Factors associated with residual urine volume preservation in patients undergoing hemodialysis for end-stage kidney disease in Kinshasa

In Kinshasa, the capital City of the Democratic Republic of the Congo, the overall prevalence of chronic kidney disease (CKD) in the general population is estimated to be 12.4%, with 0.2% of individuals with end-stage renal disease (ESRD) [1]. Late referral is the underlying cause for the admission of most patients with ESRD. For these patients, renal replacement therapy, including hemodialysis (HD), guarantees better survival and good quality of life. Among the criteria for effective HD, the preservation of residual kidney function (RKF) or residual urine volume (RUV) plays an important role in contributing to the survival and quality of life of hemodialysis patients [2‐4]. Patient demographic characteristics, comorbid disease and characteristics of dialysis treatment have been associated with a faster decline in RUV in dialysis patients. Among these factors, we identified increasing age [4, 5], female sex [4‐7], diabetes [4, 5, 8], hypertension [6, 8, 9], left ventricular hypertrophy (LVH) [10], congestive heart failure [4, 5], proteinuria [11], frequent dialysis [7, 12‐14], intradialytic hypotension [4, 5, 7, 11] and biocompatible membrane [4, 15] as relevant. In this context, the National Kidney Foundation’s Kidney Disease Outcome Quality Initiative (KDOQI) guidelines recommended the implementation of RKF preservation strategies in dialysis patients using RUV as a surrogate indicator [16, 17]. These strategies include the use of angiotensin-converting enzyme inhibitors (ACE) [18], angiotensin receptor blockers (ARB) [19] and diuretics [20]; the control of hypovolemia [21], obesity [22] and high blood pressure; the avoidance of nonsteroidal anti-inflammatory drugs, aminoglycosides and radiographic contrast agents [23]; the reduction of LVH [10]; a lower dose of dialysis [13]; the use of biocompatible dialysates and dialyzers and bicarbonate rather than acetate; and the early initiation of HD [5, 24, 25]. It is now accepted that the use of ultrapure water in HD helps preserve RKF [26]. Although the definition of the RKF varies across studies, RUV may emerge as a pragmatic alternative to calculate RKF [27]. Therefore, the search for factors that may contribute to the preservation of RUV in resource-limited settings should be a priority to improve the practice of HD. In sub-Saharan Africa, including the Democratic Republic of the Congo (DRC), there are no studies examining the predictors of RUV in patients undergoing HD. The present study assessed factors that may contribute to the preservation of residual urine volume in HD patients in Kinshasa.

This historical cohort study enrolled 250 patients with ESRD undergoing HD in two HD centers in Kinshasa (Ngaliema Medical Center and Afia Medical Center) between January 2007 and July 2013. These facilities own Fresenius 4008B, 4008 s and, since 2010, 5008 s machines and a water treatment unit with double-pass reverse osmosis, which ensures the production of ultrapure water. Patients were alternately treated with hemodialysis (HD) or hemodiafiltration (HDF). The dialyzers used were made of polysulfone or high-flux membranes such as Helixone® Plus High-flux.

All patients with ESRD admitted for HD who received at least 4 weeks of renal replacement therapy were included in this study. Socio-demographic and anthropometric parameters of interest were age, gender, weight and height. For clinical parameters, systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured before the first session of HD using an electronic sphygmomanometer. Pulse pressure (PP) was calculated as the difference between SBP and DBP. The current medical treatment and complications at the initiation of HD were recorded on the appropriate form. Twenty-four-hour urine collection was used to measure RUV before beginning HD, and a similar urine collection protocol was used on days between dialysis 6 and 12 months later. The definition of volemia combined measurements of venous pressure, blood pressure and weight gain. For the evaluation of central venous pressure, we first performed a clinical assessment by placing the patient lying at 45° from the bed plane and appreciating the visibility of the jugular vein for a clinical estimation of 5–7 cm H2O. Patients with a value greater than 7 cm H2O benefited from the measure of central venous pressure. Hypervolemia was defined as a central venous pressure ≥ 13 cm H2O [28]. If the clinical estimate was below 5 cm H2O with decreased weight and blood pressure, patients were considered hypovolemic. Left ventricular hypertrophy (LVH) was noted when the thickness of the interventricular septum and posterior left ventricular wall was greater than 11 mm at the end of diastole or when the left ventricular mass was greater than or equal to 120 g/m2 [29]. Heart failure was defined on the basis of signs of pulmonary and peripheral stasis and a systolic ejection fraction of the left ventricle < 40% [30]. Chronic tubulointerstitial nephropathy was defined as End stage renal disease with the absence or modest degree of the two principal hallmarks of glomerular and vascular diseases of the kidney: salt retention, manifested by edema and hypertension; and proteinuria, which usually is modest and less than 1 to 2 g/d in TIN.

This study assessed predictors of RUV preservation in HD patients. The mean RUV at the initiation of HD was as high as 680 ml. This is significantly elevated compared to the data reported in the CHOICE study [31]. RUV was also better preserved in the present study than in the CHOICE study.

Indeed, one year after the initiation of HD, 60% of patients in this study still yielded > 250 ml/day as opposed to 23.2% in the CHOICE study [31]. Numerous comorbidities and the more advanced ages of HD patients in Western countries could explain this difference.

When we analyze the effect of time alone, we note that the speed of the decline in RUV slowed down with time. This effect remained present even after adjustments for ACE inhibitors use, chronic tubulointerstitial nephropathy and LVH in model 2. This slowing of the decline in urine volume with time could be attributed to an improvement in the technique of dialysis with the introduction of the HDF technique in the two study centers since 2010. HDF is characterized by a small decrease in RUV, similar to peritoneal dialysis.

