Objective. In 1998, we implemented a clinical imaging protocol in which children with suspected appendicitis underwent ultrasonography (US) followed by computed tomography (CT). We sought to determine the impact of the US-CT protocol on changes in perforation and negative appendectomy rates.
Methods. Children with unequivocal presentations for appendicitis went to the operating room without entering the imaging protocol. Using a modified time series design, we analyzed a prospective and retrospective cohort of consecutive patients who were admitted from the emergency department for suspected appendicitis. The perforation and negative appendectomy rates were computed for the periods before and after implementation of the imaging protocol and adjustment for time trends was made.
Results. A total of 1338 children were identified. Eight hundred ten (60.5%) children had equivocal clinical findings. A total of 920 patients were admitted for suspected appendicitis before the protocol was implemented; 526 (57.2%) of the 920 children had appendicitis, and 186 (35.4%) of them had perforation. A total of 91 (14.7%) of 617 had negative appendectomies. After the protocol was implemented, 418 patients were admitted for suspected appendicitis; 328 (78.5%) had appendicitis with 51 (15.5%) perforated. There were 14 (4.1%) of 342 cases of negative appendectomies. After implementation of the imaging protocol, the perforation rate decreased from 35.4% to 15.5%, and the negative appendectomy rate decreased from 14.7% to 4.1%. After secular time trends were adjusted for, the imaging protocol continued to have a strong association with a reduction in perforation rate and negative appendectomy rate.
Conclusion. The implementation of an imaging protocol using US and CT resulted in a marked decrease in the perforation and negative appendectomy rates in children with suspected appendicitis.
The diagnosis of appendicitis can frequently be made from the history, physical examination, and laboratory studies obtained during the initial clinical encounter.1–3 However, in children and adolescents, accurate diagnosis is especially difficult, as the initial presentation of appendicitis may be obscure and closely mimicked by other common conditions.4–8 Delay in the diagnosis of appendicitis increases the risk of perforation, which, in turn, increases the rate of postoperative complications9 to as high as 39%, as compared with 8% in patients with uncomplicated appendicitis.10,11 Conversely, a normal appendix is unnecessarily removed in 15% to 40% of children who undergo emergency appendectomy.2,11–13
Appendiceal ultrasonography (US) has been used at many centers during the past decade to improve the accuracy of clinical diagnosis. However, reported accuracies range widely from 71% to 95%.14–19 In addition, studies have shown high false-negative rates,19,20 increased costs, and decreased overall accuracy21 with the use of sonography in patients with appendicitis. The use of computed tomography (CT) to diagnose appendicitis has increased markedly during the past 3 years with proven accuracies between 93% and 98%.18,20,22–27 Recent studies have shown improvement in patient management and decreased costs in patients who underwent CT for the diagnosis of appendicitis.20,23,28,29
A clinical protocol of US followed by CT was implemented at Children’s Hospital, Boston, in the spring of 1998 with a 94% accuracy for the diagnosis of acute appendicitis.20 The protocol was shown to improve patient management in children who underwent imaging.20,29 However, the impact of the US-CT protocol on the true clinical outcomes—the overall perforation and negative appendectomy rates—has not been previously studied. The purpose of this study was to determine the changes in perforation and negative appendectomy rates after the implementation of the US-CT imaging protocol.
The study was designed as a modified time series, involving a prospective and retrospective cohort. We identified a cohort of consecutive patients who were admitted for suspected appendicitis through the emergency department (ED) of Children’s Hospital, Boston, from January 1996 to December 1999. All patients who were admitted to the hospital for suspected appendicitis were eligible. Children’s Hospital is a large, urban pediatric teaching hospital with an ED that experiences 50 000 patient visits and a general surgical service that performs 4000 operations annually.
Patients who were evaluated between January 1996 and June 1998 and between January 1999 and December 1999 were identified through a query of the hospital database selecting patients with International Classification of Diseases, Ninth Revision, codes for appendicitis, perforated appendicitis, appendectomy, and abdominal pain. A single reviewer using a standardized data collection tool abstracted data on symptoms, clinical examination findings, radiographic studies, and pathology reports. Children who were evaluated between July 1998 and December 1998 were identified prospectively during the evaluation of the US-CT protocol. Data on symptoms, clinical examination findings, radiographic studies, and pathology reports were collected prospectively on these children.
