Whether Intravenous Contrast is Necessary for CT Diagnosis of Acute Appendicitis in Adult ED Patients? Yu-Hui Chiu, MD, Jen-Dar Chen, MD, Shih-Hao Wang, MD, Chui-Mei Tiu, MD, Chorng-Kuang How, MD, Jiun-I Lai, MD, Yi-Hong Chou, MD, Carlos Lam, MD, Ray-Jade Chen, MD Rationale and Objectives: To assess the necessity of intravenous contrast medium for abdominopelvic computed tomography (CT) diagnosis of acute appendicitis (APP) among adult patients with right lower quadrant (RLQ) abdominal pain at emergency department (ED). Materials and Methods: ED patients with clinical suspicion of APP from RLQ pain for a period of 8 months were enrolled retrospectively. Both pre- and postintravenous contrast-enhanced CT scans were performed for these patients. The visibility of vermiform appendix and specific CT findings of APP were recorded separately for noncontrast CT (NCT) and contrast-enhanced CT (CCT) images without knowledge of the patient’s identity and final diagnosis. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of CT diagnosis for APP were compared between the two groups. The ease of identifying appendix was also compared. Results: Forty-two (42.0%) of the 100 patients (55 males, 45 females; age range, 16–90 years; mean age, 49.3 years) were APP. There was no significant difference for the visibility of appendix (94% vs. 91%; P = .589) and radiological characters between the CCT and NCT groups. There were significant differences between the two groups for sensitivity (100% vs. 90.5%; P = .036), specificity (94.8% vs. 100%; P = .038), PPV (93.3% vs. 100%; P = .021), NPV (100% vs. 93.5%; P = .021), but no significant difference for accuracy (97% vs. 96%; P = 1). The appendix was easier to detect on CCT than NCT images (P = .013). Conclusion: The diagnostic sensitivity of CCT was significantly better than that of NCT. Intravenous contrast administration could also make doctors easier in indentifying appendixes. Key Words: Acute appendicitis; computed tomography; contrast medium; length of stay. ªAUR, 2013

A

cute appendicitis (APP) remains one of the leading causes of acute abdominal pain requiring surgical treatment in patients presenting to the emergency department (ED), occurring in 27.5% of surgical abdominal emergencies (1–3). Unfortunately, timely diagnosis remains clinically challenging, and the correct diagnosis is not made in at least 20% of patients with APP (4–8). Delays increase the risk of appendiceal perforation, postoperative complications and medical expenses (8,9). As helical abdominopelvic computed tomography (CT) performed in patients with clinically suspected APP has shown

high diagnostic accuracy (93% to 98%) and reduction of hospital resource utilization in many previous reports, the use of this imaging modality in diagnosing APP has increased steadily (5–8,10–13). Various methods to enhance the visibility of CT have been studied, and the protocol with intravenous contrast administration has proven to be an effective technique, which also reduced the negative appendectomy rate (1,11). However, with the increasing emphasis on the dose of radiation exposure and the risk of contrast-induced nephropathy or allergic reaction, noncontrast CT is particularly appealing in nowadays’ choice of diagnostic tests. The aim of this study was to assess the necessity of intravenous contrast medium for CT diagnosis of APP among adult patients at ED.

Acad Radiol 2013; 20:73–78 From the Department of Emergency Medicine, Wan-Fang Hospital, Taipei Medical University, Taipei, Taiwan (Y.-H.C., C.L., R.-J.C.); Department of Radiology, Taipei City Hospital, Zhongxiao Branch, Taipei, Taiwan, No. 87, Tongde Road, Nangang District, Taipei 115, Taiwan (J.-D.C.); Department of Radiology, Taipei Veterans General Hospital, Taipei, Taiwan (J.-D.C., C.-M.T., Y.-H.C.); Department of Emergency Medicine, Mackay Memorial Hospital, Taipei, Taiwan (S.-H.W.); Department of Emergency Medicine, Taipei Veterans General Hospital, Taipei, Taiwan (C.-K.H.); and National Yang-Ming University School of Medicine, Taipei, Taiwan (Y.-H.C., J.-D.C., S.-H.W., C.-M.T., C.-K.H., J.-I.L., Y.-H.C.). Received April 25, 2012; accepted July 20, 2012. Address correspondence to: J.-D.C. e-mail: [email protected] ªAUR, 2013 http://dx.doi.org/10.1016/j.acra.2012.07.007