The present study did not show an association between urine volume and age or sex. These results are contrary to those of the literature [4‐7]. However, Nechita et al., in a study of a cohort of 216 patients with stage 5 chronic kidney disease (CKD) beginning chronic HD, also failed to find a significant association between residual diuresis and age or sex [32]. It is possible that the small sample size did not reveal the influence of age or sex on the loss of urine volume. Another reason may be related to the fact that urine volume is the result of glomerular filtration, tubular reabsorption and the influence of diuretic therapy. In addition, in CKD, the concentration and dilution tubular capacity is altered as a result of tubulointerstitial fibrosis and increased tubular resistance by ADH [33, 34].

In the present study, independent predictors of urine volume were the use of ACE inhibitors, chronic tubulointerstitial nephropathy and LVH.

The beneficial role of ACE inhibitors on RKF and RUV in HD has been demonstrated in the analysis of the U.S. Renal Data System data, a study of 2211 incident dialysis patients with RKF defined by UV > 200 ml/day. This study found that the use of an ACE inhibitor was independently associated with a decreased risk of UV loss [5]. Several other studies have shown the benefit of ACE inhibitors on RKF and RUV in HD patients [35‐38]. The prospective study by Xydakis et al. proved that ACE inhibitors have a significant effect on preserving RKF in patients starting dialysis, at least during the first 12 months from the initiation of the HD, with RKF defined as residual GFR and RUV [18]. This beneficial effect of ACE inhibitors can be explained not only by the control of blood pressure parameters and by the anti-proteinuric effect but also by their action against inflammation, malnutrition and glomerular and interstitial fibrosis linked to the deleterious action of angiotensin II [38]. In addition, blood contact with the HD filter membrane causes a cascade of reactions, including activation of mononucleated cells that are responsible for several inflammatory mediators, such as IL-1_β, IL-6, TNFα, reactive oxygen species (ROS), and nitrogen monoxide, and the release of proteins from the acute phase of inflammation, such as C reactive protein [38]. The use of ACE inhibitors decreases serum TNF alpha and C reactive protein levels [39], resulting in a reduction in inflammation and oxidative stress and subsequently better preservation of RKF and RUV. Kidney Disease Outcomes Quality Initiative guidelines currently recommend the use of ACE inhibitors in the preservation of RKF and in the control blood pressure in patients with RUVs in excess of 100 ml [16].

The principal manifestations of TIN are those of tubular dysfunction. The tubulointerstitial lesions are localized either to the cortex or medulla. The extent of damage determines the severity of tubular dysfunction. Disruption of these structures, therefore, results in different degrees of nephrogenic diabetes insipidus and clinically manifests as polyuria and nocturia [33, 34]. A close correlation exists between the severity of chronic tubulointerstitial nephropathy and impaired renal tubular and glomerular function. Chronic tubulointerstitial nephropathy usually preserve the RUV more often in this study. Tubular dysfunction probably justified the preservation of RUV in patients with chronic tubulointerstitial nephropathy [34].

LVH is an important predictor of cardiovascular mortality and morbidity in dialysis patients. Hypertension, diabetes, increased body mass index, gender, age, anemia, and hyperparathyroidism have been described as risk factors for LVH in HD patients [40]. LVH is also recognized in the literature as a predictor of the loss of RUV [4, 5, 10]. LVH was negatively correlated with RUV [40]. The results of our study corroborate the information described in the literature regarding the negative effect of LVH on RUV. The mechanisms by which LVH favors altered RUV are both hemodynamic and neurohormonal. Diastolic and/or systolic dysfunction decrease renal perfusion pressure and effective ultrafiltration pressure [10] and increase renal venous pressure and water and sodium retention, contributing substantially to the decline in RUV. In addition, sympathetic stimulation, renin-angiotensin-aldosterone system activation, inflammation, oxidative stress and endothelial dysfunction also contribute to alterations of RUV by renal fibrosis [41]. In CKD, a veritable vicious circle is established between LVH and RUV. The loss of RUV favors hypervolemia and elevated blood pressure, two factors that favor LVH, which will exacerbate the loss of RUV by hypoperfusion. To combat LVH, the use of ACE inhibitors is the treatment of choice. A comparison between ACE inhibitors and other drug controls showed that ACE inhibitors cause a greater reduction in LVH in HD patients [42]. Another strategy is to increase the duration and frequency of HD sessions [43]. If this strategy has the capacity to reduce RUV in conventional HD [13, 14], this is not the case in hemodiafiltration (HDF). High-efficiency post-dilution on-line HDF reduces LVH without altering RUV [44].

Given the influence of the time spent in HD on RUV, it is necessary to make efforts to improve the practice of HD. The quality of the water treatment (ultrapure water) [26], the use of biocompatible dialysates and dialyzers [24, 25], and the prevention of peridialytic hypotension [5, 7, 11] must be taken into account.

The interpretation of the results of the present study should consider some limitations. First, the retrospective characteristic of the study design precludes the establishment of an all cause-effect relationship. Second, the small sample size did not confer much power to the statistical tests to identify additional associations. Without urine creatinine available, the accuracy of 24-h urine output in the present study could not be certified. However, despite these limitations, the present study is the first to evaluate independent predictors of RUV in Sub-Saharan Africa.

In the present study, the use of ACE inhibitors, chronic tubulointerstitial nephropathy and LVH reduction emerged as the main independent predictors of RUV preservation over time. The use of ACE inhibitors and HDF should be preferred for the optimal management of HD patients.