Unenhanced Limited CT of the Abdomen in the Diagnosis of Appendicitis in Children: Comparison with Sonography

Transcription

1 Lisa H. Lowe 1 Michael W. Penney 1 Sharon M. Stein 1 Richard M. Heller 1 Wallace W. Neblett 2 Yu Shyr 3 Marta Hernanz-Schulman 1 Received December 16, 1999; accepted after revision May 22, Department of Radiology and Radiological Sciences, Vanderbilt University Children s Hospital and Medical Center, D-1120 Medical Center North, Nashville, TN Address correspondence to M. Hernanz-Schulman. 2 Department of Pediatric Surgery, Vanderbilt University Children s Hospital and Medical Center, Nashville, TN Department of Biostatistics, Vanderbilt University Children s Hospital and Medical Center, Nashville, TN AJR 2001;176: X/01/ American Roentgen Ray Society Unenhanced Limited CT of the Abdomen in the Diagnosis of Appendicitis in Children: Comparison with Sonography OBJECTIVE. The purpose of this investigation is to determine the sensitivity, specificity, and accuracy of unenhanced limited CT of the abdomen in children with suspected appendicitis and compare these results with graded compression sonography. MATERIALS AND METHODS. Seventy-six children underwent unenhanced limited CT over a 11-month period for evaluation of suspected appendicitis. A historical cohort of 86 consecutive children who had undergone graded compression sonography was identified. Results were correlated with surgical, pathologic, chart, and clinical follow-up data. The sensitivity, specificity, accuracy, rate of alternate diagnosis, time to perform examinations, and charge at our institution were determined for unenhanced limited CT and sonography. RESULTS. Sensitivity, specificity, and accuracy for unenhanced limited CT were 97%, 100%, 99%, respectively, and were 100%, 88%, 91%, respectively, for sonography. Alternate diagnoses were suggested in 35% and 28% children without appendicitis who had unenhanced limited CT and sonography, respectively. Unenhanced limited CT required 5 min and sonography required min to perform. The charge at our institution was $408 for unenhanced limited CT and $295 for sonography. CONCLUSION. CT can be performed rapidly in children without IV, oral, or rectal contrast medium. Unenhanced limited CT and sonography are highly sensitive, specific, and accurate in the evaluation of children with suspected appendicitis. A ppendicitis affects 6% of the United States population and is the most common surgical condition of childhood [1]. Most patients are diagnosed clinically by patient history and physical examination. However, an atypical clinical presentation occurs in 30 45% of children and in as many as 45% of young females in whom gynecologic abnormalities are common [1 5]. Preoperative clinical assessment alone has yielded an accuracy ranging between 70% and 90% [4, 6, 7] but is coupled with rates of unnecessary laparotomy of between 15% and 25% [3, 8, 9]. In approximately 20% of patients, a delay in diagnosis and surgical treatment results in increased morbidity including appendiceal perforation, which occurs in 15% to 30% of patients [9 12]. Puylaert [13] introduced sonography as a noninvasive and highly accurate technique in the diagnosis of appendicitis. This initial work led to confirmation of high sensitivity, specificity, and accuracy of sonography, especially in patients with equivocal or atypical clinical findings [14 16]. The sonographic diagnosis of appendicitis can achieve an accuracy of greater than 90%, which in some studies has decreased the rate of negative findings for suspected appendicitis at laparotomy from 25% to 10% [16, 17]. However, it has become evident that the successful implementation of sonography is highly operator-dependent and requires considerable operator experience and expertise [4, 18 20]. In one study, the use of sonography actually decreased diagnostic accuracy from 86% to 77% when compared with surgical consultation [4]. More recently, CT has been extensively evaluated as an adjunct in the accurate diagnosis of patients with suspected appendicitis. These studies have included predominantly adult populations and have been largely retrospective, and the methodology has varied widely, ranging from a full abdominal study with IV and oral contrast material [10, 21 23] to a prospective targeted study evaluating only the lower abdomen with or without the aid of rectally introduced contrast material [12, 24 27]. AJR:176, January

