Journal of Gastrointestinal Surgery

, Volume 11, Issue 11, pp 1417–1422

Computed Tomography in the Diagnosis of Acute Appendicitis: Definitive or Detrimental?

  • Sandeepa Musunuru
  • Herbert Chen
  • Layton F. Rikkers
  • Sharon M. Weber
Article

DOI: 10.1007/s11605-007-0268-y

Cite this article as:
Musunuru, S., Chen, H., Rikkers, L.F. et al. J Gastrointest Surg (2007) 11: 1417. doi:10.1007/s11605-007-0268-y

Abstract

Objectives

Utilization of computed tomography (CT) scans in patients with presumed appendicitis was evaluated at a single institution to determine the sensitivity of this diagnostic test and its effect on clinical outcome.

Methods

Adult patients (age > 17 years) with appendicitis were identified from hospital records. Findings at surgery, including the incidence of perforation, were correlated with imaging results.

Results

During a 3-year period, 411 patients underwent appendectomy for presumed acute appendicitis at our institution. Of these patients, 256 (62%) underwent preoperative CT, and the remaining 155 (38%) patients did not have imaging before the surgery. The time interval between arrival in the emergency room to time in the operating room was longer for patients who had preoperative imaging (8.2 ± 0.3 h) compared to those who did not (5.1 ± 0.2 h, p < 0.001). Moreover, this possible delay in intervention was associated with a higher rate of appendiceal perforation in the CT group (17 versus 8%, p = 0.017).

Conclusions

Preoperative CT scanning in patients with presumed appendicitis should be used selectively as widespread utilization may adversely affect outcomes. The potential negative impact of CT imaging includes a delay in operative intervention and a potentially higher perforation rate.

Keywords

Appendicitis Diagnosis Perforation Imaging CT 

Introduction

Approximately 250,000 appendectomies are done per year in the USA, making it the most common emergency procedure performed by general surgeons. Despite such a large number of cases, diagnosis is often difficult. Demonstrating this fact, at the time of operation, a normal appendix is found in approximately 15–20% of cases.1 Much of the uncertainty in diagnosis occurs in women of childbearing age and in patients with atypical presentations.

The consequences of a negative appendectomy include development of intraabdominal adhesions, adverse effects of anesthesia, cost, and negative effects on quality of life from unnecessary surgery.2,3 To demonstrate the potential negative impact on these patients, the complication rate after a negative appendectomy may be as high as 6% and reoperation occurs in up to 2%.4 On the other hand, in patients with appendicitis, delaying the time to definitive intervention may lead to negative consequences including perforation. Appendiceal perforation is associated with a threefold increase in complications, including an eightfold increase in abscess formation. There are also higher rates of reoperation, sepsis, infertility, and dehiscence associated with perforated appendicitis, compared to simple appendicitis.1 Thus, a great deal of effort has been placed on making an early and accurate diagnosis, as multiple complications can occur, both in the setting of a false-negative and a false-positive diagnosis.

Computed tomography (CT) scan is playing a larger role in clarifying the clinical picture in patients with presumed appendicitis, particularly over the last decade. CT was popularized in the late 1990s after publication of a prospective trial that supported the routine use of CT with rectal contrast. This trial found that CT was 98% accurate in diagnosing appendicitis, although the technique employed is not commonly utilized.5 Another prospective randomized trial in patients with atypical symptoms of appendicitis determined that noncontrast CT scan is superior to ultrasound in diagnostic accuracy and reliability.6 Because of these studies, the use of CT for the routine diagnosis of appendicitis has markedly increased at our institution over the last decade. Therefore, we sought to evaluate the sensitivity, specificity, and positive and negative predictive value of CT, and its effect on clinical outcome at an academic teaching institution.

