Article, Radiology

Predicting need for additional CT scan in children with a non-diagnostic ultrasound for appendicitis in the emergency department

a b s t r a c t

Objective: This study aimed to determine which children with suspected appendicitis should be considered for a Computerized tomography scan after a non-diagnostic ultrasound (US) in the Emergency Department (ED). Methods: We retrospectively reviewed patients 0-18 year old, who presented to the ED with complaints of ab- dominal pain, during 2011-2015 and while in the hospital had both US and CT. We recorded Demographic and clinical data and outcomes, and used univariate and multivariate methods for comparing patients who did and didn’t have appendicitis on CT after non-diagnostic US. Multivariate analysis was performed using logistic regres- sion to determine what variables were independently associated with appendicitis.

Results: A total of 328 patients were enrolled, 257 with non-diagnostic US (CT: 82 had appendicitis, 175 no- appendicitis). Younger children and those who reported vomiting or had right lower abdominal quadrant (RLQ) tenderness, Peritoneal signs or white blood cell count N 10,000 in mm3 were more likely to have appendicitis on CT. RLQ tenderness (Odds Ratio: 2.84, 95%CI: 1.07-7.53), peritoneal signs (Odds Ratio: 11.37, 95%CI: 5.08-25.47) and WBC count N 10,000 in mm3 (Odds Ratio: 21.88, 95%CI: 7.95-60.21) remained significant after multivariate analysis. Considering CT with 2 or 3 of these predictors would have resulted in sensitivity of 94%, specificity of 67%, positive predictive value of 57% and negative predictive value of 96% for appendicitis.

Conclusions: Ordering CT should be considered after non-diagnostic US for appendicitis only when children meet at least 2 predictors of RLQ tenderness, peritoneal signs and WBC N 10,000 in mm3.

(C) 2017

Introduction

Acute appendicitis is the most common Abdominal emergency in children [1], and early diagnosis is critical in preventing perforation, ab- scess formation and postoperative complications [2]. History and phys- ical examination alone has a low sensitivity and specificity [3], so imaging may play a key role in the accurate and Prompt diagnosis of suspected appendicitis. Computerized tomography has high sensi- tivity for diagnosis of appendicitis, and has been used more frequently in the pediatric population in recent years [4]. However, due to radiation exposure [5], providers in the emergency department (ED) are prefer- ring ultrasound (US) as imaging modality [6,7]. Point of Care Ultrasound for appendicitis performed in ED has high effectiveness in clinical deci- sion making of ED physicians [8]. Although the specificity of US is high, the sensitivity is variable (50-100%) depending on the operator [9]. It is unclear which children should have advanced imaging after US is

? Meetings: Nothing.Grant or other financial support: Nothing.

* Corresponding author.

E-mail address: [email protected] (T. Nishizawa).

considered non-diagnostic when performed by the ED physician [10]. Performing selective CT only for children with non-diagnostic US [11, 12] may reduce CT utilization, but a significant proportion of these chil- dren may still have a negative test for appendicitis on CT [11]. The pur- pose of this study was to determine a method for identification of children who will benefit from a CT scan after a non-diagnostic US.

Methods

We conducted a retrospective study at an urban, tertiary-care hospi- tal in Japan with about 40,000 ED visits per year. Our hospital ethics committees approved the research protocol. We obtained data from the electronic medical record (EMR) system and the hospital electronic administrative system for the period January 1, 2011 through December 31, 2015 in order to identify patients who presented to the ED with complaints of abdominal pain and while in the hospital had both an US exam and abdominal CT.

Demographic information was automatically imported from the elec- tronic hospital administrative data system into an excel spreadsheet.

http://dx.doi.org/10.1016/j.ajem.2017.07.005

0735-6757/(C) 2017

Additional data that could not be exported as well as radiology EMR data was entered by the primary investigator (TN).

We included all patients 0-18 years of age presenting to the ED with abdominal pain that underwent abdominal US as the first Imaging study performed by an ED physician (junior resident, senior resident, or at- tending physician), and that subsequently had a CT scan. The decision to obtain imaging was determined by the ED physician. ‘Junior resi- dents’ in Japan are in residency during the first 2 years after Medical School, rotating in several departments. ‘Senior residents’ are specializ- ing in Emergency Medicine during years 3 and 4 after finishing Medical School. ‘Attending physicians’ are board-certified in Emergency Medi- cine and working at the ED for at least six years.

