The diagnosis of appendicitis in children less than 5 years old can be especially challenging [1-3]. Clinical presentations can be confusing and the diagnosis is often delayed or missed [4], resulting in a higher frequency of complications such as appendiceal perforation [5-7]. In addition, many typical features of appendicitis, such as localized right lower quadrant pain, fever, and leukocytosis, may be minimal or absent in younger children [4].

Sonography and CT criteria for appendicitis have been largely defined in adults and in children without further stratification of patients by age [8-10]. Imaging findings in the high-risk group of young children have not received much attention and are not well described. This study was undertaken to compare the imaging characteristics of younger children (< 5 years old) to those of older children and to determine the impact of a younger age on sonography and CT diagnoses and the clinical outcomes of appendiceal perforation and negative appendectomy rate.

This is a retrospective review of all consecutive patients who underwent both preoperative imaging and surgery for suspected acute appendicitis during a 45-month period between January 1, 2000, and September 30, 2003. The setting was a large, urban, pediatric teaching hospital with an emergency department with more than 50,000 annual patient visits and a surgical service that performs 3,800 operations per year. Patients were identified using a hospital database of children who were operated on for appendicitis. The hospital's institutional review board approved this study.

The diagnosis of appendicitis was confirmed pathologically in all cases. However, some patients with a perforated appendix and periappendiceal abscess were first treated medically with antibiotics or by catheter drainage before an interval appendectomy at a later time after the acute process had resolved. In some of these cases, pathology showed that the appendix had already returned to normal. Therefore, in this small group of patients, the diagnosis of appendicitis was based on clinical course, classical CT findings for appendicitis, and clinical outcome (6-month follow-up postdischarge).

Based on a retrospective review of electronic medical records by two of the authors, the following data were obtained: demographic information, types of imaging studies performed, imaging findings on sonography and CT, CT technique used, imaging interpretation, surgical and pathologic findings, and clinical course of those who underwent an interval appendectomy. Only initial interpretations by a board-certified pediatric radiologist or a second-year pediatric radiology fellow were used for this study.

Imaging was considered after initial evaluation by emergency department physicians and subsequent examination by a fourth- or fifth-year surgical resident. Sonography or CT was used as the initial imaging study depending on an institutional clinical practice guideline developed jointly by the departments of emergency medicine, surgery, and radiology. Patients who were imaged with both techniques typically underwent sonography first followed by CT if the initial study was indeterminate or negative for appendicitis. All CT studies were performed on a helical Hi-Speed Advantage single-detector scanner (GE Healthcare). Images were acquired from the iliac crests (bottom of the L3 vertebral body) through the pubic rami using helical mode at a pitch of 1.5:1, with 5-mm collimation, reconstructed at a 3-mm interval. The study was extended to the entire abdomen if any abnormality was detected on the most cephalad images.

During the study, one of three helical CT techniques was used to evaluate clinically suspected appendicitis depending on the patient's characteristics: rectal contrast material only, rectal and IV contrast material, and oral and IV contrast material. The first imaging protocol implemented in our department consisted of a limited abdominal helical CT with rectal contrast material only and was in effect from November 1999 to May 2000 [11]. Our standard protocol was modified to include the use of both rectal and IV contrast material and was in effect from June 1, 2000, to February 1, 2004 [12]. Patients unable to tolerate or retain rectal contrast material were imaged using oral and IV contrast material.

Exclusionary criteria for administration of rectal contrast material were patient age less than 4 years, developmental delay, inability to tolerate or retain rectal contrast material, presence of bloody stools, or preexisting conditions that placed patients at a high risk for intestinal perforation (inflammatory bowel disease, transplant recipients, oncology patients, or children with collagen vascular disorders). These children received oral contrast material 1 or 2 hours before imaging. All eligible patients were given IV contrast material. Patients with a contraindication to IV contrast material administration were imaged with rectal contrast material only.

Pelvic sonograms were performed using a Sequoia sonography unit (Siemens Medical Solutions) equipped with 5.0- and 7.5-MHz linear array transducers and the graded-compression technique.

