Development of a clinical decision model for thyroid nodules

Alexander Stojadinovic, George E. Peoples, Steven K. Libutti, Leonard R. Henry, John Eberhardt, Robin S. Howard, David Gur, Eric A. Elster, Aviram Nissan

Research output: Contribution to journalArticle

36 Citations (Scopus)

Abstract

Background. Thyroid nodules represent a common problem brought to medical attention. Four to seven percent of the United States adult population (1018 million people) has a palpable thyroid nodule, however the majority (>95%) of thyroid nodules are benign. While, fine needle aspiration remains the most cost effective and accurate diagnostic tool for thyroid nodules in current practice, over 20% of patients undergoing FNA of a thyroid nodule have indeterminate cytology (follicular neoplasm) with associated malignancy risk prevalence of 2030%. These patients require thyroid lobectomy/isthmusectomy purely for the purpose of attaining a definitive diagnosis. Given that the majority (7080%) of these patients have benign surgical pathology, thyroidectomy in these patients is conducted principally with diagnostic intent. Clinical models predictive of malignancy risk are needed to support treatment decisions in patients with thyroid nodules in order to reduce morbidity associated with unnecessary diagnostic surgery. Methods. Data were analyzed from a completed prospective cohort trial conducted over a 4-year period involving 216 patients with thyroid nodules undergoing ultrasound (US), electrical impedance scanning (EIS) and fine needle aspiration cytology (FNA) prior to thyroidectomy. A Bayesian model was designed to predict malignancy in thyroid nodules based on multivariate dependence relationships between independent covariates. Ten-fold cross-validation was performed to estimate classifier error wherein the data set was randomized into ten separate and unique train and test sets consisting of a training set (90% of records) and a test set (10% of records). A receiver-operating-characteristics (ROC) curve of these predictions and area under the curve (AUC) were calculated to determine model robustness for predicting malignancy in thyroid nodules. Results. Thyroid nodule size, FNA cytology, US and EIS characteristics were highly predictive of malignancy. Cross validation of the model created with Bayesian Network Analysis effectively predicted malignancy [AUC = 0.88 (95%CI: 0.820.94)] in thyroid nodules. The positive and negative predictive values of the model are 83% (95%CI: 76%91%) and 79% (95%CI: 72%86%), respectively. Conclusion. An integrated predictive decision model using Bayesian inference incorporating readily obtainable thyroid nodule measures is clinically relevant, as it effectively predicts malignancy in thyroid nodules. This model warrants further validation testing in prospective clinical trials.

Original languageEnglish (US)
Article number12
JournalBMC Surgery
Volume9
Issue number1
DOIs
StatePublished - 2009
Externally publishedYes

Fingerprint

Thyroid Nodule
Cell Biology
Fine Needle Biopsy
Neoplasms
Thyroidectomy
Electric Impedance
Area Under Curve
Unnecessary Procedures
Surgical Pathology
Bayes Theorem
ROC Curve
Thyroid Gland

ASJC Scopus subject areas

  • Surgery

Cite this

Stojadinovic, A., Peoples, G. E., Libutti, S. K., Henry, L. R., Eberhardt, J., Howard, R. S., ... Nissan, A. (2009). Development of a clinical decision model for thyroid nodules. BMC Surgery, 9(1), [12]. https://doi.org/10.1186/1471-2482-9-12

Development of a clinical decision model for thyroid nodules. / Stojadinovic, Alexander; Peoples, George E.; Libutti, Steven K.; Henry, Leonard R.; Eberhardt, John; Howard, Robin S.; Gur, David; Elster, Eric A.; Nissan, Aviram.

In: BMC Surgery, Vol. 9, No. 1, 12, 2009.