We excluded those who had [1] imaging studies due to trauma, [2] history of appendicitis, [3] US or CT documented before presenting to our ED, [4] CT performed before US, [5] US findings not found in the EMR and [6] suspected diagnoses other than appendicitis based on physical examination, clinical history and US.

Data included the following: demographic information, medical his- tory (chief complaint, duration of pain, associated symptoms), physical examination (area of tenderness, presence of peritoneal signs), labora- tory studies (White blood cell count), imaging (who performed the US, results of imaging) and final diagnosis.

Peritoneal signs were interpreted as positive if the child had at least

one of the following findings; rebound tenderness, guarding, heel drop sign, tapping pain and psoas sign. Leukocytosis was defined as N 10,000 WBC in mm3.

At our institution, the initial imaging modality used to evaluate chil- dren with suspected appendicitis was US. All US exams were per- formed by an ED physician. US exams were classified as positive for appendicitis if the ED physician documented ‘Visualization of non- compressible swollen appendix (appendix diameter N 6 mm)’, ‘Appendicolitis’, or ‘Peri-appendiceal fluid’. US exams were classified as negative if they included ‘Visualization of a compressible non- swollen appendix’. US exams were classified as inconclusive if they in- cluded statements such as ‘Non-visualization of appendix and no alter- native diagnosis rendered’ or ‘Non-specific ileocecal lymph node swelling’. The terms ‘negative US’ and ‘inconclusive US’ were consid- ered as non-diagnostic US.

CT studies were classified as positive for appendicitis if the Radiology report stated ‘Enlarged or thickened appendix’, ‘Consistent with acute appendicitis’ or ‘Consider appendicitis’. CT studies were classified as negative if the radiology report stated ‘Appendix not visualized‘ or ‘Nor- mal appendix’.

All data points were reviewed and verified by two investigators (TN and SM). Any discrepancies were reviewed by both authors to deter- mine accurate data points. No data analyses were done until the final database was locked.

Univariate and multivariate methods were used in comparing pa- tients who did and didn’t have appendicitis on CT among children with non-diagnostic US. To determine what variables were indepen- dently associated with outcomes, multivariate analysis was performed using logistic regression analysis. Values of p b 0.05 were considered statistically significant for all tests. survey responses were collected on a MS Excel 2010 (Ver 14) file and converted to International Busi- ness Machines (IBM) Statistical Package for Social Sciences (SPSS) Ver- sion 23.0 for analysis. All statistical methods were reviewed by a statistician.

Results

A total of 473 patients with complaints of abdominal pain had both US and CT during the study period. We excluded 129 patients who had

[1] imaging studies due to trauma (34, 7.2%), [2] history of appendici- tis (11, 2.3%), [3] US or CT documented before presenting to our ED (63, 13.3%), [4] CT performed before US (10, 2.1%), [5] US findings not found in the EMR (11, 2.3%) and [6] suspected other diagnoses

than appendicitis based on physical examination, clinical history and US (16, 3.4%). Thus, a total of 328 (69.3%) records were analyzed

(Fig. 1).

Demographic characteristic of the study population is presented in Table 1. Mean age was 12.2 years (3-18 years), and 184 (56%) were males. Mean BMI was 17.9 (11.8-30.8).

History: The mean Abdominal pain duration was 21 h (1-168 h), 135 patients (24.7%) reported vomiting and 28 (8.5%) had diarrhea. The most common chief complaint was Right lower quadrant abdominal pain (204, 62.2%). Other complaints were lower abdominal pain (51, 15.5%), upper abdominal pain (39, 11.9%), periumbilical pain (26, 7.9%), left lower abdominal pain (7, 2.1%) and right upper abdominal pain (1, 0.3%).

Physical examination: Two hundred fifty eight patients (78.7%) had right lower abdominal tenderness, 26(7.9%) had upper abdominal ten- derness, 24(7.3%) had lower abdominal tenderness, and 11(3.4%) had periumbilical tenderness. A few patients had tenderness in left lower area (5, 1.5%) and right upper area (2, 0.6%). Two patients (0.6%) had no significant findings on the physical examination.

