Correlation between soluble klotho and chronic kidney disease–mineral and bone disorder in chronic kidney disease: a meta-analysis

We conducted a systematic search across medical databases, including PubMed, Web of Science, EMBASE, and Cochrane Library, up to March 2023. A total of 1944 subjects or individuals from 17 studies were included in our final analysis. The correlation coefficient (r) between sKlotho and calcium was [0.14, (0.02, 0.26)], and a moderate heterogeneity was observed (I2 = 66%, P < 0.05). The correlation coefficient (r) between Klotho and serum phosphate was [− 0.21, (− 0.37, − 0.04)], with apparent heterogeneity (I2 = 84%, P < 0.05). The correlation coefficient (r) between sKlotho and parathyroid hormone and vascular calcification was [− 0.23,(− 0.29, − 0.17); − 0.15, (− 0.23, − 0.08)], with no significant heterogeneity among the studies. (I2 = 40%, P < 0.05; I2 = 30%, P < 0.05). A significant correlation exists between low sKlotho levels and an increased risk of CKD–MBD in patients with CKD. According to the findings, sKlotho may play a role in alleviating CKD–MBD by lowering phosphorus and parathyroid hormone levels, regulating calcium levels, and suppressing vascular calcification. As analysis showed that sKlotho has an important impact on the pathogenesis and progression of CKD–MBD in CKD patients. Nonetheless, further comprehensive and high-quality studies are needed to validate our conclusions.


Study design
The study has been registered in the International Database of Prospective Systematic Reviews (Registration No. CRD42022385019) and conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 20 .

Data sources and search strategy
The electronic databases PubMed, Web of Science, EMBASE and Cochrane Library were comprehensively queried for pertinent literature from their inception until March 11, 2023.Each publication's bibliography was also perused to discover any additional material related to this subject.
If necessary, we will contact the authors who conducted the investigation to obtain more information regarding the eligibility criteria in order to respond to any inquiries.The Supplemental 1 provides the search terms and a detailed search approach.

Study inclusion and exclusion criteria
Inclusion criteria: (1) Participants were adults aged at least 18 years; (2) The study type was cohort or observational; (3) The study included complete data information; (4) The publication language is English; (5) Include at least one outcome relevant to the study of interest.
Exclusion criteria: (1) Incomplete study data or statistical errors; (2) Duplicate studies; (3) Nonclinical research, including in vivo or in vitro experimental; (4) Overlapping or duplicated research populations; (5)  Case reports, abstracts, reviews, conference summaries and comments; (6) Study aims to investigate the correlation between renal klotho expression, including mRNA and protein levels, and other relevant parameters; (7)  Special participants, such as patients who have undergone kidney transplantation.Exclude patients with diabetic nephropathy; (8) The research design is not standardized.

Data extraction and quality assessment
Using a standardized form, two independent investigators (Zhongyu Fan and Xuejiao Wei) extracted the data.The following data were extracted: First author, year of publication, location, age, research index, sample size, study type, disease models and correlation coefficient (Pearson or Spearman).We transformed the published Pearson correlation coefficient into Spearman correlation coefficient in order to better use the data because the latter are unaffected by the logarithmic modification.In addition, by applying the Fisher's transformation, the correlation coefficients are converted into approximately normally distributed Z-values 21 .The standard deviation (standard error, SE) of Z is then calculated, followed by the inverse Fisher's transformation of the Z-values to obtain the correlation coefficient and confidence intervals (CI).If data is unavailable or incomplete, the corresponding authors will be contacted to obtain the relevant data.The disagreement in data extraction were resolved by consulting a third reviewer.Two authors (Xiaoyu Zhu and Kun Yang) evaluated the research quality and risk of bias using the Newcastle-Ottawa Scale (NOS) 22 .Studies that received a score of 7 or higher were deemed to have a low risk of bias and were classified as high-quality studies.

Meta-analysis and statistical analysis
Data analysis was conducted using Review Manager 5.4 software (Cochrane Collaboration, Copenhagen, Denmark).When I 2 < 50%, a fixed-effect model was employed.When I 2 ≥ 50%, a random-effect model was used 23 .In the presence of significant heterogeneity, sensitivity and subgroup analyses were conducted to examine the sources of heterogeneity.A sensitivity analysis was conducted, excluding each study sequentially, to ensure the consistency of the outcomes.We visually examined the probability of publication bias by using funnel plots.A two-tailed test with P < 0.05 was considered statistically significant.

Results
After searching PubMed, Embase, Cochrane Library and Web of Science, a total of 2382 publications that were relevant to the topic were selected (PubMed: n = 535; Embase: n = 648; Cochrane Library: n = 71; Web of Science: n = 1128).
In these studies, a total of 788 duplicates were removed.Through screening titles and abstracts, a total of 627 publications were excluded.After reading the full texts, an additional 122 articles were excluded due to various reasons, resulting in a final selection of 17 studies for our meta-analysis.The procedure for selecting articles is showed in Fig. 1.The risk of bias in included literature is displayed in Fig. 2.