2 Lowe et al. At our institution, we have successfully used sonography for approximately 6 years. More recently, we have used CT as an alternative diagnostic method. To optimize the length of time to perform CT and to approximate the noninvasiveness of sonography, our examinations were performed without IV or enteric contrast material and were targeted to the lower abdomen and pelvis. Our goal in this report is to determine the sensitivity, specificity, and accuracy of unenhanced limited CT in children with suspected appendicitis and to compare and correlate these results with similar sonographic data. Materials and Methods 1 Unenhanced limited CT was performed in 76 consecutive children with suspected appendicitis over a period of 11 months, from November 1997 to September A historical cohort of 86 consecutive children who had undergone graded compression sonography from January 1997 to October 1997 were retrospectively identified from computerized hospital records and served as control subjects. Twelve children were excluded from the study because of our inability to confirm that they had not subsequently developed appendicitis (4 underwent unenhanced limited CT; 8 sonography) leaving 72 children (36 boys and 36 girls) and 78 (35 boys and 43 girls) children in the unenhanced limited CT and sonography study groups, respectively. The children in the unenhanced limited CT group ranged in age from 11 months to 19 years 8 months (mean, 11.5 years; median, 13 years), and those in the sonography group ranged in age from 3 months to 17 years 10 months (mean, 9.3 years; median, 9 years). The only children undergoing unenhanced limited CT and sonography were those with atypical or equivocal clinical history and physical findings. Children were referred by pediatric surgeons, emergency physicians, and attending and community pediatricians. Sonography was performed routinely before November 1997, whereas unenhanced limited CT was the study performed after November To eliminate potential selection bias originating in the type of study requested during the transition period, 10 patients who underwent sonographic examination after October 1997 were not included in the group of historical controls. Unenhanced limited CT examinations were performed without administration of sedation or IV, oral, or rectal contrast material on one of two scanners (Tomoscan AV, Phillips Medical Systems, Shelton, CT; or Somatom Plus, Siemens Medical Systems, Iselin, NJ). The imaging parameters were the following: 5-mm collimation, 200 ma, 120 kvp, and 1:1 table pitch from the inferior edge of the liver to the pubic symphysis. Images collimated to a 3-mm thickness through the region of the appendix were obtained at the discretion of the on-site radiologist (a third-year resident or the attending pediatric radiologist). All final interpretations were reported by the attending pediatric radiologist. Positive diagnostic criteria included an enlarged appendix (>6 mm), periappendiceal inflammation, phlegmon, or abscess (Figs. 1 and 2). An appendicolith alone was not considered diagnostic unless accompanied by other signs of appendiceal enlargement or inflammation. Negative diagnostic criteria included absence of an enlarged appendix, right lower quadrant inflammation, phlegmon, and abscess (Figs. 3 and 4). Visualization of a normal appendix was not considered requisite for a negative diagnosis. Sonograms were obtained using a 5- to 7.5-MHz linear transducer on a sonography unit (SSA 270A; Toshiba America Medical Systems, Tustin, CA) by or in the presence of an attending pediatric radiologist. Positive diagnostic criteria included a noncompressible appendix measuring larger than 6 mm, periappendiceal 2 Fig year-old boy with acute appendicitis. Axial CT image shows enlarged appendix (curved arrow) with surrounding periappendiceal inflammation (straight arrows). Fig year-old boy with perforated appendicitis and phlegmon. Axial CT image reveals inflammatory mass (open arrows) in right lower quadrant containing appendicolith (solid arrow). Fig year-old girl with normal appendix. Axial CT image shows retrocecal appendix (arrow) without evidence of inflammation to suggest appendicitis. 3 4 Fig year-old girl with normal appendix. Axial CT image shows normal appendix (arrow) is filled with air. 32 AJR:176, January 2001