One hundred thirty seven patients (41.8%) had peritoneal signs – Re-

bound: 76 (27.9%), Guarding: 21(9.7%), heel drop sign: 47(19.7%), tap- ping pain: 35(15.5%), Psoas sign: 31(14%).

Laboratory: The mean WBC was 11,931 (6300-28,800) with 127 (39.6%) having leukocytosis (WBC N 10,000 in mm3).

Point of Care US: US for 183 patients (55.8%) were performed by ju- nior residents, 27 (8.2%) by senior residents and 118 (36%) by ED at- tending physician. Exams for 71 patients (21.6%) were interpreted by the physician doing the exam as confirming a diagnosis of appendicitis (69 had swollen appendix, 10 had stercolith and 4 had appendiceal fluid), 240 (73.2%) were interpreted as inconclusive (237 with no sig- nificant findings, and 3 had non-specific findings), and 17 (5.2%) were interpreted as no appendicitis. US had a sensitivity of 43.4%, specificity of 95.6%, positive predictive value of 88.7% and negative predictive value of 68.1%.

CT: Of 328 patients, 145 (44.2%) had appendicitis on CT.

Among 257 patients with a non-diagnostic US (Table 2), younger children were significantly more likely to have appendicitis on CT (p

= 0.012). Patients who reported vomiting or had right low abdominal quadrant (RLQ) tenderness, peritoneal signs or leukocytosis were more likely to have appendicitis on CT (p b 0.05). There were no statis- tical differences for gender, BMI, presence of fever, diarrhea, or duration of abdominal pain.

We then conducted a multivariate analysis and included age, vomiting, diarrhea, RLQ tenderness, peritoneal signs and leukocytosis in the model (Table 3). Only RLQ tenderness (Odds Ratio: 2.84, 95%CI: 1.07-7.53), peritoneal signs (Odds Ratio: 11.37, 95%CI: 5.08-25.47) and leukocytosis (Odds Ratio: 21.88, 95%CI: 7.95-60.21) were consid- ered significant predictors of appendicitis. Nagelkerke R square was

0.594. Percent of correct classification was 64.6% for appendicitis and 92.3% for no appendicitis. There were no statistical differences in these two groups for age, vomiting, diarrhea.

Non-diagnostic US in conjunction with each of the three predictors (RLQ tenderness, peritoneal signs and leukocytosis) had a sensitivity of 85.4%, 70.7%, 91.5% for appendicitis on CT, specificity of 34.7%,

82.3%, 64.9%, positive predictive value of 38.0%, 65.2%, 56.0% and nega-

tive predictive value of 83.6%, 85.7%, 94.0% respectively.

With combinations of the three predictors retained in the multivar- iate model (RLQ tenderness, peritoneal sign, leukocytosis), we devel- oped profiles for patients in whom US was non-diagnostic. Table 4 describes profiles of patients who had appendicitis on CT. One hundred seventeen (96%) of patients who met 0 or 1 of these features had no ap- pendicitis on CT, while 5 patients (4%) had appendicitis. Based on the prevalence of appendicitis cases in each of the subgroups corresponding to these profiles, the following decision rule was devised to select pa- tients who may benefit from a CT (Fig. 2). If patients with non- diagnostic US meet at least 2 predictors, a CT should be considered.

Fig. 1. Flow of study.

This decision rule has a sensitivity of 94%, specificity of 67%, PPV of 57% and NPV of 96%.

Limitations

Our study has several limitations. This is a retrospective, single- center non-randomized study, where decisions on type of imaging mo- dality were left to the clinician, and practice variation may play a role in our findings. However, we included results from over 50 physicians and findings represent practice in many other EDs. Also, it is possible that some children were discharged after an US and went to another facility where they were diagnosed with appendicitis. Patients who presented with acute appendicitis at a different institution after being discharged with no appendicitis on CT would have been incorrectly categorized as negative CT in our analysis. However, it is our practice to tell patients with abdominal pain who we decide to discharge that we cannot rule out appendicitis when abdominal pain after being discharged. Our cen- ter is the largest in the city, so most children will be seen in our center

again if they are diagnosed with appendicitis at another facility after being discharged. Furthermore, studies from North America suggest that overweight children are likely to have non-diagnostic US [13,14]. However, the average Body Mass Index (BMI) in our cohort was 17.9, likely not affecting US sensitivity in this cohort.

