Background
Lung cancer is the leading cause of death from cancer, resulting 1.38 million people deaths each year [
1]. Its 5-year survival rate is still as low as 15%, and it is poor while compared with those in high incidence of other cancer [
2]. Previous studies pointed out that lung cancer is the most common cancer among men and women, and both developed and developing countries bear a huge social and economic burden [
3]. Previous publications proved that both genetic and environment factors were related to lung cancer risk [
4‐
7]. Furthermore, trace-heavy elements also played a significant role on human health and disease [
8,
9], as well as lung cancer [
10].
Magnesium is one of the trace elements in our bodies, and to date, some papers had been published to investigate the association between serum magnesium levels and lung cancer risks. Two papers [
11,
12] reported a higher of serum magnesium level in cases with lung cancer, while six papers [
13‐
18] found a lack of significant association. Conversely, three papers [
19‐
21] suggested that it is lower in lung cancer cases when compared to the healthy controls. Therefore, the aim of this study was to identify and synthesize all citations evaluating the relationship between serum magnesium levels and lung cancer risk.
Methods
Study selection
A comprehensive literature search was conducted in platforms of PubMed, WanFang, China National Knowledge Internet (CNKI), and SinoMed databases up to December 31, 2017. Free words adopted were as follows: “magnesium” or “Mg” combined with “lung cancer” or “lung carcinoma” without restrictions. Reference lists of the studies retrieved were also examined to find any additional study potentially unidentified. The course of study selection was completed by two investigators independently. Any resulting discrepancies were resolved by a third reviewer.
The inclusion criteria were as follows: (i) having a prospective design or a case-control design or a cross-sectional study; (ii) evaluating the association between serum magnesium levels and risk of lung cancer; (iii) reporting mean and standard deviation (SD) of magnesium levels (or sufficient data to compute them) both in lung cancer patients and healthy controls; and (iv) studies published in English language or Chinese language. If more than one article referred to the same populations, only the study that included the most lung cancer cases or the latest publication was included.
Data extraction and quality assessment of studies
Two investigators independently extracted the following data: (1) first author’s last name; (2) publication year; (3) study design; (4) country; (5) number of lung cancer cases and participants; (6) sex of cases; (7) age range or mean age of the cases; (8) mean and SD of magnesium levels both in lung cancer patients and healthy controls; and (9) method used for detection of magnesium. Any resulting discrepancies were resolved by a third reviewer.
The methodological quality of studies was evaluated independently by two researchers using the Newcastle-Ottawa Quality Assessment Scale [
22]. The three components were as follows: (1) patient selection (4 points); (2) comparability (2 points); and (3) outcome (3 points) for a total score of 9 points.
Statistical analysis
Standardized mean difference (SMD) and their 95% confidence interval (CI) were calculated for relationship between serum magnesium levels and lung cancer risk. A random effect model was used in our meta-analysis [
23]. The heterogeneity among studies was evaluated with
I2 and Q tests. [
24].
P < 0.05 in Q test and
I2 > 50% indicated statistically significant heterogeneity [
25]. Meta-regression was adopted to assess the between-study heterogeneity. Egger’s regression asymmetry test [
26] and funnel plot [
27] were used to visually examine publication bias on study outcome. Statistical analyses were performed using STATA version 12.0 (Stata Corporation, College Station, TX, USA). A two-sided
P < 0.05 was defined as statistical significance.
Discussion
In this study, we assessed the association between serum magnesium levels and risk of lung cancer. We did not find a positive association between serum magnesium levels and lung cancer risk. Through our subgroup analysis, we further found no significant association among Asian and European populations. Significant heterogeneity between studies observed in this meta-analysis should be considered as a major limitation of these findings; however, heterogeneity was mainly related to strength of the association rather than the direction of risk estimate, suggesting overall promising findings on the outcome investigated in the present study.
Two previous prospective cohort studies concluded that higher category of dietary magnesium intake had no significant association on lung cancer risk among German population and China population [
28,
29]. However, a report [
30] had been resulting that higher magnesium levels in drinking water could reduce the risk of lung cancer deaths in women. To our knowledge, no comprehensive analysis had been published to assess the serum magnesium levels on lung cancer risk. In our study, we did not find significant association of lower serum magnesium levels in patient with lung cancer. However, level of magnesium in other disease may be in the normal range, and that magnesium can have an effect on this disease [
31].
The existence of heterogeneity among the studies, which is common in meta-analyses [
32], may affect the pooled results. Meta-regression was performed to find the potential covariates (publication year, case number, geographic location, sex, and different methods to detect magnesium levels) which may cause this high heterogeneity. However, no covariate was found to significantly contribute to heterogeneity. In our study, most of the included studies obtained nonsignificant association between serum magnesium levels and lung cancer risk. Only one study [
12] reported that serum magnesium level in patient with lung cancer is extremely higher than that in healthy controls. We reviewed the article again and confirmed the data exacted from the study; no error was made. Sensitivity analysis was performed, and no study had essential effect to the significant between-study heterogeneity and the whole result. On the other hand, we used a random effect model to combine the results. As we all know, random effect model had wider rage about 95%CI than fix effect model and could obtain more accurate results. Furthermore, only three studies [
12,
13,
17] reported the types and staging of lung cancer, which may also be a factor on the between-study heterogeneity. Therefore, studies with detailed information of types and staging of lung cancer are wanted to further explore this association.
Some advantages existed in our study. Firstly, a comprehensive literature search was performed to investigate the relationship between serum magnesium levels and lung cancer risk. Secondly, most of the included studies involved large numbers of patients and healthy controls, and this may strengthen the power of the pooled results. Thirdly, there was no significant publication when tested by Egger and funnel plot, which indicates that our results are stable.
The present study has some limitations. Firstly, the individual studies may have failed to control for potential confounders, which may introduce bias in an unpredictable direction. Secondly, ten of 11 studies were from Asia, and 9 were from China, and thus, more related researches from other countries are wanted to verify the association between geographic location and lung cancer risk.
Conclusions
Based on the obtained results, we concluded that serum magnesium levels may have no significant association in patients with lung cancer. As we experienced some limitations in our study, such as more studies were from Asia, further studies are wanted to confirm this finding.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (
http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.