Abstract
Acute appendicitis is among the commonest surgical diseases faced by an emergency clinician. A prompt diagnosis is mandatory to reduce morbidity and mortality, which may be difficult for junior doctors who first see the patient in the emergency department. This study was conducted to assess the efficacy of Alvarado score as a tool to improve the accuracy of diagnosis of appendicitis. We did a prospective pre- and post-test study in 381 patients with right iliac fossa pain in the emergency department of BP Koirala Institute of Health Sciences. Alvarado scoring was used for the post-test study. We found that the Alvarado scoring system is an effective diagnostic method, which decreases overdiagnosis of appendicitis in the emergency department, and at the same time reduces the risk of missing acute appendicitis cases. It is inexpensive and easy to implement by junior doctors with good results.
Introduction
Junior house officers with less than 12 months' experience still staff the emergency departments of hospitals in most parts of Nepal. Delays in diagnosis or a wrong diagnosis can lead to problems in the management of an emergency. Appendicitis is a typical emergency condition that can be used to flag the accuracy of the diagnosis of the emergency department staff, as many patients present atypically and it is, therefore, imperative to diagnose and manage them early in order to avoid a high rate of complications.
In 1986, Alvarado developed a scoring system based on the important clinical parameters of acute appendicitis 1 and several studies have subsequently reported the importance of this scoring system in the diagnosis of appendicitis and also in order to exclude other differential diagnoses of right iliac fossa pain. However, the usefulness of this diagnostic method in the hands of inexperienced doctors in hospitals in developing countries has not so far been assessed.
The present study examines the way in which the application of the Alvarado scoring system has affected the diagnosis of acute appendicitis and the subsequent correlation with the clinical, operative and histopathological diagnoses.
Patients and methods
The term ‘pre-test’ was used to cover the ability of doctors to diagnose appendicitis before the scoring system had been introduced; ‘post-test’ covers their ability after the introduction of the scoring system.
The Alvarado score was introduced to the junior doctors after 100 patients with right iliac fossa pain underwent appendicectomy. They were shown how to complete the proforma. When they had been trained, the following 100 cases undergoing appendicectomy were recorded. The diagnosis of acute appendicitis using the Alvarado score was correlated with peroperative findings, histopathology and the diagnosis at the time of discharge.
Two sets of questionnaires were used by junior doctors during the emergency, one for pre- and the other for post-test. The variables studied were: name; sex; age; date; and initial diagnosis after the evaluation of the clinical features. The post-test questionnaire also recorded the clinical signs, symptoms and laboratory parameters used to calculate the Alvarado score (Table 1). The diagnoses made during the emergency were then compared with the final diagnoses which had been made based on the peroperative findings of those cases that subsequently underwent appendicectomy, histopathological findings and the final diagnosis on discharge by the operating surgeon. An appendix which was either inflammed with features of oedema, ulceration, gangren, or grossly inflammed cells with areas of necrosis microscopically was considered as appendicitis. 2
Alvarado scoring system
Statistical methods
The difference in diagnosis in pre- and post-test groups was analysed by Epi Info 2000 using 2x2 tables for sensitivity, specificity, positive predictive value and negative predictive value.
Results
The total number of patients in the study were 381, including 210 cases for pre-test analysis and 171 cases for post-test analysis.
Pre-test analysis
An analysis was performed on 210 patients with right iliac fossa pain. The youngest patient was 6 years old and the oldest was 84. The mean age of patients in this group was 30.8. The majority of patients were aged between 11 and 40 peaking in the 21–30 age group. There were more males than females in the sample.
Of 210 patients, the initial diagnosis of appendicitis was made in 85 (40.4%), possible cases of appendicitis in 96 (45.8%) and some other diagnoses in 29 (13.8%). In the initial appendicitis diagnosis group, the final diagnosis of appendicitis was made for 56 (65.8%) and another diagnosis in 29 (34.2%). For the group who were diagnosed as possibly having appendicitis, 50 (52.1%) did have appendicitis and 46 (47.9%) did not. Of the group who were diagnosed as not having appendicitis, the final diagnosis of appendicitis was made in seven (24.1%).
In the pre-test group, who did not use the Alvarado score, the sensitivity of the duty medical officer and the diagnostic accuracy for ‘definite sure appendicitis’ was 49.6% (95% confidence interval [CI]: 40.5%–58.6%), specificity 70.1% (95% CI: 60.4%–78.3%), positive predictive value 65.9% (95% CI: 55.3%–75.1%) and negative predictive value 54.4% (95% CI: 45.7%–62.9%); the percentage for accurate diagnoses was 59.04%.
Post-test analysis
An analysis was performed on 171 patients of right iliac fossa pain. The clinical parameters included in the Alvarado scoring system were taken into consideration. The youngest patient in this group was 6 years old and the oldest was 90. The mean age of patients in this group was 28.96.
In 171 patients, an initial diagnosis of appendicitis was made in 116 (67.8%) patients, possible appendicitis in 18 (10.5%) and another diagnosis in 37 (21.7%). In the group who were initially diagnosed as having appendicitis, the final diagnosis was appendicitis for 111 (95.6%) patients and five (4.4%) had another diagnosis. In the group who diagnosed a possible appendicitis, the final diagnosis was appendicitis in seven (38.8%) and 11 (61.2%) had another diagnosis. The remaining 37 were subsequently found not to have appendicitis.