Discussion

We found that for children with suspected appendicitis and a non- diagnostic US exam by an emergency physician, CT is likely needed only when children have at least two of the following; RLQ tenderness, peritoneal signs or WBC N 10,000 in mm3. If we had applied this decision rule to our study group, we could have avoided about 65% of CT scans that were negative for appendicitis with a much lower False negative rate of 4% (Fig. 3).

Acute appendicitis is the most common cause of abdominal surgical interventions in children [1]. Early diagnosis may help prevent perfora- tion, abscess formation, and postoperative complications [2]. Clinical

Table 1 Demographic and clinical information on study cohort and comparison of children with appendicitis and no appendicitis on CT scan (US = Ultrasound, CT = Computerized Tomography, BMI = Body Mass Index, WBC = White Blood Cell count, CI = Confidence Interval).

Total (n = 328)

Appendicitis on CT (n = 145)

No appendicitis on CT (n = 183)

95% CI

p-Value

Age, yr (range)

12.24 (3-18)

11.37 (3-18)

12.84 (3-18)

11.84-12.63

0.000

Gender male, n (%)

184 (56.1)

90 (62.1)

94 (51.4)

0.052

BMI, (range)

17.9 (11.8-30.8)

17.5 (11.8-30.8)

18.6 (13.0-30.1)

17.35-18.34

0.094

Chief complaint

Right lower pain, n (%)

204 (62.2)

100 (69)

104 (56.8)

Lower pain, n (%)

51 (15.5)

28 (19.3)

23 (12.6)

Upper pain, n (%)

39 (11.9)

11 (7.6)

28 (15.3)

Other pain, n (%)

34 (10.3)

6 (4.1)

28 (15.3)

History

Duration, hours (range)

21 (1-168)

18 (1-96)

23 (1-168)

18.57-23.31

0.559

Vomiting, n (%)

135 (24.7)

54 (37.2)

81 (14.8)

0.000

Diarrhea, n (%)

28 (8.5)

9 (6.2)

19 (10.4)

0.179

US

Junior performed, n (%)

183 (55.8)

77 (53.1)

106 (57.9)

Senior performed, n (%)

27 (8.2)

11 (7.6)

16 (8.7)

Attending performed, n (%)

118 (36)

57 (39.3)

61 (33.3)

US positive, n (%)

71 (21.6)

63 (43.4)

8 (4.4)

US inconclusive, n(%)

240 (73.2)

81 (55.9)

159 (86.9)

US negative, n (%)

17 (5.2)

1 (0.7)

16 (8.7)

Laboratory

WBC, in mm3 (range) 11,931 (6300-28,800) 15,200 (6300-28,000) 9315 (3900-28,800) 11,427.47-12,435.32 0.000

prediction scores, such as Alvarado [15] and PAS score [16], have been used in the clinical setting, and the addition of imaging has improved ac- curacy of diagnosis [17].

Imaging

ED physicians are using point of care US more frequently as an imag- ing modality to evaluate appendicitis [7]. Diagnostic accuracy of US has been reported to be quite variable and heavily dependent on operator experience [18], but we found it to be a helpful adjunct diagnostic tool in our cohort. The high rate of non-diagnostic US reported in our study is consistent with that reported in the literature because US alone done by ED physician has been found to have relatively low sen- sitivity (44%) [6]. In our cohort, majority (56%) of US exams were

performed by junior residents and the low sensitivity we found in our cohort may be associated with our rigorous assessment of residents’ competency in point of care ultrasound, including evaluation of their as- sessment of appendicitis. This is similar to Emergency Medicine pro- grams in North America where bedside US is considered as one of the essential procedural skills [19].

A recent study has shown that children with equivocal US may be amenable to observation [20], but the study is limited by using CT as their primary end-point rather than finding if patients had appendicitis. A meta-analysis of diagnostic performance of CT mainly in adults or adolescents demonstrated superior sensitivity (91%) and specificity (90%) for appendicitis compared to US (78%, 83% respectively) [21], but radiation exposure potentially leading to leukemia and brain tumors in children has been a concern [5]. Several algorithms support reduction

Table 2

Demographic and clinical information on study cohort and comparison of children with appendicitis and no appendicitis on CT scan among children with non-diagnostic US (US = Ultra- sound, CT = Computerized Tomography, WBC = White Blood Cell count, CI = Confidence Interval).