MATERIALS AND METHODS Study Population

This retrospective, noninterventional study was conducted in the ED of a 2700-bed tertiary care medical center located in Taipei, Taiwan, with an ED annual census of approximately 80,000. The institutional review board approved the study. One hundred seventy-three patients (age $18 years) who presented to our ED with right lower quadrant abdominal pain, 73

CHIU ET AL

Academic Radiology, Vol 20, No 1, January 2013

Figure 1. A flow chart of enrolling patients. APP, acute appendicitis; CT, computed tomography; RLQ, right lower quadrant.

clinical suspicion of APP, and subsequently received helical abdominopelvic CT scans between July 2008 and February 2009 were retrospectively included. Both pre- and postintravenous contrast-enhanced CT scans were performed according to this institution’s protocol unless contraindicated. Excluded were those patients who took oral contrast medium (n = 56), lacked intravenous contrast administration (n = 11), or were diagnosed as APP but unproven surgically (n = 6) (Fig 1). The negative appendectomy rate at this institution was 15.8% (14). All patients were followed for more than 6 months. According to the clinical, radiological, surgical, and pathologic results, the diagnosis of each patient was retrospectively classified as either APP or not. All the cases of APP were surgically proven (n = 42). Those without APP were proven either by surgery (n = 9) or by clinical course and radiological findings (n = 41). Uneventful clinical follow-up for minimum of 2 weeks was considered as an acceptable reference standard for the exclusion of APP. The 9 patients with negative CT findings and without specific clinical diagnoses, received no antibiotics and no appendectomy in 6-month follow-ups, and they were excluded from the diagnosis of APP. The main CT features of APP included an abnormal appendix, periappendiceal inflammation, and changes in the cecal apex (5,15–19). Data Collation and CT Analysis

Chart review was performed for the following items of each patient: age, gender, routine blood test values, surgical records, and pathological findings. All these patients received CT scans by a multidetector CT scanner (Philips Brilliance 40, Israel, Tel Aviv) with collimation 32
1.25 mm, rotation time 0.75 seconds, pitch 0.906, matrix: 512
512, KV 120, and mA 250 to 300 depending on the patient’s body size under the use of tube current modulation software (D-DOM, Philips). The noncontrast CT for lower abdomen and postcontrast CT for whole abdomen were performed for all these patients. All the images were reconstructed with slice thickness 5 mm and reconstruction interval 5 mm for axial and coronal display. The visibility of vermiform appendix and specific CT findings of APP, including maximum outer appendiceal diameter, appendiceal wall thickening, appendiceal wall enhancement after intravenous contrast medium, appendicolith, periappendiceal inflammation, and cecal wall thickening, 74

TABLE 1. Etiologic Factors (n = 100) Etiology Patients with acute appendicitis Unruptured appendix Ruptured appendix Patients with non-acute appendicitis Obstetric and gynecologic diseases Pelvic inflammatory disease Tuboovarian abscess Ovarian cyst rupture Ovarian tumor Diverticulitis Negative findings Enterocolitis Acute cholecystitis Mesenteric adenitis Ureteral stone Intestinal obstruction Others Colon carcinoma Hollow organ perforation Pancreatitis Pelvic spindle cell sarcoma Retroperitoneal abscess

n (%) 42 (42.0%) 30 (71.4%) 12 (28.6%) 58 (58.0%) 12 (20.7%) 7 (12.1%) 3 (5.2%) 1 (1.7%) 1 (1.7%) 9 (15.5%) 9 (15.5%) 8 (13.8%) 4 (6.9%) 4 (6.9%) 4 (6.9%) 3 (5.2%) 5 (8.6%) 1 (1.7%) 1 (1.7%) 1 (1.7%) 1 (1.7%) 1 (1.7%)

these findings were recorded separately for noncontrast CT (NCT) images and contrast-enhanced CT (CCT) images by two experienced radiologists (J.D.C., C.M.T.) who did not know the initial CT reports or final diagnoses during imaging review. We also classified the ease of identifying appendix into score 1 (the vermiform appendix could be indentified within 1 minute), and score 2 (longer than 1 minute). Statistical Analysis

