thirty-two patients were actually admitted after CT results were known (excluding the two patients in whom the management plan before CT was not recorded). Thus, the net effect of CT scanning was that 10 (23.8%) of 42 patients who would have been admitted to the hospital were sent home. In addition, six patients who would have been admitted for observation (n = 5) or sent home with medical treatment (n = 1) went on to have immediate surgery after CT.
|
A comparison of the most likely diagnosis provided by the ordering physician and the patent's diagnosis after CT is provided in Table 3. In 27 patients (47.4%), the diagnoses before and after CT were congruent. However, in 30 patients (52.6%), CT was discordant with the initial clinical presentation. In 15 patients (26.3%), the CT findings were normal or "near normal," effectively excluding any disorder. In the remaining 15 patients (26.3%), CT provided an alternative diagnosis to the one initially suspected on clinical grounds.
|
Follow-up was available for 44 (77.2%) of 57 patients and included surgery or pathology in 24 patients (54.5%), other imaging studies in six (13.6%), and clinical follow-up in 14 (31.8%). Three of the 14 patients with clinical follow-up died, one from a leaking thoracoabdominal aneurysm, one from metastatic ovarian carcinoma, and one from metastatic cervical carcinoma. The mean clinical follow-up for the remaining 11 patients was 81 ± 97 days (range, 3-339 days).
CT enabled the correct diagnosis in 41 (93.2%) of 44 patients for whom follow-up was available. In two patients (4.5%), CT provided a false-positive diagnosis. In one patient, CT documented a thickened wall in the transverse and descending colon; a colonic biopsy showed only prominent lymphoid aggregates, without evidence of colitis. In a second patient, CT showed inflammatory changes around the appendix in the right lower quadrant; appendectomy yielded a normal appendix. In one patient CT provided a false-negative diagnosis. That patient was a 31-year-old man with acute right lower quadrant pain. The CT findings were interpreted as normal (the appendix was visualized). However, because of high clinical suspicion, the patient had an appendectomy that revealed retrocecal appendicitis (Fig. 1A,1B). The sensitivity and specificity for CT were 96.9% and 83.3%, respectively.
|
|
Follow-up was not available in 13 patients (22.8%). In seven (53.8%) the CT findings were abnormal, and in six (46.2%) the CT findings were interpreted as normal. Diagnoses made on CT scans for which there was no follow-up included a thickened gallbladder wall, pelvic fluid collection in a postoperative patient, hydronephrosis (n = 2), distended appendix with pelvic fluid, Crohn's disease, and diverticulosis. Data were adjusted for verification bias to correct for the fact that patients with abnormal CT findings are more likely to undergo follow-up imaging, biopsy, or surgery than patients having normal CT findings. When the data were adjusted for verification bias, the sensitivity and specificity for all pathologic findings were 96.0% and 86.5%, respectively.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
For example, Figure 2A,2B,2C,2D shows the images of a 44-year-old woman who presented to the emergency department after 24 hr of nonspecific periumbilical abdominal pain and fever. The pain had localized to her right lower quadrant at the time of presentation. Abdominal CT was performed and showed an inflamed appendix with signs of perforation but no abscess formation. The patient was admitted for observation and was given IV vancomycin, metronidazole, and ceftriaxone. Within 24 hr her pain and fever had resolved, and she was begun on a a therapeutic regimen of oral alatrofloxacin mesylate (Trovan; Pfizer, New York, NY) for 14 days and discharged. Four weeks later she underwent follow-up CT that showed complete resolution of the periappendiceal inflammation (Fig. 2D). The patient then underwent laparoscopic appendectomy, performed on an outpatient basis, 4 months later. Had CT not been performed in the emergency department, it is likely that the patient would have gone to immediate surgery, which may have required an open surgical procedure and possible delayed bowel closure because of the periappendiceal inflammation. In contrast, this patient spent less than 24 hr in the hospital and had her surgery performed at her convenience. No hospitalization was required, reducing the cost of the appendectomy.
|
|
|
|
Despite the broad use of imaging in the evaluation of the acute abdomen, few studies have attempted to assess the impact of imaging on the diagnosis and management of patients with acute abdomen. Prospective studies assessing the value of conventional abdominal or chest radiographs, or both, in patients with acute abdomen found a change of management in 4-10% [9, 10]. Although early retrospective assessment of abdominal CT in the acute abdomen showed little diagnostic yield [11], another early retrospective report showed the ability of CT to identify unsuspected findings leading to a change in clinical management in 22 (27.5%) of 80 patients with active Crohn's disease [12]. Similar results were reported with sonography [13, 14]; early prospective assessment of the role of sonography [15] in acute abdomen showed a 19-23% yield of unsuspected findings that influenced patient management [16, 17].
In a prospective evaluation of the impact of CT on the diagnosis and treatment of patients with acute abdomen, Taourel et al. [18] reported that CT results modified the primary therapeutic strategy in 30% of cases and prevented unnecessary laparotomies in seven of 40 patients. In a later study, Taourel et al. [19] reported that prospective CT showed a change in treatment in 12 (21.0%) of 57 patients with suspected acute small-bowel obstruction. A larger retrospective series by a different group of researchers assessing the impact of CT in the acute abdomen initially treated without surgery showed a change in management in 25 (27.8%) of 90 patients on the basis of CT results [20].
More recently, tailored limited CT protocols [21] have been advocated [22] to increase the sensitivity and specificity of CT over conventional radiography, particularly in the diagnosis of appendicitis [23, 24] and ureteral calculi [25]. Limited CT examinations tailored to confirm or exclude a specific suspected cause of acute abdomen may reduce the examination time and cost as well as the need for IV contrast material, while improving accuracy for a specific diagnosis. However, a risk exists of premature narrowing of the diagnostic focus in clinical settings. In patients in whom the clinical presentation is not specific for a single cause, the ultimate diagnosis may be delayed, resulting in an increased risk of medicolegal action.
