Study selection
We retrieved a total of 308 articles from Medline (n = 39), Embase (n = 121), PubMed (n = 103), and Web of Science (n = 45), and obtained three studies from the relevant references. After excluding 154 duplicates, the titles and abstracts of the remaining studies were reviewed, which led to the exclusion of 113 unrelated articles. Following exclusion of review articles and studies analyzing patient data from the same institution over the same time-period as well as irrelevant and unavailable data, 15 articles were finally included (Figure 1). A total of 4377 patients with GC underwent gastrectomy and lymphadenectomy in the included studies. Seven of the studies were performed in China, five in Japan, two in South Korea, and one in Germany. All studies contained at least one risk factor for SHLN metastasis. The general characteristics and quality assessments of the included studies are listed in Table 1.
Age
Five studies with a total of 1823 GC patients were included for analysis. The number of GC patients with No. 10 LN+ was 159, 179 in patients < 60 years and patients > 60 years, respectively (OR = 0.90, 95% CI = 0.54–1.48, I2 = 63%, p = 0.67). We observed no significant differences between the two groups (Figure 2a), suggesting that age is not correlated with SHLN metastasis. Owing to study heterogeneity, a sensitivity analysis was conducted by eliminating individual studies. Notably, heterogeneity was significantly decreased when the study of Aoyagi et al. (2010) was removed (I² = 18%).
Sex
Thirteen studies (including 2888 males and 1132 females) were used for gender analysis. No significant differences were observed between male and female groups (OR = 0.88, 95% CI = 0.72–1.07, I2 = 24%, p = 0.19; Figure 2b), indicating that gender is not a risk factor for SHLN metastasis.
Tumor size
Five studies were included in analysis of tumor size. Due to differences in statistical analyses of tumor diameter, we divided the information into two groups (> 5 cm vs. < 5 cm and > 10 cm vs. < 10 cm). No significant differences were evident between the > 10 cm and < 10 cm groups, and studies had high heterogeneity (OR = 0.58, 95% CI = 0.21–1.58, I2 = 71%, p = 0.28). However, the > 5 cm group exhibited a markedly higher incidence of SHLN metastasis, with no significant study heterogeneity (OR = 4.89, 95% CI = 2.98–8.03, I2 = 0%, p < 0.01; Figure 2c).
Tumor location
Eleven studies involving 341 patients with tumors located in the greater curvature (Gre) and 2809 with tumors in other locations were included in the analysis of tumor location. Differences between tumors within the greater curvature and those found elsewhere were significant (OR = 3.10, 95% CI = 1.92–5.02, I2 = 54%, p < 0.01; Figure 2d). The results suggest that Gre location of tumors is a risk factor for SHLN metastasis. In the sensitivity analysis, heterogeneity did not change significantly upon elimination of individual studies.
Lauren’s type
Four articles containing 929 patients evaluated Lauren’s type, which included both diffuse and intestinal subtypes. After combining the data, significant differences were evident between diffuse and intestinal types (OR = 2.91, 95% CI = 1.84–4.59, I2 = 0%, p < 0.01; Figure 2e), indicating that diffuse type is associated with an increased incidence of SHLN metastasis.
Borrmann type
Nine articles included information regarding Borrmann type. Based on the Borrmann classification system, GCs are divided into four types (I, II, III, and IV). We combined types I–III and compared the data with type IV. The overall heterogeneity of the two groups was small and type IV was associated with a significantly increased risk of SHLN metastasis (OR = 2.49, 95% CI = 1.84–3.37, I2 = 0%, p < 0.01; Figure 2f).
Histological differentiation
Twelve studies provided data on histological differentiation. We set poorly differentiated and undifferentiated types as the exposure groups and moderately-differentiated and well-differentiated types as the control groups. Subsequently, studies were individually removed for sensitivity analysis. Our results showed no significant changes in heterogeneity and the poorly differentiated and undifferentiated types were associated with significantly higher risk of SHLN metastasis (OR = 2.29, 95% CI = 1.80–2.92, I2 = 25%, p < 0.01; Figure 2g).
Depth of invasion
Data regarding depth of invasion were included in 10 studies. T3 and T4 were set as the exposure group and T1 and T2 as the control group. There was no significant heterogeneity in either group, and T3 and T4 groups exhibited a markedly higher rate of SHLN metastasis than T1 and T2 groups (OR = 6.39, 95% CI = 4.04–10.12, I2 = 1%, p < 0.01; Figure 2h).
Lymph node metastasis
Nine studies provided information on lymph node metastases, including N1, N2, and N3. N1 and N2–3 were grouped separately. After the data were combined, GC patients with N2 or N3 exhibited a significantly increased risk of SHLN metastasis (OR = 6.96, 95% CI = 4.64–10.44, I2 = 44%, p < 0.01; Figure 2i). Owing to heterogeneity, we conducted a sensitivity analysis by eliminating individual studies and found that heterogeneity disappeared when the study of Huang (2009) was removed (I2 = 0%).
Distant metastasis
Three articles, including 1272 patients provided data on distant metastases. After the data were combined, no heterogeneity was evident, and distant metastasis was associated with a significantly higher rate of SHLN metastasis (OR = 8.66, 95% CI = 5.53–13.56, I² = 0%, p < 0.01; Figure 2j).
Neurological, vascular, and lymphatic invasion
We excluded several studies reporting combined data on blood vessels and lymphatics. Two studies included data on neurological invasion. Upon combination of data from both studies, no significant differences between the SHLN+ group and the SHLN- group were evident (OR = 1.72, 95% CI = 0.98–3.03, I2 = 16%, p = 0.06; Figure 2k). Three studies examined vascular invasion. When the data were combined, differences between the SHLN+ group and the SHLN- group were significant (OR = 2.57, 95% CI = 1.21–5.47, I² = 0%, p = 0.01; Figure 2l). Three articles provided information on lymphatic invasion. Upon combination of the data, we observed a significant difference between the two groups (OR = 3.41, 95% CI = 1.81–6.44, I² = 0%, p < 0.01; Figure 2m). Data from our meta-analysis collectively suggest that the presence of vascular and lymphatic invasion is a significant risk factor for SHLN metastasis.
TNM stage
Eight articles reported TNM stage that was classified into four subtypes (I, II, III, and IV). We combined types I–III into a single group for comparison with type IV. The heterogeneity of the two groups was small and type IV was associated with a significantly increased risk of SHLN metastasis (OR = 22.70, 95% CI = 11.57–44.56, I² = 0%, p < 0.01; Figure 2n).
Other groups with positive lymph node metastasis
Four studies referred to other regional lymph node metastases, which were associated with SHLN metastasis. Their combined values and effects are listed in Table 2. The results showed that other regional lymph nodes, with the exception of No. 5 LN (p = 0.14) and No. 8a LN (p = 0.10), are associated with SHLN metastasis. Sensitivity analysis conducted by changing the effect model revealed no significant heterogeneity.
Publication bias
Publication bias was assessed only when more than 10 studies were included in the risk factor analysis. We observed no obvious asymmetry in the Funnel plot of histological differentiation (Figure 3). Similarly, other aggregated data did not exhibit publication bias.