Materials and Methods

Four hundred and eleven adult patients (age > 17 years) underwent emergency appendectomy for presumed appendicitis at the University of Wisconsin Hospital and Clinics over a 3-year period from January 2002 through December 2004. These patients were identified by ICD-9 codes. Medical records were retrospectively reviewed to assess whether CT scans were utilized for preoperative diagnosis. Patients were analyzed for demographic variables such as age and gender. In addition, each patient’s medical record was reviewed to evaluate white blood cell count, time interval from emergency room to operating room, laparoscopic versus open procedure, operating room time (defined as incision to closure), and presence or absence of appendiceal perforation. Pathology results for each specimen were reviewed. Perforation was defined as either gross perforation found at the time of operation and/or microperforation discovered on histological exam. This study was approved by the University of Wisconsin Institutional Review Board.

Our policy was to utilize abdominal and pelvis CT scans with intravenous and oral contrast. However, if patients had renal insufficiency, patients either underwent prehydration or the intravenous contrast was withheld. During the time of the study, there was no institutional policy dictating which patients received CT scans. In general, patients were first evaluated by emergency room (ER) physicians followed by surgical residents. CT scans may have been ordered by the ER physician or the surgical team. Patients with negative CT scans underwent operative intervention if there was a high clinical suspicion of appendicitis. During the time of the study, patients with appendicitis underwent operation at the time of the next available operating room.

Statistical Analysis

Statistical analyses between groups were performed with analysis of variance (ANOVA) using Statistical Package for the Social Sciences (SPSS) software (SPSS Inc., version 14.0). Data are represented as mean ± SEM. Sensitivity was defined as the number of cases of appendicitis correctly diagnosed by CT divided by the total number of cases of appendicitis. Specificity was defined as the number of cases without appendicitis divided by the number of negative tests obtained. Indeterminate CT scans were categorized as negative, as they did not assist with clinical decision-making. Positive predictive value was calculated from the number of positive cases of appendicitis cases diagnosed by CT compared to the total number of positive CT scans. Negative predictive value was calculated from the number of negative cases of appendicitis cases diagnosed by CT compared to the total number of negative CT scans. Significance was defined as a p value < 0.05.

Results

Patient Demographics

Of the 411 appendectomy patients, 256 (62%) had a preoperative CT, and the remaining 155 (38%) patients did not have imaging before their operation. The median age was older for the CT group compared to the non-CT group (Table 1). The majority of the patients were male, but a higher percentage of females underwent CT imaging. The mean white blood cell (WBC) counts at presentation to the emergency room (ER) were similar between the two groups (Table 1).
Table 1

Patient Demographics

 

Number of Patients

Mean Age

Gender

WBC (×1,000)

Male

Female

CT

256

37 ± 1

127

129

13.6 ± 0.3

No CT

155

31 ± 1

103

52

14.4 ± 0.3

p Value

 

<0.001

0.001

NS

Mean ± SEM

NS Not significant, WBC white blood cell

Operative Results

Comparisons were made between the number of laparoscopic versus open appendectomies in each group. There was a significant difference with more laparoscopic procedures performed in the non-CT group (Table 2). The mean operating time for both CT and non-CT patients was 1.2 h.
Table 2

Operative Results

 

OR Time (Hours)

Percent Laparoscopic

CT

1.2 ± 0.1

88%

No CT

1.2 ± 0.1

96%

p Value

NS

0.003

Mean ± SEM

NS Not significant, OR operating room

The final pathology in the corresponding groups is found in Table 3. Overall, CT had a sensitivity of 92% (208:225) and a specificity of 68% (21:31). The positive and negative predictive values were 91% (208:229) and 37% (10:27), respectively.
Table 3

Pathological Results

 

 

Pathology

 

Positive

Negative

Indeterminate

CT scan

 

+

208

19

2

 

7

7

 
 

+/−

10

3

 

No CT

 

133

22

 

(+) Appendicitis, (−) normal appendix, (+/−) indeterminate

The negative appendectomy rate for patients diagnosed with appendicitis on CT was 8% (19:227). For all patients who underwent appendectomy without preoperative imaging, the negative appendectomy rate was 14% (22:155), which was not significantly different (p = 0.09). In addition, 7 of 14 patients with negative CT scans were ultimately found to have appendicitis; thus, the negative appendectomy rate in patients with negative CT scans was 50% (p < 0.001, compared to those with CT scans positive for appendicitis). When evaluating both negative and indeterminate scans together, the negative appendectomy rate was 37% (17:27, p < 0.001).