The association between sKlotho and calcium
Figure 3 presents a meta-analysis of the correlation between sKlotho and Ca.Seven included studies demonstrated a positive correlation between sKlotho and Ca, while one study showed a negative correlation.There was moderate heterogeneity among the eight studies (I 2 = 66%, P < 0.05), so a random-effect meta-analysis was conducted, and the pooled correlation coefficient(r) and its 95% CI were [0.14, (0.02,0.26)], indicating a significant positive correlation between sKlotho and Ca.
Sensitivity analysis by removing any study from this study did not affect the results, suggesting the stability and reliability of the random-effects calculations.
As shown in the funnel plot (Supplemental Fig. 1), the distribution is asymmetric, indicating the presence of publication bias.
To further investigate the underlying causes of the considerable heterogeneity observed in these studies, subgroup meta-analyses were performed, stratifying the analysis based on disease model (Pre-dialysis or Dialysis), average age (≥ 60 years or < 60 years), sample size (N ≥ 100 or N < 100), and study quality (score ≥ 7 or < 7).The results of the subgroup meta-analyses are shown in Table 2 and Supplemental Figs.2a-d.In all subgroups, the positive correlation remained significant.Subgroup age exhibited statistical apparent heterogeneity (I 2 = 93.7%,P < 0.05).Therefore, age was considered as potential source of heterogeneity among the studies .

The association between sKlotho and phosphate
Eight studies were evaluated to assess the relationship between sklotho and phosphate.Seven studies found a positive correlation between sklotho and P, whereas one found an inverse correlation.The studies were found to have a high degree of heterogeneity (I 2 = 84%, P < 0.05).So a random-effect meta-analysis was performed, and the pooled correlation coefficient(r) and its 95% CI were [-0.21,(-0.37,-0.04)],demonstrating a remarkable negative connection between sKlotho and P (Fig. 4).Sensitivity analysis conducted on these eight articles showed that none of them significantly influenced the results of the meta-analysis, indicating strong stability using random-effects calculations.
The symmetry observed in the funnel plot suggests without presence of publication bias in the literature included in this study.(Supplemental Fig. 3).
Furthermore, the subgroup analysis results demonstrated that disease model possible contributors to the observed heterogeneity (I 2 = 82.7%,P < 0.05) (Table 3 and Supplemental Figs.4a-d).
As evidenced by the funnel plot, there is an asymmetrical distribution, suggesting the existence of publication bias.(Supplemental Fig. 5).
There was no signifcant heterogeneity between subgroups in analyses of disease model, average age, sample size and study quality (P > 0.05) (Table 4 and Supplemental Fig. 6a-d).

