The accuracy of emergency medicine and surgical residents in the diagnosis of acute appendicitis

Transcription

1 American Journal of Emergency Medicine (2010) 28, Original Contribution The accuracy of emergency medicine and surgical residents in the diagnosis of acute appendicitis You Hwan Jo MD a, Kyuseok Kim MD, PhD a,, Joong Eui Rhee MD a, Tae Yun Kim MD a, Jin Hee Lee MD a, Sung-Bum Kang MD b, Duck-Woo Kim MD b, Young Hoon Kim MD c, Kyoung Ho Lee MD c, So Yeon Kim MD c, Christopher C. Lee MD d, Adam J. Singer MD d a Department of Emergency Medicine, Seoul National University Bundang Hospital, Gyeonggi-do , Korea b Department of Surgery, Seoul National University Bundang Hospital, Gyeonggi-do , Korea c Department of Radiology, Seoul National University Bundang Hospital, Gyeonggi-do , Korea d Department of Emergency Medicine, Center for Internal Emergency Medicine, Stony Brook University Hospital, Stony Brook, NY 11794, USA Received 28 January 2009; revised 17 March 2009; accepted 19 March 2009 Abstract Objective: This study was conducted to compare the diagnostic accuracy for acute appendicitis between emergency medicine residents (EMRs) and surgical residents (SRs). Methods: We conducted a prospective cohort study of adult patients with right lower quadrant pain. Each patient was evaluated by an EMR and an SR, and physicians predicted the probability of appendicitis into 4 groups from highest (group 1) to lowest (group 4). The diagnostic accuracies of EMR and SR for the diagnosis of appendicitis were compared by constructing receiver operating characteristics curves. In each case, an Alvarado score was calculated and a computed tomography (CT) scan of the abdomen and pelvis was performed, and their diagnostic accuracies were also compared with the predicted probabilities. Results: Of a total 191 patients, 120 underwent surgery, and the negative appendectomy rate was 6.8%. There was a significant correlation between the predicted probabilities of EMR and SR. The areas under the curve for EMR and SR were and 0.657, which were not statistically different. The areas under the curve of the Alvarado score and the CT were and 0.978, respectively. The diagnostic accuracy of the CT scan was significantly higher than those of the Alvarado score and the resident-predicted probabilities. Conclusion: In patients with right lower quadrant abdominal pain who have already been evaluated by EMR, consultation evaluation by SR does not appear to improve clinical diagnostic accuracy, and routine performance of CT before surgical consultation should be considered for these patients Elsevier Inc. All rights reserved. Corresponding author. Tel.: ; fax: address: dremkks@snubh.org (K. Kim) /$ see front matter 2010 Elsevier Inc. All rights reserved. doi: /j.ajem

2 Accuracy of EMR and SR in diagnosis of acute appendicitis 1. Introduction Acute appendicitis is one of the most common surgical emergencies presenting to the emergency department (ED) [1,2]. Although early diagnosis and surgical intervention are mandatory to prevent complications, clinical evaluation such as history, physical examination, and laboratory tests are not always accurate. It has been reported that the accuracy of clinical diagnosis for acute appendicitis ranges from 71% to 97% [1-3]. To reduce the number of negative appendectomies in patients without appendicitis, diagnostic methods such as the Alvarado score, ultrasonography, and computed tomography (CT) have been evaluated [4-6]. Traditionally, surgeons have been thought to be more skilled and experienced in the diagnosis of acute appendicitis than other physicians [7,8]. Therefore, early surgical consultation is recommended to evaluate patients with suspected acute appendicitis to minimize complications such as perforation or misdiagnosis. However, early surgical evaluation, although ideal, may not always be available. Furthermore, few reports have compared the diagnostic accuracies of emergency physicians and surgeons in acute appendicitis. This study was conducted to compare the accuracy of diagnosing acute appendicitis between emergency medicine residents (EMRs) and surgical residents (SRs). We also compared the diagnostic accuracy of clinical evaluation to those of the Alvarado score and the CT findings. We hypothesized that the diagnostic accuracies of EMRs and SRs would be similar and lower than that of CT imaging. 