Outcomes

The time interval from the patient’s first contact in the emergency room to the operating room start time was significantly longer in the CT group as compared to the non-CT group (Table 4). In addition, the rate of perforation (based on final pathology and intraoperative observation) was significantly greater in the imaging group [17% (43:256) versus 8% (13:155), p = 0.017].
Table 4

Outcomes

 

Time from ER to OR (Hours)

Percent Perforation (All specimens)

Percent Perforation (Appendicitis only)

CT

8.2 ± 0.3

17%

19%

No CT

5.1 ± 0.2

8%

10%

p Value

<0.001

0.017

0.033

Mean ± SEM

ER Emergency room, OR operating room

Because more patients in the CT group actually had appendicitis [89% (225:254) versus 86% (134:155) in the non-CT group], we compared perforation rates in those with pathologically proven appendicitis. In this group, the perforation rate remained significantly elevated in the CT group [19% (42:225) versus 10% (13:133), p = 0.033].

Discussion

The diagnosis of acute appendicitis is often not straight forward. Imaging studies, including CT scans, have been employed at increasing rates over the last decade in an attempt to improve diagnostic accuracy. In this study, the use of CT scan for diagnosis of appendicitis and its effect on clinical outcome was retrospectively reviewed at our institution. The two major findings of this investigation are (1) there was no significant difference in the negative appendectomy rate between those that had preoperative imaging and those that did not and (2) there was a significantly longer time to operation in patients who had preoperative CT scan, and this was associated with an increased rate of appendiceal perforation.

Although initial reports on the use of CT scans in patients with appendicitis concluded that CT should be used routinely in all patients suspected to have appendicitis, more recent reports suggest that a selective approach is likely more beneficial.1,5,6 As these more selective approaches to preoperative imaging have been employed, it has become clear that CT scanning is beneficial for the diagnosis of appendicitis in patients with atypical presentations and in women of childbearing age.7 However, this is not without increased cost, radiation exposure, and a potential delay in time to definitive treatment. To demonstrate this, even in studies that utilized a selective approach to imaging with the use of an institutional pathway, CT was obtained in 529 cases but only a minority (97, 18%) actually had appendicitis.1 It is clear that we still need to make progress in clinically assessing patients with presumed appendicitis, and, even when CT is used selectively, the cost of making an accurate diagnosis remains high. Although institution of a clinical pathway in one study, Antevil et al., led to a substantial decrease in the number of negative appendectomies (from 16 to 4%, p < 0.001), the issue of improving patient selection for CT remains a problem.

The negative appendectomy rate in the present study was 11% (29:256), which is consistent with the national average. The negative appendectomy rate for patients receiving CT imaging was 8% compared to the non-CT group rate of 14%, which was not a significant difference (p = 0.09). Importantly, of the 14 patients with CT scans determined to be negative, seven of these patients were found to have appendicitis at surgery, resulting in a 50% false negative rate. Thus, this reiterates the importance of relying on clinical findings even in the setting of a negative CT.

Although not examined in this study, it is clear that prolonged time from first symptoms to definitive operation increases the risk of rupture in patients with appendicitis.8,9 In fact, the risk of rupture increases approximately 5% for each ensuing 12-h period after 36 h.8 In addition, multiple studies, including our own, have found that utilization of preoperative CT scan leads to a delay in definitive treatment.8,9,10, 11, 12 Some studies have found that obtaining a CT results in a delay to operative intervention as great as 6–12 h compared to patients that did not have preoperative imaging.8,13

