Integrating MR Imaging Into the Clinical Workup of Pregnant Patients Suspected of Having Appendicitis Is Associated With a Lower Negative Laparotomy Rate

Abstract

Acute appendicitis is the most common nonobstetric condition requiring surgery in pregnant women. Early surgery based on an initial evaluation results in an increased negative laparotomy rate. However, prolonged observation to improve diagnostic accuracy can lead to increased perforation rates. Imaging should lead to a simultaneous decrease in negative laparotomies and perforations by allowing clinicians to determine earlier and with more certainty whether to perform surgery. Ultrasonography (US) and magnetic resonance imaging (MRI) offer radiation-free alternatives to computed tomography (CT). This retrospective study was performed to determine whether integrating MRI into the assessment of right lower quadrant pain in pregnant patients was associated with improved outcomes measured by the negative laparotomy and perforation rates. The parturients underwent laparotomy or MRI because appendicitis was suspected. Of 187 patients who met initial criteria, 82 had surgery according to the study protocol. A search of the radiology information system yielded 217 MRI examinations; 40 patients had surgery (32 for suspicion of appendicitis and 8 for other conditions). Thus, 32 patients underwent both MRI and surgery, and the study population included 267 unique patients. No contrast material was administered for the MRI. Criteria suggesting appendicitis included an appendix diameter of greater than 6 mm; wall thickness greater than 2 mm; fluid-filled lumen; and surrounding edema, fluid, or abscess. Magnetic resonance imaging reports were categorized as being positive, negative, or equivocal for appendicitis. Charts were reviewed for patients who did not undergo surgery. The frequency of presurgical MRI was used to determine a cutoff between pre-MRI (1996–2003) and post-MRI (2004–2011) cohorts. Negative laparotomy and perforation rates were calculated for both cohorts. Negative laparotomy rate was defined as the percentage of surgeries for presumed appendicitis tabulated as negative laparotomies. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated and adjusted for the prevalence of appendicitis in the pregnant population. Seven MRI examinations were performed between 1996 and 2003, but between 2004 and 2011, 210 MRI examinations were performed. For 1996 to 2011, 50 positive appendectomies, 32 negative appendectomies, and 12 perforations were recorded. The negative laparotomy rate was 55% (17/31) in the pre-MRI cohort and 29% (15/ 51) in the post-MRI group (P = 0.02), representing a 47% decrease in the negative laparotomy rate ([55%–29%]/55%). The perforation rate was 21% (3/14) and 26% (9/35) in the pre- and post-MRI cohorts, respectively (P > 0.99), thus statistically unchanged. The appendix was visualized in 70 (32%) of 217 patients. The sensitivity, specificity, positive predictive value, and negative predictive value of MRI were 89%, 97%, 74%, and 99%, respectively. In the pre-MRI cohort, US was performed in 26 (84%) of 31 patients who had surgery. Three US examinations were true positive; none were false positive. In the post-MRI cohort, US was performed in 41 (80%) of 51 patients who had surgery. Twelve US examinations were true positive, but none were in patients who had MRI. The 2 cohorts did not differ in the use of US (P = 0.4). Magnetic resonance imaging without oral contrast material is a valuable diagnostic tool for evaluating pregnant patients suspected of having appendicitis.