The Study Of Vascular Calcification In Peritoneal Dialysis Patients

Objective1.To explore the prevalence of abdominal aortic calcification and associated factors of abdominal aortic calcification score(AACS)in peritoneal dialysis(PD)patients.2.To investigate the predictors of AACS progression in PD patients.3.To investigate the correlation between AACS and cardiac structure and function of PD patients4.To explore the association of AACS with all-cause mortality as well as cardiovascular disease(CVD)outcomes in PD patients.Methods1.Eligible PD patients in Renji Hospital,Shanghai Jiao Tong University School of Medicine from July 2011 to July 2014 were enrolled.Abdomen lateral Xray was used to determine baseline AACS for each patient at enrollment.The incidence of abdominal aortic calcification was evaluated.A multifactorial ordered logistic regression model was used to study AACS related factors.Criteria for inclusion in the multivariate model were P<0.1 in univariate Logistic regression analysis and variables regarded as clinically significant.2.A second AACS was determined on a repeated abdomen lateral X-ray obtained after 24 months of follow-up.Progression of abdominal aortic calcification was defined as an increase in AACS compared with the bassline.A multivariate binary logistic regression model was used to explore the independent predictors for abdominal aortic calcification progression.Criteria for inclusion in the multivariate model were P<0.1 in univariate Logistic regression analysis and variables regarded as clinically significant.3.Cardiac structure and function of all individuals were accessed using echocardiography at recruitment,including left atrial diameter,left ventricular end-diastolic diameter,left ventricular end-systolic diameter,left ventricular posterior wall thickness,interventricular septal thickness,and left ventricular ejection fraction,and peak mitral ring diastolic velocity in early(E')and peak diastolic velocity in late(A')were measured by tissue Doppler imaging.The left ventricular mass index(LVMI)was determined by the formula of Devereux with correction for body surface area.Left ventricular hypertrophy(LVH)was diagnosed with LVMI(29)115g/m2 in male and(29)95g/m2 in the female.Mitral annulus E'/A'(27)1 was used as the definition of left ventricular diastolic dysfunction.Spearman correlation analysis and Logistic regression model were used to explore the relationship between these cardiac structures and functional indicators and abdominal aortic calcification.A multivariate Logistic model was used to analyze the related factors of LVH.4.All patients were prospectively followed up until death,PD dropout,or to the end of the study(September 1st,2019).Cumulative incidences of all-cause mortality as well as cardiovascular outcomes among three groups were estimated using a competing risk model and compared through Gray test.The competing risk regression model was used to evaluate the association of AACS and all-cause mortality,cardiovascular events and CVD mortality.Results1.Two hundred and ninety-two PD patients were enrolled in this study.The cohort consisted of 160 males(54.8%)with the mean age of 57.1 ± 15.2 years and median PD vintage 28.4(IQR 12.0,57.8)months.Among them,seventy-five patients(25.7%)had diabetes,and ninety-four(32.2%)had a history of CVD.The abdomen lateral X-ray showed that 167(57.2%)subjects in this cohort had abdominal aortic calcification.and their average AACS was 2.0(0.0,6.0).Patients were divided into 3 groups based on the tertiles of AACS(Low AACS group,AACS=0,n=125;Medium AACS group,0(27)AACS?4,n=72;High AACS group,4(27)AACS?24,n=95).Multivariate order logistic regression analysis showed that longer age(OR 1.081,95% CI 1.056-1.107,P(27)0.001)and PD vintage(OR 1.012,95% CI 1.004-1.019,P=0.003),diabetes(OR 2.554,95% CI 1.415-4.609,P=0.002)and CVD history(OR 1.919,95% CI 1.108-3.325,P=0.020)were independent risk factors of escalating AACS in