January 29, 2008
The Use of a Computed Tomography Scan to Rule Out Appendicitis in Women
By Lopez, Peter P Cohn, Stephen M; Popkin, Charles A; Jackowski, Julie; Michalek, Joel E
Diagnosing appendicitis continues to be a difficult task for clinicians. The use of routine CT scan has been advocated to improve the accuracy of diagnosing appendicitis. When compared with the use of clinical examination alone, CT scan was not significantly different with regard to making the diagnosis of appendicitis in women of childbearing age. The use of computed tomography in making the diagnosis of appendicitis has become the current standard of practice in most emergency rooms. In women of childbearing age, with possible appendicitis, we prospectively compared clinical observation alone (OBS) to appendiceal CT scan with clinical observation (CT). Ninety women (OBS: 48, CT: 42) with questionable appendicitis and an Alvarado Score ranging from two to eight were prospectively randomized. A true positive study/exam resulted in a laparotomy that revealed a lesion requiring operation (confirmed by pathology). A true negative exam/study did not require operation. Hospital stay (OBS = 1.9 +- 1.6 vs CT = 1.3 +- 1.4 days) and charges (OBS = $9,459 +- 7,358 vs CT = $9,443 +- 8,773) were similar. The OBS group had an accuracy of 93 per cent, sensitivity of 100 per cent, and a specificity of 87.5 per cent. The CT group had an accuracy of 93 per cent, sensitivity of 89.5 per cent, and specificity of 95.6 per cent. Although this study is too small to statistically establish equivalence, the data suggest that a CT scan reliably identifies women who need an operation for appendicitis and seems to be as good as clinical examination. DIAGNOSING APPENDICITIS continues to be a difficult task for physicians. This is particularly true in women of childbearing age. Many female patients who are suspected of having appendicitis are found to have other conditions, such as pelvic inflammatory disease, acute cholecystitis, urinary tract infections, ovarian cysts, and gastroenteritis.1 Not surprisingly, negative laparotomy rates for women of childbearing age are reported to be as high as 34 to 45 per cent.2,3
Patients and Methods
All female patients between the ages of 18 and 45 years presenting to the Surgical Emergency Room at Jackson Memorial Hospital with the possible diagnosis of acute appendicitis between November 1999 and February 2001 and from March 2003 to December 2004 were evaluated for inclusion in the study. The University of Miami institutional review board approved the experimental protocol for both times of the study. All patients gave informed consent. Patients were excluded if they were awaiting interval appendectomy, were unable to receive intravenous contrast for computed tomography, were HIV positive, had known inflammatory bowel disease, or were pregnant.
Those included in the study were evaluated by the surgical team, consisting of a junior and senior surgical resident and an in-house surgical attending, before the diagnostic work-up. Each patient received an Alvarado clinical score using the following grading system:14 symptoms (migration) (1), anorexia (1), nausea/vomiting (1), signs (tenderness right lower quadrant (RLQ) (2), rebound (1), elevated temp (1), laboratory (leukocytosis) (2), and left shift in polymorphonuclear leukocyte (PMN) (1), for a total of 10.
Using the Alvarado score, we categorized patients in one of three groups. Group 1 had a score greater than eight with a very high probability of having appendicitis. These patients had no further work-up, but proceeded to surgery, with pathological confirmation of appendicitis. Group 2 had a score of less than two with a very low probability of having appendicitis. These patients had no further surgical evaluation except for telephone follow-up to confirm the absence of appendicitis. Group 3 had a score of two to eight with an intermediate probability of having appendicitis. Patients were randomized to either clinical assessment alone or clinical evaluation with appendiceal CT. Randomization was done using a computer and the results were placed in numbered envelopes, opened sequentially when a patient met entry criteria.
All the patients assigned to the clinical assessment (OBS) were admitted to the surgical service and monitored with serial physical examinations and laboratory studies until there was resolution of symptoms, another diagnosis was confirmed (e.g., Pelvic Inflammatory Disease, gastroenteritis, urinary tract infection), or the patient underwent appendectomy for high clinical suspicion. All patients were seen and examined by the surgical team, including an attending surgeon, before discharge. Those patients who were discharged or transferred without having an appendectomy were followed up at 1 week with a telephone call.
Similarly, the patients in the CT arm underwent a CT scan followed by surgery or were discharged after being admitted for a period of observation. All patients who were observed after having a negative CT scan were discharged after resolution of their symptoms without having an appendectomy. These patients were followed up with a 1-week telephone call.
