Evaluating Appendicitis Scoring Systems Using a Prospective Pediatric Cohort
Presented at the Pediatric Academic Societies meeting, May 2006, San Francisco, CA.
Received 24 May 2006; received in revised form 20 July 2006, 14 September 2006, 15 November 2006, 14 December 2006 and 19 December 2006; accepted 20 December 2006. published online 12 April 2007.
Study objective
This article evaluates the performance of the previously published Alvarado and Samuel appendicitis scoring systems in a prospectively identified pediatric cohort.
Methods
A prospective cohort of patients, aged 3 to 21 years, being evaluated for appendicitis was enrolled during 20 consecutive months at a large, urban, pediatric hospital. Study forms were completed by pediatric emergency medicine attending physicians before imaging or surgery. Final diagnosis was determined by pathology or follow-up telephone call. Each score was tested as proposed by the original authors. Sensitivity, specificity, and predictive values were calculated for all patients and prepubertal patients younger than 10 years.
Results
Five hundred eighty-eight patients were studied. The median age was 11.9 years (interquartile range [IQR] 8.5; 14.9 years). Thirty-four percent of patients had appendicitis. An Alvarado score greater than or equal to 7 yielded a sensitivity of 72% (95% confidence interval [CI] 66% to 78%), specificity 81% (76% to 84%), negative predictive value (NPV) 85% (81% to 89%), and positive predictive value (PPV) 65% (59% to 72%). A Samuel score greater than or equal to 6 yielded a sensitivity of 82% (77% to 87%), specificity 65% (60% to 70%), NPV 88% (84% to 91%), and PPV 54% (48% to 60%). When analysis was limited to patients younger than 10 years, n=206, an Alvarado score greater than or equal to 7 yielded a sensitivity of 73% (62% to 84%), specificity 80% (73% to 86%), NPV 89% (83% to 94%), and PPV 58% (45% to 69%). A Samuel score greater than or equal to 6, when patients younger than 10 years were considered, yielded a sensitivity of 77% (66% to 87%), specificity 65% (56% to 72%), NPV 88% (82% to 94%), and PPV 45% (35% to 55%). Receiver operator characteristic curves had an area under the curve of 0.83 (95% CI 0.79 to 0.86) (Alvarado) and 0.81 (95% CI 0.78 to 0.85) (Samuel).
Conclusion
Although the Alvarado and Samuel scores provide measurably useful diagnostic information in evaluating children with suspected appendicitis, neither method provides sufficient PPV to be used in clinical practice as the sole method for determination of the need for surgery.
aDepartment of Medicine, Children’s Hospital Boston, Harvard Medical School, Boston, MA
bDivision of Emergency Medicine, Children’s Hospital Boston, Harvard Medical School, Boston, MA
cDivision of Pediatric Emergency Medicine, Morgan Stanley Children’s Hospital of New York, Columbia University, New York, NY.
Address for correspondence: Carisa Schneider, MD. Department of Medicine, Children’s Hospital Boston, 300 Longwood Ave, Boston, MA 02115. Tel (617) 355-6000
Supervising editor: David M. Jaffe, MD
Author contributions: AK and RB conceived the study and designed the trial. AK and RB supervised the conduct of the trial and data collection. AK and RB undertook recruitment of patients and managed the data, including quality control. RB provided statistical advice on study design, RB takes responsibility for all statistical calculations in the manuscript, CS and RB analyzed the data, CD drafted the manuscript, and all authors contributed substantially to its revision. CD, AK and RB take responsibility for the paper as a whole.
Funding and support: The authors report this study did not receive any outside funding or support.
Publication dates: Available online March 26, 2007.
ABSTRACT