Descriptive results were reported as number and percentage for categorical data and mean value  standard deviation (SD) for continuous data when appropriate. The statistic characters, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of CT diagnosis for APP, were compared between NCT and CCT groups by two-sample Z test for proportions. The radiological characters, appendiceal diameter (>8 mm), appendiceal wall thickening, presenting appendicolith, periappendiceal

Academic Radiology, Vol 20, No 1, January 2013

IV CONTRAST CT FOR APPENDICITIS

TABLE 2. Comparison of Radiological Data between NCT and CCT Images NCT Patients with This Sign/No. of Patients Appendiceal diameter (>8 mm) Appendiceal wall thickening Presenting appendicolith Periappendiceal inflammation Cecal wall thickening

CCT Patients with This Sign/No. of Patients y

Odd Ratio (OR)

95% Confidence Interval (CI)

P Value

42/91*

47/94

0.08

0.03–0.24

.34

42/91*

46/94y

0.20

0.10–0.39

.63

22/91* 42/100

19/94y 45/100

0.04 0.06

0.01–0.13 0.02–0.19

.25 .51

45/100

45/100

0.23

0.13–0.42

1.00

*Vermiform appendix could not be identified in 9 cases: 2 in patients with acute appendicitis and 7 in patients without. Vermiform appendix could not be identified in 6 cases: 1 in patients with acute appendicitis and 5 in patients without.

y

Figure 2. A 75-year-old female with acute appendicitis demonstrated on both noncontrast computed tomography (left) and contrast-enhanced computed tomography (right) images. Both images were scored as 1.

inflammation, and cecal wall thickening were compared between the two groups by McNemar’s chi-square test. The ease of identifying appendixes between the two groups was also compared by McNemar’s chi-square test. All statistical analyses were completed with SPSS version 18.0 (SPSS Inc., Chicago, IL), and two-tailed P value less than .05 was considered statistically significant. RESULTS One hundred patients with clinical suspicion of APP met our study criteria and were included for further analysis. The mean age of the patients was 49.3 years (SD  20.54; range 18–90 years). There were 55 men (55.0%) and 45 women (45.0%). Forty-two patients (42.0%) were diagnosed with APP and 58 patients (58.0%) without. The final diagnoses of those patients without APP included the following: obstetric and gynecological diseases in 12 cases (20.7%), negative cases in 9 (15.5%), diverticulitis in 9 (15.5%), enterocolitis in 8 (13.8%), urinary tract diseases in 4 (6.9%), mesenteric adenitis in 4 (6.9%), acute cholecystitis in 4 (6.9%), intestinal obstruction in 3 (5.2%), and others (n = 5). The etiologic factors of the APP and non-APP groups were summarized in Table 1. The comparison of radiological characteristics between the NCT and CCT images were summarized in Table 2. Vermiform appendix could not be identified in nine images of the NCT group and six images of the CCT group. Among the 6 patients, whose appendixes were not made out in both NCT

