Randomized trial of bioelectrical impedance analysis versus clinical criteria for guiding ultrafiltration in hemodialysis patients: effects on blood pressure, hydration status, and arterial stiffness.

BACKGROUND: Chronic fluid overload is common in maintenance hemodialysis (HD) patients and is associated with severe cardiovascular complications, such as arterial hypertension, left ventricular hypertrophy, congestive heart failure, and arrhythmia. Therefore, a crucial target of HD is to achieve the so-called dry weight; however, the best way to assess fluid status and dry weight is still unclear. Dry weight is currently determined in most dialysis units on a clinical basis, and it is commonly defined as the lowest body weight a patient can tolerate without developing intra-dialytic or inter-dialytic hypotension or other symptoms of dehydration. One of the most promising methods that have emerged in recent years is bioelectrical impedance analysis (BIA), which estimates body composition, including hydration status, by measuring the body's resistance and reactance to electrical current. Our objective was to study the effect BIA-guided versus clinical-guided ultrafiltration on various cardiovascular disease risk factors and markers in HD patients. MATERIALS AND METHODS: We included 135 HD patients from a single center in a prospective study, aiming to compare the long-term (12 months) effect of BIA-based versus clinical-based assessment of dry weight on blood pressure (BP), pulse wave velocity (PWV), and serum N-terminal fragment of B-type natriuretic peptide (NT-proBNP). The body composition was measured using the portable whole-body multifrequency BIA device, Body Composition Monitor-BCM(®) (Fresenius Medical Care, Bad Homburg, Germany). RESULTS: In the "clinical" group there were no changes in BP, body mass index (BMI), and body fluids. The PWV increased from 7.9 ± 2.5 to 9.2 ± 3.6 m/s (P = 0.002), whereas serum NT-proBNP decreased from 5,238 to 3,883 pg/ml (P = 0.05). In the "BIA" group, BMI and body volumes also did not change; however, there was a significant decrease in both systolic BP, from 144.6 ± 14.7 to 135.3 ± 17.8 mmHg (P < 0.001), and diastolic BP, from 79.5 ± 9.7 to 73.2 ± 11.1 mmHg (P < 0.001). In this group, PWV also decreased from 8.2 ± 2.3 to 6.9 ± 2.3 m/s (P = 0.001) and NT-proBNP decreased from 7,552 to 4,561 pg/ml (P = 0.001). CONCLUSION: BIA is not inferior and possibly even better than clinical criteria for assessing dry weight and guiding ultrafiltration in HD patients.

Regression of left ventricular hypertrophy in hemodialyzed patients is possible: a follow-up study.

BACKGROUND: Few studies have addressed the description of serial changes in left ventricular mass (LVM) and relevant risk factors. The aims of our study were to describe trends in left ventricular (LV) structure and function derived from echocardiographic measurements over a 10-year period in Fresenius Nephrocare Dialysis Center in Iasi and to compare the results with those obtained on a smaller group 4 years ago. METHODS: Three hundred and thirty-four hemodialyzed patients were enrolled at baseline, between January 1999 and March 2009. Echocardiography was performed at inclusion and several times for each patient during this period, until the end of the study. Mean values of the biochemical parameters (hemoglobin, serum proteins, calcium, phosphate) at the time of the echocardiographic examination were calculated and included in the final analysis. RESULTS: Outcome in dialysis was 70.5% alive at the end of the study. The most important improvement was observed in LV mass index: at the 4th echocardiography, the mean LVMi was 144.8 vs. 156.0 g/m(2) at the 2nd echocardiographic examination vs. 167.2 g/m(2) at the first echocardiographic examination (mean decrease 3.34 ± 9.6 g/m(2)/month). Significant results were obtained by comparing LVMi only in patients with all 4 echocardiographies: left ventricular hypertrophy regression was statistically significant, from 172.7 g/m(2) at the 1st echocardiography to 146.0 g/m(2) at the 4th, i.e. 15.4% reduction of LVMi. Delta LVMi significantly correlated only with changes in hemoglobin (P < 0.05).There was a significant regression of the relative wall thickness from an average of 0.46 to 0.42 (P < 0.05). CONCLUSION: Our study proves that regression of LVH in hemodialyzed patients is possible and constitutes a must-achieve objective in dialysis centers.

The relationship between chronic volume overload and elevated blood pressure in hemodialysis patients: use of bioimpedance provides a different perspective from echocardiography and biomarker methodologies.

