In this study, serum dp-ucMGP levels were positively associated with the AAC score by abdominal lateral X-rays in PD patients. In addition, serum dp-ucMGP levels was the independent predictor of vascular calcification in PD patients.
Similar to results observed in patients with CKD and on maintenance hemodialysis, we found that PD patients exhibited markedly elevated serum dp-ucMGP levels compared with the healthy controls(22, 23). Next, our findings confirmed that the serum dp-ucMGP levels were also significantly higher in PD patients with AAC than PD patients without AAC. Also, the possible association between dp-ucMGP and vascular calcification was repeatedly investigated in populations with high atherogenic status, such as elderly, uremic, HD, diabetics, atherosclerotic, and hypertensive patients(12). In agreement with the results published by Schurgers et al., who reported a positive association (both in univariate and multivariate analysis) between dpucMGP levels and the calcification score in 107 CKD patients (including 40 HD patients). Our present study found that the serum dp-ucMGP levels increased as the AAC worsened, indicating the positive association of AAC severity with increased serum dp-ucMGP levels in PD patients, which was further verified by correlation analysis. The area under the ROC curve of dp-ucMGP is 0.9227. And the sensitivity and the specificity are 85.53% and 92.31%, respectively, when the cut-off point is 6.99 ng/mL, indicating that the level of serum dp-ucMGP has a good reference value for evaluating the degree of vascular calcification.
Dp-ucMGP is the precursor of MGP and acts as inactive form of MGP. Several studies have demonstrated that MGP plays a central role in the inhibition of calcification and preventing the deposition of calcium in the vascular matrix, which requires vitamin K-dependent carboxylation(23, 24). Furthermore, dp-ucMGP is a widely accepted surrogate biomarker of vitamin K adequacy, and studies have showed that oral vitamin K supplementation increases carboxylation of MGP and lowers plasma dp-ucMGP(14, 23). Whereas the high dp-ucMGP levels are strongly suggestive for vascular vitamin K insufficiency to a level that the calcification-inhibitory activity of MGP is impaired, which may contribute to the extremely high risk for arterial calcification in hemodialysis patients(13, 25). Besides, dp-ucMGP was also identified as a survival marker, with lower levels heralding a better prognosis and decreased vascular calcification in previous studies(11, 19, 26).
Chronic inflammatory plays an important role in the pathogenesis of vascular calcification, which involves the activation of macrophages, release of proinflammatory cytokines, and the differentiation of vascular smooth muscle cells into osteoblasts(27, 28). Different studies have suggested a direct association of MGP with inflammation, furthermore, MGP is expressed by stimulated monocytes and macrophages, showing involvement in the endogenous inflammatory response mechanism of monocytes and macrophages(28, 29).
A previous study found that lower circulating dp-cMGP levels was associated with more inflammation and was predictive of higher mortality in hemodialysis patients. Patients with lower dp-cMGP levels were also associated with more calcifications as denoted by a composite semiquantitative calcification score(19). Unsurprisingly, our findings confirmed serum hs-CRP in PD patients with AAC was significantly higher than that of the PD patients without AAC, and a significant correlation between AACS and serum hs-CRP. However, we found no significant correlation between serum dp-ucMGP and hs-CRP in PD patients, possibly due to the different race, sample sizes, and the methodologies used to evaluate arterial calcification.
Vascular calcification is a complex process involving many factors. Until now, risk factors for VC, including older age, longer dialysis vintage, diabetes, hypertension, dyslipidemia, and smoking are confirmed both in CKD cohorts and non-CKD cohorts(30, 31). Moreover, hyperphosphatemia, elevated calcium-phosphate product, and hyperparathyroidism are well-recognized predictors of VC in dialysis patients. Consistently, our results showed that the PD-AAC group had higher BMI, longer dialysis vintage, serum corrected calcium, phosphate, calcium phosphorus product, iPTH, UA, and hs-CRP, but lower serum ALB than the PD-nonAAC group. We also found that the AAC score was positively associated with age, BMI, corrected calcium, ALP, and iPTH, while negatively correlated with ALB. The AAC risk factors identified in our study included age, dialysis vintage, corrected calcium, Ca-P product, ALP, iPTH, hs-CRP, and dp-ucMGP. These findings indicated that increased circulating calcium load in PD patients is associated with elevated vascular calcification. Multiple studies suggested that MGP acts as a calcification-inhibitor by inhibition of calcium-phosphate precipitation, the formation of matrix vesicles, the formation of apoptotic bodies, and trans-differentiation of vascular smooth muscle cells(32). Consistently, we also compared the correlation between dp-ucMGP and several clinical parameters, and found that serum dp-ucMGP was significantly associated with phosphate, calcium phosphorus product, ALP and iPTH.
We found that 59.4% PD patients suffered from abdominal aortic calcification, in accordance with published data in Asia population(33). Compared to hemodialysis, peritoneal dialysis patients have better residual renal function and electrolyte balance. Besides, Ca, phosphate, iPTH and bone turnover seem to play a lesser role in vascular calcification among PD patients compared with HD patients(34). Thereby, the incidence of calcification in the membrane PD group is lower than that in hemodialysis group(35, 36). Additionally, the association of Ca, phosphate and iPTH with mortality among PD patients were somewhat different from those among HD patients(37, 38). The Ca-P balance between hemodialysis and peritoneal dialysis needs to be further explored, as well as the association between vascular calcification and dialysis modality.
There are still several limitations in this study. First, the sample size is small, and all samples are from the same region. There might be some selection bias, and the results from various areas and different samples might differ. Second, this study is cross-sectional and we can not figure out the causal relationship between AAC and dp-ucMGP. Third, the lack information of phosphate intake and medication when collecting patient information leads to inevitable confounders. In addition, due to lack of follow-up information, the relationship between AAC progression and dp-ucMGP remains unknown. Larger multicenter studies are needed in the future.
In conclusion, our study implies the positive association between serum dp-ucMGP and AAC score in PD patients. Higher serum dp-ucMGP level is independently associated with AAC in PD patients. These findings indicate that dp-ucMGP could be used as a potential indicator of VC for further investigation.