Diagnostic accuracy of the Alvarado score versus modified two-parameter model (RLQ Tenderness plus Leukocytosis) in acute appendicitis

Background: The Alvarado score is a commonly used tool for stratifying patients suspected of having appendicitis. However, its complexity and subjective components can limit reproducibility. A simpler two-parameter model combining right lower quadrant (RLQ) tenderness and leukocytosis has been proposed as a more practical alternative.
Objective: To evaluate the diagnostic accuracy of a modified two-parameter model versus the full Alvarado score in patients with suspected acute appendicitis.
Methods: A retrospective cohort study was conducted at Queen Alia Military Hospital, Jordan, from January 2019 to January 2025. Medical records of 500 adult patients who underwent appendectomy for suspected appendicitis were reviewed. Data extracted included Alvarado score components, complete blood count results, intraoperative findings, and histopathological diagnosis. The modified two-parameter model was defined by leukocytosis (WBC >10,000/µL) plus RLQ tenderness. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC) were calculated. ROC curves were compared using DeLong's test.
Results: Among 500 patients (mean age 28.6±12.4 years, 56% male), histopathology confirmed appendicitis in 452 cases (90.4%). The Alvarado score (cutoff ≥8) showed a sensitivity of 92.0%, specificity of 85.4%, and AUC of 0.91 (95% CI: 0.88–0.94). The two-parameter model showed a sensitivity of 89.6%, specificity of 81.3%, and AUC of 0.89 (95% CI: 0.86–0.92). DeLong's test showed no statistically significant difference between the two AUCs (p=0.18). The two-parameter model correctly classified 86.2% of cases, while the full Alvarado score correctly classified 88.4%. In the intermediate-risk subgroup (Alvarado 5–7, n=150), the two-parameter model maintained good discriminatory ability (AUC=0.85, 95% CI: 0.79–0.91).
Conclusion: The modified two-parameter model (RLQ tenderness + leukocytosis) has diagnostic accuracy equivalent to the full Alvarado score for acute appendicitis. Its simplicity and ease of implementation make it a practical alternative for rapid risk stratification, especially in resource-limited or high-volume settings.