3 CT Versus Sonography for Appendicitis Diagnosis Fig year-old boy with abdominal pain and appendicitis. Axial CT image shows appendicolith (arrow) in right lower quadrant. Periappendiceal inflammation was not appreciated because of paucity of fat in this patient. hyperechogenicity indicating adjacent inflammation, the presence of a shadowing appendicolith, and right lower quadrant phlegmon or abscess. Hyperemia of the appendix when the appendix was borderline in size ( mm) was used as an adjunctive criterion to increase confidence in a positive diagnosis. Negative diagnostic criteria included a compressible right lower quadrant without an enlarged appendix, right lower quadrant inflammation, phlegmon, or abscess. The original reports of all studies were used in data tabulation because our purpose is to evaluate the accuracy of these studies at the time of clinical impact. All unenhanced limited CT examinations were interpreted by an attending pediatric radiologist, one of the authors; all sonographic examinations were performed or supervised and interpreted by a pediatric radiologist. Collection and correlation of CT and sonography reports with follow-up data were performed by one of the authors who was unaware of the study results. Positive results of unenhanced limited CT and sonographic examinations were confirmed by surgical reports, pathologic data, and medical records. Negative imaging results for appendicitis were confirmed surgically and pathologically or by medical record review, clinical follow-up, and telephone follow-up at 6 months or more from the time of imaging. The sensitivity, specificity, and accuracy of unenhanced limited CT and sonography were calculated and subjected to statistical analysis. Significance was determined between the sonography and unenhanced limited CT groups by evaluating the 95% confidence interval (CI), which was calculated using binomial proportions for the difference between experimental and control groups. The rate of alternate diagnoses, the time required to perform each examination, and the charge for unenhanced limited CT and sonography were compared. To determine the significance between alternate diagnoses in the two study groups, data were analyzed using the Fisher s exact test for categoric variables. All tests of significance were twosided, and differences were considered statistically significant when the p value was less than Statistical software (version 7.0; SAS Institute, Cary, NC) was used for all analyses. The institutional review board at our institution approved this study. Results Among the 72 patients who underwent unenhanced limited CT examination (n = 72), 35 patients had true-positive findings; 36, true-negative findings; zero, false-positive findings; and one, false-negative finding. In the child with the false-negative finding, an appendicolith was seen surrounded by fluid-filled loops of small bowel, but neither the appendix nor inflammation of mesenteric fat was seen (Fig. 5). The diagnosis of appendicitis was suspected but was not definitively made at the initial report, and so this case was classified as a false-negative. Among the 78 patients who underwent sonographic examination, 20 patients had truepositive findings; 51, true-negative findings; seven, false-positive findings; and zero, falsenegative findings. The seven cases of falsepositive diagnoses are described as follows. Two cases involved children in whom an enlarged noncompressible tubular structure that was believed to be an appendix (8.0 and 7.4 mm) was detected. Two cases involved children who had an enlarged hyperemic appendix (7.0 and 8.0 mm). One patient had an appendix that was enlarged (6.2 mm) and associated with free fluid. In another patient, an enlarged appendix (8.0 mm) was associated with a right lower quadrant abscess, the cause of which was not determined by surgical or pathologic evaluation. The seventh case of a false-positive diagnosis involved a child in whom the appendix was not visualized but who had a right lower quadrant abscess. At surgery, a normal appendix was seen and the abscess was related to the adjacent small bowel. Pathology measurements were available by report in four patients. Two appendices that measured sonographically as 8.0 mm were described pathologically as 7.0 and 8.0 mm; a third appendix measured sonographically as 6.2 mm was described pathologically as 6 mm. The fourth appendix, which was in the patient with the right lower quadrant abscess, was not visualized sonographically and was described pathologically as 6.0 mm. The sensitivity, specificity, and accuracy of unenhanced limited CT (including 95% CI) were 97% (95% CI, %), 100%, and 99% (95% CI, %), respectively. Sonography had a sensitivity, specificity, and accuracy of 100%, 88% (95% CI, 80 96%), and 91% (95% CI, 85 