Endoscopic retrograde appendicography: an effective diagnostic method for acute appendicitis

Background and Aim: Appendicography had been used in the diagnosis of chronic appendicitis. To our knowledge, the role of endoscopic retrograde appendicography for the diagnosis of acute appendicitis remains unknown. The aim of this study was to evaluate the role of endoscopic retrograde appendicography for the diagnosis of acute appendicitis. Patients and Method: Patients with suspected acute appendicitis between December 2013 and November 2015 at Second Aliated Hospital of Harbin Medical University underwent endoscopic retrograde appendicography. The ndings and complication were analyzed retrospectively. Results: Thirty-three patients (20 men and 13 women, average age 44 ± 18 years) with suspected acute appendicitis were studied. Acute appendicitis was ruled out by normal endoscopic retrograde appendicography in 24% and conrmed in 69.6% (23). In 2 patients (8%) appendiceal orice cannulation failed. Colonoscopic ndings in acute appendicitis were mucosal hyperemia and edema of appendiceal orice (83%), outpouring of pus from the appendiceal orice (74%), and swollen cecal mucosa (61%). Appendicograpic ndings were either normal or in acute disease showed diffuse lumenal dilation (diameter: 0.8 ± 0.4 mm), partial stenosis (43%), stiffness or inexibility (87%) and lling defects (22%). There were no complications during or after follow-up for a median of 13 months (IQR: 9-24 months). Conclusions: Endoscopic retrograde appendicography appears to be a reliable and safe method to conrm or exclude the diagnosis of acute appendicitis and prevent unnecessary appendectomy.


Introduction
The preoperative diagnosis of acute appendicitis (AA) is primarily based on clinical evaluations that is often only resolved by appendectomy. The rate of false positive diagnoses of acute appendicitis has remained unchanged following the introduction of new imaging techniques such as multidetector enhanced computed tomography (CT) or ultrasonography (US) [1,2]. While false positive tests to negative appendectomy , false negative diagnoses may result in delayed management and complications of the disease. A more accurate diagnostic method is needed especially when the initial diagnostic approach is not de nitive.
Colonoscopic appendicography has been reported as a useful technique in differentiate indistinct abdominal symptoms [3]. Endoscopic retrograde appendicitis therapy (ERAT) was rst reported in 2012 as a new procedure for management of acute appendicitis [4]. The rst step in ERAT is endoscopic retrograde appendicography (ERA) performed to con rm the diagnosis of acute appendicitis. The aim of this study was to systematically evaluate the e cacy of endoscopic retrograde appendicography in the diagnosis of acute appendicitis.
Consecutive patients with suspected acute uncomplicated appendicitis were offered endoscopic retrograde appendicography and ERAT at the Second A liated Hospital of Harbin Medical University (a tertiary care center in China) between December 2013 and November 2015. Inclusion criteria included: patients clinically suspected to have acute appedicitis (i.e., Alvarado scores ≥ 5) [5], and who were reluctant to undergo operative appendectomy but consented to receive ERAT. Exclusion criteria included those with perforated appendicitis or periappendiceal abscess con rmed by non-enhanced CT (16detector-row) or US. In preparation for endoscopic retograde appendicography, bowel preparation was done using either 2L polyethylene glycol electrolyte solution or low-pressure cleansing enemas (300-500 mL per enema) given ve times. For patients with mild and moderate symptoms the oral prep was given 4-6 hour before the procedure. For clinically severe cases or patients with anorexic or nauseous/vomiting low-pressure cleansing enemas (300-500 mL per enema) were recommended approximately 30 minutes prior to endoscopy so as not to delay treatment of appendicitis.
The study was performed according to the principles of the Declaration of Helsinki, and was approved by the Institutional Review Board of the Second A liated Hospital of the Harbin Medical University. The written informed consent was obtained from the patients before the study.

Technique procedures
Endoscopic retrograde appendicography and ERAT were performed as described previously. Brie y, endoscopic retrograde appendicography involves an colonoscope (CF-H260 or GIF-Q260J, Olympus, Japan) with an attached transparent cap introduced. After nding Gerlach's valve in the cecum, the colonoscope was positioned close to the appendiceal ori ce. Gerlach's valve was pushed aside using the transparent cap so that the tip of a catheter (OE-104-2225DL, EndoFlex, Germany) could be wedged into the appendiceal ori ce. A 0.035 inch guide wire (Boston Scienti c, US) was then inserted into the appendiceal lumen over the catheter and placed deep into the lumen under uoroscopic guidance.
Decompression of the appendiceal lumen was achieved by suction using a syringe (5 mL) attached to the catheter. A soluble contrast agent was then infused to ll the appendix while being monitored by uoroscopy to check the location, length, shape, content and exibility of the appendix. Successful ERAT was de ned as resolution of abdominal pain and normalization of the leukocyte count after treatment. If abdominal pain was not relieved within 48 hours after ERAT, emergency appendectomy was performed.
The reference standard for con rming acute appendicitis was de ned as symptoms resolution after successful ERAT with subjects remaining recurrence-free during follow-up. In the cases of failure ERAT, or recurrence after ERAT, subsequent appendectomy and histological ndings were consistent with acute appendicitis. Patients without appendicitis included those for whom another disease was diagnosed or who uneventfully recovered after corresponding treatment based on the current guidelines .
The primary outcome measures were the performance characteristics (sensitivity, speci city, positive and negative predict values) of endoscopic retrograde appendicography in the patients with suspected acute appendicitis. The diagnostic features of endoscopic retrograde appendicography are shown in Table 1.
The diagnosis of acute appendicitis was made when the endoscopic retrograde appendicography features of the patient matched (1) or (2) (3) as shown in Table 1.
The secondary outcome measures included complications, appendiceal cannulation rate, colonoscopy and endoscopic retrograde apendicography ndings and the accuracy of CT and US.
Telephone follow-up was conducted to assess for recurrent symptoms and long-term complications until the end of the study period. If any symptoms were present, such as abdominal pain, fever or other digestive symptoms, the patient was recommended to return for further examination, laboratory tests and US or CT, as necessary.

