Thyroid Nodules as a Risk Factor for Thyroid Cancer in Patients With Graves' Disease

A Systematic Review and Meta-Analysis of Observational Studies in Surgically Treated Patients

Anastasios Papanastasiou; Konstantinos Sapalidis; Dimitrios G. Goulis; Nikolaos Michalopoulos; Evangelia Mareti; Stylianos Mantalovas; Isaak Kesisoglou

Disclosures

Clin Endocrinol. 2019;91(4):571-577. 

In This Article

Results

Initially, 1240 observational studies were identified through the literature search. When the duplicates were removed, 1088 remained. Seven studies fulfilled the inclusion criteria and included in the meta-analysis (n = 2582). The study selection process is demonstrated in Figure 1. All the included studies were retrospective. The frequency of thyroid cancer among them varied from 3.8[9] to 29.2%.[10] The characteristics of the included studies are summarized in Table 1. Additionally, the study design and the indications of surgery of each study are presented in Table 2 and Table S1, respectively.

Figure 1.

PRISMA study flow diagram

The prevalence of thyroid malignancy in the included studies was 11.5% (297/2582). Thyroid nodules were identified in 968 patients (37.5%). The patients with GD were classified into two groups depending on the preoperative detection of at least one thyroid nodule or not. The prevalence of thyroid cancer in these groups was 22.2% and 5.1%, respectively, being higher in the '≥1 nodule' group (OR 5.3, 95% CI 2.4-11.6, I 2 83%) (Figure 2).

Figure 2.

Meta-analysis of the association between preoperatively detected thyroid nodules and the incidence of thyroid cancer in patients with Graves' disease [Colour figure can be viewed at wileyonlinelibrary.com]

A sensitivity analysis was performed without patients with malignant or suspicious cytology. One of the observational studies was excluded because of nonavailable data. From the remaining studies, 22 patients in total were removed (with malignant or suspicious cytology), assuming that they had thyroid cancer. The remaining patients were included in this sensitivity analysis (n = 2019). At least one thyroid nodule was detected in 647 patients (32%). The prevalence of thyroid cancer in patients with at least one thyroid nodule was 20.7%, while in those without any thyroid nodule was 5% (OR 4.02, 95% CI 1.24-12.99, I 2 89%) (Figure 3).

Figure 3.

Sensitivity analysis of the association between preoperatively detected thyroid nodules and the incidence of thyroid cancer in patients with Graves' disease without malignant or suspicious cytology [Colour figure can be viewed at wileyonlinelibrary.com]

A subgroup analysis was conducted, comparing the prevalence of thyroid cancer between patients with a solitary nodule and those with multiple nodules. Four out of the seven observational studies were included (n = 630), as the remaining did not contain relevant data. A solitary thyroid nodule was identified in 118 patients (18.7%) and multiple nodules in 512 patients (81.3%). The prevalence of thyroid cancer was 34 (28.8%) and 93 (18.3%) patients, respectively. The meta-analysis of the four observational studies showed no difference between the two subgroups (OR 1.4, 95% CI 0.9-2.3, I 2 0%) (Figure 4).

Figure 4.

Subgroup analysis, according to the number of nodules, of the association between preoperatively detected thyroid nodules and the incidence of thyroid cancer in patients with Graves' disease [Colour figure can be viewed at wileyonlinelibrary.com]

Regarding the quality assessment, six of the studies were considered as 'moderate risk', while one was considered as 'serious risk', due to confounding bias and selection of participants violations. A detailed presentation of the systematic error risk assessment of all included studies is presented in Table S2.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....