CT Technique and Interpretation
These patients had CT of the abdomen and pelvis with a helical GE HI Speed Advantage Scanner (GE Medical Systems, Milwaukee, WI) after oral administration of 900 mL 2.1 per cent barium sulfate suspension at least 2 hours before scanning. No rectal contrast was administered and all patients received a 125 mL intravenous bolus of Optiray(R)320 (Mallinckrodt Ine, St. Louis, MO). Seven millimeter sections were obtained from the lower lungs through the pelvis at a pitch of 1.5.13 No complications were associated with this radiologic protocol. The scans were interpreted by a senior radiology resident and/or a radiology attending according to previously recorded CT criteria for confirming appendicitis.13
Outcome Measures and Statistical Analysis
The primary endpoint was the presence or absence of appendicitis (confirmed by pathology). We also recorded the presence or absence of gangrenous appendicitis, abscess, and/or perforation. The length of hospital stay and charges were noted in all patients. With 75 patients/group, this study would have a power of 83 per cent to detect a 20 per cent difference in the percentage of patients misclassified (CT: 10%, OBS: 30%), assuming two-sided testing with a significance level of five per cent. Means were contrasted with t tests; if the assumptions of the test were violated, then a Wilcoxon test was used. Percentages were contrasted with Fisher's exact test. All statistical testing was 2-sided with a significance level of five per cent and SAS Version 9.1 for Windows (SAS Institute, Cary, NC) was used throughout.
From November 1999 to February 2001 and again from March 2003 to December 2004, a total of 95 women of childbearing age were considered for inclusion in the study. None of the study patients presented with an Alvarado score >8 on initial evaluation. There were two patients with an Alvarado Score <2 who were excluded from the study. Additionally, three patients refused randomization during this study period and thus, 90 patients composed our study population.
Forty-eight patients were randomized to OBS group and 42 to the CT group. In the OBS arm, two patients were excluded for protocol violations leaving 46 patients for final analysis (Fig. 1). The two protocol violations were because these two patients received a CT scan after being randomized to the observation group.
There were no differences (Table 1) between the two groups with regard to mean age (OBS: 28.8 years, CT: 27.9 years, P = 0.62), mean Alvarado score (OBS: 5.8, CT: 5.9, P = 0.85), number of perforations (OBS: O, CT: 0,P= 1.0), median white blood cell count (OBS: 13.4, CT: 12, P = 0.5), median length of stay (OBS: 1 day, CT: 1 day, P = 0.22), and median hospital charges (OBS: $10,900, CT: $10,100, P = 0.3).
The number of patients, accuracy, sensitivity, specificity, positive predictive value, and negative predictive value are displayed in Table 2. The accuracy of the OBS and CT groups were 93 per cent and 93 per cent, respectively. The two groups were not significantly different with regard to accuracy (OBS: 93%, CT: 93%, P = 1.0), sensitivity (OBS: 100%, CT: 89.5%, P = 0.21), specificity (OBS: 87.5%, CT: 95.6%, 0.61), positive predictive value (OBS: 88%, CT: 94.4%, P = 0.63), or negative predictive value (OBS: 100%, CT: 91.7%, P = 0.49). Fig. 1. Cohort diagram.
TABLE 1. Patient Characteristics
One patient had a CT scan interpreted as negative. This patient went to the operating room based on clinical assessment and was found to have appendicitis, confirmed by pathology.
A follow-up phone call was attempted at 1 week in all patients who were discharged home without a diagnosis of appendicitis (true negative patients) in both the CT and OBS groups. There were 21 true negative patients in the OBS group, with seven patients who could not be reached. Thirteen were successfully contacted and reported no additional symptoms indicative of appendicitis. One patient did complain of new abdominal pain and was readmitted to our hospital for a hysterectomy. In the CT arm, 12 of the 22 patients could not be contacted. Ten patients in this group reported no further abdominal discomfort.
In the overall study population, there were no bowel perforations, abdominal abscesses, or gangrenous appendices. There was also no mortality.
The use of routine CT scan has been advocated to improve the accuracy of diagnosing appendicitis.12, 15 Because an earlier prospective randomized study done by Hong et al.13 did not show any added benefit of routine CT use in all patients who presented with right lower quadrant pain, we hypothesized that routine CT scans might specifically benefit women of childbearing age, a patient group historically plagued by low diagnostic accuracy rates.1, 8 This study is the first randomized prospective trial examining the routine use of CT in diagnosing appendicitis specifically in women of childbearing age. Our results conclude that CT scanning was not significantly different from clinical assessment by a general surgeon and that both modalities accurately identified women who need an operation for acute appendicitis (OBS: 93%, CT: 93%). In addition, we found that hospital length of stay and hospital charges were similar with either the use of clinical examination alone or CT scanning.
