Misdiagnosis of Appendicitis and the Use of Diagnostic Imaging - 21/08/11

Doi : 10.1016/j.jamcollsurg.2005.04.039

David R. Flum, MD, MPH ^†,^§,^⁎ : FACS, Timothy D. McClure, MS ^†, Arden Morris, MD, MPH , Thomas Koepsell, MD, MPH ^,^‡,^§
Robert Wood Johnson Clinical Scholars Program, University of Washington, Seattle, WA
^† Department of Surgery, University of Washington, Seattle, WA
^‡ Department of Epidemiology, University of Washington, Seattle, WA
^§ Department of Health Services, University of Washington, Seattle, WA.

Correspondence address: David R Flum, MD, MPH, FACS, Department of Surgery, University of Washington, Health Sciences Center, Box 356410, Seattle WA 98195-6410.

Résumé

Background

CT and ultrasound (US) are increasingly recommended to establish the diagnosis of appendicitis, but population-based rates of misdiagnosis (negative appendectomy [NA]) have not improved over time. The objective of this study was to determine the relationship between CT/US and NA in common practice.

Study design

Using data from the Group Health Cooperative on all patients undergoing appendectomy between 1980 and 1999, a longitudinal study was conducted to determine the frequency of NA over time and a case-control, medical record–based study of a subset of patients from the 1990s was conducted to determine the accuracy of CT/US.

Results

Of 4,058 patients undergoing appendectomy (mean age 31 ± 18.6 [SD] years, 49.6% women), 631 (15.5%) had an NA. The overall incidence of NA remained stable over time at 1.5/10,000 patient-years, as did the age and gender adjusted rate (incident rate ratio 0.95, 95% CI 0.97, 1.01). In 1999, nearly 40% of patients had either CT or US. The aggregate sensitivity of these tests was only 74.2% (95% CI 65.7, 83.7), with a positive predictive value of 95.1% (95% CI 91.5, 96.8). CT scans were 88.3% sensitive, with 97.2% positive predictive value (95% CI 92.9, 100). Ultrasounds were 69.5% sensitive, with a positive predictive value of 94.1% (95% CI 89.6, 96.4). More than one in five tests obtained in patients with NA were positive for appendicitis (21.7% for CT and 20.8% for US).

Conclusions

The rate of NA was unchanged over time despite the introduction and use of CT/US, and this appeared to be related to the inconsistent performance characteristics of the tests. This study cautions against overreliance on CT/US in diagnosing appendicitis and emphasizes the need for test benchmarking in routine practice before establishing protocols for presumed appendicitis.

Le texte complet de cet article est disponible en PDF.

Abbreviations and Acronyms : GHC, IRR, NA, PPV, US

Plan

Methods

Study design

Setting

Data sources and characteristics

Competing Interests Declared: None.
The Robert Wood Johnson Foundation provided generous funding for this project.
The views expressed in this article are those of the authors and not necessarily those of the Robert Wood Johnson Foundation or the University of Washington.