2. Methods 2.1. Study design and setting We conducted a prospective cohort study of patients with right lower quadrant abdominal pain. This study was approved by the institutional review board of our hospital. Informed consent was obtained from all patients. This study was conducted at an urban, tertiary care ED with an annual census of from August 2006 to October Study subjects We enrolled consecutive patients who presented to the ED with pain to the right lower quadrant of the abdomen [9]. Patients who were younger than 15 years or transferred from other hospitals whose diagnosis had been already confirmed were excluded Study protocol and data acquisitions All patients were initially evaluated by an intern, and if the patient had right lower quadrant pain, immediate consultation with an EMR and an SR was performed for all patients. Despite evidence that timely analgesia may actually improve the diagnosis, in this case, pain control was not performed before surgical evaluation. In addition, all clinical evaluations by EMR and SR were performed before any laboratory tests or CT imaging of the abdomen and pelvis. Physicians were asked to predict the probability that the patients had acute appendicitis into 1 of 4 groups (group 1, 2, 3, and 4) corresponding to clinical probabilities of 80% to 100%, 60% to 79%, 40% to 59%, and 20% to 39%, respectively. Standardized data collection forms were used throughout the study. The separate forms of clinical probability of acute appendicitis were completed by EMRs and SRs before any laboratory and radiologic studies, and physicians were blinded to another's impressions on each case. Another data collection form included demographic information, symptoms and signs of the patients, and laboratory test results. The calculated Alvarado score was not included in the data collection forms, but all of the elements of the Alvarado score were included. The Alvarado score ranges from 0 to 10 from least probable to most probable [4]. After obtaining the clinical probability, abdominal CT imaging (Brilliance, Phillips Medical Systems, Cleveland, Ohio) was performed on patients unless patients had contraindications such as pregnancy, renal insufficiency, and allergy to the contrast medium. Intravenous nonionic contrast material (Ultravist 370, Schering, Berlin, Germany) was infused at a dose of 2 ml/kg. Oral or rectal contrast material was not administered. All CT images were interpreted by 3 board-certified body imaging radiologists, and the CT scans were scored on a 5-point scale from grade 1 (normal appendix) to grade 5 (definite appendicitis). The disposition of the patients was determined by an SR supervised by a board-certified surgeon after considering the history, physical examination, laboratory tests, and the CT results Outcomes The primary outcome was the comparison of diagnostic accuracies between EMR and SR. Final diagnosis of appendicitis was based on surgical pathology for patients who underwent surgical exploration. For patients who did not undergo exploration, a structured telephone follow-up was performed 3 months after the ED visit. Patients who did not require surgery within 3 months after discharge were considered not to had have appendicitis on the initial ED visit. Secondary outcomes included the diagnostic accuracies of the Alvarado scores and the CT imaging Data analysis 767 The linear regression analysis was used to examine the correlation between the predicted probabilities of EMR and SR. The diagnostic accuracy for each group of probability