The association between sKlotho and vascular calcification
There are six articles discussing the connection between sKlotho level and vascular calcification.Five studies discovered an inverse correlation between klotho and correlation, while one investigation indicated a positive correlation.The results of the meta-analysis reveal that there is inconspicuous heterogeneity among the studies www.nature.com/scientificreports/using the fixed effect model (I 2 = 30%, P < 0.05).The pooled correlation coefficient(r) and its 95% CI were [-0.15, (-0.23, − 0.08)] (Fig. 7).
Strong stability in this study is indicated by the sensitivity analysis done on these six publications, which revealed that none of them significantly affected the findings of the meta-analysis.
The identified symmetrical pattern in the funnel plot suggests no publication bias remain in the contained literature (Supplemental Fig. 7).To the best of our knowledge, our study is the first to conduct a meta-analysis on the correlation between sKlotho levels and CKD-MBD in patients with CKD.In this article, we investigated the correlation between sklotho and calcium, phosphorus, parathyroid hormone and vascular calcification.Positive correlation between sKlotho and calcium, as well as negative correlation with phosphorus, parathyroid hormone, and vascular calcification were observed, all of which were statistically significant.Although moderate heterogeneity was observed in the study of the correlation between sklotho and calcium and phosphorus (I 2 = 57%, P < 0.05; I 2 = 57%, P < 0.05), the correlation still exists in sensitivity analysis, indicating that the conclusion obtained might be considered relatively reliable.
Acting as an autocrine or paracrine enzyme, Klotho modulates the process through downregulation of the renal sodium phosphate cotransporter NaPi-2a expression 46 .This action leads to a decrease in urine phosphorus reabsorption, effectively countering hyperphosphatemia 49 .The soluble form of Klotho, acting independently, enhances calcium reabsorption while diminishing calcium loss in the kidneys through the regulation of calciumselective channels 50,51 .Additionally, it has been observed that Klotho increases calcium absorption in urine by stabilizing the TRPV5 calcium channel, regardless of the presence of FGF23 52 .The analysis conducted in our study unveiled a significant discovery: a negative correlation between klotho and serum phosphorus, along with a positive correlation between klotho and serum calcium.These correlations were statistically significant, consistent with the aforementioned research findings.
The presence of sKlotho has been identified in parathyroid tissue, and research has revealed that the excessive depletion of the Klotho-FGF receptor complex in patients with end-stage renal illness is linked to the overproduction of PTH and secondary hyperparathyroidism (SHPT) 53,54 .This investigation observed a negative correlation between sKlotho and PTH, suggesting that sKlotho exerts an inhibitory influence on PTH production.
In vitro studies have demonstrated that the presence of sKlotho impedes the process of sodium-dependent phosphate uptake and mineralization, often triggered by hyperphosphate 55 .According to recent research, people with MHD who have high levels of sKlotho had a 61% lower risk of cardiovascular events and cardiovascular mortality when compared to those who have low levels of sKlotho 56 .sKlotho suppresses vascular calcification (VC) by obstructing the Wnt/β-catenin signaling pathway and preventing the development of vascular osteoblasts 57 .The absence of Klotho has been associated with the occurrence of soft tissue calcification in patients with CKD 58 .sKlotho has also been found to enhance phosphaturia, preserve glomerular filtration, and directly inhibit phosphate absorption by vascular smooth muscle, thereby improving vascular calcification 46,50,59 .According to a prior study, patients with higher levels of sKlotho showed a decreased risk of serum creatinine doubling 28 .
According to the results presented above, it becomes evident that sKlotho has the ability to lower serum phosphorus and PTH levels, regulate blood calcium levels, and inhibit vascular calcification through various mechanisms, regardless of the presence or absence of FGF23.The meta-analysis reached a unanimous conclusion, which is consistent with the presented findings.Soluble klotho is merely one of numerous factors impacting mineral metabolism.Many other variables, including specific medications, involvement in inflammation, and dietary factors, influence serum phosphorus, parathyroid hormone, and calcium levels.Our analytical approach cannot comprehensively eliminate these confounding variables.
In the light of the subgroup analysis obtained, we identified that the study's heterogeneity concerning sklotho and serum calcium primarily stems from age.In the subgroup analysis investigating the correlation between sklotho and serum phosphorus, we noted that both age and disease models serve as the main sources of heterogeneity.Hence, the presence of heterogeneity might be attributable to factors such as age and the patients' dialysis status.Not surprisingly, a correlation exists between Klotho and lifespan, as supported by research indicating a decline in sKlotho levels with advancing age 14,62 .This suggests that the expression of sKlotho is regulated by the aging process.Earlier studies have established a significant link between reduced sKlotho levels and unfavorable outcomes among those undergoing dialysis 63 .The impact of sKlotho levels on clinical outcomes could potentially diminish as individuals age.A prospective association between sKlotho and the aging process emerges.Existing research has indicated that chronic renal failure profoundly affects gene expression and kidney function in MHD www.nature.com/scientificreports/patients 64 .Taken together, it is plausible that age and dialysis status may act as prominent confounding variables and contributors to heterogeneity.This meta-analysis has certain limitations.Firstly, the sample sizes of participants in several of the included studies is relatively small, despite the high overall number of patients engaged in this meta-analysis.Consequently, even though we applied a random-effects model for our research, it may result in an underestimation of our overall results.For all these reasons, further research is necessary to validate our findings.Secondly, this analysis relies on published studies, which often report positive and significant outcomes.Studies with negative or non-significant results are more prone to rejection.Additionally, certain results related to our analysis are still in the process of being accepted, potentially leading to publication bias.Thirdly, some of the studies included in our analysis only presented relevant indicators as positive or negative correlations, without providing r-values or the raw data required to calculate correlation coefficients.Fourthly, some of the outcomes exhibit heterogeneity, and although this can be partially reduced through subgroup or sensitivity analyses, potential confounding factors may still exist due to differences in the study populations and sklotho measurement methods.Nevertheless, since the number of included studies is limited, more detailed subgroup analyses cannot be carried out, which could influence the precision and usefulness of the results.
In summary, the findings of our study point to a robust association between sKlotho and CKD-MBD in patients with CKD, leading to a notable alleviation of mineral metabolism challenges within this population.However, it is crucial to recognize that this meta-analysis is subject to certain limitations.The outcomes of this study offer a positive outlook on the potential utility of sKlotho as a feasible biomarker for patients with CKD-MBD.Nevertheless, further comprehensive and rigorous investigations are imperative to validate our findings.

Figure 1 .
Figure 1.Screening procedures for the study.

Figure 2 .
Figure 2. Inclusion of literature quality assessment bias risk summary figure.

Figure 4 .
Figure 4. Forest plots of the summary r with effect estimate pooled r (95% CI) for the association between sKlotho level and P.

Figure 7 .
Figure 7. Forest plots of the summary r with effect estimate pooled r (95% CI) for the association between sKlotho level and VC.

Table 1 .
Characteristics of the studies included.

Table 2 .
Subgroup analysis results of sKlotho level and Ca.

Table 3 .
Subgroup analysis results of sKlotho level and P.