Using the Alvarado score, in the post-test group the sensitivity of the duty medical officer's diagnosis of ‘sure appendicitis’ was 94.1% (95% CI: 88.3%–97.1%), specificity 90.6% (95% CI: 79.7%–95.9%), positive predictive value 95.7% (95% CI: 90.3%–98.1%), negative predictive value 87.3% (95% CI: 76.0%–93.7%) and the accuracy of diagnosis was 93%.
None of the patients who were discharged as not having appendicitis came with complications such as appendicular abscess or peritonitis.
The presumptive diagnoses for the pre-test (210) and post-test (171) patients are shown in Table 2.
Presumptive diagnosis for the pre-test and post-test patients
Summary of the analysis
The summary of the analysis is shown in Table 3. A test of proportion was used to compare the various diagnostic characteristics of pre- and post-test analysis. In the case of frequencies less than 5, the Fisher's exact test was used.
Summary table for diagnosis of acute appendicitis
A positive predictive value is the probability that a patient with a positive test result really does have the condition for which the test was conducted.
A negative predictive value is the probability that a patient with a negative test result really is free of the condition for which the test was conducted.
Discussion
Acute appendicitis is mainly a disease of adults – the mean age of sufferers is between 20 and 30. 3,4 The disease does not have any particular predilection for either sex. 3–5 In this study, the mean age of the patients in the pre-test and post-test groups were 30.78 and 28.96, respectively, with a male to female ratio of 55.6% to 44.4%.
Patients suffering from acute appendicitis require prompt diagnosis and management, because of its relatively high rate of complications 6–8 and further morbidity and mortality. 9,10 Therefore, quality decision-making in the management of acute appendicitis can be evaluated in terms of the postoperative normal appendix rate or diagnostic accuracy as judged by the operative findings, histopathology report and the discharge diagnosis.
Various diagnostic modalities have been used in order to diagnose acute appendicitis. These include ultrasonography; computed tomography and clinical evaluation; with varying rates of sensitivity, specificity, and diagnostic accuracy. 11–15 Before the advent of scoring systems in the diagnostic evaluation of acute appendicitis, the rate of postoperative normal appendix rates were quite high because the appendicectomies were performed with the view to minimizing the incidence of perforation and higher mortality. 16,17 With time, the belief that it would be possible to reduce morbidity and mortality was being questioned due to the feared complications of appendicitis, especially when the cost of the operation had been taken into consideration. Thus, various scoring systems, using various clinical parameters and laboratory values, came into existence and have undergone review processes from time to time.
A postoperative normal appendix was found in 7% in a pre-test analysis and 5% in a post-test analysis of the total of operated patients. The postoperative normal appendix in patients undergoing laparotomy for right iliac fossa pain ranged from 3% to 17.5% in other studies. 17,18 According to different authors, the acceptable negative appendicectomy rate, before the advent of scoring system, ranged from 10% to 30%. 16,17,19,20 The postoperative normal appendix rate in our study in the pre-test group was well below the acceptable value.
The duty medical officer diagnosed appendicitis in 65.8% cases without using any scoring system. In our study, the false positive rate was 34.2% and the false negative rate was 52.1%. There are various studies, both comparative as well as non-comparative, which have investigated the diagnostic accuracy of various scoring systems, but there are few studies of cases diagnosed by duty medical officers which did not use any scoring system as a control group. Considering the high false positive rate and false negative rate, potentially resulting in subsequent morbidity in the form of normal appendicectomy, the need for an appropriate scoring system using different clinical and easily available laboratory data is required.
In a previous study, it was found that no patient with a score below 5 had appendicitis 17 but in our study four cases (11%) proved to have appendicitis even with an Alvarado score of 4 or 5. With a score of 3 or less it can confidently be said that there is no appendicitis. If the score is between 4 and 5, the patient needs observation and a review after 4–6 h and a re-evaluation of the score. We operated on patients with a score of 6 or more.
The Crnogorac et al. study found that the Alvarado score has a high sensitivity of 87%, average specificity of 60% and high diagnostic value of 82.7%. 21 In the Chan et al. study, the positive and negative predictive values of Alvarado score were 77% and 97.6%, respectively. 22 Denizbasi et al. found that the sensitivity and specificity of the Alvarado score were 95.4% (95% CI: 91%–99%) and 45.7% (95% CI: 37.4%–54%), respectively. 5
The higher sensitivity and specificity found in our study is because ‘sure appendicitis’ and ‘suspected appendicitis’ have been distinguished. When ‘sure appendicitis’ and ‘suspected appendicitis’ were combined the sensitivity was 99.6% (95% CI: 96.1%–100.0%) and the specificity was 69.4% (95% CI: 56.2%–80.1%) which is closer to the results reported in other studies.
Comparing the pre-test and post-test groups, there was an improvement in diagnostic certainty from ‘suspected (?) appendicitis’ towards ‘sure appendicitis’ or ‘other diagnosis’. Sensitivity increased from 49.6% to 94.1% and specificity increased from 70.1% to 90.6%. Both the positive predictive value and the negative predictive value also improved significantly.
We concluded that the Alvarado score is a non-invasive, safe diagnostic procedure, which is simple, reliable and repeatable. It can be used in all conditions and is an inexpensive and uncomplicated supportive diagnostic method. The implementation of the use of the Alvarado score was shown to increase the certainty with which junior doctors made a diagnosis of acute appendicitis in the emergency department. This increase in performance will accrue benefits for patient care. It is recommended that the Alvarado score be used as a matter of policy in emergency departments. In cases where there is a paucity of diagnostic tools, clinical re-evaluations every 4–6 h are strongly recommended.