Our study shows that appendicitis in young children is not rare, occurring in approximately 9% of patients with appendicitis at our institution. Consequently, there is need for a high index of suspicion for appendicitis in this population. Unlike older children, children less than 5 years old can present with nonspecific signs and symptoms, even in the presence of diffuse peritonitis [3, 4]. Other authors have shown that at this young age appendicitis can present with the triad of abdominal pain, tenderness, and vomiting [2]. However, these authors also reported that all cases had a perforated appendix at presentation. A major factor contributing to the difficulty and delay of diagnosis in these children is their inability to communicate and localize disease precisely [1, 4].

In addition, signs and symptoms commonly overlap with those present in other common conditions occurring in this age group, such as viral illnesses, intussusception, gastroenteritis, constipation, and mesenteric adenitis. Often these patients have already been evaluated by a primary care physician and have received treatment before being seen in the emergency department. In this study, the diagnosis of appendicitis was initially missed in four patients less than 5 years old. Each had a confusing and atypical clinical course and required up to three visits to the emergency department before the correct diagnosis was made. As a result of delays in diagnosis, the frequency of appendiceal perforation is higher in younger children [4].

This study confirms this higher risk of perforation in younger children. Our data also show that the risk for perforation appears to be inversely proportional to the age of the patient such that the predicted probability of perforation, based on logistic regression modeling, was almost 60% for a 3-year-old, 50% for a 5-year-old, less than 30% for a 10-year-old, and 10% for a 17-year-old. This higher risk was reflected in the higher frequency of abscesses on both CT and sonography examinations. In addition, echogenic fat on sonography was more commonly seen in younger patients. This finding is thought to be related to edema of the periappendiceal fat and is likely to be due to a higher frequency of transmural inflammation in this group.

Because of the high frequency of appendiceal perforation in younger children, a higher percentage of patients less than 5 years old had an interval appendectomy (13.4%) compared with children 5 years old or above (4.7%). As a result, imaging played an important additional role in this group by mapping the extent of the disease in preparation for planning of imaging-guided abscess treatment.

We found a higher number of false-positive imaging examinations in younger children. Yet, indeterminate studies were not an important clinical problem in this age group. The reasons for these age-related differences remain unclear.

CT was the imaging technique most frequently used, either alone or in combination with sonography for the initial evaluation of children with appendicitis, regardless of patient age. This practice probably relates to the difficulty in excluding early appendicitis by sonography. Lack of visualization of a normal appendix is a known limitation of sonography, and there are large institutional differences. Sivit et al. [18] reported visualization of 49% of normal appendixes using sonography, whereas in our study, a normal appendix could only be shown with sonography in 2% of cases [15]. A recent study by Kaiser et al. [19] shows that the imaging strategy of sonography followed by CT in patients with negative or indeterminate sonography studies had a significantly higher diagnostic accuracy by receiver operating characteristic analysis, compared with sonography alone and clinical evaluation alone.

We were concerned that imaging in younger children would be less accurate for detecting appendicitis than in older children. Our study was reassuring in that the imaging features in younger children are not appreciably different from those in older children. An interesting finding was the lack of indeterminate studies in younger children and the slightly higher frequency of false-positive studies in younger children. We hypothesize that imaging findings in younger children were more likely to be interpreted as definitive because of the higher frequency of more advanced appendicitis and the higher frequency of alternative diagnoses in the younger age group.

The limitations of this study are its retrospective design and a cohort limited to only those children with pathologic proof of appendicitis. Our research design did not allow us to determine the utility of imaging in the larger group of children with suspected appendicitis nor could we determine the relative frequency of young children with suspected appendicitis referred for imaging. We provided specificity values for imaging studies. However, these values can be misleading since they were derived from the small number of false-negative studies performed in the selected patients who underwent a laparotomy.

We conclude that appendicitis should be included in the practical differential diagnosis of young children with abdominal pain and that the imaging findings in this population are the result of a high frequency of appendiceal perforation. More work is needed to develop effective triage mechanisms for better identification of younger patients with appendicitis.

Address correspondence to G. A. Taylor ([email protected]).