PD patients.2.A total of two hundred and ten patients underwent the second abdomen lateral X-ray after follow-up of 24 months,including 114 males(54.3%)with mean age of 58.4 ± 14.7 years and median PD vintage 30.0(16.5,58.5)months,and 50 patients(23.8%)had diabetes.The second AACS after 24 months was significantly higher than the baseline [2.0(0.0,8.0)vs.1.0(0.0,3.5),P(27)0.001].Abdominal aortic calcification progressed in 99 patients(47.1%).Multivariate Logistic regression model showed that age(OR = 1.032,95% CI 1.003-1.061,P = 0.029),longer dialysis duration(OR = 1.013,95% CI 1.001-1.026,P = 0.037),elevated phosphorus(OR = 2.599,95% CI 1.134-5.958,P = 0.024),and the presence of abdominal aortic calcification at baseline(OR = 7.241,95% CI 3.206-16.352,P(27)0.001)were important risk factors predicting the progression of abdominal aortic calcification in PD patients.3.A total of 262 patients were enrolled in this part of the study,including 145 males(55.3%),with a mean age of 57.2 ± 15.4 years and median dialysis vintage 27.2(13.0,55.3)months,and 68 patients(26.0%)had diabetes,243(92.7%)had hypertension.154 patients(58.8%)had abdominal aortic calcification in this cohort,which is independently related with left atrial diameter(OR=1.178,95%CI 1.081-1.284,P=0.019),left ventricular enddiastolic diameter(OR=0.896,95%CI 0.804-0.998,P=0.046),and left ventricular diastolic dysfunction(OR=2.534,95%CI 1.022-6.281,P=0.045).After adjusted for cofounders like age,sex,BMI,diabetes,prior history of CVD,systolic blood pressure,pulse pressure,albumin,calcium,phosphorus,intact parathyroid hormone and hemoglobin,abdominal aortic calcification was found to be independently correlated with LVH(OR=1.986,95%CI 1.048-3.765,P=0.035).4.After the median follow-up of 47.5(24.4,61.2)months,95 patients(32.5%)died,and causes of death included lethal CVD strike(n=52),peritonitis(n=11),other infectious disease(pulmonary infection,biliary tract infection,etc)(n=19),gastrointestinal bleeding(n=5),gastrointestinal perforation(n=1),and unknown causes(n=7).The estimated cumulative all-cause mortality incidences and CVD-related mortality were significantly lower in patients of the Low AACS group than in their counterparts of Medium and High AACS group(Gray= 53.78,P(27)0.001;Gray= 22.35,P(27)0.001,respectively).Subdistribution hazards models found that AACS was independently predictable of all-cause mortality(SHR 2.983,95% CI 1.610-5.520,p(27)0.001)as well as CVD-related mortality(SHR 2.419,95% CI 1.111-5.270,p=0.026).CVD events occurred in 75(35.7%)patients during the follow-up,including acute coronary syndrome(n=15),acute congestive heart failure(n=8),cerebral infarction(n=14),cerebral hemorrhage(n=19),sudden cardiac deaths(n=22)and peripheral vascular disease(n=2).Patients of the Low AACS group had significantly lower cumulative incidences of CVD events(Gray= 28.68,P(27)0.001).After adjusted for cofounders,AACS remained as an independent predictor of CVD events(SHR=3.039,95%CI 1.519-6.080,P=0.002).Conclusions1.Abdominal aortic calcification was common in PD patients.Age,PD vintage,diabetes,and preexisting CVD were associated with higher AACS in the present cohort.2.Progression of abdominal aortic calcification was seen in a considerable proportion of PD patients over the two-year follow-up.Patients who were older,or with longer dialysis duration,or with the preexisting abdominal aortic calcification were at higher risk of abdominal aortic calcification progress while managing phosphorus levels might contribute to slow this process.3.Abdominal aortic calcification is related to the increase of LVMI and a decrease of left ventricular diastolic function and is also found to be an independent correlative factor for LVH in PD patients.4.AACS could predict all-cause mortality,CVD-related mortality as well as CVD events occurrence in this population.It thus might be a safe and feasible method to identify PD patients with adverse outcomes.