The Risk Factors For Intradialytic-hypotension And Its Impacts On The Outcomes Among Maintenance Hemodialysis Patients:a Prospective Cohort Study

Part Ⅰ A Study on the Risk Factors for Intradialytic-hypotension among Maintenance Hemodialysis PatientsObjectiveTo assess the risk factors of intradialytic-hypotension (IDH) among maintenance hemodialysis(MHD) patients and to provide clinical evidence for the prevention and treatment of IDH.MethodsWe recruited293patients who underwent hemodialysis during Jan.2009to Dec.2010. Intradialytic blood pressure was monitored during a3-month period. IDH was defined as an event characterized by a sudden drop in systolic BP more than20mmHg or in mean artery pressure (MAP) more than10mmHg associated with clinical events and need for nursing interventions. Logistic regression analysis was used to assess the risk factors for IDH.Results1、A total of293patients were recruited. Male61.4%. with an age range49～68. The primary diseases for end-stage renal disease were glomerulonephritis68.3%, diabetic nephropathy10.2%, hypertensive nephropathy8.2%, and etc.2、The incidence rate of IDH was39.9%.176patients with no-IDH (<1/10hypotensive events/3months) served as controls.80patients with o-IDH (>1/10but<1/3hypotensive events/3months) and37patients with f-IDH (>1/3hypotensive events/3months) were identified among all293patients of our dialysis center.3、Multivariate logistic regression analysis showed that age, ultrafiltration rate, serum NT-proBNP, serum albumin, serum β2MG and AoRD were associated with IDH among MHD patients.4、We used ultrafiltration volum/body weight (UFV/W) to evaluate IDH, the area under the ROC curve (AUC) of which was0.706(95%CI:0.64～0.77, P<0.01). The cut-off value of UFV/W for IDH was4.33%, with a sensitivity of53.2%and a specificity of76.5%.5、We used serum NT-proBNP to evaluate IDH, the area under the ROC curve (AUC) of which was0.762(95%CI:0.69～0.83,P<0.01). The cut-off value of serum NT-proBNP for IDH was1746.5pg/ml, with a sensitivity of88.6%and a specificity of51.1%. Furthermore, we used serum NT-proBNP to evaluate f-IDH, the area under the ROC curve (AUC) of which was0.654(95%CI:0.55～0.76, P<0.01). The cut-off value of serum NT-proBNP for f-IDH was8106.5pg/ml, with a sensitivity of41.2%and a specificity of86.3%.Conclusion1、Elderly, high ultrafiltration rate, high level of serum NT-proBNP, high level of serum β2NMG, hypoalbuminemia and shorter AoRD are independent risk factors for IDH among MHD patients.2、Serum NT-proBNP has a high sensitivity for evaluating IDH and a high specificity for evaluating f-IDH. Part Ⅱ A Prospective Cohort Study on the prognosis of Intradialytic-hypotension among Maintenance Hemodialysis PatientsObjectiveTo acknowledge the prognosis of intradialytic-hypotension among maintenance hemodialysis(MHD) patients for the prevention and treatment of IDH.MethodsWe recruited144patients who underwent hemodialysis during Mar.2009to May.2009. Intradialytic blood pressure was monitored during a3-month period. IDH was defined as an event characterized by a sudden drop in systolic BP more than20mmHg or in mean artery pressure (MAP) more than10mmHg associated with clinical events and need for nursing interventions. Dialysis-related information was collected such as complications, prognosis and etc. Kaplan-Meier method, log-rank test, logistic regression and Cox regression analyses were performed to examine the association between IDH and survival, using a follow-up through31May2011.Results1、A total of144patients were recruited. Male59.0%, with an average age58.02±14.91. The primary diseases for end-stage renal disease were glomerulonephritis60.4%, diabetic nephropathy11.8%, hypertensive nephropathy7.6%, and etc.2、The incidence rate of IDH was36.1%. Patients with o-IDH and f-IDH accounted for25.7%and10.4%respectively.3、During the two-year follow-up,21patients died, with a mortality rate8.0%per year,8.86per100person-year.12patients were censored during the two-year observation period.21deaths during the follow-up included7cerebrovascular events,5cardiovascular events,5infectious events,3tumor events and2events for sudden death.4、There was no difference of overall mortality rate, cardiovascular mortality rate and cerebrovascular mortality rate between the groups. And there was no difference of overall mortality rates between groups using Kaplan-Meire survival curve.5、The multivariate Cox regression model indicated that IDH increased the risk of death(HR=3.030,95%CI:1.036～8.860, P=0.043). We examined the relationship between BP and mortality using another multivariate Cox regression model. The decline in SBP and DBP were significant and independent risk factors for two-year mortality. The hazard ratio for death was3.429(95%CI:1.157～10.465) when the decline in SBP was analyzed in increments of20mmHg. and was3.139(95%CI:1.079～9.131) when the decline in DBP was analyzed in increments of10mmHg.6、The multivariate Cox regression model indicated that there was no significant difference of cardiovascular survival between groups.7、We divided IDH-prone patients into2groups according to the ultrafiltration volum/body weight. The high ultrafiltration-rate group had higher BMI. We compared the cardiac function at the beginning with that at the end of the study by echocardiography. The ejection fraction in the high ultrafiltration-rate group decreased. In low ultrafiltration-rate group, the multivariate Cox regression model indicated that the higher ejection fraction was. the lower two-year mortality was. We used EF to evaluate survival, the area under the ROC curve (AUC) of which was0.894(95%CI:0.86～0.97, P<0.01). The cut-off value of EF for survival was62.5%, with a sensitivity of80.8%and a specificity of83.3%.8、Among the CGN patients, the multivariate Cox regression model indicated that IDH increased the risk of overall mortality(HR=3.775,95%CI:1.074～13.272).9、Among the DN patients, the multivariate Cox regression model showed IDH was related to neither overall mortality nor cardiovascular mortality.10、339events happened overall during the follow-up:16cerebrovascular events,45cardiovascular events,42infectious events, and ect.11、The more frequently IDH happened, the more events of hospitalization and overall events took place individually. Logistic analysis showed IDH was the risk factor for gastrointestinal event (OR=6.209,95%CI:1.599～24.106,P=0.008). A positive relationship was found between the decline in SBP and cerebrovascular event. Gastrointestinal event had a negative relationship with the pre-dialysis BP and the lowest intradialytic BP.12、Logistic analysis showed cerebrovascular event was the risk factor for poor outcome (OR=2.854,95%CI:1.211～6.724, P=0.016) in MHD patients.Conclusion1、IDH is a significant and independent risk factor for two-year overall mortality. The lower intradialytic BP is monitored, the higher overall mortality rate is.2、Some of the IDH-prone patients have cardiac insufficiency, leading to poor outcome, while other IDH-prone patients have pretty good cardiac function. Due to long-term over-ultrafiltration, the latter patients might develop cardiac insufficiency.3、The more frequently IDH happens, the more events of hospitalization and overall events take place individually. Cerebrovascular event is the risk factor for poor outcome in MHD patients.4、The active prevention and treatment of IDH will greatly improve the prognosis of the MHD patients. Those IDH-prone patients who are young, in good nutritional status and have barely good cardiac-function should pay close attention to the control of intre-dialytic weight gain, thus to prevent or reduce the incidence of IDH, protecting cardiac-function. Those IDH-prone patients who have cardiac-insufficiency should take measures such as increasing the frequency of hemodialysis, lowering ultrafiltration-rate or extension dialysis hours for each single time, thus to prevent or reduce the incidence of IDH.