Acute Abdomen in Pediatric Patients Admitted to the Pediatric Emergency Department

Transcription

1 Pediatr Neonatol 2008;49(4): ORIGINAL ARTICLE Acute Abdomen in Pediatric Patients Admitted to the Pediatric Emergency Department Yu-Ching Tseng 1,2, Ming-Sheng Lee 2, Yu-Jun Chang 3, Han-Ping Wu 4,5 * 1 Division of Pediatric Emergency Medicine, Changhua Christian Hospital, Changhua, Taiwan 2 Department of Pediatrics, Changhua Christian Hospital, Changhua, Taiwan 3 Laboratory of Epidemiology and Biostatistics, Changhua Christian Hospital, Changhua, Taiwan 4 Department of Pediatrics, Buddhist Tzu Chi General Hospital, Taichung Branch, Taichung, Taiwan 5 Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan Received: Dec 14, 2007 Revised: May 9, 2008 Accepted: Aug 28, 2008 KEY WORDS: abdominal pain; abdomen, acute; child; emergency service, hospital Background: Acute abdomen in children is a serious condition frequently encountered in the pediatric emergency department (ED). This study aimed to analyze the clinical spectrum of acute abdomen, and to investigate the prevalence of various etiologies in different age groups of children admitted to the pediatric ED. Methods: From 2005 to 2007, we retrospectively recruited 3980 consecutive pediatric patients who presented to the pediatric ED suffering from acute abdominal pain. Of these patients, 400 were identified as having acute abdomen. These patients were then divided into traumatic and non-traumatic groups, and also divided into four age groups: infant, preschool-age, school-age and adolescent. Differences between the traumatic and non-traumatic groups in the prevalence, clinical presentations, laboratory and imaging findings, and hospital courses were analyzed statistically. Results: In the non-traumatic group (n = 335), the most common etiology in infants was incarcerated inguinal hernia (14/31, 45.1%), followed by intussusception (13/31, 41.9%), while acute appendicitis was the major cause in children older than 1 year (68.7%). In the traumatic group (n = 65), the major cause of acute abdomen was traffic accidents (76.9%). The liver was the most frequently injured organ, followed by the spleen. The mortality rate was highest in patients with multiorgan injury. In both groups, bowel loop dilation and local ileus were the two most common findings demonstrated by plain film X-rays. Children in the traumatic group who underwent abdominal computed tomography (CT) scans all showed positive findings for their diagnoses. Patients with bowel perforation or obstruction had the longest durations of hospitalization in the non-traumatic group, while those with multiorgan injury had the longest duration in the traumatic group. Conclusion: The etiology of acute abdomen varied depending on the age of the patient. Acute appendicitis was the most common cause of acute abdomen in children older than 1 year of age, followed by traumatic injury. Abdominal CT scanning was a useful diagnostic imaging modality in patients with both traumatic and nontraumatic abdominal pain. *Corresponding author. Department of Pediatrics, Buddhist Tzu Chi General Hospital, Taichung Branch, No. 66, Section 1, Fongsing Road, Tanzih Township, Taichung County 42743, Taiwan. arthur1226@gmail.com 2008 Taiwan Pediatric Association

2 Acute abdomen in children and adolescents Introduction Acute abdominal pain is one of the most common problems in children admitted to the pediatric emergency department (ED), and often presents a diagnostic dilemma for primary clinicians. Acute abdominal pain in patients presenting to the ED is often diagnosed as a disorder that does not require surgical intervention, such as acute gastroenteritis, functional digestive disorders or constipation. However, abdominal emergencies requiring surgery must be picked up by primary pediatric physicians in the pediatric ED because of their potentially life-threatening risks. The symptoms and signs that suggest acute abdomen include rebounding pain, involuntary guarding or rigidity, abdominal distention and diffuse tenderness. 