Total (n = 257)

Appendicitis on CT (n = 82)

No appendicitis on CT (n = 175)

95% CI

p-Value

Age, yr (range)

12.45 (3-18)

11.60 (5-18)

12.85 (3-18)

11.99-12.90

0.012

Gender male, n (%)

139 (54.1)

48 (58.5)

91 (52.0)

0.327

BMI, n (range)

17.79 (11.8-30.8)

17.36 (11.8-30.8)

18.60 (13-30.1)

17.13-18.45

0.083

Chief complaint

Right lower pain, n (%)

150 (58.4)

54 (65.9)

96 (54.9)

Lower pain, n (%)

41 (16)

18 (22)

23 (13.1)

Upper pain, n (%)

35 (13.6)

7 (8.5)

28 (16)

Other, n (%)

31 (12)

3 (3.7)

28 (16)

History

Duration, hours (range)

21 (1-168)

18 (1-72)

23 (1-168)

18.43-24.18

0.671

Vomiting, n (%)

54 (21)

29 (35.4)

25 (14.3)

0.000

Diarrhea, n (%)

21 (8.2)

4 (4.9)

17 (9.7)

0.187

Physical examinations

Right lower tenderness, n (%)

190 (73.9)

70 (85.4)

120 (68.6)

Lower tenderness, n (%)

21 (8.2)

6 (7.3)

15 (8.6)

Upper tenderness, n (%)

26 (10.1)

2 (2.4)

24 (13.7)

Other tenderness, n (%)

20 (7.8)

4 (4.9)

16 (9.1)

Peritoneal signs, n (%)

89 (34.6)

58 (70.7)

31 (17.7)

0.000

US

Junior performed, n (%)

149 (58)

46 (56.1)

103 (58.9)

Senior performed, n (%)

23 (8.9)

9 (11)

14 (8)

Attending performed, n (%)

85 (33.1)

27 (32.9)

58 (33.1)

US inconclusive, n (%)

240 (93.4)

81 (98.8)

159 (90.9)

US negative, n (%)

17 (6.6)

1 (1.2)

16 (9.1)

Laboratory

WBC b 10,000 in mm3, n (%)

116 (46.4)

7 (8.5)

109 (64.9)

WBC 10,000 in mm3?, n (%)

134 (53.6)

75 (91.5)

59 (35.1)

Table 3

Multivariate analysis of independent predictors for appendicitis on CT scan (CI = Confi- dence Interval).

Factor

p-value

Odds Ratio (95% CI for Odds Ratio)

Age

0.798

0.987 (0.894-1.090)

Vomiting

0.250

1.691 (0.690-4.144)

Diarrhea

0.104

0.287 (0.053-1.295)

Right lower tenderness

0.036

2.842 (1.073-7.526)

Peritoneal signs

0.000

11.373 (5.078-25.474)

Leukocytosis

0.000

21.879 (7.950-60.210)

in unnecessary CT. A European study revealed a 64% reduction in esti- mated radiation dose in adult patients if standard-dose (SD) CT was only used when low-dose (LD) CT was inconclusive for appendicitis [22]. In a European prospective study at an ED, the sensitivity of Mag- netic Resonance Imaging (MRI) for the diagnosis of appendicitis in chil- dren was 100%, but cost and access are barriers to its use [23].

A multicenter study with N 1000 patients with acute abdominal pain, imaging with US and selective CT in cases with negative or inconclusive US demonstrated 94% sensitivity for appendicitis [24]. In a retrospective study in a pediatric ED, Rajesh et al. achieved a reduced rate of CT in about half of the children in whom the US findings were definitive [11]. Similarly, Wagenaar et al. reported selective CT after US by trained ED staff and significant monetary savings [25]. Moreover, the imaging strategy of CT after inconclusive US can maintain diagnostic accuracy for appendicitis among children [26]. Despite staged imaging protocols commonly used for children with non-diagnostic US for appendicitis [27], many children still undergo unnecessary CT scans. In our study, 68% of children with non-diagnostic US and CT scan had no appendicitis, similarly to the rate (74%) in a suburban pediatric ED in the US [11].

