Ultrasonographic criteria for diagnosis of cervical lymph node metastasis of squamous cell carcinoma in the oral and maxillofacial region

Citation
H. Yusa et al., Ultrasonographic criteria for diagnosis of cervical lymph node metastasis of squamous cell carcinoma in the oral and maxillofacial region, J ORAL MAX, 57(1), 1999, pp. 41-48
Citations number
30
Categorie Soggetti
Dentistry/Oral Surgery & Medicine
Journal title
JOURNAL OF ORAL AND MAXILLOFACIAL SURGERY
ISSN journal
02782391 → ACNP
Volume
57
Issue
1
Year of publication
1999
Pages
41 - 48
Database
ISI
SICI code
0278-2391(199901)57:1<41:UCFDOC>2.0.ZU;2-Y
Abstract
Purpose: This retrospective study was done to establish an objective criter ion for sonographic diagnosis of cervical lymph nude metastasis of squamous cell carcinoma in the oral and maxillofacial region. Patients and Methods: A quantitative criterion for differentiating metastat ic from nonmetastatic nodes was statistically calculated using logistic reg ression analysis of 106 cervical lymph nodes in 41 patients. A prospective study was then conducted to ascertain the reliability of the criterion for 82 nodes in another 23 patients. Results: The following equation was obtained from the retrospective study: Y = 1/1 + e(-lambda) (lambda = -7.5709 + 4.2423 X1 + 1.9354 X2 + 0.5507 X3) Y represents the predicted value, lambda, also representing a predicted val ue, was defined as negative for nonmetastatic and positive for metastatic n odes; X1, signifying the existence of an echogenic hilus, was defined as 0 for presence and 1 for absence; X2, signifying the internal echo pattern, w as defined as 0 for homogeneous and 1 for heterogeneous; X3, signifying the minimal diameter, was the actual measured value (mn). This equation showed a sensitivity of 82%, a specificity of 91%, and an accuracy of 88% in the prospective study. Conclusion: The equation calculated by logistic regression analysis was a r eliable criterion for diagnosing cervical lymph node metastasis.