1 However, acute abdomen may not be easily diagnosed in young children based on these clinical presentations because of their poor ability to express themselves. Abdominal surgical emergencies can also be induced by trauma. Internal organ involvement resulting from traumatic insults in children can differ markedly from those in adults. Moreover, acute abdomen in children of different ages may have different etiologies and clinical courses, and may produce different laboratory test results and imaging findings. Different etiologies of acute abdomen can show different distributions in different months, but the reasons for this remain unclear. In the present study, we analyzed the etiologies, laboratory test results, imaging findings and clinical courses of pediatric patients with acute abdomen. We also analyzed the prevalence of various etiologies of acute abdomen in different age groups and in different months, in children admitted to the pediatric ED. 2. Patients and Methods 2.1. Patient population From July 2005 to June 2007, we retrospectively re cruited 3980 consecutive patients aged younger than 18 years who presented to the ED at the Changhua Christian Hospital, Taiwan, with acute abdominal pain. Among the 3980 patients, 400 children were identified as having been diagnosed with acute abdomen. These 400 patients were divided into two groups: a non-traumatic group and a traumatic group. The non-traumatic group comprised patients with acute abdomen unrelated to trauma but requiring surgical intervention. The traumatic group comprised patients who presented to the ED with acute abdominal pain caused by traumatic injury Methods Demographic information on the patients was obtained from hospital chart records. The patients ages, genders, chief complaints, and clinical symptoms and signs were recorded on our data sheet. We also noted the etiologies of acute abdomen, hospital duration, laboratory test results (white blood cell [WBC] count, C-reactive protein [CRP], total neutrophil count, and manual band count), imaging findings (abdominal radiography, computed tomography [CT] and ultrasonography), the time of admission, and their condition at discharge. In the traumatic group, the location of the trauma, and additional laboratory test results, including hemoglobin (Hb), amylase, lipase, and aspartate aminotransferase/ alanine aminotransferase, as well as the type of trauma, were also recorded. All patients in our study were divided into four age groups: infant (1 month 1 year), preschool-age (2 6 years), school-age (7 12 years) and adolescent (13 18 years). The etiologies of acute abdomen in the non-traumatic group were divided into six major categories: (1) acute appendicitis, (2) hollow organ perforation, (3) incarcerated inguinal hernia, (4) intussusception, (5) intestinal obstruction other than due to the diseases described above, and (6) ovarian torsion. In the traumatic group, the locations of the injuries were divided into six categories: (1) liver, (2) spleen, (3) kidney, (4) pancreas, (5) intestine/colon, and (6) multiple organs (more than one organ involved) Statistical analysis Differences in the prevalence, clinical symptoms and signs, laboratory test results, imaging findings, hospital courses and the lengths of hospitalization between the traumatic and the non-traumatic groups were analyzed by Fisher s exact test and the Kruskal Wallis test. A p value of < 0.05 was considered significant. 3. Results Acute abdomen had a non-traumatic etiology in 335 of the 400 patients, and was due to trauma in 65 patients. There were 270 boys and 130 girls who ranged in age from 1 month to 18 years (mean age, 10.6 ± 5.7 years). The case distributions in all patients and in the traumatic group are shown in Figure 1. The demographics, clinical presentations, laboratory test results, imaging findings and hospital courses in children with acute abdomen are presented in Table 1.

3 128 Y.C. Tseng et al A B Appendicitis 64.0% Trauma 16.3% Perforation 3.5% Hernia, incarcerated 7.5% Torsion 1.3% Intussusception 6.3% Intestinal obstruction 1.3% Multiple organs 44% Intestine & colon 8% Liver 22% Spleen 18% Kidney 5% Pancreas 3% Figure 1 Case distribution of patients with (A) acute abdomen, and (B) distribution of the different locations of organ injuries involved in the traumatic group. Percentage Month Appendicitis (n = 256) Perforation (n = 14) Hernia, incarcerated (n = 30) Intussusception (n = 25) Intestinal obstruction (n = 5) Torsion (n = 5) Trauma (n = 65) Figure 2 Case distribution based on different months in our study period Demographics The most common etiology of acute abdomen in infants was incarcerated inguinal hernia (14/31, 45.1%), followed by intussusception (13/31, 41.9%). These etiologies were uncommon in school-age and adolescent children. In contrast, acute appendicitis was the major cause of surgical abdomen in children older than 1 year in the non-traumatic group (68.7%). Based on the pathology reports of resected appendices