97%), respectively. The difference between the sensitivity, specificity, and accuracy of unenhanced limited CT and sonography was not significant (p > 0.05). Excluding patients with the correct diagnosis by imaging with appendicitis (true-positive cases), unenhanced limited CT and sonography suggested an alternative diagnosis in 13 (35%) of 37 and in 16 (28%) of 58 patients, respectively (Table 1). This difference was not statistically significant (p > 0.05). Unenhanced limited CT required 5 min to perform, and sonography required min once scanning was initiated. At our institution, there was no delay in performing the unenhanced limited CT examination, whereas sonographic studies after routine working hours were delayed because of the need to call in the attending radiologist given the high operator-dependence of this modality. At our institution, the charge for unenhanced limited CT is $408 and that of sonography is $295. Discussion This study shows that the sensitivity, specificity, and accuracy of unenhanced limited CT are excellent in the evaluation of children with suspected appendicitis. Unenhanced limited CT in pediatric patients compares favorably with sonography and with previous reports of routine contrast-enhanced CT in patients of all ages, in which sensitivities have consistently ranged between 87% and 100% [3, 10, 11, 21, 24 28]. Our limited CT protocol did not include contrast material administration for several reasons. First, the addition of oral contrast material may have required placement of an orogastric tube in some children and would have delayed the time to diagnosis, because the ingested contrast material needs time to travel to the right lower quadrant. In some patients, passage of oral contrast into the colon may require several hours, and the length of time is unpredictable. Second, the addition of IV contrast material would have added an invasive component to the examina- AJR:176, January

4 Lowe et al. TABLE 1 Location and Diagnosis Alternate Diagnoses in Children Suspected of Having Appendicitis Unenhanced Limited CT (n = 13) a Diagnosed as bowel wall thickening of greater than 3 mm. tion and would have delayed the study because of placement of an IV line, which is often preceded by application of topical analgesic creams in children. Finally, contrast material administered rectally would have introduced additional delay, and its implementation in some children would have necessitated the use of restraining devices or the use of sedation. Because the colon and the cecum are usually readily identifiable on CT by location, morphology, and contents, we chose not to administer enteric contrast material to our patients. The results of our series show that in our patients, this additional step was unnecessary. The successful application of sonography to diagnose appendicitis is highly operator-dependent and requires considerable training and experience [4, 18, 19]. This point is emphasized by the wide variability of reported sonographic diagnostic sensitivities ranging from 45% to 100% [4, 13, 14, 17, 19, 27, 29]. In our series, the sensitivity of sonography was 100% probably because all scans were obtained by or with the assistance of the attending pediatric radiologist. Most examinations, especially those in patients without appendicitis, required a great deal of time and patience necessary to calm an apprehensive or uncooperative child with tense abdominal musculature, which can preclude sustained application of graded compression. Similar to the experience of others, our level of comfort in a negative diagnosis on sonography was considerably less when compared with our level of comfort in a negative diagnosis on CT because a normal appendix is much more difficult to visualize confidently on sonography (<5%) than on CT (50 100%) [3, 11, 30]. Location and Diagnosis Sonography (n = 16) Urinary Tract 4 Urinary tract 1 Renal or ureteral stone 2 Hydronephrosis 1 Pyelonephritis 1 Hydronephrosis 1 Gastrointestinal tract 4 Gastrointestinal tract 8 Mesenteric nodes 2 Mesenteric nodes 4 Gastroenteritis a 2 Gastroenteritis a 1 Pancreatitis 1 Cholecystitis 1 Constipation 1 Gynecologic system 5 Gynecologic system 7 Ovarian cyst 4 Ovarian cyst 6 Ovarian dermoid 1 Ovarian dermoid 1 It is interesting that in six of seven sonograms with false-positive findings, the appendix was larger than 6.0 mm, with five of six being 7.0 mm or larger. Thus, six (10.3%) of 58 children who did not have appendicitis had an appendix that was larger than 6.0 mm, which is the accepted