In fact, the one case of false-negative CT findings in our study occurred in a 40-year-old man who presented with nonspecific abdominal pain and hematuria (Fig. 1A,1B). The emergency department physician ordered CT urography, which was performed without oral contrast material. No renal or ureteral calculi were identified, and appendicitis was not seen on the CT scan. However, because of high clinical suspicion, the patient was taken to the operating room, and a laparoscopic appendectomy was attempted. Retrocecal appendicitis was identified, and because the appendix had adhered to the retroperitoneum, the laparoscopic procedure had to be converted to open surgery. In retrospect, several signs suggestive of appendicitis can be seen on this CT scan (Fig. 1A,1B). We suspect that, had CT been performed with oral contrast material (and not tailored to visualize the renal system), the appendicitis would likely have been evident prospectively.
One of the advantages of performing this study at our institution was the availability of a computer order entry system [26]. This system allowed us to collect a wide range of data before the actual time the physician ordered the CT scan. Because this system is automated, we could collect this information on an ongoing basis without additional paperwork. The expanded use of computer order entry systems should facilitate similar research at other institutions and should facilitate documentation of the value of CT and other imaging techniques.
Our study had several limitations. The first is that of selection bias, in that not all patients presenting to the emergency department with acute abdominal pain underwent CT scanning, and we have not investigated the outcome of patients who did not undergo abdominal CT. This lack of data precludes performing a formal cost-effectiveness analysis. A second limitation is that because the use of e-mail allowed free text entry for collecting the physician's level of diagnostic certainty after CT, we do not have the physician's level of certainty after CT for 11 patients. Similarly, the diagnosis before CT was absent for two patients in whom the free-text option "other" was chosen, but no diagnosis was provided. If we were to repeat this study, all data entry points would be mandatory.
Despite these limitations, we believe that our study provides clear evidence of the value of CT performed in the emergency department for patients with acute abdominal pain. Under many managed care contracts, the cost of imaging studies is included in the outpatient imaging budget. These budgets simply calculate the cost to the insurer of performing the imaging studies and negate the potential "downstream" cost savings that an imaging test may provide.
The results of our study lend support to the view that the cost of imaging performed in the emergency department should be considered in the context of the patient's total care. The true value of CT performed in the emergency department lies not only in its ability to provide a correct diagnosis, but also in its ability to increase the emergency department physician's level of certainty and to exclude alternative diagnoses. The use of CT both facilitates more timely surgical intervention and reduces the number of patients requiring hospital admission. In our opinion, negative findings on CT can almost always allow emergency physicians to confidently exclude many (if not all) serious cases of acute abdominal pain.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  E. Lazarus, W. W. Mayo-Smith, M. B. Mainiero, and P. K. Spencer CT in the Evaluation of Nontraumatic Abdominal Pain in Pregnant Women Radiology, September 1, 2007; 244(3): 784 - 790. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B. MacKersie, M. J. Lane, R. T. Gerhardt, H. A. Claypool, S. Keenan, D. S. Katz, and J. E. Tucker Nontraumatic Acute Abdominal Pain: Unenhanced Helical CT Compared with Three-View Acute Abdominal Series Radiology, October 1, 2005; 237(1): 114 - 122. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S Master, G. F Longstreth, and A. L Liu Results of computed tomography in family practitioners' patients with non-acute abdominal pain Fam. Pract., October 1, 2005; 22(5): 474 - 477. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Lewis, G. A. Banet, M. Blanda, F. M. Hustey, S. W. Meldon, and L. W. Gerson Etiology and Clinical Course of Abdominal Pain in Senior Patients: A Prospective, Multicenter Study J. Gerontol. A Biol. Sci. Med. Sci., August 1, 2005; 60(8): 1071 - 1076. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. R. Birchard, M. A. Brown, W. B. Hyslop, Z. Firat, and R. C. Semelka MRI of Acute Abdominal and Pelvic Pain in Pregnant Patients Am. J. Roentgenol., February 1, 2005; 184(2): 452 - 458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Raptopoulos, G. Katsou, M. P. Rosen, B. Siewert, S. N. Goldberg, and J. B. Kruskal Acute Appendicitis: Effect of Increased Use of CT on Selecting Patients Earlier Radiology, February 1, 2003; 226(2): 521 - 526. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Alobaidi and A. Shirkhoda Value of Bone Window Settings on CT for Revealing Appendicoliths in Patients with Appendicitis Am. J. Roentgenol., January 1, 2003; 180(1): 201 - 205. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S Ng, C. J E Watson, C. R Palmer, T. C. See, N. A Beharry, B. A Housden, J A. Bradley, and A. K Dixon Evaluation of early abdominopelvic computed tomography in patients with acute abdominal pain of unknown cause: prospective randomised study BMJ, December 14, 2002; 325(7377): 1387 - 1387. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. H. Gottlieb, T. C. La, E. N. Erturk, J. L. Sotack, S. L. Voci, R. G. Holloway, L. Syed, I. Mikityansky, A. T. Tirkes, R. Elmarzouky, et al. CT in Detecting Urinary Tract Calculi: Influence on Patient Imaging and Clinical Outcomes Radiology, November 1, 2002; 225(2): 441 - 449. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Ahn, W. W. Mayo-Smith, B. L. Murphy, S. E. Reinert, and J. J. Cronan Acute Nontraumatic Abdominal Pain in Adult Patients: Abdominal Radiography Compared with CT Evaluation Radiology, October 1, 2002; 225(1): 159 - 164. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N Raby The role of CT in acute abdominal pain Imaging, May 1, 2001; 13(2): 112 - 123. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
