Utility of serum and urine leucine-rich alpha-2 glycoprotein 1 (LRG1) as predictors of appendicitis and complicated appendicitis in children.

PURPOSE

Leucine rich alpha-2 glycoprotein 1 (LRG1) has emerged as a promising biomarker for appendicitis, especially in pediatric patients. However, the currently available data are sparse, and the biomarker must be validated in more settings and compared to standard inflammatory markers. We aimed to evaluate the diagnostic and discriminative utility of serum and urine LRG1 in children with other causes of abdominal pain (no appendicitis) versus appendicitis, and uncomplicated versus complicated appendicitis.

METHODS

The study design was prospective including children ≤ 15 years with suspected appendicitis. Blood and urine samples were collected at the time of clinical evaluation at the Pediatric Emergency Department and analyzed for concentrations of LRG1. Appendicitis diagnosis and severity were determined through histopathological examination and intraoperative findings. Group comparisons were carried out using Kruskal-Wallis test with post hoc Dunn-Bonferroni tests for pairwise comparisons. Associations between LRG1 and other laboratory and clinical variables and the odds of appendicitis and complicated appendicitis were assessed by univariate and multivariable logistic regression analyses. Diagnostic (no appendicitis versus appendicitis) and discriminative (uncomplicated versus complicated appendicitis) performance were evaluated through Receiver Operating Characteristic (ROC) curves with analyses of Areas Under the Curve (AUC). Optimal cutoffs were generated using Youden's index, and diagnostic and predictive values were calculated and compared.

RESULTS

172 children were included. 132 (77%) had appendicitis and 56 (42%) of these had complicated appendicitis. The median age was 10 (IQR 8-12) years and 98 (57%) were boys. Serum concentrations of LRG1 did not differ significantly between the groups. Urine LRG1 was significantly higher among children with complicated appendicitis and no appendicitis compared to children with uncomplicated appendicitis (p < 0.001). In the logistic regression analysis of all children with suspected appendicitis, increased serum LRG1 was associated with a decreased odds of appendicitis (OR 0.96 [95% CI 0.93-0.99], p = 0.008). This association remained after adjustment for age, sex and symptom duration (aOR 0.95 [0.92-0.98], p = 0.003). Urine LRG1 was not associated with the odds of appendicitis. Neither serum nor urine LRG 1 were significantly associated with the odds of complicated appendicitis. When it comes to diagnosing appendicitis, both serum and urine LRG1 had AUC values of 0.39. However, urine LRG had a specificity of 95% and a PPV of 83%. The discriminative performance of serum LRG1 was poor, but the AUC for urine LRG1 of 0.65 was better than the ones for leukocytes, neutrophils and neutrophil percentages. Still, it was lower than the AUCs for C-reactive protein (CRP) and Appendicitis Inflammatory Response (AIR) score. Urine LRG1 has a high specificity and PPV for all cases of appendicitis, and correctly identifies cases of complicated appendicitis to a greater extent than some of the currently available inflammatory markers. Still, the regression analyses show no significant associations between urine LRG1 and appendicitis and complicated appendicitis in children.

CONCLUSION

In contrast to previous studies, in this cohort serum LRG1 was associated with decreased odds of appendicitis, shedding some doubt over the clinical utilization of serum LRG1 as a biomarker for appendicitis in children.