current upper limit of normal. In two patients, the appendix measured 8.0 mm on sonography and was confirmed pathologically to be 7.0 and 8.0 mm. An alternate diagnosis was made in 13 (18%) of 72 patients on unenhanced limited CT and in 16 (21%) of 78 patients on sonography. The rates of alternate diagnoses made using unenhanced limited CT and sonography are not significantly different and are comparable with those cited in the literature [14, 21, 24, 26, 31]. Excluding cases correctly diagnosed as appendicitis (true-positive cases) using imaging, unenhanced limited CT, and sonography suggested an alternate diagnosis in 13 (35%) of 37 and 16 (28%) of 58 cases, respectively. When the alternate diagnoses were divided by organ system, unenhanced limited CT detected urinary tract disease in four (31%) of 13 patients, which was more than the one (6%) of 16 cases detected with sonography. This difference was not statistically significant (p > 0.05). Urolithiasis was diagnosed in two (15%) of 13 patients undergoing unenhanced limited CT, and in none of the patients undergoing sonography. Unenhanced limited CT suggested a gynecologic or gastrointestinal alternate diagnosis in five (38%) and four (31%) of 13 children, respectively; at sonography, seven (44%) and eight (50%) of 16 children had an alternate diagnosis that was gynecologic or gastrointestinal, respectively. There was no statistical difference between unenhanced limited CT and sonography for the detection of gynecologic and gastrointestinal alternate diagnoses (p > 0.05). The ionizing radiation exposure involved in CT versus sonography is an obvious and valid concern. Patients who undergo unenhanced limited CT do not need to undergo abdominal radiography because CT is a radiographic modality for abdominal imaging, and we believed abdominal radiography is unnecessary. The collective effective dose equivalent from an abdominal radiograph is 0.56 msv, and it is 1.11 msv for a full CT scan of the abdomen [32, 33]. In these cases, as in all diagnostic studies, any increase in radiation exposure needs to be weighed in the balance of diagnostic accuracy, rapidity of study performance, and diagnostic confidence level. Although the charge for unenhanced limited CT versus sonography may vary among institutions and geographic locations, at our institution the charge for unenhanced limited CT ($408) is more than that of sonography ($295). Despite the fact that charge and cost are not equivalent, the cost of accurate imaging is minor when juxtaposed to the cost of unnecessary laparotomy, delayed surgery in positive cases, and delayed identification of other conditions requiring treatment. This has been shown in one series of 100 patients in whom accurate exclusion of appendicitis and diagnosis of appendicitis or other conditions that required treatment resulted in a net hospital cost savings of $44,731 [12]. The goal of imaging in the evaluation of patients with suspected appendicitis is to provide a sensible balance between obtaining the highest possible diagnostic accuracy while considering the degree of invasiveness, radiation dose, patient discomfort, and the factors of time and cost. Our study shows that both unenhanced limited CT and sonography are accurate imaging modalities in children with suspected appendicitis. Because of its lack of associated radiation exposure, sonography performed by a radiologist with the appropriate experience and expertise is highly desirable. The choice of which type of study to perform is likely to depend on the available resources and personnel at various institutions and the clinical history of the patient, physical examination findings, laboratory data, and differential diagnostic possibilities. Although absolute statements regarding the most appropriate modality in 34 AJR:176, January 2001

5 CT Versus Sonography for Appendicitis Diagnosis individual patients are inadvisable, our data suggest that the expediency of unenhanced limited CT and its sensitivity, specificity, accuracy, lack of operator-dependence, and lack of invasiveness render it a valuable tool in the evaluation of children with a clinically equivocal diagnosis of appendicitis. In conclusion, unenhanced limited CT can be performed rapidly and is able to achieve equal accuracy to full abdominal CT with IV and enteric contrast material. Unenhanced limited CT and sonography of the abdomen are highly sensitive, specific, and accurate in the evaluation of children with suspected appendicitis. References 1. Federle MP. Focused