from appendectomies, 15.6% (40/256) of patients with appendicitis presented with early appendicitis, 64.1% (164/256) with suppurative or gangrenous changes, and 20.3% (52/256) with perforated appendicitis. All cases of ovarian torsion, which is a gynecological abdominal emergency, were diagnosed in children of more than 12 years of age. In addition, 21 cases of incarcerated inguinal hernia were right-sided (70%), seven cases were left-sided (23.4%), and two cases involved both sides (6.6%). Incarcerated inguinal hernias were most prevalent in preschool-aged children (50%) and infants (46.7%), and least prevalent in adolescents (3.3%). During the study period, 95 pediatric patients were diagnosed with intussusception, including 91 cases with ileo-colic type, three with ileo-ileo-colic type and one with ileo-cecal-colic type. Surgical reductions were performed in 21 cases of ileo-colic type intussusception and in all cases of ileo-ileocolic and ileo-cecal-colic type intussusception. Intestinal obstruction was noted in five patients in the non-traumatic group. The etiologies were intraabdominal abscess (n = 2), foreign body ingestion (olive) (n = 1), patent omphalomesenteric duct with malrotation (n = 1) and adhesion ileus (n = 1); perforated appendicitis was excluded in the causes of intra-abdominal abscesses. The first case of intra-abdominal abscess was caused by perforation following acute gastroenteritis, and the abscess culture showed Bacteroides ovatus and Escherichia coli coinfection. The culture results from the other case of intra-abdominal abscess revealed Enterobacter cloacae, E. coli, and methicillin-resistant Staphylococcus aureus. In the case of foreign body ingestion, the location of the olive obstruction was 50 cm away from the ileocecal valve. The case of adhesion ileus was due to a previous surgical complication (laparoscopic appendectomy), and was unresponsive to decompression treatment. Figure 2 shows the monthly distribution of non-traumatic cases. Analysis of the monthly distribution of cases revealed that the peak of appendicitis occurred in September, while the peak of intussusception occurred in February. Most patients in the traumatic group were adolescents (n = 41, 63.1%). The characteristics of the

4 Acute abdomen in children and adolescents 129 Table 1 Demographic data, clinical presentations, laboratory test results, imaging findings, and hospital courses in patients with acute abdomen* Appendicitis Perforation Intussusception incarcerated Torsion obstruction Trauma p Hernia, Intestinal Age < Infant 1 (0.4) 1 (7.1) 14 (46.7) 13 (52.0) 2 (40.0) 0 0 Preschool-age 28 (10.9) 6 (42.9) 15 (50.0) 11 (44.0) 2 (40.0) 0 12 (18.5) School-age 113 (44.1) 2 (14.3) 0 1 (4.0) 1 (20.0) 0 12 (18.5) Adolescent 114 (44.5) 5 (35.7) 1 (3.3) (100) 41 (63.1) Gender Male 165 (64.5) 13 (92.9) 25 (83.3) 17 (68) 5 (100.0) 0 45 (69.2) Female 91 (35.5) 1 (7.1) 5 (16.7) 8 (32) 0 5 (100) 20 (30.8) Clinical symptoms and signs Fever 147 (57.4) 8 (57.1) 0 6 (24) 2 (40) Vomiting 167 (65.2) 9 (64.3) 1 (3.3) 8 (32) 4 (80) 2 (40) 1 (1.5) Diarrhea 56 (21.8) 9 (64.3) 0 2 (8) 3 (60) Abdominal fullness 5 (1.9) 5 (35.7) 1 (3.3) 1 (4) 1 (20) 0 2 (3.0) RLQ tenderness 253 (98.8) 1 (7.1) (60) 0 < Diffuse 3 (1.2) 13 (92.9) 0 1 (4) 3 (60) 0 Intermittent pain (88) 0 2 (40) 0 Rebounding pain 246 (96) 14 (100) (40) 1 (20) Muscle guarding 9 (3.5) 4 (28.5) Imaging studies Abdominal X-ray 243 (94.9) 14 (100) 13 (43.3) 20 (80) 5 (100) 1 (20) 12 (18.5) < Negative finding 101 (41.6) 1 (7.1) 8 (62.5) 5 (25) 0 1 (100) 8 (66.7) < Bowel dilatation 55 (22.6) 0 3 (23.1) 7 (35) (8.3) Local ileus 31 (12.8) 1 (7.1) 1 (7.7) 4 (20) 3 (60) 0 2 (16.7) Air-fluid level 1 (0.4) 0 1 (7.7) 0 1 (20) 0 0 Fecalith 54 (22.2) 1 (7.1) Free air 1 (0.4) 11 (78.6) (20) 0 1 (8.3) Crescent sign (20) Ultrasonography 56 (21.9) 3 (21.4) (60) 4 (80) 40 (61.5) < Accurate diagnosis 41 (73.2) 2 (66.7) 25 (100) 2 (66.7) 4 (100) 31 (77.5) Abdominal CT 61 (23.8) 4 (28.6) 0 2 (8) 3 (60) 0 64 (98.5) < Accurate diagnosis 60 (98.4) 4 (100) 2 (100) 3 (100) 64 (100) Laboratory tests WBC ( 10 3 /mm 3 ) 15.2 ± ± ± ± ± ± ± Neutrophil count (%) 81.0 ± ± ± ± ± ± ± 15.4 < Band count (%) 8.5 ± ± ± ± ± ± CRP (mg/dl) 7.81 ± ± ± ± ± ± ± ICU admission 4 (1.6) 2 (14.3) (66.2) ICU admission days 3.5 ± ± ± Hospitalization days 4.9 ± ± ± ± ± ± ± Mortality (4.6) *Data are presented as n (%) or mean ± standard deviation. RLQ = right lower quadrant; CT = computed tomography; WBC = white blood cell count; CRP = C-reactive protein; ICU = intensive care unit.