Clinical signs

Combining clinical variables with imaging can help diagnose appen- dicitis [28]. Clinical prediction score such as the Alvarado score [15] and the PAS score [16] have been used to identify patients with appendicitis. These scores include clinical history, physical examination and blood tests.

Our multivariate analysis model in children with non-diagnostic US suggests that leukocytosis (WBC N 10,000 in mm3), RLQ tenderness and peritoneal signs were significantly different between children that were found later to have appendicitis on a CT scan and those that their CT showed no appendicitis.

RLQ tenderness is a staple finding in appendicitis, especially when tenderness is at McBurney’s point [29]. Similar to Assefa et al [30], we found 85% of patients with appendicitis and a non-diagnostic US had RLQ tenderness on exam. The specificity was only 34.7% in our study be- cause there were other diseases that caused RLQ tenderness such as gas- troenteritis, diverticulitis and ovarian hemorrhage. This low specificity may have resulted in unnecessary CT scans or negative appendectomies [31].

Peritoneal signs were classically associated with appendicitis, but they generated small increase in the likelihood of appendicitis [32]. Only 60% of ED physicians thought rebound tenderness was valuable sign in diagnosing appendicitis since the sensitivity was relatively low [33]. In our study, the sensitivity of peritoneal signs was 70%. This rela- tive low sensitivity meant physical findings did not rule out appendicitis.

Fig. 2. The clinical decision rule (US = Ultrasound, CT = Computerized Tomography, WBC

= White blood cell counts, RLQ = Right Low Quadrant).

Leukocytosis alone has limited predictive value for appendicitis, with a recent study reporting that 21% of negative appendectomies in chil- dren had an elevated Neutrophil count at presentation [34]. Our findings were consistent with previous studies that found reduced radiation ex- posure for children when combining laboratory results with US [35]. Brian et al. reported the NPV of non-diagnostic US was 97% in patients with WBC b 7500 in mm3, and 96% in patients with WBC b 11,000 in mm3 [10]. Although there is no consensus in the literature about opti- mal WBC threshold values for maximizing predictive value, the cutoff established in our study is consistent with the Alvarado and the PAS score [15,16]. We showed high NPV (94%), using the cutoff of WBC b 10,000 in mm3.

Combination US and clinical signs

We found three predictors with a high NPV, especially when US was non-diagnostic. Some studies reported higher NPV through the combi- nation of several clinical variables and US findings [28]. In a multicenter prospective study in the Netherlands, Leeuwenburgh et al. developed a simple decision rule with high NPV (94%, 95%CI = 87% to 98%) for adult patients with suspected appendicitis and non-diagnostic US [36]. Clini- cal predictors were male gender, migration of abdominal pain to the

Table 4

Profiles of patient characteristics associated with acute appendicitis among patients with non-diagnostic US (WBC = White Blood Cell counts, RLQ = Right Low Quadrant).

WBC N 10,000

Peritoneal sign

RLQ tenderness

Appendicitis

No appendicitis

Total

0 or 1 characteristics, n (%)

5 (4)

117 (96)

122

117 (96)

122

2 or 3 characteristics, n (%)

77 (57)

58 (43)

135

58 (43)

135

Total, n (%)

82 (32)

175 (68)

257

175 (68)

257

Fig. 3. The changes in the number of CT scans by using our clinical decision rule . Ordered CT scans (n) after non-diagnostic US.

RLQ, vomiting and WBC count higher than 12,000 in mm3. Non- diagnostic US and 0 or 1 predictor meant that no CT or MRI were needed with ability to discharge 60% of patients [36].

Proposed guidelines

Based on appendicitis prediction factors, we suggest that children with suspected appendicitis on clinical evaluation, and a non- diagnostic US in the ED will only be considered for a CT scan when they have at least two predictors. Sensitivity (94%) and NPV (96%) for appendicitis are much higher than US alone (sensitivity: 43%, NPV: 68%), and comparable to that of performing selective CT after US (sensi- tivity 99%) [11]. Using this highly sensitive rule, we could have avoided 117 (67%) CT scans that were negative for appendicitis in our cohort with a much lower false negative rate (4%) compared to previously sug- gested (12%) [36].