One area not examined in this retrospective study is the type of contrast utilized for the CT and whether the accuracy of CT is dependent on route of contrast administration. There are conflicting opinions as to whether the use of intravenous, oral, and/or rectal contrast will result in the most accurate images (Table 5). The original prospective study evaluating the accuracy of CT for the diagnosis of appendicitis utilized rectal contrast only, and other prospective randomized trials concluded that the use of rectal contrast only compared to triple contrast (intravenous, oral, and rectal) resulted in decreased delay to definitive surgery, decreased perforation rate, and a decrease in contrast-related morbidity without any compromise in diagnosis.5,14 In spite of this, the routine practice employed by many institutions is to utilize oral and intravenous contrast for the diagnosis of appendicitis, likely because of the fact that this also allows for assessment of other areas of intra-abdominal pathology. Tables 5 and 6 summarize recent studies and the sensitivity, specificity, and accuracy for CT imaging of appendicitis.
Table 5

Summary of Recent Prospective, Randomized Trials Evaluating Use of CT for Diagnosis of Appendicitis

Study

Type of Study

Number of Patients

Type of Contrast

Sensitivity (%)

Specificity (%)

Accuracy (%)

Negative Appendectomy Rate (%)

Walker15

Prospective randomized

63

No CT (PE only) OR PO and/or IV contrast

100

79

89

19

65 (CT)

Rectal

94

100

96

5

Mittal14

Prospective randomized

52

Triple contrast (PO, IV, rectal)

97

86

92

8.3

39

Rectal only

88

100

92

7.7

Hong12

Prospective randomized

68

None

100

73

90

NA

97(CT)

PO, IV

91

93

92

NA

CT Computed tomography group, PO oral contrast, IV intravenous contrast, NA not available

Table 6

Summary of Recent Retrospective and Prospective Nonrandomized Trials Evaluating Use of CT for Diagnosis of Appendicitis

Study

Type of Study

Number of Patients

Type of Contrast

Sensitivity

Specificity

Accuracy

Negative Appendectomy Rate (%)

Torbati16

Prospective nonrandomized

250

None, PO, IV, rectal

92%

97%

96%

7.8

Hershko17

Prospective nonrandomized

198 (CT)

PO, IV

91%

92%

91%

16

in’t Hof18

Prospective nonrandomized

103

None

95.4%

100%

95%

NA

Rao5

Prospective nonrandomized

100

Rectal

98%

98%

98%

NA

Lee11

Retrospective

766 (total)

NA

83

31.7

74.9

15.7

47 (CT)

NA

83.8

40

74.5

NA

Fuchs19

Retrospective

42

None

   

11.9

182

PO, IV

99%

96%

97%

6.3

Present study

Retrospective

155

None

   

14

256 (CT)

PO, IV

92%

68%

88%

8

CT Computed tomography group, PO oral contrast, IV Intravenous contrast, NA not available

The major limitation of our study, as well as many of the cited studies, is its retrospective design. Another limitation is that only patients undergoing abdominal exploration for appendicitis were included. CT may have benefited patients with suspected acute appendicitis but who were successfully managed non-operatively after a negative CT, or who were found to have other intraabdominal explanations for their abdominal pain based on findings on CT.

In conclusion, preoperative CT scanning in patients with suspected appendicitis should be used selectively as widespread utilization may adversely affect outcomes. We believe that CT imaging does have a role in the diagnosis of acute appendicitis, particularly in patients with atypical presentation and in women of childbearing age with unusual symptoms. The routine use of CT scan to evaluate patients suspected of having acute appendicitis will result in unnecessary exposure to contrast and radiation in a large number of patients and delay in operation intervention. Therefore, it should be discouraged.

Copyright information

© The Society for Surgery of the Alimentary Tract 2007

Authors and Affiliations

  • Sandeepa Musunuru
    • 1
  • Herbert Chen
    • 1
  • Layton F. Rikkers
    • 1
  • Sharon M. Weber
    • 1
  1. 1.Department of SurgeryUniversity of WisconsinMadisonUSA

Personalised recommendations