5 130 Y.C. Tseng et al injuries and the organs involved are presented in Table 2. Traumatic injuries involved more than one organ in 44% of these patients. The liver was the organ most frequently involved in abdominal trauma (21.5%), followed by spleen laceration (18.5%). The most common cause of injury was traffic accidents (76.9%), and the least common was child abuse (1.5%) Clinical presentations and laboratory tests The clinical presentations and laboratory data are shown in Table 1. Patients with appendicitis commonly presented with rebounding pain (n = 246, 96%), vomiting (n = 167, 65.2%), fever (n = 147, 57.4%), and diarrhea (n = 56, 21.8%). All patients with intestinal perforation presented with rebounding pain. In laboratory tests, WBC counts, total neutrophil counts and CRP levels showed significant differences among children with different etiologies of acute abdomen (p < 0.01). Mean WBC counts, neutrophil counts and CRP levels were all highly elevated in patients with acute appendicitis and intestinal perforation. However, CRP levels and band forms in the differential counts were relatively low in patients with incarcerated hernia and intussusception. The characteristics of patients in the traumatic group are shown in Table 2. There were no significant difference in mean Hb values and WBC counts among the groups with different organ injuries Imaging studies In our study, plain radiographs were available for 77% (308/400) of the patients. Bowel loop dilation and local ileus were the two most common findings on plain radiographs in children with acute abdomen. Fecaliths accounted for only 22.2% of patients with appendicitis. After excluding patients with perforations due to appendicitis, there were 14 patients with intestinal perforations. The locations of the perforations included the gastrium (n = 4), the small bowel (n = 1), the duodenum (n = 2) and the cecum (n = 7). In 78.6% (11/14) of these patients, free air was demonstrated on abdominal radiographs (Figure 3A). In addition, abdominal CT scans were taken in 33.5% (134/400) of the patients in our study. Most of the traumatic patients (98%) received abdominal CT examinations, and the definitive diagnoses in all of these patients were based on the CT findings. Moreover, 32% (131/400) of the patients underwent ultrasound examinations. In our series, the patients with intussusception were all diagnosed using bedside abdominal sonogram (100%) (Figure 3B), and two of them also underwent abdominal CT examinations in order to exclude pathologic leading points, because of recurrent episodes Hospital course and outcome In the non-traumatic group, patients with bowel perforation and obstruction had the longest durations of hospitalization, while those with incarcerated inguinal hernia had the shortest duration. Six patients in this group were admitted to the pediatric intensive care unit (PICU) because of perforated appendicitis (n = 4) or intestinal perforation (n = 2). The average length of stay in the PICU for patients with perforated appendicitis was 4.9 ± 3.3 days, and that for patients with intestinal perforation was 4.5 ± 0.7 days. There were no mortalities in this group. In the traumatic group, three patients died; two of them had injuries involving multiple organs and one had intestinal perforation. Patients with multiple organ injuries and those with liver lacerations had the longest durations of hospitalization, and patients with spleen lacerations had the shortest duration of hospitalization and ICU stay. 4. Discussion Abdominal pain is one of the most common complaints in childhood. 