Children were defined as having equivocal presentations for appendicitis when they had a history, physical examination, or laboratory evaluation that was concerning but not classic for acute appendicitis in addition to either undergoing US or CT imaging or being admitted to the hospital for an observation period.
Before US-CT Protocol
Before the spring of 1998, all children with suspected appendicitis were evaluated in the ED by a senior surgical resident, who was in the fourth or fifth postgraduate year, under the direct supervision of a pediatric surgical attending. Children with unequivocal clinical presentations for appendicitis underwent appendectomy without imaging. Children with equivocal clinical presentations for appendicitis were admitted to the surgical service for a period of inpatient observation and serial physical examinations with or without imaging studies performed in the ED. The imaging studies included pelvic ultrasonography and/or abdominopelvic CT scans with intravenous and oral contrast. During early 1998, pelvic CT scans with rectal contrast began to be used at our institution for the diagnosis of acute appendicitis.
After US-CT Protocol
From July to December 1998, we prospectively evaluated the US-CT imaging protocol for acute pediatric appendicitis. All children were evaluated by the senior surgical resident, who was in the fourth or fifth postgraduate year, under the direct supervision of an attending pediatric surgeon in the ED. Children with unequivocal clinical presentations underwent appendectomy without imaging. Children with equivocal presentations for appendicitis were initially evaluated with pelvic US. When the US was positive for appendicitis, the patient underwent appendectomy without additional imaging. If the sonographer was unable to visualize the appendix or when the sonogram was equivocal or inconclusive, the patient underwent CT with rectal contrast (Fig 1). Details of this study are published elsewhere.20 After the study period, children with equivocal clinical presentations continued to be evaluated with the US-CT protocol. By the end of 1999, some children, however, began to have only CT with rectal contrast performed for the diagnosis of acute appendicitis.
Final Diagnosis and Patient Follow-up
Final clinical outcomes were determined by pathologic reports in patients who underwent appendectomy and by 2-week telephone follow-up in children who were identified prospectively and managed nonoperatively. Patients who were identified retrospectively and were admitted for observation, did not undergo surgery, and did not return to the hospital after discharge were considered not to have appendicitis.
Negative Appendectomy and Perforation Rates
Negative appendectomy rate was defined as the number of normal appendices identified at pathology over all appendectomies performed. Perforation rate was defined as the number of perforated appendices identified at pathology over all cases of pathologically proven appendicitis. Negative appendectomy and perforation rates for the periods before and after July 1, 1998, were computed for all children, for equivocal cases, and by gender and age group (≤5 years, 6–10 years, 11–13 years, and ≥14 years).
Adjustment for Time Trends
The data were then analyzed with adjustment for time trends using multivariate logistic regression, testing for the effect of the US-CT imaging protocol on perforation and negative appendectomy rates, while adjusting for secular time trends. Contained in the model were the dichotomous imaging variable (before or after protocol) and year as a trend variable with 4 possible values (1996, 1997, 1998, or 1999).
The data were analyzed with descriptive statistics. Differences in age were evaluated by independent sample 2-tailed Student t test. Differences in gender and negative appendectomy and perforation rates were evaluated with the 2-tailed χ2 statistic. The effect of secular time trends on negative appendectomy and perforation rates was evaluated by multivariate logistic regression models using forward-stepwise selection. All calculations were performed with SPSS for Windows, version 7.5 (SPSS Inc, Chicago, IL).
A total of 1338 children who were admitted to the hospital with suspected appendicitis were identified between 1996 and 1999. The mean age of the entire cohort was 11.0 ± 4.3 years (range: 3 days–22 years). There were 598 (44.7%) female patients. Patient demographics by year are shown in Table 1. There was no difference among the years in either patient gender or age. A total of 810 (60.5%) of the 1338 children had equivocal clinical findings; 385 (47.5%) of these were male patients. The mean age of the patients with equivocal findings was 11.0 ± 4.6 years.
