Clinical factors predicting the need for biopsy in children with cervical lymphadenopathy

Abstract

Background: Diagnosing lymphadenopathy in children can be challenging for clinicians, as it may represent a manifestation of various conditions, including granulomatous diseases and malignancies. The cervical region is frequently affected in cases of peripheral lymphadenopathy. While surgical biopsy can provide a definitive diagnosis, it is not always necessary.

Objectives: This study aimed to characterize pediatric patients with cervical lymphadenopathy who underwent surgical biopsy and to identify clinical factors associated with lymph nodes that require further management (LNFM), such as those with granulomatous or malignant features.

Methods: A retrospective review was conducted on 87 pediatric cases of cervical lymphadenopathy that underwent surgical biopsy. A descriptive analysis was performed. Receiver operating characteristic (ROC) analysis was used to identify the lymph node size most indicative of LNFM. Logistic regression analysis was conducted to determine independent predictors of LNFM, granulomatous lymph nodes, and malignant lymph nodes.

Results: Final diagnoses revealed 37 cases (42.5%) as non-LNFM, 36 cases (41.4%) as granulomatous, and 14 cases (16.1%) as malignant lymph nodes. ROC analysis identified 2 cm as the optimal lymph node size threshold for predicting LNFM, balancing sensitivity and specificity. Multivariate logistic regression revealed that submandibular location (overlying skin redness (OR), 11.9; 95% confidence intervals (CI), 2.5–55.4; P = 0.002) and abnormal chest x-ray findings (OR, 20.7; 95% CI, 2.1–201.5; P = 0.009) were independent predictors of LNFM. Subgroup analysis further showed that OR, 8.0; 95% CI, 1.1–58.8; P = 0.04) and submandibular location (OR, 9.7; 95% CI, 1.3–72.1; P = 0.027) were significant predictors of granulomatous lymph nodes. No significant predictors were found for malignant lymph nodes.

Conclusion: Submandibular location and abnormal chest x-ray findings serve as key predictors of LNFM. Lymph nodes exceeding 2 cm in size are more likely to fall into the LNFM category. OR and submandibular involvement suggest granulomatous pathology. Most supraclavicular nodes were classified as LNFM and were frequently malignant, underscoring the need for careful assessment of this area. These findings may support clinical decision-making prior to biopsy in cases of cervical lymphadenopathy.