and CCT images, 1 patient was finally diagnosed as APP, 1 cholecystitis, 1 intestinal obstruction, 1 pancreatitis, 1 pelvic inflammatory disease, and 1 retroperitoneal abscess. In the 3 patients with recognized appendixes only in CCT images, 1 patient was APP, 1 intestinal obstruction, and 1 pelvic inflammatory disease. Mean appendiceal diameter was 8.73 mm (SD  3.66; range, 3.3–19.3 mm) in patients with NCT images, whereas 8.97 mm (SD  3.51; range, 3.4–18.9 mm) in patients with CCT images. There was no statistically significant difference between these two groups regarding the appendiceal diameter >8 mm, appendiceal wall thickening, presence of appendicolith, periappendiceal inflammation, or cecal wall thickening (P > .05) (Figs 2–4). The contrast of statistics with two-sample Z test for proportions between the NCT and CCT groups were shown in Table 3. The visibility of appendix was not significantly different between the two groups (P = .589). There were statistically significant differences of CT diagnosis for APP in sensitivity (P = .036), specificity (P = .038), PPV (P = .021), and NPV (P = .021). The diagnostic accuracy was also not statistically significant different (P = 1.000). With respect to the ease of identifying appendix, 72 cases were classified as score 1 with 28 cases as score 2 in NCT group. In CCT group, 82 cases were categorized as score 1 and 18 cases as score 2. There was statistically significant difference between NCT and CCT images regarding the ease of identifying appendixes by McNemar’s chi-square test (OR = 0.13, 95% CI = 0.04–0.48, P = .013) (Figs 2–5). 75

CHIU ET AL

Academic Radiology, Vol 20, No 1, January 2013

Figure 3. A 51-year-old male with negative finding. Both noncontrast computed tomography (left) and contrast-enhanced computed tomography (right) images were scored as 1.

Figure 4. A 30-year-old female with APP. Non-appendicitis interpreted on noncontrast computed tomography images with score 2 (left) and appendicitis on contrast-enhanced computed tomography images with score 1 (right).

DISCUSSION When evaluating patients with acute abdominal pain, many intraabdominal diseases may mimic APP clinically (16,20,21). In our study, 42 of the 100 patients were diagnosed as APP and 58 patients were with other etiologies. Gastrointestinal tract diseases accounted for 31 (53.4%) of these 58 patients that clinically mimicked APP, including diverticulitis (n = 9), enterocolitis (n = 8), acute cholecystitis (n = 4), mesenteric adenitis (n = 4), intestinal obstruction (n = 3), colon carcinoma (n = 1), hollow organ perforation (n = 1), and pancreatitis (n = 1). Women of child-bearing age had great risk of misdiagnosis, and 12 (20.7%) patients were related to obstetric and gynecologic diseases, such as pelvic inflammatory disease (n = 7), tuboovarian abscess (n = 3), ovarian cyst rupture (n = 1), and ovarian tumor (n = 1). In addition, 4 (6.9%) patients were diagnosed with urinary tract diseases and 9 (15.5%) had no final diagnosis. The etiologies were similar to previous studies (1,16,21). Nowadays, there are various CT protocols for the diagnosis of APP, including different combination of intravenous, oral, or rectal contrast and nonenhanced manners. The optimal techniques and role of CT remain controversial and varied among institutions. Intravenous contrast administration has been demonstrated to aid in diagnosing APP in patients with mild and perforated appendicitis and lacked mesenteric fat (1–3,10,11,22,23). In our study, the overall visibility of appendix (91% vs. 94%) in NCT and CCT groups showed no statistical difference (P > .05). The CT characters of APP also revealed no statistically significant difference 76

TABLE 3. Comparison of Statistic Data between NCT and CCT Images

Visibility Sensitivity Specificity PPV NPV Accuracy

NCT (%)

CCT (%)

Z

P Value

91.0 90.5 100.0 100.0 93.5 96.0

94.0 100.0 94.8 93.3 100.0 97.0

0.54 2.92 2.07 2.31 2.31 0.00

.589 .036 .038 .021 .021 1.000

CCT, contrast-enhanced computed tomography; NCT, noncontrast computed tomography; NPV, negative predictive value; PPV, positive predictive value.

between these two groups in appendiceal diameter, appendiceal wall thickening, presenting appendicolith, periappendiceal inflammation, and cecal wall thickening (P > .05). There were still 3 of 100 our patients with diffusely dilated small-bowel loops, whose appendixes were recognized only in CCT images by additional information of appendiceal wall enhancement. In the past, NCT has been reported with an extremely high diagnosing accuracy (90% to 97%) and sensitivity (87% to 98%) for APP. Some investigators even suggested that contrast administration is not necessary (3,22,24,25). Our results of NCT groups were similar with 90.5% sensitivity, 100% specificity, and 96% accuracy. The sensitivity (100%), specificity (94.8%), and accuracy (97%) of CCT groups in our study were also comparable with the sensitivity (94% to 100%) and specificity (92% to 97%) reported in previous