Research output: Contribution to journalArticle

Stojadinovic, A, Peoples, GE, Libutti, SK, Henry, LR, Eberhardt, J, Howard, RS, Gur, D, Elster, EA & Nissan, A 2009, 'Development of a clinical decision model for thyroid nodules', BMC Surgery, vol. 9, no. 1, 12. https://doi.org/10.1186/1471-2482-9-12
Stojadinovic A, Peoples GE, Libutti SK, Henry LR, Eberhardt J, Howard RS et al. Development of a clinical decision model for thyroid nodules. BMC Surgery. 2009;9(1). 12. https://doi.org/10.1186/1471-2482-9-12
Stojadinovic, Alexander ; Peoples, George E. ; Libutti, Steven K. ; Henry, Leonard R. ; Eberhardt, John ; Howard, Robin S. ; Gur, David ; Elster, Eric A. ; Nissan, Aviram. / Development of a clinical decision model for thyroid nodules. In: BMC Surgery. 2009 ; Vol. 9, No. 1.
@article{495654a6fc1b4109b20b3ff06afbefd7,
title = "Development of a clinical decision model for thyroid nodules",
abstract = "Background. Thyroid nodules represent a common problem brought to medical attention. Four to seven percent of the United States adult population (1018 million people) has a palpable thyroid nodule, however the majority (>95{\%}) of thyroid nodules are benign. While, fine needle aspiration remains the most cost effective and accurate diagnostic tool for thyroid nodules in current practice, over 20{\%} of patients undergoing FNA of a thyroid nodule have indeterminate cytology (follicular neoplasm) with associated malignancy risk prevalence of 2030{\%}. These patients require thyroid lobectomy/isthmusectomy purely for the purpose of attaining a definitive diagnosis. Given that the majority (7080{\%}) of these patients have benign surgical pathology, thyroidectomy in these patients is conducted principally with diagnostic intent. Clinical models predictive of malignancy risk are needed to support treatment decisions in patients with thyroid nodules in order to reduce morbidity associated with unnecessary diagnostic surgery. Methods. Data were analyzed from a completed prospective cohort trial conducted over a 4-year period involving 216 patients with thyroid nodules undergoing ultrasound (US), electrical impedance scanning (EIS) and fine needle aspiration cytology (FNA) prior to thyroidectomy. A Bayesian model was designed to predict malignancy in thyroid nodules based on multivariate dependence relationships between independent covariates. Ten-fold cross-validation was performed to estimate classifier error wherein the data set was randomized into ten separate and unique train and test sets consisting of a training set (90{\%} of records) and a test set (10{\%} of records). A receiver-operating-characteristics (ROC) curve of these predictions and area under the curve (AUC) were calculated to determine model robustness for predicting malignancy in thyroid nodules. Results. Thyroid nodule size, FNA cytology, US and EIS characteristics were highly predictive of malignancy. Cross validation of the model created with Bayesian Network Analysis effectively predicted malignancy [AUC = 0.88 (95{\%}CI: 0.820.94)] in thyroid nodules. The positive and negative predictive values of the model are 83{\%} (95{\%}CI: 76{\%}91{\%}) and 79{\%} (95{\%}CI: 72{\%}86{\%}), respectively. Conclusion. An integrated predictive decision model using Bayesian inference incorporating readily obtainable thyroid nodule measures is clinically relevant, as it effectively predicts malignancy in thyroid nodules. This model warrants further validation testing in prospective clinical trials.",
author = "Alexander Stojadinovic and Peoples, {George E.} and Libutti, {Steven K.} and Henry, {Leonard R.} and John Eberhardt and Howard, {Robin S.} and David Gur and Elster, {Eric A.} and Aviram Nissan",
year = "2009",
doi = "10.1186/1471-2482-9-12",
language = "English (US)",
volume = "9",
journal = "BMC Surgery",
issn = "1471-2482",
publisher = "BioMed Central",
number = "1",

}

TY - JOUR

T1 - Development of a clinical decision model for thyroid nodules

AU - Stojadinovic, Alexander

AU - Peoples, George E.

AU - Libutti, Steven K.

AU - Henry, Leonard R.

AU - Eberhardt, John

AU - Howard, Robin S.

AU - Gur, David

AU - Elster, Eric A.