2 Although most of these complaints arise from self-limiting conditions, abdominal pain might herald a surgical or medical emergency. 3,4 This retrospective study aimed to help primary pediatricians to better understand the clinical spectrum of acute abdomen in children, including the demographics, clinical presentations, laboratory test results, imaging findings, and clinical courses. This research also provides detailed information on the epidemiologic variations of the disease in children presenting to the pediatric ED. In the non-traumatic group, intussusception and incarcerated inguinal hernia were the major causes of acute abdomen in children 1 year old, and acute appendicitis was the major etiology of surgical abdomen in children > 1 year old. According to other investigations, intussusception occurs frequently between the ages of 3 months and 5 years, with 60% of cases occurring in the first year of life and a peak incidence at 6 11 months of age. 5 A 9-year survey recorded the following anatomic sites of intussusception found during open reduction: 19 (82.5%) in the ileo-colic area, three (13%) in the ileo-cecal region, and one (4.5%) was ileo-ileo-colic. 6 In our study, only 26.3% of patients with intussusception required surgery. In addition, 21 cases with ileocolic type intussusception (23.1%), three ileo-ileocolic type (100%), and one ileo-cecal-colic type underwent surgery (100%). Based on our analysis, we found that surgical reductions were needed in all cases of ileo-ileo-colic and ileo-cecal-colic type

6 Acute abdomen in children and adolescents 131 Table 2 Characteristics of patients in the traumatic group* Liver Spleen Kidney Pancreas Intestine, colon Multiple organs p (n = 14) (n = 12) (n = 3) (n = 2) (n = 5) (n = 29) Age Infant Preschool-age 7 (50.0) (20) 4 (13.8) School-age 3 (21.4) 3 (25.0) (40) 4 (13.8) Adolescent 4 (28.6) 9 (75.0) 3 (100) 2 (100) 2 (40) 21 (72.4) Gender Male 8 (57.1) 9 (75) 3 (100) 2 (100) 4 (80) 19 (65.5) Female 6 (42.9) 3 (25) (20) 10 (34.5) Imaging studies Abdominal X-ray 2 (14.3) 2 (16.7) 1 (33.3) 0 2 (40) 5 (17.2) Negative finding 1 (50) 1 (50) 1 (100) 1 (50) 4 (80) Bowel dilatation 1 (50) Local ileus 0 1 (50) 0 1 (50) 0 Free air (20) Ultrasonography 9 (64.3) 8 (66.7) 2 (66.7) 0 4 (80) 17 (58.6) Accurate diagnosis 9 (100) 7 (87.5) 1 (50) 1 (25) 13 (76.5) Abdominal CT 14 (100) 12 (100) 3 (100) 2 (100) 4 (80) 29 (100) Accurate diagnosis 14 (100) 12 (100) 3 (100) 2 (100) 4 (100) 29 (100) Laboratory tests WBC ( 10 3 /mm 3 ) 12.5 ± ± ± ± ± ± Neutrophil count (%) 61.0 ± ± ± ± ± ± Band count (%) 10.3 ± ± ± ± ± CRP (mg/dl) 9.40 ± ± Hb (g/dl) 11.2 ± ± ± ± ± ± Amylase (U/L) ± ± ± ± ± ± Lipase (U/L) 20.0 ± ± ± ± ± AST (U/L) ± ± ± ALT (U/L) ± ± ± ± ± Urgent surgery 2 (14.3) 4 (33.3) (80.0) 13 (44.8) ICU admission 10 (71.4) 10 (83.3) 1 (33.3) 1 (50) 2 (40) 19 (65.5) ICU admission days 9.9 ± ± ± ± ± ± Hospitalization days 12.3 ± ± ± ± ± ± Mortality *Data presented as n (%) or mean ± standard deviation. CT = computed tomography; WBC = white blood cell count; CRP = C-reactive protein; Hb = hemoglobin; AST = aspartate aminotransferase; ALT = alanine aminotransferase; ICU = intensive care unit.