In summary, CT scan should be considered after non-diagnostic point of care US for appendicitis only when children meet at least 2 pre- dictors out of RLQ tenderness, peritoneal signs and WBC N 10,000 in mm3. This decision rule provides high sensitivity and NPV, and unneces- sary CT may be avoided.

Author contribution

TN, SM, RG and HH conceived the study, designed the trial. SM, RG and HH supervised the conduct of the trial and data collection. TN, SM, RG, and HH undertook recruitment of patients and managed the data, including quality control. RG provided statistical advice on study design and analyzed the data; SM, RG and HH chaired the data oversight com- mittee. TN drafted the manuscript, and all authors contributed substan- tially to its revision. TN takes responsibility for the paper as a whole.

Conflicts of interest

Nothing.

Acknowledgement

We would like to thank Mr. Boris Kuzeljevic for his dedicated work on the statistical portion of this work.

References

  1. Addiss DG, Shaffer N, Fowler BS, et al. The epidemiology of appendicitis and appen- dectomy in the United States. Am J Epidemiol 1990;132(5):910-25.
  2. Rothrock SG, Pagane J. Acute appendicitis in children: emergency department diag- nosis and management. Ann Emerg Med 2000;36(1):39-51.
  3. Bundy David G, Byerley Julie S, Allen Liles E, et al. Does this child have appendicitis? JAMA 2007;298(4):438-51.
  4. Tsze Daniel S, Asnis Lisa M, Merchant Roland C, et al. Increasing computed tomogra- phy use for patients with appendicitis and discrepancies in pain management be- tween adults and children: an analysis of the NHAMCS. Ann Emerg Med 2012; 59(5):395-403.
  5. Pearce MS, Salotti JA, Little MP, et al. Radiation exposure from CT scans in childhood and subsequent risk of leukemia and brain tumours: a retrospective cohort study. Lancet 2012;380(9840):499-505.
  6. Fathi M, Hasani SA, Zare MA, et al. Diagnostic accuracy of emergency physician per- formed graded Compression ultrasound study in acute appendicitis: a prospective study. J Ultrasound 2014;18(1):57-62.
  7. Lam SH, Grippo A, Kerwin C, et al. Bedside ultrasonography as an adjunct to routine evaluation of acute appendicitis in the emergency department. West J Emerg Med 2014;15(7):808-15.
  8. Gungor F, Kilic T, Akyol KC, et al. Diagnostic value and effect of bedside ultrasound in acute appendicitis in the emergency department. Acad Emerg Med 2017;24(5):578-86.
  9. Doria AS, Moineddin R, Kellenberger CJ, et al. US or CT for diagnosis of appendicitis in children and adults? A meta-analysis. Radiology 2006;241(1):83-94.
  10. Cohen B, Bowling J, Lipskar A, et al. The non-diagnostic ultrasound in appendicitis: is a non-visualized appendix the same as a negative study? J Pediatr Surg 2015;50(6): 923-7.
  11. Krishnamoorthi R, Ramarajan N, Wang NE, et al. Effectiveness of a staged US and CT protocol for the diagnosis of pediatric appendicitis: reducing radiation exposure in the age of ALARA. Radiology 2011;259(1):231-9.
  12. Aspelund G, Fingerest A, Gross E, et al. Ultrasonography/MRI versus CT for diagnos- ing appendicitis. Pediatrics 2014;133(4):586-93.
  13. Keller C, Wang NE, Imler DL, et al. Predictors of nondiagnostic ultrasound for appen- dicitis. J Emerg Med 2017;52(3):318-23.
  14. Schuh S, Man C, Cheng A, et al. Predictors of non-diagnostic ultrasound scanning in children with suspected appendicitis. J Pediatr 2011;158(1):112-8.
  15. Alvarado A. A practical score for the early diagnosis of acute appendicitis. Ann Emerg

    Med 1986;15(5):557-64.

    Goldman RD, Carter S, Stephens D, et al. Prospective validation of the pediatric ap- pendicitis score. J Pediatr 2008;153(2):278-82.