BACKGROUND: Chronic volume overload is very frequent in hemodialysis (HD) patients and is directly associated with hypertension, increased arterial stiffness, left ventricular hypertrophy (LVH), heart failure and ultimately with higher mortality and morbidity. One major issue is that presently there are very few comparative studies of the various methods (clinical, bioimpedance, inferior cava vein diameter (ICV) and Brain Natriuretic Peptide (NT-proBNP)) for volume status evaluation and their correlation with cardiovascular disease. METHODS: In 160 patients treated by chronic HD in our center, euvolemic according to clinical assessment, we performed evaluation of volume status through bioimpedance spectroscopy (BIS), ICV and NT-proBNP, as well as echocardiography, to estimate the left ventricle structure and function. RESULTS: Despite appearing clinically euvolemic, severe fluid overload, as defined by a relative tissue hydration (RTH)--i.e. fluid overload over extracellular water ratio (FO/ECW)--above 15% was found in 25.6% of patients. Four categories of patients were considered according to pre-HD BP and BIS values. Forty-five percent of patients (group A) had a reasonable control of BP and volume (SBP < 150 mmHg and RTH < 15%), 29.3% (group B) were classified as hypertensive (SBP > 150 mmHg and RTH < 15%), 16.7% (group C) had high blood pressure and marked volume expansion, (SBP > 150 mmHg and RTH > 15%), while 9% (group D) had SBP < 150 mmHg despite RTH > 15%. Assuming that BIS is the most accurate and validated method to assess hydration status, we calculated the positive predictive value for ICV-based evaluation--18%, with a sensitivity of 67% and an important proportion of false negative cases (45%). NT-proBNP was even less accurate: PPV of only 26%, with a sensitivity of 60% and a specificity of only 45% and an extremely high proportion of false positive cases (73%). Group A patients had the best cardio-vascular profile: lowest LV mass and NT-proBNP levels. CONCLUSION: Using multi-frequency body impedance spectroscopy, we found a large group of hypertensive and/or fluid-overloaded patients despite apparently being at "dry weight" on clinical evaluation and a marked discrepancy between clinical appearance and fluid status. Of the 4 different methods, assuming BCM "gold standard", there were major disagreements and discrepancies between the other three methodologies. BCM is a valuable and simple bed-side tool for the correct management of BP and risk stratification in HD patients as it allows for excellent discriminators of more abnormal cardiac and vascular profiles.

Serial echocardiographic changes in patients on hemodialysis: an evaluation of guideline implementation.

BACKGROUND: Few studies have addressed the description of serial changes in left ventricular mass (LVM) and relevant risk factors. All these studies were initiated before the implementation of EBPG or K/DOQI guidelines. The aims of our study were to prospectively describe trends in left ventricular (LV) structure and function, evaluate risk factors for progression of LVM derived from serial echocardiographic measurements starting January 2003, 6 months after full implementation of EBPG in our unit. METHODS: One hundred seventy-six patients were enrolled at baseline, between 1 January 2003 and 1 October 2004; 33 patients were excluded from analysis due to poor echocardiographic window, 14 patients died and 26 were transplanted or transferred during the follow-up period of minimum 12 months. One hundred and three patients were included in the final analysis (mean age 51 years, mean follow-up 24.1 +/- 14.4 months). Echocardiography was performed at inclusion and at the end of study. Biochemical, blood pressure (BP) and medication data were collected and the means of monthly values were used. RESULTS: At baseline, 86.4% of the patients had left ventricular hypertrophy (LVH) (56.3% concentric hypertrophy, 30.1% eccentric hypertrophy, 6.8% concentric remodeling and only 6.8% normal LV geometry), 25.6% had systolic dy-sfunction and 50.5% had abnormal LV volume index (LVVI). Similar data were recorded at follow-up (82.5%, 44.7%, 37.9%, 7.7%, 9.7%, and 19.5%, respectively). Baseline left ventricular mass index (LVMI) significantly correlated with hemoglobin (Hb) and total protein level. LVMI at follow-up correlated to Hb, SBP, PP, mean level of serum phosphate, calcium x phosphate product and cholesterol. Independent predictors for LVMI (multiple regressions) were anemia and mineral metabolism markers. In our population, 62.1% of the patients had a regression of LVMI, with a mean decrease in LVMI of 12 g/m 2 (1.7 +/- 11.7 g/m 2 /month) over more than 12 months of guideline implementation. Regressors had a significant improvement of anemia, serum phosphate level and calcium x phosphate product (p<0.05). CONCLUSION: Our study suggests that a holistic interventional approach, targeting various pathogenic mechanisms, as per guidelines, can elicit at least a partial regression in LV structural and functional abnormalities in hemodialysis patients.