Statistical analyses
Results were expressed as mean ± standard deviation (SD) or median (inter-quartile range, IQR). Quantitative variables were compared between the two groups using the t-test or the Mann-Whitney U test, as appropriate. Categorical variables were compared between the two groups using the Fisher's exact test. A two-tailed P value <0.05 was considered to be statistically signi cant. Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) version 17.0 for Windows.

Results
The 33 consecutive patients (20 men and 13 women, average age 44 ± 18 years) were enrolled. The patient characteristics of patients are shown in Table-2. In patients with clinically suspected acute appendicitis, the median of Alvarado scores was 7 (IQR: 5.5-8); 25 of the 33 patients (75.7%) were diagnosed as de nite acute appendicitis. In 8 patients (24.2%) acute appendicitis was excluded on the basis of reference standards.
All the patients with con rmed acute appendicitis by endoscopic and endoscopic retrograde appedicography recovered uneventfully. The diagnosis of acute appendicitis in the 2 patients in whom the appendical or ce could not be canculated were con rmed on the basis of appendectomy and histology. We included these 2 failure cases in the equivocal ERA results. During follow-up (median: 13 months; IQR: 9-24 months), 2 (8%) of the patients with acute appendicitis subsequently underwent appendectomy for recurrent abdominal pain and were histologically diagnosed with acute appendicitis.
The diagnoses of the 8 patients without acute appendicitis are shown in Table 4. All 8 patients uneventfully recovered after corresponding treatment and no recurrences occurred during follow-up. One attempted endoscopic retrograde appendicography had failed because the appendiceal ori ce was blocked by a malignant tumor. We also included the patient with carcinoma in the equivocal negative results. Overall, in 8 patients (24%) acute appendicitis was de nitely ruled out through endoscopic retrograde appendicography.
Overall, endoscopic retrograde appendicography correctly diagnosed 23 patients with acute appendicitis and ruled it out in 8 cases; 2 cases had equivocal results due to failure appendix cannulation. If the equivocal results were included, the sensitivity and speci city of ERA were 92% and 83%, respectively. The positive predict value and negative predict value of ERA were 92% and 83% respectively. Furthermore, there were no complications during a median follow-up period of 13 months (IQR:9-24 months) .

Comparison with CT and ultrasound
In the study, CT scans were performed in 24 patients and US in 17 before endoscopic retrograde appedicography; both CT scan and US were performed in 7 patients. With US, 6 patients (35%) were incorrectly diagnosed and 2 other patients (12%) had equivocal results. Four patients (17%) were misdiagnosed and 8 patients (33%) had equivocal results using CT scans. Even if the equivocal results of CT and US were excluded from the calculations, the sensitivity and speci city of CT was 88% and 71% respectively. The sensitivity and speci city of US was 73% and 50% , respectively. The sensitivity and speci city of the endoscopic retrograde appendicography results were 100% and 100%, respectively (when equivocal results were included , 92% and 89% respectively) (