3 768 Y.H. Jo et al. Table 1 Baseline characteristics of the patients Total (N = 191) Age (y) 37.3 ± 16.7 Male 87 (45.6%) Symptoms Migration to the right lower quadrant 95 (49.7%) Anorexia 89 (46.6%) Nausea/vomiting 101 (52.9%) Signs Tenderness in the right lower quadrant 178 (93.2%) Rebound tenderness 84 (44%) Temperature ( C) 36.8 ± 0.7 Temperature 7.3 C 28 (14.7%) White blood cell count (/μl) ± White blood cell count /μL 126 (66.0%) Segmented neutrophil (%) 76.8 ± 11.4 Segmented neutrophil 75% 121 (63.4) Data are expressed mean ± SD or number (percentage) as appropriate. was compared using the χ 2 or Fisher exact test as appropriate. We performed receiver operating characteristic curve analysis to compare the diagnostic characteristics of EMR, SR, the Alvarado score, and the CT scan. The area under the curve (AUC) was calculated and a univariate Z test was used to compare the AUC as described by Hanley and McNeil [10]. Statistical analyses were conducted using SPSS software version 13.0 (SPSS Inc, Chicago, Ill). Medcalc (MedCalc Software, Belgium) was used to compare the AUC. A P value of less than.05 was considered statistically significant. Table 3 Comparison of diagnostic accuracy between emergency medicine and SRs Group EM (N = 191) Surgery (N = 191) P 1 30/37 (81.1) 32/41 (78.1) /89 (66.3) 47/76 (61.8) /41 (34.2) 22/45 (48.9) /24 (33.3) 10/29 (34.5).930 Data are expressed as number of appendicitis cases/number of total patients (percentage). EM indicates emergency medicine. symptoms, and 93% of patients had tenderness in the right lower quadrant of the abdomen. The signs and symptoms used to calculate the Alvarado score are shown in Table 1. Of a total of 191 patients, 120 (62.8%) patients underwent surgical exploration and 71 (37.2%) were discharged home. Of the 120 patients who underwent exploration, one patient was diagnosed with an ovarian torsion before operation and the surgery was performed by a gynecologist. The remaining 119 patients went to the operating room with the impression of acute appendicitis. Of these patients, 111 (93.2%) were confirmed to have appendicitis by pathologic findings, and one was found to have a mucinous tumor. Telephone followup was completed on all patients who did not undergo operation, and there was no additional case of acute appendicitis within the 3-month follow-up period. 3. Results During the study period, 278 consecutive patients with pain to the right lower quadrant of the abdomen were enrolled. Of these patients, 87 were not evaluated by an SR before CT imaging and were excluded. Therefore, 191 patients were included into the final analysis. Their mean age was 37.3 ± 16.7 years and 87 patients (45.6%) were male. Nausea and vomiting were the most common presenting Table 2 Comparison of the probability groups between EMR and SRs Group Surgery (N = 191) Total Emergency medicine (N = 191) Total Fig. 1 Receiving operating curve characteristics of EMRs, SRs, the Alvarado score, and the CT scan for diagnosing acute appendicitis. The area AUC of the CT is (95% confidence interval, ) and is significantly higher than those of EMRs, SRs, and the Alvarado score (0.698 [95% CI, ], [95% CI, ], and [95% CI ], respectively; P b.001).

4 Accuracy of EMR and SR in diagnosis of acute appendicitis Fig. 2 The Alvarado scores in patients with or without acute appendicitis. During the study period, second-, third-, and fourth-year EMRs and second- and third-year SRs participated in the study. The clinical probabilities of acute appendicitis predicted by EMR and SR are presented in Table 2. The linear regression revealed a correlation coefficient of (P b.001), and there was a significant correlation for the clinical diagnosis of acute appendicitis between EMR and SR. The diagnostic accuracies in each probability category were not significantly different between EMR and SR (P N.05 for all, Table 3). The receiver operating characteristic analysis revealed that the overall diagnostic accuracy was not different between EMR and SR (P =.33, Fig. 1). Alvarado scores were calculated in all enrolled patients, and the mean score was 5.9 ± 2.0 (Fig. 2). The AUC of the Alvarado score was 0.735, which was similar to those of EMR and SR (P =.44 and 0.14, respectively; Fig. 1). A CT scan of the abdomen and pelvis was performed in 187 patients (Fig. 3). Of the 4 patients who did not undergo CT imaging, 3 underwent ultrasonography because of pregnancy and the remaining patients had no imaging study. The accuracy of CT imaging in the diagnosis of appendicitis was superior to those of EMRs, SRs, and the Alvarado score (P b.001, Fig. 1). 