Arterial wave reflections and mortality in haemodialysis patients--only relevant in elderly, cardiovascularly compromised?

BACKGROUND: Chronic kidney disease (CKD) patients have a 3-30-fold increased risk of death compared with the general population. This mortality difference is even more pronounced in younger subjects. Two markers of aortic stiffness--aortic pulse wave velocity (PWV) and augmentation index (AIx)--have been prospectively related to all-cause and cardiovascular (CV) mortality in end-stage renal disease (ESRD) populations. The aims of our study were first, to confirm the important deleterious effect of arterial stiffness in uraemia and second, to assess the impact on survival of increased AIx in a relatively young non-diabetic dialysis population, with minimal CV disease. METHODS: Ninety-two patients (mean age 42.6 +/- 11.2 years) were included in the study and followed for a period of 61 +/- 25 months. None of the patients had diabetes mellitus, and only 3.3% had prior history of CV disease. AIx was determined by applantation tonometry using a SphygmoCor device (AtCor, PWV Inc., Westmead, Sydney, Australia). RESULTS: Mean AIx in our study population was 19.9 +/- 13.7%; other significant haemodynamic parameters were: systolic blood pressure (SBP) 129 +/- 24 mmHg, pulse pressure 35.3 +/- 17.5 mmHg with 27.2% of the study population receiving angiotensin-converting enzyme inhibitors (ACE-I). On univariate analysis, in our group AIx correlated with: body weight (P < 0.001), radial SBP (P < 0.001) and haemoglobin levels (P < 0.05). There was no correlation between AIx and any of the echocardiographic parameters. In the stepwise multiple regression analysis, the only independent predictors for AIx were weight (P < 0.001), SBP (P < 0.001) and haemoglobin (P < 0.05) with the model explaining 33% of the AIx variability (adjusted R(2) = 0.33). During the follow-up period, 15 deaths were recorded. In the Cox analysis (P = 0.014; chi square 20.7 for the model) the only independent predictors for all-cause mortality were age (P = 0.001), left ventricular mass index (P = 0.032) and ACE-I therapy (P = 0.039) while AIx did not reach statistical significance. There was no difference in patients' survival when divided by AIx tertiles, assessed by the log rank test (P = 0.78). CONCLUSION: Our results fail to support the notion that an increased effect of wave reflections on central arteries is a strong and independent predictor of mortality in all ESRD patients on haemodialysis. The effect of arterial wave reflections might be in fact dependent on patient age and concurrent comorbidity status.

Increased arterial stiffness in children on haemodialysis.

BACKGROUND: Measures of aortic stiffness--aortic pulse wave velocity (PWV) and augmentation index (AIx)--have been shown to be powerful predictors of survival in adult haemodialysis (HD) patients. Very few data have been reported regarding arterial stiffness in paediatric renal populations. METHODS: PWV and aortic AIx were determined from contour analysis of arterial waveforms recorded by applanation tonometry using a SphygmoCor device in 14 children on HD (age = 14.1 years) and in 15 age, height matched children controls. RESULTS: Pre-HD AIx (29.7 +/- 15.4%) and PWV (6.6 +/- 1.0 m/s) were significantly higher compared with children controls (8.3 +/- 8.0% and 5.4 +/- 0.6 m/s, respectively, P < 0.0001). The only significant difference between normal and HD children was BP level: 103/61 vs 114/72 mmHg, P < 0.05. In children of HD patients, a multiple linear regression model including BP, age, height, weight, Ca and P levels as independent variables accounted for 57% of the variability in AIx. Dialysis had no impact on AIx (post-HD: 28.5 +/- 12.7%) or on PWV (post-HD: 6.7 +/- 0.8 m/s). CONCLUSIONS: We show, in this first-ever report of increased arterial stiffness in children on dialysis, that end-stage renal disease is associated with abnormalities in arterial wall elastic properties, comparable with adult levels, even in childhood. Most importantly, the absence of a discernible amelioration with dialysis implies that purely structural and not functional alterations lie behind the increased arterial stiffness.