AU - Nissan, Aviram

PY - 2009

Y1 - 2009

N2 - Background. Thyroid nodules represent a common problem brought to medical attention. Four to seven percent of the United States adult population (1018 million people) has a palpable thyroid nodule, however the majority (>95%) of thyroid nodules are benign. While, fine needle aspiration remains the most cost effective and accurate diagnostic tool for thyroid nodules in current practice, over 20% of patients undergoing FNA of a thyroid nodule have indeterminate cytology (follicular neoplasm) with associated malignancy risk prevalence of 2030%. These patients require thyroid lobectomy/isthmusectomy purely for the purpose of attaining a definitive diagnosis. Given that the majority (7080%) of these patients have benign surgical pathology, thyroidectomy in these patients is conducted principally with diagnostic intent. Clinical models predictive of malignancy risk are needed to support treatment decisions in patients with thyroid nodules in order to reduce morbidity associated with unnecessary diagnostic surgery. Methods. Data were analyzed from a completed prospective cohort trial conducted over a 4-year period involving 216 patients with thyroid nodules undergoing ultrasound (US), electrical impedance scanning (EIS) and fine needle aspiration cytology (FNA) prior to thyroidectomy. A Bayesian model was designed to predict malignancy in thyroid nodules based on multivariate dependence relationships between independent covariates. Ten-fold cross-validation was performed to estimate classifier error wherein the data set was randomized into ten separate and unique train and test sets consisting of a training set (90% of records) and a test set (10% of records). A receiver-operating-characteristics (ROC) curve of these predictions and area under the curve (AUC) were calculated to determine model robustness for predicting malignancy in thyroid nodules. Results. Thyroid nodule size, FNA cytology, US and EIS characteristics were highly predictive of malignancy. Cross validation of the model created with Bayesian Network Analysis effectively predicted malignancy [AUC = 0.88 (95%CI: 0.820.94)] in thyroid nodules. The positive and negative predictive values of the model are 83% (95%CI: 76%91%) and 79% (95%CI: 72%86%), respectively. Conclusion. An integrated predictive decision model using Bayesian inference incorporating readily obtainable thyroid nodule measures is clinically relevant, as it effectively predicts malignancy in thyroid nodules. This model warrants further validation testing in prospective clinical trials.

AB - Background. Thyroid nodules represent a common problem brought to medical attention. Four to seven percent of the United States adult population (1018 million people) has a palpable thyroid nodule, however the majority (>95%) of thyroid nodules are benign. While, fine needle aspiration remains the most cost effective and accurate diagnostic tool for thyroid nodules in current practice, over 20% of patients undergoing FNA of a thyroid nodule have indeterminate cytology (follicular neoplasm) with associated malignancy risk prevalence of 2030%. These patients require thyroid lobectomy/isthmusectomy purely for the purpose of attaining a definitive diagnosis. Given that the majority (7080%) of these patients have benign surgical pathology, thyroidectomy in these patients is conducted principally with diagnostic intent. Clinical models predictive of malignancy risk are needed to support treatment decisions in patients with thyroid nodules in order to reduce morbidity associated with unnecessary diagnostic surgery. Methods. Data were analyzed from a completed prospective cohort trial conducted over a 4-year period involving 216 patients with thyroid nodules undergoing ultrasound (US), electrical impedance scanning (EIS) and fine needle aspiration cytology (FNA) prior to thyroidectomy. A Bayesian model was designed to predict malignancy in thyroid nodules based on multivariate dependence relationships between independent covariates. Ten-fold cross-validation was performed to estimate classifier error wherein the data set was randomized into ten separate and unique train and test sets consisting of a training set (90% of records) and a test set (10% of records). A receiver-operating-characteristics (ROC) curve of these predictions and area under the curve (AUC) were calculated to determine model robustness for predicting malignancy in thyroid nodules. Results. Thyroid nodule size, FNA cytology, US and EIS characteristics were highly predictive of malignancy. Cross validation of the model created with Bayesian Network Analysis effectively predicted malignancy [AUC = 0.88 (95%CI: 0.820.94)] in thyroid nodules. The positive and negative predictive values of the model are 83% (95%CI: 76%91%) and 79% (95%CI: 72%86%), respectively. Conclusion. An integrated predictive decision model using Bayesian inference incorporating readily obtainable thyroid nodule measures is clinically relevant, as it effectively predicts malignancy in thyroid nodules. This model warrants further validation testing in prospective clinical trials.

UR - http://www.scopus.com/inward/record.url?scp=69449101715&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69449101715&partnerID=8YFLogxK

U2 - 10.1186/1471-2482-9-12

DO - 10.1186/1471-2482-9-12

M3 - Article

C2 - 19664278

AN - SCOPUS:69449101715

VL - 9

JO - BMC Surgery

JF - BMC Surgery

SN - 1471-2482

IS - 1

M1 - 12

ER -