appendix CT technique: a commentary. Radiology 1997;202: Abu-Yousef MM, Bleicher JJ, Maher JW, Urdaneta LF, Franken EA Jr, Metcalf AM. High-resolution sonography of acute appendicitis. AJR 1987;149: Birnbaum BA, Jeffrey RB Jr. CT and sonographic evaluation of acute right lower quadrant abdominal pain. AJR 1998;170: Ford RD, Passinault WJ, Morse ME. Diagnostic ultrasound for suspected appendicitis: does the added cost produce a better outcome? Am Surgeon 1994;60: Siegel MJ. Acute appendicitis in childhood: the role of US. Radiology 1992;185: Wade CDRDS, Morrow SE, Balsara ZN, Burkhard TK, Goff WB. Accuracy of ultrasound in the diagnosis of acute appendicitis compared with surgeon s clinical impression. Arch Surg 1993;128: Zielke A, Hasse C, Sitter H, Rothmund M. Influence of ultrasound on clinical decision making in acute appendicitis: a prospective study. Eur J Surg 1998;164: Dunn EL, Moore EE, Elerding SC, Murphy JR. The unnecessary laparotomy rate for appendicitis: can it be decreased? Am Surg 1982;48: Skaane P, Amland PF, Nordshus T, Solheim K. Ultrasonography in patients with suspected acute appendicitis: a prospective study. Br J Radiol 1990;63: Balthazar EJ, Rofsky NM, Zucker R. Appendicitis: the impact of computed tomography imaging on negative appendectomy and perforation rates. Am J Gastroenterology 1998;3: 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: Rao PM, Rhea JT, Novelline RA, Mostafavi AA, McCabe CJ. Effect of computed tomography of the appendix on treatment of patients and use of hospital resources. N Engl J Med 1998;338: Puylaert JBCM. Acute appendicitis: US evaluation using graded compression. Radiology 1986;158: Sivit CJ, Newman KD, Boenning DA, et al. Appendicitis: usefulness of US in diagnosis in a pediatric population. Radiology 1992;185: Rossi P, Covarelli P, Mosci F, Bisacci R, Sensi B, Moggi L. Ultrasonography in the management of acute appendicitis. Surg Endosc 1996;10: Ramachandran P, Sivit CJ, Newman KD, Schwartz MZ. Ultrasonography as an adjunct in the diagnosis of acute appendicitis: a 4-year experience. J Pediatr Surg 1996;31: Garcia Pena BM, Mandl KD, Kraus SJ, et al. Ultrasonography and limited computed tomography in the diagnosis and management of appendicitis in children. JAMA 1999;282: Skaane P, Schistad O, Amland PF, Solheim K. Routine ultrasonography in the diagnosis of acute appendicitis: a valuable tool in daily practice? Am Surg 1997;63: Pohl D, Golub R, Schwartz GE, Stein HD. Appendiceal ultrasonography performed by nonradiologists: does it help in the diagnostic process? J Ultrasound Med 1998;17: Zielke A, Hasse C, Sitter H, Kisker O, Rothmund M. Surgical ultrasound in suspected acute appendicitis. Surg Endosc 1997;11: Balthazar EJ, Birnbaum BA, Yee J, Megibow AJ, Roshkow J, Gray C. Acute appendicitis: CT and US correlation in 100 patients. Radiology 1994;190: Friedland JA, Siegel MJ. CT appearance of acute appendicitis in childhood. AJR 1997;168: Jabra AA, Shalaby-Rana EI, Fishman EK. CT of appendicitis in children. J Comput Assist Tomogr 1997;21: Balthazar EJ, Megibow AJ, Siegel SE, Birnbaum BA. Appendicitis: prospective evaluation with high-resolution CT. Radiology 1991;180: Malone AJ, Wolf CR, Almed AS, Melliere BF. Diagnosis of acute appendicitis: value of unenhanced CT. AJR 1993;160: Funaki B, Grosskreutz SR, Funaki CN. Using unenhanced helical CT with enteric contrast material for suspected appendicitis in patients treated at a community hospital. AJR 1998;171: Lane MJ, Liu DM, Huynh MD, Jeffrey RB Jr, Mindelzun RE, Katz DS. Suspected acute appendicitis: nonenhanced helical CT in 300 consecutive patients. Radiology 1999;213: Rhea JT, Rao PM, Novelline RA, McCabe CJ. A focused appendiceal CT technique to reduce the cost of caring for patients with clinically suspected appendicitis. AJR 1997;169: Jahn H, Mathiesen FK, Neckelmann K, Hovendal CP, Bellstrom T, Gottrup F. Comparison of clinical judgment and diagnostic ultrasonography in the diagnosis of acute appendicitis: experience with a score-aided diagnosis. Eur J Surg 1997;163: Incesu L. Diagnosis of acute appendicitis in children. AJR 1995;165: Jain KA, Quam JP, Ablin DS, Gerscovich EO, Shelton DK. Imaging findings in patients with right lower quadrant pain: alternative diagnoses to appendicitis. J Comput Assist Tomogr 1997;21: National Council on Radiation Protection and Measurements. Exposure of the U.S. population from diagnostic medical radiation. 1989:report no. 100: Rothenberg LN, Pentlow KS. Radiation dose in CT. RadioGraphics 1992;12: AJR:176, January