769 before and after study. In a study of Denizbasi and Unluer [12], the accuracy of EMR and SR using the Alvarado score was similar. However, no comparison was performed of the clinical impressions independent of the Alvarado score. Kharbanda et al [13] compared pediatric emergency physicians to SRs, and they reported no difference in the ability to clinically predict appendicitis. However, all study subjects were children, and all patients were first evaluated by an attending emergency physician, not by an EMR [13]. To the best of our knowledge, our study is the first report to compare the clinical diagnostic accuracy for acute appendicitis between EMR and SR in adult patients. In the present study, there was a significant correlation between the predicted probabilities of EMR and SR. A similar study that assessed the agreement for the diagnosis of appendicitis was conducted by Kharbanda et al [13], and the correlation coefficient of the study was The Alvarado score is a scoring system for the diagnosis of acute appendicitis [4]. Alvarado reported that patients with a score of 7 or higher had a 93% chance of acute appendicitis [4]. However, following studies reported lower diagnostic accuracy and suggested that the Alvarado score should be used in combination with imaging studies [9,12,14,15]. In the present study, the diagnostic accuracy of the Alvarado score was similar to the clinical diagnosis of EMR and SR. Several studies have demonstrated the benefit of CT imaging in the diagnosis of acute appendicitis [6,9,16-22]. The accuracy of the CT has been reported to be greater than 95% and may reduce the rates of negative appendectomy, delayed perforation, and unnecessary observation [9,17,21-23]. The AUC of the CT imaging in our study was 0.978, which is similar to other studies. The negative appendectomy rate in our study was 6.8%. We believe that the clinical diagnosis and the surgeon's decision to operate may be altered by the CT scan, but we did not compare the treatment decisions before and after CT imaging in the current study. It is possible that routine use of CT may delay time to operation 4. Discussion In this prospective study, the clinical diagnostic accuracy for acute appendicitis in patients with right lower quadrant pain did not differ between EMR and SR. In addition, the accuracy of CT imaging was highest compared to the clinical diagnosis and the Alvarado score. Few studies have compared the diagnostic accuracies of EMR and SR in patients presenting with suspected acute appendicitis. In one retrospective study, the negative appendectomy and perforation rates were similar for SRs and non-srs under the supervision of attending emergency physicians [11]. However, this study was a retrospective, Fig. 3 The grades of the CT scan in patients with or without acute appendicitis.

5 770 Y.H. Jo et al. thereby increasing the perforation rate. However, Kim et al [21] reported that the times to operation whether CT was performed or not were not different, and perforation rate of our study was only 4.5% (5 cases). 5. Limitations This study has several limitations. First, this study was conducted at one institution and may not generalize to other institutions. Second, 87 patients were excluded because of not being evaluated by SR before CT scan, and these patients might influence the results. Third, we did not assess the accuracy of the attending physicians. We compared EMR to SR with different years of training in the diagnosis of acute appendicitis and did not control for the amount of the resident's prior experience in diagnosing acute appendicitis. Therefore, we cannot comment on the accuracy of attending surgeons and emergency physicians. However, because the initial evaluation of many patients in academic settings is performed by a resident, we chose to compare the clinical impressions between EMR and SR. Fourth, although the consultation with an EMR and an SR was performed immediately and simultaneously, the evaluation was done by an EMR and then by an SR, and we did not measure the time intervals from initial evaluation by an EMR to the evaluation by an SR. Thus, it is possible that the clinical evaluation changes during this interval. However, the surgical evaluation was always performed after that of the EMR and no pain analgesics were given. Therefore, the effect of time interval between EMR and SR evaluation on the impression of diagnosis would be insignificant. 