The results of our CT scans were based on scanning the abdomen and pelvis after oral and intravenous contrast with a 7 mm slice thickness. The accuracy using this method to rule out or confirm appendicitis was 93 per cent, which compares favorably to other CT series.4, 8, 12, 15 We believe that the 7 mm thickness of the CT slice sufficiently covers the abdomen and pelvis, however the selection of a narrower slice collimation could enhance the detection of the appendix and therefore increase the sensitivity (89.5%) of the CT scans.16 In this study, patients had a complete CT of the abdomen and pelvis to rule out other etiologies for their pain. Noncontrast CT scans have been advocated by Malone et al.5 to expedite the patient's work-up in the emergency room but with a slight sacrifice in accuracy. Our protocol does not use rectal contrast due to concerns of increased time and the technical assistance required. Other studies using rectal contrast report variable sensitivities, ranging from 84 to 96 per cent11, 17 and increased accuracy (98%).12
Patients who were found not to have appendicitis by CT scan or clinical observation had a follow-up phone call at 1 week postdischarge. These follow-up calls were made at 1 week to confirm the true negatives on patients that were discharged. However, our success rate for getting in touch with these patients after discharge at 1 week was less than ideal. Of the 43 true negative patients in the study, we were only able to contact 24 patients by phone for a follow-up of only 55 per cent. Based on the demographic pattern of our patients, those doing well after discharge rarely return to clinic for their follow-up appointments, thus it was not surprising that we could not get better follow-up by phone. It is possible that some of the 19 true negative patients that we were unable to contact after discharge had appendicitis, which was missed by OBS or CT. However, we believe that this number is zero or low because most of these patients use the county hospital exclusively to receive their medical care. We believe that these patients, if they became sick, would theoretically return to our hospital.
This trial was started in November 1999 and was accruing patients up until February of 2001 when the lead author was finishing his fellowship and leaving the institution. The data at that time was presented at the Southwestern Surgical Clinical Congress in San Diego in April 2001. The study at the time of the presentation was underpowered secondary to not having enough patients enrolled in the study. The lead author returned to the University of Miami in late 2002 and the study was restarted in March of 2003 to enroll enough patients. The protocol for the study was approved for both time periods by the institution's institutional review board. The study was stopped the second time after entry of a total of 95 patients. The primary reason for halting the trial the second time in December of 2004 was our inability to accrue patients. This was due to the emergency medicine physicians routinely obtaining an abdominal CT scan before consulting the surgical service when evaluating patients with acute right lower quadrant abdominal pain during the last time period of our study. Thus, patients were not available to enter into our trial at the time of consultation, as they had already undergone CT radiologic imaging. It is often stated that "a head CT is worth a room full of neurologists." In speaking to our emergency medicine colleagues, as well as general surgeons throughout the US, it seems that the current medical culture supports the notion that "an abdominal CT scan is worth a room full of general surgeons." Surgeons have come to expect the CT results available at the time of consultation for evaluation of acute appendicitis. Abdominal examination by a qualified surgeon is no longer the initial phase in the definitive workup of a patient with right lower quadrant pain who presents to the emergency department.
As a result of not reaching our precalculated sample size, this study is too underpowered to detect a statistical difference. However, from the previous study by Hong et al.,13 a subgroup of women patients of childbearing age was shown to have no difference between CT scan and clinical exam when diagnosing appendicitis. Their results in this patient population are similar to our findings. In Hong's study, a subset of 48 women of childbearing age were prospectively evaluated following the same protocol as our study and had similar patient characteristics. We believe that this reinforces our findings that either CT scanning or clinical assessment by a surgeon can be used successfully as a screening modality to rule in or out appendicitis in women of childbearing age. In our current study, the negative appendectomy rate was 12 per cent (3 false positives out of 25 clinically examined), which is lower than the negative appendectomy rate reported in the literature for female patients of childbearing age (19-40%).7, 8, 15 If clinicians are unavailable or unable to provide a low negative appendectomy rate with clinical assessment alone, then abdominal CT scans seem to be a valuable tool to aid in making the diagnosis of appendicitis.12, 15, 18, 19
A routine diagnostic test that can accurately and swiftly diagnose appendicitis in women of childbearing age has been desired by clinicians for some time. In the evaluation of right lower quadrant abdominal pain, CT scan seems to be equivalent to clinical examination by a general surgeon without increasing hospital charges or length of stay. This prospective randomized trial found that the use of computed tomography in women of childbearing age who presented with right lower quadrant was not significantly different from clinical assessment by an experienced clinician in accurately identifying patients who require an operation for appendicitis. Although this study is too small to statistically establish equivalence, these data suggest that a CT scan examination reliably identifies women who need an operation for appendicitis and seems to be as good as clinical examination.
1. Rothrock SG, Green SM, Dobson M, et al. Misdiagnosis of appendicitis in nonpregnant women of childbearing age. J Emerg Med 1995;13:1-8.
2. Deutsch A, Zelikovsky A, Reiss R. Laparoscopy in the prevention of unnecessary appendectomies: A prospective study. Br J Surg 1982;69:336-7.