7 132 Y.C. Tseng et al A B C D Figure 3 Imaging findings in patients with acute abdomen: (A) free air under both hemidiaphragms (arrow) on erect abdominal radiograph; (B) ultrasonography shows intussusception: target sign (arrow); (C) axial abdominal computed tomography shows twisted and engorged mesenteric vessels in a whirl pattern (arrow), which is diagnostic of bowel volvulus; (D) blunt abdominal trauma with liver laceration (arrow). intussusceptions, compared to 23.1% (21/91) of cases of ileo-colic type intussusception. Primary pediatricians should therefore pay particular attention to patients who are diagnosed with ileo-ileo-colic type or ileo-cecal-colic type intussusception, because these patients are more likely to require surgical reduction. A study in North America showed that 60% of cases of incarcerated inguinal hernia occurred during the first year of life, with a male predominance (female/male: 1/6), and that it occurred more often on the right side (2:1). 5 Another study reported that nearly 50% of incarcerated inguinal hernias occurred before the age of 6 months. 1 In our study, however, 53% (16/30) of patients with incarcerated inguinal hernias were older than 1 year. This may reflect a difference between our pediatric ED and those in previous reports. The female:male ratio in our study was 1:5, and the ratio of right-sided to left-sided hernias was almost 3:1 (21:7). Our findings revealed that the incidences of appendicitis were similar in school-age children and adolescents (44.1% vs. 44.5%) and there was a female:male ratio of 7:13 (91/165), the perforation rate in our study was 20.3% (52/256). Other studies have reported a slight male predominance for appendicitis, with a peak incidence at 9 12 years old, 7,8 and a perforation rate of about 30 65%. 5 The lower perforation rate in our study might reflect the high level of availability of medical care in this residential area in Taiwan, and the fact that most doctors are alert to the implications of right lower quadrant abdominal pain. Past experience has revealed that the seasonal incidence of intussusception peaks in spring and autumn, 9 and that the correlation between intussusception and adenovirus infection is high. In this study, we found that cases of intussusception requiring surgery peaked in February, but no obvious seasonal difference could be found. Our previous studies failed to reveal any monthly patterns of distribution of other diseases in the non-traumatic group; however, our survey suggests that more cases of appendicitis occurred in September, while more intestinal perforation occurred in June. Larger sample sizes are needed to confirm any seasonal disease patterns. In the traumatic group, liver and spleen lacerations were the two most common indications for surgery (21.5% and 18.5%, respectively), suggesting that the spleen is injured nearly as frequently as the liver (Figure 3D). Previous studies have shown the spleen to be the most commonly injured intra-abdominal

8 Acute abdomen in children and adolescents 133 organ in adults who sustained abdominal trauma, while blunt liver trauma was the most common fatal abdominal injury. 7,10,11 The spleen and liver are both fragile intra-abdominal organs, which can easily be injured in a crash. We also noted that the patients with spleen lacerations had the shortest duration in the ICU and shortest duration of hospital stay in our study, while patients with liver lacerations had the longest durations. Although the mean Hb level did not differ significantly among patients in the traumatic group, it may still be an important means of evaluating blood loss. 7,10,11 For example, one patient with multiorgan injury suffered from massive bleeding and had the lowest level of Hb (1 g/dl) in our survey. Also, a patient with liver laceration had an Hb level of 4.3 g/ dl, and a patient with intestinal perforation had an Hb level of 3.8 g/dl. Intestinal perforation and liver laceration both carry high risks of massive bleeding. 10,11 Lacerations of other organs such as the spleen, kidney or pancreas were associated with the lowest Hb levels, of 10.2, 10.2, and 12.1 g/dl, respectively. The spleen, kidney and pan creas may be less likely to produce massive internal bleeding than the liver and intestine. In our study, imaging studies performed in children with acute abdomen included plain radiographs, contrast studies, abdominal ultrasound, and abdominal CT scans. 7,12 In the non-traumatic group, bowel loop dilation and local ileus were the most common findings on plain abdominal films. However, an accurate diagnosis could not be made in most of these patients based on plain films. One prospective study reported that the presence of prior abdominal surgery, foreign body ingestion, abnormal bowel sounds, abdominal distension or peritoneal signs were 93% sensitive and 40% specific for confirming diagnostic or suggestive radiographs in patients with major disease; in the absence of any of these five clinical features, the sensitivity and specificity of plain film may be low. 13 We therefore suggest that patients with non-traumatic abdominal pain should undergo plain film examinations when these risk factors are present. In the traumatic group, however, only 18.5% (12/65) of patients underwent plain film examinations, and most findings were negative (66.7%, 8/12). Plain film examinations therefore appear to be less useful as a screening tool in pediatric patients with traumatic abdominal pain. In comparison, ultrasonography is a better imaging modality for establishing a diagnosis in patients with acute abdomen. 