The objective of this study was to examine the use of WBC count and polymorphonuclear leukocyte differential (PMN%) for improving the predictive value of ultrasound (US) in children with suspected appendicitis.

We conducted a retrospective cohort study of children undergoing US for suspected appendicitis between January 1, 2010 and December 31, 2012 at a single children's hospital (n = 845). Negative (NPV) and positive predictive values (PPV) for appendicitis were calculated for common constellations of US findings and compared with and without the use of laboratory thresholds (WBC >9 × 10³/μL and PMN% >65% for PPV; WBC ≤9 × 10³/μL and PMN% ≤65% for NPV).

Fifty-one percent of US were considered “equivocal” (ie, appendix incompletely visualized, no primary or secondary signs, or presence of fluid only) and NPV increased significantly for this cohort using laboratory thresholds (41.9% vs 95.8%; p < 0.001). Primary signs of appendicitis, without secondary signs, were documented in 18% of examinations, and the PPV associated with this cohort increased from 79.1% to 91.3% (p < 0.001) using laboratory thresholds. Secondary signs with or without primary signs were documented in 24% of examinations, and laboratory thresholds improved the PPV in this cohort from 89.1% to 96.8% (p < 0.001). Guidelines recommending against the use of CT for very high-risk and low-risk categories (NPV >95% and PPV >95%) on the basis of combined US and laboratory data could have reduced the number of CTs by 27.1% (101 of 373) during the study period.

The incorporation of WBC count and PMN% can substantially improve the predictive value of US in the diagnosis of suspected appendicitis in children.