Before US-CT Protocol
A total of 920 patients were admitted for suspected appendicitis before July 1998. The mean age of the children was 11.0 ± 4.3 years. There were 491 (53.4%) male patients. A total of 526 (57.2%) of the 920 patients had appendicitis. There were 186 cases (35.4%) of perforated appendicitis and 91 cases (14.7%) of negative appendectomies (Table 2). A total of 404 (44%) of 920 children went directly to the operating room without imaging studies.
Of the 810 children with equivocal presentations, 516 (63.7%) were evaluated before July 1998. The mean age of this cohort was 11.1 ± 4.7 years; 232 (45%) were male patients. A total of 174 (33.7%) of the 516 children had appendicitis. There were 60 cases (34.5%) of perforated appendicitis and 40 cases (18.7%) of negative appendectomies (Table 2).
Table 3 shows the gender differences in rates of perforated appendicitis and negative appendectomy for the children before the imaging protocol was implemented. A total of 116 (36.1%) of 321 male and 70 (34.1%) of 205 female patients had perforated appendicitis before implementation of the imaging protocol. Thirty-eight of 359 male (10.6%) and 53 (20.5%) of 258 female patients had a negative appendectomy before the implementation of the protocol. Perforation and negative appendectomy rates by patient age are shown in Table 4.
After US-CT Protocol
A total of 418 patients were admitted for suspected appendicitis after July 1998. The mean age of the children was 10.9 ± 4.1 years. There were 249 (59.6%) male patients; 328 (78.5%) of the 418 children had appendicitis. There were 51 cases (15.5%) of perforated appendicitis and 14 cases (4.1%) of negative appendectomies (Table 2). A total of 124 (30%) of 418 children went directly to the operating room without imaging studies.
After July 1998, 294 (36.3%) of the 810 children with equivocal presentations were evaluated; 153 (52%) were male patients. Their mean age was 11.0 ± 4.3 years; 212 (72.1%) of the 294 had appendicitis. There were 37 cases (17.5%) of perforated appendicitis and 7 cases (3.2%) of negative appendectomies (Table 2).
Thirty-three (16%) of 206 male and 18 (14.8%) of 122 female patients were found to have had perforated appendicitis after the imaging protocol was implemented. Eight (3.7%) of 214 male and 6 (4.7%) of 128 female patients were found to have had a negative appendectomy (Table 3). Table 4 depicts perforation and negative appendectomy rates by patient age.
Compared with the time period before July 1998, the perforation rate in children with suspected appendicitis after July 1998 decreased from 35.4% to 15.5% (P < .001). In addition, there was a substantial decrease in the negative appendectomy rate from 14.7% to 4.1% (P < .001). Figure 2 depicts the trend in perforation and negative appendectomy rates by 6-month periods. Although there seems to be a secular trend in the decrease in perforation and negative appendectomy rates before the imaging protocol was implemented, this decrease corresponds to a trend in increased use of CT and ultrasound during this time period, even before the protocol was fully implemented (Fig 3). Note that the use of ultrasound began to decrease after the prospective study on the accuracy of the imaging protocol was completed. Because the accuracy of CT had proved to be better than that of ultrasound, several surgeons began to use CT alone in selected cases.
For patients who had equivocal presentations of appendicitis, the perforation rate decreased from 34.5% to 17.5% (P < .001) and the negative appendectomy rate decreased from 18.7% to 3.2% (P < .001; Table 2). Male patients had a significant decrease in perforation rate from 36.1% to 16% (P < .001). They were also found to have a decrease in negative appendectomy rate from 10.6% to 3.7% (P = .004). Female patients also had a significant decrease in perforation rate from 34.1% to 14.8% (P < .001) as well as in negative appendectomy rate from 20.5% to 4.7% (P < .001; Table 3).
Children ≤5 years of age did not have a significant decrease in either the perforation rate (P = .10) or negative appendectomy rate (P = .67) after the protocol was fully implemented. Children between the ages of 6 and 10 did have a significant decrease in both the perforation rate (P < .001) and negative appendectomy rate (P = .001) as did children between 11 and 13 years (P = .006 and P = .005, respectively) and children 14 years or older (P = .01 and P = .003, respectively; Table 4). After secular time trends were adjusted for, the imaging protocol continued to be associated with a reduction in the perforation rate (P < .001) and the negative appendectomy rate (P = .02) in the cohort of children.