Discussion
Acute appendicitis is one of the common causes of acute abdominal pain. Although the classical symptoms of acute appedicitis is periumbilical pain migrating to the right lower quadrant, the presentation is rarely typical and diagnostic errors are common resulting in both negative appendectomies or appendical perforations. This study systematically investigated the value of endoscopic retrograde appendicography in the diagnosis of acute appendicitis and showed a high diagnostic yield in patients with acute abdominal pain. Possibly more importantly was the ability to prevent unnecessary operations and preserve the normal appendix.
Previous studies have reported colonoscopy as a useful diagnostic method of acute appendicitis in the patients with atypical presentation of abdominal pain or with non-diagnostic imaging studies [6,7]. Chang et al. [7] described the colonoscopic features of acute appendicitis, including hyperemia and bulging at the appendiceal ori ce area with surrounding mucosal edema and drainage of pus from the appendcieal ori ce. Our study con rms the value of these features. However, some patients with acute appendicitis did not have any of these colonoscopic features. In our study, although there were 17 patients (74%) with pus discharge from the appendiceal ori ce, only 4 (23%) had spontaneous pus discharge. In 13 (77%) pus discharge only occurred after successful cannulation of appendiceal ori ce.
Moreover, one patient presented with normal mucosa around the area of appendiceal ori ce which is consistent with the fact that in some patients the in ammatory changes may affect only the distal appendix [8]. Thus, colonoscopy alone is unable to reliably predict which patients have an abnormal appendiceal lumen. In contrast, appendicography combined with colonoscopy allowed acquisition of clear images both of appendiceal ori ce and lumen for providing more accurate diagnosis.
The advantage of endoscopic retrograde appendicography over traditional barium enema lies in higher success rate of lling appendix. The diagnosis of acute appendicitis by barium enema is mainly based on non-lling of the full appendix [9]. However, 15-23% of normal appendices fail to ll [10,11] such that failed lling of the appendix is not a reliable sign for acute appendicitis [12]. On the contrary, with a remarkable success (91%) of lled appendix, endoscopic retrograde appendicography easily identi ed the presence of lumen dilatation, partial stenosis, lack of exibility and intraluminal lling-defects and thus can reliably con rm the diagnosis of acute appendicitis.
Another advantage of endoscopic retrograde appendicograpgy over CT or US is that it provides radiological imaging of appendix with more objectivity, irrespective of shape and position of appendix within abdominal cavity. Although some studies have reported high sensitivity and speci city of CT and US [13][14][15], Wilson et al. described the equivocal rates of CT (28%) and US (75%) [16], which is consistent with the result of CT (33%) in the present study. However, the equivocal rate of US (12%) was lower than the previous study, maybe due to the small sample size. US is more operator dependent, relying both on the US technician and the interpretation by the radiologist. Even though CT is more objective, the diagnostic accuracy is also limited by technical and interpretative pitfalls due to the shape and position of appendix inside the abdomen [17]. Actually, clinicians have reported potential negative appendectomy rates of 22% due to CT [18]. Another reason for relative lower accuracy of CT and US in our study is that many cases underwent endoscopic retrograde appendicograpgy because of equivocal results of CT scan or US. Moreover, CT exposes the patient to more ionizing radiation than radiography and is associated with a risk of radiation-induced cancer.
Another advantage of endoscopic retrograde appendicography compared to CT or US lies in differential diagnosis of abdominal pain mimicking acute appendicitis. During the procedures, the endoscopist can visualize the entire colon or appendix and make histological diagnosis by biopsy. In the present study, endoscopic retrograde appendicography identi ed 4 patients with ileocecal lipoma, ulcerative colitis, colon and appendix carcinoma and typhlitis. However, further study is necessary to obtain a full comparison of the diagnostic accuracy of endoscopic retrograde appendicography, CT and US.
Previous studies have also reported that the patients with an atypical presentation were diagnosed as acute appendicitis using colonoscope [19]. However, colonoscopy was not widely used for diagnosis of acute appendicitis, because the conventional wisdom is that colonoscope may aggravate abdominal pain and induce complications. However, there were no complications in this study. In contrast, the results showed that endoscopic retrograde appendicography can be safely performed on the patients with suspected acute uncomplicated appendicitis.
Endoscopic retrograde appendicography procedure can be performed in the outpatient department. Initially, we admitted patients after undergoing ERAT as this was a preliminary study. Now, in China more than 3000 cases of ERAT (using ERA as a diagnostic tool) have been preformed and it is recognized that it is not necessary to admit the patients to hospital. The procedure can also be performed in outpatient settings which is currently widely practiced in China. This is a preliminary study with a limited sample size, a prospective controlled randomized trial is necessary to provide more valuable information for clinical implication of endoscopic retrograde appendicography. However, with the development of colonoscopic technology, we believe this method will become the most reliable method of diagnosis for acute appendicitis in the future.

Conclusion
Our study suggests that endoscopic retrograde appendicography can accurately diagnosis and exclude acute appendicitis. Endoscopic retograde appendicography provides the physicians with clear images of both appendiceal ori ce and lumen which other methods cannot do. However, further study is needed to compare the e cacy of ERA with other radiological examinations.

Abbreviations
Endoscopic retrograde appendicography (ERA); Endoscopic retrograde appendicitis therapy (ERAT); Acute appendicitis(AA); Computed tomography (CT); Ultrasonography (US); Barium enema (BE) Declarations Ethics approval and consent to participate The present study was undertaken under the guidence of the ethics comittee of The First A liated Hospital of Zhengzhou University. The authors have declared that no ethical con icts. The written informed consent for participation in the study was obtained where participants are children (under 16 years old) from their parent or guardian.   Data are numbers (%) and mean ± deviation(SD). * Pus discharge includes spontaneously discharge in 4 patients and discharge after appendiceal intubation in 12 patients. CA, contrast agents  respectively (when equivocal results were included , 53% and 42% respectively). The sensitivity and specificity of the ultrasonography (US) results were 73% and 50% , respectively (when equivocal results were included , 73% and 17% respectively). The sensitivity and specificity of the endoscopic retrograde appendicography (ERA) results were 100% and 100%, respectively (when equivocal results were included , 92% and 89% respectively)