6. Conclusions The diagnostic accuracy for acute appendicitis of EMR and SR in patients with right lower quadrant pain did not differ, and the clinical diagnosis of appendicitis between EMR and SR was significantly correlated. The diagnostic accuracy of CT imaging was significantly higher than those of the clinical diagnosis and the calculated Alvarado scores. These findings suggest that if an EMR suspects acute appendicitis, early surgical evaluation might not be mandatory, and CT imaging should be considered for a proper diagnosis. References [1] Andersson RE, Hugander A, Ravn H. Repeated clinical and laboratory examinations in patients with an equivocal diagnosis of appendicitis. World J Surg 2000;24: [2] Graffeo CS, Counselman FL. Appendicitis. Emerg Med Clin North Am 1996;14: [3] John H, Neff U, Kelemen M. Appendicitis diagnosis today: clinical and ultrasonic deductions. World J Surg 1993;17: [4] Alvarado A. A practical score for the early diagnosis of acute appendicitis. Ann Emerg Med 1986;15: [5] Sivit CJ, Newman KD, Boenning DA, et al. Appendicitis: usefulness of US in a pediatric population. Radiology 1992;185: [6] Gwynn LK. The diagnosis of acute appendicitis: clinical assessment versus computed tomography evaluation. J Emerg Med 2001;21: [7] Kosloske AM, Love CL, Rohrer JE, et al. The diagnosis of appendicitis in children: outcomes of a strategy based on pediatric surgical evaluation. Pediatrics 2004;113: [8] Sooriakumaran P, Lovell D, Brown R. A comparison of clinical judgment vs the modified Alvarado score in acute appendicitis. Int J Surg 2005;3: [9] Kim K, Rhee JE, Lee CC, et al. Impact of helical computed tomography in clinically evident appendicitis. Emerg Med J 2008; 25: [10] Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from same cases. Radiology 1983;148: [11] Liu CC, Lu CL, Yen DH, et al. Diagnosis of appendicitis in the ED: comparison of surgical and nonsurgical residents. Am J Emerg Med 2001;19: [12] Denizbasi A, Unluer EE. The role of the emergency medicine resident using the Alvarado score in the diagnosis of acute appendicitis compared with the general surgery resident. Eur J Emerg Med 2003; 10: [13] Kharbanda AB, Fishman SJ, Bachur RG. Comparison of pediatric emergency physicians' and surgeons' evaluation and diagnosis of appendicitis. Acad Emerg Med 2008;15: [14] McKay R, Shepherd J. The use of the clinical scoring system by Alvarado in the decision to perform computed tomography for acute appendicitis in the ED. Am J Emerg Med 2007;25: [15] Ohmann C, Yang Q, Franke C. Diagnostic scores for acute appendicitis. Abdominal pain study group. Eur J Surg 1995;161: [16] Balthazar EJ, Megibow AJ, Siegel SE, et al. Appendicitis: prospective evaluation with high-resolution CT. Radiology 1991;180:21-4. [17] Bendeck SE, Nino-Murcia M, Berry GJ, et al. Imaging for suspected appendicitis: negative appendectomy and perforation rates. Radiology 2002;225: [18] Rao PM, Rhea JT, Novelline RA, et al. Helical CT technique for the diagnosis of appendicitis: prospective evaluation of a focused appendix CT examination. Radiology 1997;202: [19] Rao PM, Rhea JT, Novelline RA, et al. Effect of computed tomography of the appendix on treatment of patients and use of hospital resources. N Engl J Med 1998;338: [20] Wijetunga R, Tan BS, Rouse JC, et al. Diagnostic accuracy of focused appendiceal CT in clinically equivocal cases of acute appendicitis. Radiology 2001;221: [21] Kim K, Lee CC, Song KJ, et al. The impact of helical computed tomography on the negative appendectomy rate: a multi-center comparison. J Emerg Med 2008;34:3-6. [22] Sun JS, Noh HW, Min YG, et al. Receiver operating characteristic analysis of the diagnostic 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: [23] Peck J, Peck A, Peck C, et al. The clinical role of noncontrast helical computed tomography in the diagnosis of acute appendicitis. Am J Surg 2000;180:133-6.