3. Bongard F, Landers DV, Lewis F. Differential diagnosis of appendicitis and pelvic inflammatory disease: A prospective analysis. Am J Surg 1985;150:90-6.
4. Balthazar EJ, Megibow AJ, Siegel SE, Birnbaum BA. Appendicitis: Prospective evaluation with high-resolution CT. Radiology 1991;180:21-4.
5. Malone AJ, Wolf CR, Maimed AS, Melliere BF. Diagnosis of acute appendicitis: Value of unenhanced CT. AJR Am J Roentgenol 1993;160:763-6.
6. Rhea JT, Rao PM, Novelline RA, McCabe CJ. A focused CT technique to reduce the cost of caring for patients with clinically suspected appendicitis. AJR Am J Roentgenol 1997; 169:113-8.
7. Schuler JG, Shortsleeve MJ, Goldenson RS, et al. Is there a role for abdominal computed tomographic scans in appendicitis? Arch Surg 1998;133:373-7.
8. Rao PM, Rhea JT, Novelline RA, et al. Helical CT technique for the diagnosis of appendicitis: Prospective evaluation of a focused CT examination. Radiology 1997;202:139-44.
9. Stroman DL, Bayouth CV, Kuhn JA, et al. The role of computed tomography in the diagnosis of acute appendicitis. Am J Surg 1999;178:485-9.
10. Choi YH, Fischer E, Hoda SA, et al. Appendiceal CT in 140 cases: Diagnostic criteria for acute and necrotizing appendicitis. Clin Imaging 1998;22:252-71. 11. Lane MJ, Liu DM, Huynh MD, et al. Suspected acute appendicitis: Nonenhanced helical CT in 300 consecutive patients. Radiology 1999;213:341-6.
12. 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;338:141-6.
13. Hong JH, Cohn SM, Ekeh P, et al. A prospective randomized study of clinical assessment versus computed tomography in the diagnosis of acute appendicitis. Surg Infect (Larchmt) 2003 ;4: 231- 9.
14. Alvarado A. A practical score for the early diagnosis of acute appendicitis. Ann Emerg Med 1986;15:557-64.
15. Walker S, Haun W, Clark J, et al. The value of limited computed tomography with rectal contrast in the diagnosis of acute appendicitis. Am J Surg 2000;180:450-5.
16. Weltman DI, Yu J, Krumenmacker J Jr, et al. Diagnosis of acute appendicitis: Comparison of 5- and 10-mm CT sections in the same patient. Radiology 2000;216:172-7.
17. Heaston DR, McClellan JS, Heaston DK. Community hospital experience in 600+ consecutive patients who underwent unenhanced helical CT for suspected appendicitis. AJR 2000; 174:53.
18. Balthazar EJ, Rofsky NM, Zucker R. Appendicitis: The impact of computed tomography imaging on negative appendectomy and perforation rates. Am J Gasstroenterol 1998;93:768-71.
19. Rao PM, Rhea JT, Rattner DW, et al. Introduction of appendiceal CT: Impact on negative appendectomy and perforation rates. Ann Surg 1999;229:344-9.
PETER P. LOPEZ, M.D., F.A.C.S.,* STEPHEN M. COHN, M.D.,* CHARLES A. POPKIN, M.D.,[dagger] JULIE JACKOWSKI, R.N.,[dagger] JOEL E. MICHALEK, Ph.D.,* THE APPENDICITIS DIAGNOSTIC GROUP[dagger]
From the * Department of Surgery, University of Texas Health Science Center, San Antonio, Texas and the
[dagger] DeWitt Daughtry Family Department of Surgery, University of Miami Miller School of Medicine,
For the Appendicitis Diagnostic Group: Margaret Brown, MSN, Fahim Habib, M.D., Jeffrey Augenstein, M.D., Erik Barquist, M.D., Patricia Byers, M.D., Carl I. Schulman, M.D., Enrique Ginzburg, M.D., Mauricio Lynn, M.D., Mark McKenney, M.D., Nicholas Namias, M.D., David Shatz, M.D., Danny Sleeman, M.D., S. Morad Hameed, M.D., Robb R. Whinney, D.O., Louis Pizano, M.D., Peter Ekeh, M.D., Suzanne LeBlang, M.D., and Martin Newman, M.D.
Address correspondence and reprint requests to Peter P. Lopez, M.D., F.A.C.S., Assistant Professor of Surgery, Department of Surgery, University of Texas Health Science Center, 7703 Floyd Curl Drive, MC 7842, San Antonio, TX 78229. E-mail: Lopez @ uthscsa.edu.
Copyright Southeastern Surgical Congress Dec 2007
(c) 2007 American Surgeon, The. Provided by ProQuest Information and Learning. All rights Reserved.