Academic Radiology, Vol 20, No 1, January 2013

IV CONTRAST CT FOR APPENDICITIS

Figure 5. A 21-year-old male with mesenteric adenitis. Normal appendix was showed on both noncontrast computed tomography images, which were scored as 2 (left) and contrast-enhanced computed tomography images, which were scored as 1 (right).

studies with only IV contrast material (26,27). Although overall the accuracy had no significant difference (P = 1.000) between NCT and CCT groups in our study, the sensitivity of CCT was significantly better than that of NCT (100% vs. 90.5%; P = .036). In patients presenting with abdominal pain suspecting for APP, the ultimate goal of CT imaging is to make a prompt diagnosis and reduce the delay of appendectomy. We think performing CT with intravenous contrast material would be still necessary for patients suspected APP. In addition, the management of patients with suspected APP by emergency physicians includes observation, serial laboratory tests, and diagnostic images. Emergency physicians need to establish a convincing diagnosis before consulting surgeons. Although some investigators showed high accuracy with NCT alone, many emergency physicians or radiologist are not as familiar with or comfortable interpreting nonenhanced images. In our study, the ease of identifying appendixes in CCT group was indeed easier than NCT group (P = .013). In today’s crowded EDs, the difficulty level and consuming time of physicians’ interpretations would influence patients’ length of stay or disposition, and the use of hospital resources (3,9,11). Our study was limited by several factors, such as its retrospective nature and patient selection bias. We only included patients with right lower quadrant abdominal pain and clinical suspicion of APP. Patients with atypical appendiceal positions, such as retrocecal, pericolic gutter, retroileal, or pelvic or retroperitoneal appendicitis, may have different presentations who might not be included into this study (28,29). Because a part of patients did not receive surgical exploration, the final diagnoses may not be true in some cases. Few patients, who were diagnosed as other etiologies of acute right lower abdominal pain and treated with antibiotics, might still have the possibility of getting APP. In addition, the interpretation differences among physicians or the amount and distribution of patients’ fat may also affect the results. However, the aim of our study was to determine the relative agreement and compare the two protocols. In conclusion, the diagnostic sensitivity of CCT showed significantly superiority than that of NCT. Intravenous contrast administration could also increase ease in indentifying appendixes. Performing CT with intravenous contrast material for diagnosing APP would be necessary for adult ED patients, which might improve the care of ED patients by decreasing the time to diagnosis and disposition by promoting

initially appropriate management to reduce the use of hospital resources.