  16. Sun JS, Noh HW, Min YG, et al. Receiver operating characteristic analysis of the diag- nostic performance of a computed tomographic examination and the Alvarado score for diagnosing acute appendicitis: emphasis on age and sex of the patients. J Comput Assist Tomogr 2008;32(3):386-91.
  17. Mittal MK, Dayan PS, Macias CG, et al. Performance of ultrasound in the diagnosis of ap- pendicitis in children in a multicenter cohort. Acad Emerg Med 2013;20(7):697-702.
  18. Amini R, Adhikari S, Fiorello A. Ultrasound competency assessment in emergency medicine residency programs. Acad Emerg Med 2014;21(7):799-801.
  19. Ramarajan N, Krishnamoorthi R, Gharahbaghian L, et al. Clinical correlation needed: what do emergency physicians do after an equivocal ultrasound for pediatric acute appendicitis? J Clin Ultrasound 2014;42(7):385-94.
  20. Van Randen A, Bipat S, Zwinderman AH, et al. Acute appendicitis: meta-analysis of diagnostic performance of CT and graded compression US related to prevalence of disease. Radiology 2008;249(1):97-106.
  21. Poletti PA, Platon A, De Perrot T, et al. Acute appendicitis: prospective evaluation of a diagnostic algorithm integrating ultrasound and low-dose CT to reduce the need of standard CT. Eur Radiol 2011;21(12):2558-66.
  22. Thieme ME, Leeuwenburgh MM, Valdehueza ZD, et al. Diagnostic accuracy and pa- tient acceptance of MRI in children with suspected appendicitis. Eur Radiol 2014; 24(3):630-7.
  23. Lameris W, van Randen A, van Es HW, et al. Imaging strategies for detection of ur- gent conditions in patients with acute abdominal pain: diagnostic accuracy study. BMJ 2009;338:b2431.
  24. Wagenaar AE, Tashiro J, Wang B, et al. Protocol for suspected pediatric appendicitis limits computed tomography utilization. J Surg Res 2015;199(1):153-8.
  25. Chien M, Habis A, Glynn L, et al. Staged imaging pathway for the evaluation of pedi- atric appendicitis. Pediatr Surg Int 2016.
  26. Srinivasan A, Servaes S, Pena A, et al. Utility of CT after sonography for suspected ap- pendicitis in children: integration of a clinical scoring system with a staged imaging protocol. Emerg Radiol Feb 2015;22(1):31-42.
  27. Blitman NM, Anwar M, Brady KB, et al. Value of focused appendicitis ultrasound and Alvarado score in predicting appendicitis in children: can we reduce the use of CT? AJR Am J Roentgenol 2015;204(6):W707-12.
  28. Smith MP, Katz DS, Lalani T, et al. ACR appropriate criteria right lower quadrant pain suspected appendicitis. Ultrasound Q 2015;31(2):85-91.
  29. Assefa Z, YG/yesuse A. Acute appendicitis in children admitted to Zewditu memorial hospital. Ethiop Med J 2014;52(4):189-95.
  30. Bendeck SE, Nino-Murcia M, Berry GJ, et al. Imaging for suspected appendicitis: neg- ative appendectomy and Perforation rates. Radiology 2002;225(1):131-6.
  31. Yeh Benson. Does this adult patient have appendicitis? Ann Emerg Med 2008; 52(3):301-3.
  32. Mc Cabe K, Babl FE, Dalton S. Management of children with possible appendicitis: a survey of emergency physicians in Australia and New Zealand. Emerg Med Australas 2014;26(5):481-6.
  33. Bates MF, Khander A, Steigman SA, et al. Use of white blood cell count and negative appendectomy rate. Pediatrics 2014;133(1):e39-44.
  34. Anandalwar SP, Callahan MJ, Bachur RG, et al. Use of white blood cell count and polymorphonuclear leukocyte differential to improve the predictive value of ultra- sound for suspected appendicitis in children. J Am Coll Surg 2015;220(6):1010-7.
  35. Leeuwenburgh MM, Stockmann HB, Bouma WH, et al. A simple clinical decision rule to rule out appendicitis in patients with non-diagnostic ultrasound results. Acad Emerg Med 2014;21(5):488-96.

Leave a Reply

Your email address will not be published. Required fields are marked *