Introduction

Chronic kidney disease (CKD) is characterized by an incidence of adverse cardiovascular events higher than in general population, such as left ventricular hypertrophy, coronary diseases, arrhythmia, and other cardiovascular events [1] (full text) [2]. They have the same underlying factor that is arterial hypertension as cause and/or expression of renal disease in its different stages. Thus, hypertension treatment is the first step to treat renal disease and the principal aim in these patients.

In the last decade blood pressure variability (BPV) measured during a follow-up of chronic kidney disease (CKD) patients with hypertension [3] [4] [5] (full text) [6] (full text) or hemodialysis patients [7] (full text) [8] [9] (full text) has received a great attention.

Also our group focused his interest on this issue and studied 374 subjects with CKD stages 3 and 4 from January 1^st 2006 to December 31^st 2010 in a historical prospective analysis to evaluate the relationship between BPV and mortality and dialysis start [10] (full text). At the end of follow-up 131 patients were still alive and not yet on dialysis; we defined these patients long survivors (LS).

Our aim was to to understand the relationship between BPV and mortality and/or dialysis initiation in LS subjects with CKD.

Materials and Methods

We conducted a historical prospective observational multicentric study in 131 subjects still alive at 31^st December 2010, when ended a our previous study that was published on Nephrology Dialysis Transplantation [10] (full text). Details of materials and methods are described in detail elsewhere (see reference below) [10] (full text).

Statistical Analysis

Data are expressed as mean ± SD (standard deviation); the statistical analysis is described elsewhere [10] (full text). We stratified the included patients in quartiles to identify factors associated to BPV. BPV quartiles were compared with variance analysis and chi-squared tests.

We used Kaplan-Meier curves for patients' and renal survival analysis. We adjusted our analyzes for age, sex, diabetes, renal function, blood pressure, hemoglobin, phosphate, and low-protein diet.

Statistical significance was set at p<0.05.

Results

After 60 months of observation [10] (full text), 131 patients were still alive and not yet on dialysis treatment; we defined these patients LS. They had a mean age of 72±7 years, 59% of them were male and 18% were diabetics. Tabella 1 shows demographic, clinical and laboratory baseline data of our population studied from January 2006 and of the LS patients. LS subjects have been observed for other 36 months. LS were younger (p<0.01) and had a lower blood pressure compared to the original population. Also, they had significantly lower creatinine levels (p<0.019), phosphate levels (p<0.05) , and higher hemoglobin levels (p<0.05).

Tabella 2 shows data of LS patients according to quartiles of BPV. There were no statistical differences in blood pressure, sieric sodium, albumin and hemoglobin levels; the number of anti-hypertensive drugs was significantly lower from the first to the fourth BPV quartile (p<0.05).

During a follow-up of 80.7±13.4 months, 63 patients (48.1%) died and 49 (37.4%) started dialysis treatment. In this group of patients, 28 of them died after dialysis initiation.

Kaplan-Meier curves showed a significant association between BPV and mortality risk (Figura 1), as already shown during the first 60 months of follow-up of the population included in 2006 [10] (full text). In contrast to what happened to dialysis initiation [10] (full text), there was a significant association between BPV and progression of CKD in LS patients .

To estimate patients' and renal death, we adjusted our Cox models for several demographic, case-mix and laboratory variables (Tabella 3). Despite the adjustments, the association between BPV and mortality was present (>HR: 1.061; 95% CI: 1.035–1.093; P = 0.001) and also between BPV and dialysis initiation (HR 1.049; 95% CI: 1012–1.74; P = 0.001) (Tabella 3).

Discussion

The discussion of our previous paper [10] (full text) started with the following sentence: “The major finding of the current study is the independent association of SBPV and the risk of death but not of CKD progression among CKD patients not receiving dialysis. Though limited by the small sample size and by the lack of fatal events after dialysis initiation, these results suggest that the excessive risk of death associated with SBPV is not altered by dialysis initiation”.

The present data evidence that BPV is not only associated with mortality but it is also an important risk factor for dialysis initiation in a longer follow-up.

Thus, our study evidences that population referring to nephrological outpatient clinics has a reduced incidence of mortality and dialysis initiation: these patients are younger and have lower levels of blood pressure and phosphate, and higher levels of hemoglobin without therapeutic intervention and higher stages of CKD, but also significantly lower values of BPV.

Further prospective studies are needed to understand factors involved in the relationship between BPV and patients' and renal survival, but also to understand how to reduce BPV in CKD subjects [11] (full text) [12] [13] (full text) [14].

In conclusion, our data from Long Survivors confirm that BPV may be used to stratify cardiovascular mortality risk e renal death risk in CKD patients.