REFERENCES 1. Karam AR, Birjawi GA, Sidani CA, et al. Alternative diagnoses of acute appendicitis on helical CT with intravenous and rectal contrast. Clin Imaging 2007; 31:77–86. 2. Anderson BA, Salem L, Flum DR. A systematic review of whether oral contrast is necessary for the computed tomography diagnosis of appendicitis in adults. Am J Surg 2005; 190:474–478. 3. Tamburrini S, Brunetti A, Brown M, et al. Acute appendicitis: diagnostic value of nonenhanced CT with selective use of contrast in routine clinical settings. Eur Radiol 2007; 17:2055–2061. 4. Zyluk A, Ostrowski P. An analysis of factors influencing accuracy of the diagnosis of acute appendicitis. Pol Przegl Chir 2011; 83:135–143. 5. Pickhardt PJ, Lawrence EM, Pooler BD, et al. Diagnostic performance of multidetector computed tomography for suspected acute appendicitis. Ann Intern Med 2011; 154:789–796. 6. Kim K, Rhee JE, Lee CC, et al. Impact of helical computed tomography in clinically evident appendicitis. Emerg Med J 2008; 25:477–481. 7. Kim K, Lee CC, Song KJ, et al. The impact of helical computed tomography on the negative appendectomy rate: a multi-center comparison. J Emerg Med 2008; 34:3–6. 8. Rao PM, Rhea JT, Novelline RA, et al. Effect of computed tomography of the appendix on treatment of patients and use of hospital resources. N Engl J Med 1998; 338:141–146. 9. Lin KH, Leung WS, Wang CP, et al. Cost analysis of management in acute appendicitis with CT scanning under a hospital global budgeting scheme. Emerg Med J 2008; 25:149–152. 10. Schuur JD, Chu G, Sucov A. Effect of oral contrast for abdominal computed tomography on emergency department length of stay. Emerg Radiol 2010; 17:267–273. 11. Hlibczuk V, Dattaro JA, Jin Z, et al. Diagnostic accuracy of noncontrast computed tomography for appendicitis in adults: a systematic review. Ann Emerg Med 2010; 55:51–59. 12. Bendeck SE, Nino-Murcia M, Berry GJ, et al. Imaging for suspected appendicitis: negative appendectomy and perforation rates. Radiology 2002; 225:131–136. 13. Torbati SS, Guss DA. Impact of helical computed tomography on the outcomes of emergency department patients with suspected appendicitis. Acad Emerg Med 2003; 10:823–829. 14. Lu CL, Liu CC, Fuh JL, et al. Irritable bowel syndrome and negative appendectomy: a prospective multivariable investigation. Gut 2007; 56: 655–660. 15. Moteki T, Horikoshi H. New CT criterion for acute appendicitis: maximum depth of intraluminal appendiceal fluid. AJR Am J Roentgenol 2007; 188: 1313–1319. 16. Rao PM, Mueller PR. Clinical and pathologic variants of appendiceal disease: CT features. AJR Am J Roentgenol 1998; 170:1335–1340. 17. Rao PM, Rhea JT, Novelline RA. Sensitivity and specificity of the individual CT signs of appendicitis: experience with 200 helical appendiceal CT examinations. J Comput Assist Tomogr 1997; 21:686–692. 18. Rao PM, Rhea JT, Novelline RA, et al. Helical CT technique for the diagnosis of appendicitis: prospective evaluation of a focused appendix CT examination. Radiology 1997; 202:139–144. 19. Ives EP, Sung S, McCue P, et al. Independent predictors of acute appendicitis on CT with pathologic correlation. Acad Radiol 2008; 15: 996–1003.

77

CHIU ET AL

20. Shin LK, Jeffrey RB. Sonography and computed tomography of the mimics of appendicitis. Ultrasound Q 2010; 26:201–210. 21. Kruszka PS, Kruszka SJ. Evaluation of acute pelvic pain in women. Am Fam Phys 2010; 82:141–147. 22. Hill BC, Johnson SC, Owens EK, et al. CT scan for suspected acute abdominal process: impact of combinations of IV, oral, and rectal contrast. World J Surg 2010; 34:699–703. 23. Hekimoglu K, Yildirim UM, Karabulut E, et al. Comparison of combined oral and i.v. contrast-enhanced versus single i.v. contrast-enhanced MDCT for the detection of acute appendicitis. JBR-BTR 2011; 94: 278–282. 24. Funaki B. Nonenhanced CT for suspected appendicitis. Radiology 2000; 216:916–918.

78

Academic Radiology, Vol 20, No 1, January 2013

25. Lane MJ, Katz DS, Ross BA, et al. Unenhanced helical CT for suspected acute appendicitis. AJR Am J Roentgenol 1997; 168:405–409. 26. Anderson SW, Soto JA, Lucey BC, et al. Abdominal 64-MDCT for suspected appendicitis: the use of oral and IV contrast material versus IV contrast material only. AJR Am J Roentgenol 2009; 193:1282–1288. 27. Lee KH, Kim YH, Hahn S, et al. Computed tomography diagnosis of acute appendicitis: advantages of reviewing thin-section datasets using sliding slab average intensity projection technique. Invest Radiol 2006; 41:579–585. 28. Cole MA, Maldonado N. Evidence-based management of suspected appendicitis in the emergency department. Emerg Med Pract 2011; 13:1–29. 29. Flum DR, Koepsell T. The clinical and economic correlates of misdiagnosed appendicitis: nationwide analysis. Arch Surg 2002; 137: 799–804.