The US-CT imaging protocol was associated with a dramatic reduction in the perforation and negative appendectomy rates in children who were evaluated for suspected appendicitis. No previous clinical scales or radiographic interventions have ever been associated with such a profound decrease in either the perforation or negative appendectomy rates in children with suspected appendicitis.
Our current investigation determined that concomitant with the use of the US-CT imaging protocol, the perforation rate in patients at Boston Children’s Hospital significantly decreased from 35.4% to 15.5% and the negative appendectomy rate also significantly decreased from 14.7% to 4.1%. In addition, the perforation and negative appendectomy rates were lowered in all patient subgroups except for the very young children. The early diagnosis of appendicitis in patients <5 years of age is especially difficult to make, as many of these children often present to the ED with perforation of the appendix.30 However, the US-CT imaging protocol continues to be used in this patient population.
Children who are evaluated in the ED with signs and symptoms suspicious for appendicitis are often observed to improve preoperative accuracy. However, this increase in observation time, although it may decrease the negative appendectomy rate, may increase the perforation rate and its associated complications as a result of the delay before necessary appendectomy. Hence, expedited diagnosis in children with acute appendicitis is essential. In addition, rapidly diagnosing children without appendicitis with an alternative diagnosis and avoiding potential hospitalization saves both the patient and the parent undue stress and provides a decrease in hospital resources. In fact, we have shown that the US-CT protocol was associated with a decrease in the hospitalization rate of children who were suspected of having appendicitis and subsequently were found not to have appendicitis. Before the imaging protocol, 57.2% of admitted patients had appendicitis, whereas after the protocol, 78.5% of admitted patients were found to have appendicitis.
A mention should be made regarding the number of unequivocal patients before and after initiation of the imaging protocol. One might think that there should have been the same number of patients with unequivocal findings before and after the protocol. However, in actuality, the percentage of patients with unequivocal presentations for appendicitis was 40% before the protocol and 30% after the protocol. We believe that this difference was caused by several factors. First, the accuracy of CT for appendicitis was found to be so high with little to no risk that many surgeons would order a CT scan even in some cases considered to be unequivocal for appendicitis. Also, many parents after hearing the media laud CT scans for appendicitis would request them before their child went to the operating room. The increased use of imaging is reflected in the significant decrease in negative appendectomy rate during the study period.
There were a few limitations with this investigation. First, historical controls were used; however, the prospective and retrospective cohorts seem to be comparable, as both the perforation and negative appendectomy rates remained relatively constant during the 6-month blocks until the initiation of the protocol when both markedly decreased. Second, we evaluated only patients who were admitted to the hospital after ED evaluation for suspected acute appendicitis. This was because, before the protocol, all patients who were evaluated by the surgeon in the ED were admitted to the hospital. For consistency, we also chose to evaluate only patients who were admitted to the hospital after the imaging protocol was initiated. In addition, the outcomes that we evaluated (negative appendectomy and perforation rates) can be assessed only in patients who are admitted. Hence, this selected cohort seems appropriate for the investigation that was performed.
We have, for the first time, shown a dramatic reduction in the perforation and negative appendectomy rates in children with suspected appendicitis associated with an imaging protocol highly sensitive and specific for the diagnosis. Because of improved patient management with the use of the imaging protocol, the results of this study strongly support the use of the US-CT protocol for children with signs and symptoms suspicious, but not diagnostic for, acute appendicitis.
- Received November 19, 2001.
- Accepted June 18, 2002.
- Reprint requests to (B.M.G.P.) Miami Children’s Hospital, 3100 SW 62nd Ave, Miami, FL 33155. E-mail:
Presented in part at the Society for Pediatric Research; May 14, 2000; Boston, MA.
- ↵Scholer SJ, Pituch K, Orr DP, Ditus RS. Clinical outcomes of children with acute abdominal pain. Pediatrics.1996;98 :680– 685
- ↵Garcia Peña BM, Taylor GA, Fishman SJ, Mandl KD. Costs and effectiveness of ultrasonography and limited computed tomography for diagnosing appendicitis in children. Pediatrics.2000;106 :672– 676
- Copyright © 2002 by the American Academy of Pediatrics