14,15 In our study, the diagnostic rates for intussusception and ovarian torsion were both 100%, while the rates for appendicitis and intestinal obstruction were 73.2% and 66.7%, respectively. The accuracy of ultrasonography for diagnosing pediatric appendicitis was relatively low, compared with a previous study in adults, which showed that ultrasonography had 96% sensitivity and 94% specificity for establishing a diagnosis of appendicitis. 14,16,17 A lack of compliance and the relatively small size of the appendix in children may be the primary reasons for this discrepancy. Ultrasonography may, however, be helpful in trauma patients. More than half of our patients received sonogram examinations (61.5%, 40/65), and the accurate diagnosis rate was 77.5% (31/40). Bedside abdominal sonograms in trauma patients can also be useful for tracking persistent intra-abdominal bleeding. However, many factors can affect the accuracy of ultrasonography, including obesity, free air, scar tissue, or physician s personal experience. Primary physicians in the pediatric ED therefore need a better diagnostic tool when doubt exists about the need for surgical intervention. In our study, both plain film X-rays and abdominal ultrasonography had lower diagnostic accuracies than abdominal CT scans in both the traumatic and non-traumatic groups. We therefore suggest that abdominal CT can provide important data for rapid diagnosis during the course of management of a child with suspected acute abdomen, and propose that abdominal CT should be used by primary emergency clinicians for the early differentiation of surgical emergencies in patients whose conditions cannot be confirmed by plain film examinations or abdominal ultrasonography in the ED. In conclusion, the etiology of acute abdomen varied depending on the age of the patient. Acute appendicitis was the most common cause of acute abdomen in children older than 1 year of age, followed by traumatic injury. Abdominal CT was a useful diagnostic imaging modality in patients with both traumatic and non-traumatic abdominal pain. References 1. Leung AK, Sigalet DL. Acute abdominal pain in children. Am Fam Physician 2003;67: Mason JD. The evaluation of acute abdominal pain in children. Emerg Med Clin North Am 1996;11: Erkan T, Cam H, Ozkan HC, et al. Clinical spectrum of acute abdominal pain in Turkish pediatric patients: a prospective study. Pediatr Int 2004;46: Lin KC, Wu HP, Huang CY, Lin CY, Chang CF. Discriminant analysis of serum inflammatory biomarkers which differentiate pediatric appendicitis from other acute abdominal diseases. Acta Paediatr Taiwan 2007;48: Maureen MC, Ghazala QS. Abdominal pain in children. Pediatr Clin N Am 2006;53: Eshel G, Barr J, Heyman E, et al. Intussusception: a 9-year survey ( ). J Pediatr Gastroenterol Nutr 1997;24: Fleisher GR, Ludwig S, Henretig FM, Ruddy RM, Silverman BK, eds. Textbook of Pediatric Emergency Medicine. Philadelphia: Lippincott Williams & Wilkins, 2006:

9 134 Y.C. Tseng et al 8. Wu HP, Huang CY, Chang YJ, Chou CC, Lin CY. Use of changes over time in serum inflammatory parameters in patients with equivocal appendicitis. Surgery 2006;139: Ho WL, Yang TW, Chi WC, Chang HJ, Huang LM, Chang MH. Intussusception in Taiwanese children: analysis of incidence, length of hospitalization and hospital costs in different age groups. J Formos Med Assoc 2005;104: Watanabe T, Kimura W, Yano M, Hachiya O, Ma J, Fuse A. A drainage operation for injury to the pancreas preserved the distal pancreas, spleen and the residual stomach after distal gastrectomy: a case report. Hepatogastroenterology 2007;54: Zwingmann J, Schmal H, Südkamp NP, Strohm PC. Injury severity and localisations seen in polytraumatised children compared to adults and the relevance for emergency room management. Zentralbl Chir 2008;133: Scholer SJ, Pituch K, Orr DP, Dittus RS. Test ordering on children with acute abdominal pain. Clin Pediatr 1999;38: Rothrock SG, Green SM, Hummel CB. Plain abdominal radiography in the detection of major disease in children: a prospective analysis. Ann Emerg Med 1992;21: Davies AH, Mastorakou I, Cobb R, Rogers C, Lindsell D, Mortensen NJ. Ultrasonography in the acute abdomen. Br J Surg 1991;78: Douglas CD, Macpherson NE, Davidson PM, Gani JS. Randomised controlled trial of ultrasonography in diagnosis of acute appendicitis, incorporating the Alvarado score. BMJ 2000;321: Rothrock SG, Pagane J. Acute appendicitis in children: emergency department diagnosis and management. Ann Emerg Med 2001;36: Wu HP, Lin CY, Chang CF, Chang YJ, Huang CY. Predictive value of C-reactive protein at different cutoff levels in acute appendicitis. Am J Emerg Med 2005;23:60 6.