In reply

Abstract

We thank Dr. Schwartz for his comments about our recently published article1Hlibczuk V. Dattaro J.A. Jin Z. et al.Diagnostic accuracy of noncontrast computed tomography for appendicitis in adults: a systematic review.Ann Emerg Med. 2010; 55: 51-59Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar; we agree with many of his comments. As expected in any systematic review, none of the included studies were without flaws (see quality assessment figure).1Hlibczuk V. Dattaro J.A. Jin Z. et al.Diagnostic accuracy of noncontrast computed tomography for appendicitis in adults: a systematic review.Ann Emerg Med. 2010; 55: 51-59Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar Inconsistency of reporting inconclusive computed tomography (CT) results among studies was clearly stated as an important limitation of our systematic review. Unfortunately, the majority of studies did not separately report data for patients with inconclusive or indeterminate CT results, which would have allowed for calculation of interval likelihood ratios (LRs).2Brown M. Reeves R. Interval likelihood ratios: another advantage for the evidenced-based diagnostician.Ann Emerg Med. 2003; 42: 292-297Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar In the only study within our systematic review that separately reported data on inconclusive CT scans, 24% of the scans were considered inconclusive, with an associated LR=0.9.3Tamburrini S. Brunetti A. Brown M. et al.Acute appendicitis: diagnostic value of nonenhanced CT with selective use of contrast in routine clinical settings.Eur Radiol. 2007; 17: 2055-2061Crossref PubMed Scopus (32) Google Scholar These results support a clinical approach whereby an equivocal CT interpretation by the radiologist neither increases nor decreases the pretest probability of appendicitis. When data for the calculation of interval LRs are available, we disagree with Dr. Schwartz's suggestion to dichotomize the data into a simple 2×2 table. This traditional approach results in loss of valuable information and distortion of the sensitivity and specificity values for CT images that are clearly positive or negative.2Brown M. Reeves R. Interval likelihood ratios: another advantage for the evidenced-based diagnostician.Ann Emerg Med. 2003; 42: 292-297Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar We also stated that the majority of studies were prone to differential verification bias, which may have inflated estimates for test accuracy. However, in the design of diagnostic test studies, methods used to limit one form of bias may result in the development of another type of bias. Dr. Schwarz's comments about the study by in't Hof et al highlight this problem. Although the use of laparoscopy on all subjects eliminated the risk of differential verification bias, the high prevalence of appendicitis (84%) indicates the study was likely biased toward patients with a more severe presentation compared with the typical emergency department population presenting with abdominal pain.4in't Hof K.H. van Lankeren W. Krestin G.P. et al.Surgical validation of unenhanced helical computed tomography in acute appendicitis.Br J Surg. 2004; 91: 1641-1645Crossref PubMed Scopus (31) Google Scholar The purpose of our meta-analysis was to assess the diagnostic accuracy of noncontrast CT for the diagnosis of appendicitis in the adult population. We did not seek to directly compare noncontrast CT to intravenous contrast CT studies. The results of our systematic review reveal that noncontrast CT has excellent sensitivity and specificity for the diagnosis of appendicitis. We did not suggest that intravenous contrast CT studies be abandoned; however, we did recommend that noncontrast CT be considered an alternative to contrast CT, depending on the individual patient's condition and circumstances. As stated in our conclusion, a noncontrast CT protocol may be particularly attractive for patients with contraindications to contrast, such as those at risk for contrast-induced nephropathy.1Hlibczuk V. Dattaro J.A. Jin Z. et al.Diagnostic accuracy of noncontrast computed tomography for appendicitis in adults: a systematic review.Ann Emerg Med. 2010; 55: 51-59Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar Use of Contrast With CT in Patients Suspected of Having Appendicitis: A Contrasting OpinionAnnals of Emergency MedicineVol. 56Issue 4PreviewThe use of contrast for abdominal computed tomography (CT) in patients with suspected appendicitis has been the subject of considerable debate and is discussed in 2 recent articles in Annals.1,2 Both support the use of noncontrast CT (no oral or intravenous [IV] contrast). However, CT with IV and oral contrast has an established track record, and a high level of evidence is needed before abandoning contrast. Careful examination of the existing studies does not support this change in CT protocol. Full-Text PDF