The global incidence of papillary thyroid carcinoma (PTC) has increased substantially during past decade, with 4.4% annual percent increase in the United States from 1974 to 2013 (1–3). This has been driven largely by the rise in papillary thyroid microcarcinoma (PTMC), which is defined as PTC measuring ≤1 cm in greatest dimension (2, 3). Although the majority of PTMCs are indolent with <0.5% thyroid-cancer related death, some (1–5%) may have locoregional recurrence, which is still a major concern for patients and clinicians (4). Mounting studies reported that lateral lymph node metastasis (LLNM) was associated with locoregional recurrence for PTMC (4–6). The revised American Thyroid Association (ATA) guidelines in 2015 also considered around 20% PTC patients with LLNM would have structural recurrence in the future (7).

It is well-accepted that lateral neck dissection (LND) is only recommended for PTMC patients when LLNM is diagnosed by preoperative evaluation such as physical examination, radiological imaging, and/or fine needle aspiration (FNA) (7). Prophylactic LND is not recommended because no reliable clinicopathological features are identified which can differentiate the subset of high-risk PTMC, which are more likely accompanied with microscopic lymph nodes and the potential of progressing to clinical LLNM. But some researchers still believed early treatment for PTMC with LLNM might be appropriate as long as high-risk PTMC patients could be distinguished by clinical and/or ultrasonic risk factors (8).

Moreover, active surveillance could be recommended as the first-line treatment for low-risk PTC patients (7). But around 3.8% patients under active surveillance will have novel LLNM, which is also a major concern for patients and clinicians (9, 10). So it is necessary to identify patients at high-risk for LLNM, which helps the enrollment of low-risk PTMC in active surveillance protocols (9, 10).

Accordingly, several studies have published on the preoperative clinicopathologic risk factors of LLNM for PTMC (11, 12). However, there have been no consensus on this, and the subject remains debatable. Considering the low incidence of LLNM in PTMC, a single study with small patient number may draw an unreliable conclusion with a considerable bias. Therefore, we performed a systematic review to evaluate the clinicopathologic and ultrasonic predictive factors for LLNM in PTMC.

According to the guidelines proposed by the preferred reporting items for systematic reviews and meta-analyses (PRISMA in Supplementary Material) statement, we conducted this systematic review and meta-analysis.

To the best of our knowledge, this is the first meta-analysis to assess clinical and ultrasonic risk factors for LLNM in PTMC. It was found that younger age, male gender, ETE, multifocal tumors, larger tumor size, and positive CLNM were clinical risk factors for LLNM in PTMC. Additionally, >25% tumor contact with thyroid capsule, calcification, and upper location were ultrasonic risk features for LLNM. In contrast, HT and other US features (such as composition, echo, margin, shape) were not considered as risk factors of LLNM in these patients.

High heterogeneity with an I² >50% was found in the analysis of age, sex, ETE, multifocality, HT, >25% contact with thyroid capsule, calcification as well as margin, and moderate heterogeneity with an I² >30% was found when analyzing tumor location. Interestingly, after the removal of one study from the analysis, similar results were confirmed, but the heterogeneity was decreased remarkably in the analysis of age, sex, multifocal, HT, contact>25%, calcification, and margin. Possible reasons for high heterogeneity were presented as Supplementary File 2. In contrast, the I² values for the analysis of ETE and tumor location were not changed significantly in the sensitivity analysis.

Younger age was identified to be associated with aggressiveness of PTC, such as vascular invasion and lymph node metastases (26). However, there is mounting evidence that younger PTC patients have a better disease-specific survival although they present high-risk clinical-pathological features (27–29). For PTMC under active surveillance, younger age was also associated with more probability of disease progression like tumor enlargement by ≥3 mm or novel appearance of nodal metastasis (30). The specific mechanism of age-dependent tumorigenesis and tumor progression has not been intensively investigated. Tumor surveillance by the immune system, which varies according to age, may be the reason for the difference in prognosis between younger and older patients (31).

Similarly, cervical lymph node metastases and ETE were more frequent in PTC patients with multifocality (32, 33). A meta-analysis from Australia summarized 21 papers and concluded the multifocality was an independent risk factor for recurrence, tumor progression, and aggressiveness (34). However, Wang et al. retrospectively reviewed 2,638 PTC patients from 11 medical centers from 6 countries and found that tumor multifocality has no independent risk prognostic value in clinical outcomes of PTC (35). This result was also validated on a series of 89,680 patients from SEER database (35). Recently, a study from Israel compared 690 PTC patients using propensity score matching analysis and demonstrated multifocality was not an independent prognostic factor for long-term outcomes of PTC (36). Multifocal PTC can represent either an intraglandular spread from a single primary tumor or multiple independent foci accompanied by intrathyroidal metastasis (37–39). These different clonal origins of multifocal tumor have distinct growth patterns, which may explain discrepancy between the above-discussed studies (40).

Furthermore, another novelty of this study is the first meta-analysis for ultrasonic features of LLNM in PTMC. A >25% tumor contact with thyroid capsule is related with the degree of ETE, which is also associated with LLNM of PTMC. Moreover, microcalcification(s) at US can also predict a diagnosis of malignancy, clinically cervical lymph node metastases, and prognosis, as reported previously (41, 42). In addition, an increasing number of studies have reported that upper tumor location was related with LLNM in PTC and PTMC no matter whether central neck had metastatic lymph node or not (11, 20, 43, 44). The presence of a direct lymphatic drainage from the upper third of thyroid to the neck may explain this finding (44, 45).

By meta-analyzing populations from different studies, the present study was able to evaluate the risk factors of LLNM for PTMC in a larger study sample and to adjust the results for the presence of some confounding factors. Except for ETE and tumor location, high heterogeneity of other risk factors was compensated by sensitivity analysis. The results of ETE and tumor location should be interpreted with caution because of moderate heterogeneity, and further study is needed to confirm the corresponding results.

This meta-analysis has some potential limitations. First, patients in the 14 studies were predominantly Asian. Whether ethnicity plays a role in LLNM remains unknown. Accordingly, risk factors of LLNM in PTMC from other races need further evaluation. Second, the limited number of studies hindered the implementation of subgroup and meta regression analysis. Third, the retrospective and non-randomized nature of all studies included in the analysis might be considered a source of bias. This provided associative, not causal, evidence, and mandates caution when interpreting these results. In future trials, randomized controlled trials of a higher methodological quality are needed to improve the quality of evidence.

Taken together, several risk factors for LLNM in PTMC patients which are readily available in clinical settings were identified in our systematic review and meta-analysis. The identification of high-risk patients is of utmost importance to plan a more cautious and frequent evaluation of lateral neck in the post-operative course. Moreover, high-risk ultrasonic features also need to be considered during selection of PTMC for active surveillance.

All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.