End-stage renal disease (ESRD) affects 723,413 patients in the United States; the majority of these patients undergo renal replacement therapy via hemodialysis (HD; U.S. Renal Data System, 2016). Patients on HD have increased mortality averaging 17% per patient year; cardiovascular disease accounts for more than half of all known causes of death in this population (U.S. Renal Data System, 2016). Much of the mortality is associated with fluid overload (FO; Agarwal, 2010; Chazot et al., 2012; Wizemann et al., 2009; Zoccali et al., 2017), which contributes to cardiovascular morbiditiy and mortality via longstanding pressure and volume overload contributing to hypertension and leading to left ventricular hypertrophy (LVH; Cader, Ibrahim, Paul, Gafor, & Mohd, 2014; Xu et al., 2013). Yet, FO is difficult to measure, and many patients have significant pulmonary congestion despite the absence of pulmonary rales or pedal edema (Torino et al., 2016; Zocalli et al., 2013).
HD patients have no residual renal function; they lack the ability to excrete sodium and water consumed throughout the day, resulting in fluid-related weight gains (Clark-Cutaia, Ren, Hoffman, Snetslaar, & Sevick, 2013). Thus, sodium restriction has been a standard treatment recommendation for decades. Lower sodium intake is associated with less interdialytic weight gain (Clark-Cutaia et al., 2014), improved intradialytic blood pressure, and survival (Maduell & Navarro, 2000). Dietary restriction of sodium appears to improve blood pressure, LVH, and intradialytic hypotension (Kayikcioglu et al., 2009; Maduell & Navarro, 2000). As a result, sodium restriction is a universal recommendation for ESRD management. The National Kidney Foundation recommendation of 2,400 mg per day is consensus based, rather than evidence based. The Dietary Guidelines for Americans 2010 recommend restriction of dietary sodium intake to 1,500 mg per day for persons with hypertension and/or chronic kidney disease, although little empiric evidence exists to support this recommendation. In addition, the Institute of Medicine failed to find sufficient evidence of either harm or benefit from sodium restriction at either level of restriction (Clark-Cutaia, Sommers, Anderson, & Townsend, 2016). Little is known about the appropriate degree of sodium restriction nor the clinical outcomes resulting from different levels of sodium restriction (National Kidney Foundation, 2015).
Bioimpedance spectroscopy (BIS)—a noninvasive instrument used to estimate body composition by applying a small alternating current at various frequencies—provides a way to accurately assess FO. Bioimpedance measures the opposition to the flow of electrical current through tissues. In the human body, only water with its dissolved electrolytes will conduct a current. Using validated estimation equations, BIS provides data on body composition, including fat mass and fat free mass, extracellular water (ECF), intracellular water (ICF), and total body water (TBW; Davenport, 2013; Davies & Davenport, 2014; Khalil, Mohktar, & Ibrahim, 2014). The goal of HD is to reduce both ECF and TBW, as well as remove excess solutes and waste.
Multiple cohort studies have demonstrated that FO as measured by BIS correlates with all-cause mortality in patients on HD (Chazot et al., 2012; Kim et al., 2015; Machek, Jirka, Moissl, Chamney, & Wabel, 2010). Zoccali et al. (2013) published a multicenter prospective cohort study of patients on HD who had FO measured by BIS and demonstrated it was associated with cardiovascular outcomes and mortality independent of blood pressure tertile. In addition, patients with greater degrees of FO experienced poorer outcomes (Zoccali et al., 2017). Randomized controlled trial (RCT) data have shown that ultrafiltration prescription enhanced by BIS measurement improved LVH, blood pressure, and arterial stiffness in one study and all-cause mortality in another. Relatively small changes in FO drove these results, and BIS was able to detect these differences (Hur et al., 2013; Onofriescu et al., 2014).
Theoretically, patients with lower dietary sodium intake would have lower ECF volume. ECF-to-ICF ratios have been implicated as a risk factor for cardiovascular and all-cause mortality in patients undergoing HD (Lin et al., 2003). Bioimpedance ECF and ECF% reflect hypervolemia (Arslan & Tuncalp, 2017; Kotanko, Levin, & Zhu, 2008). Aslan and Tuncalp (2017) found a reduction in sodium intake resulted in a reduction in ECF in patients that completed diet and fluid education. Whether sodium intake has an influence on FO in patients with ESRD on HD, as well as over what period of time, is not known. We had two study aims:
- to determine the feasibility of assessing the impact of sodium restriction on body fluid composition as measured by BIS among patients with three levels of sodium intake and
- to determine if there are statistical and clinical differences in BIS measures across sodium intake groups.
We piloted a double-blinded RCT to assess the feasibility of our procedures and to determine the impact of dietary sodium intake restriction on HD patient outcomes. The complete study design and methods were previously described elsewhere (Clark-Cutaia et al., 2016).
The 42 participants were recruited from three urban DaVita dialysis centers associated with a tertiary acute care academic institution. Inclusion criteria were persons of ≥21 years of age, who had the ability to read and write English, were undergoing maintenance HD, and had been on therapy for at least 3 months. Exclusion criteria were persons who intended to move out of the area or change HD centers within 6 months, had terminal illness or life expectancy of less than 12 months, planned to receive a living donor transplant in the study period, had cognitive impairment, were unable to provide informed consent, had New York Heart Association Class III or IV heart failure, had an internal defibrillator or pacemaker, and/or were pregnant. Because this was a pilot study, we excluded Spanish-speaking participants.
The study was approved by the institutional review board, and potential participants provided written informed consent and Health Insurance Portability and Accountability Act (HIPAA) release forms. Participants were randomly assigned to one of three sodium intake groups at baseline: 1,500 mg of daily sodium, 2,400 mg of daily sodium, or a control group with unrestricted daily sodium (these participants selected from a restaurant style menu and their sodium intake calculated based on consumption). Participants were admitted to an inpatient research unit for 5 days and 4 nights. The inpatient research unit Nutrition Core created an isocaloric menu that varied only in sodium content to administer the randomized dietary sodium intake (mg/day). Remaining food items on each tray were collected, blended, and analyzed for remaining sodium content, which was then deducted from the dietary sodium intake total. All meals and snacks were provided by the Nutrition Core.
Demographic and health status data were obtained through self-report using a questionnaire adapted from previous studies conducted by the coinvestigators. Patients were transported to and from their regularly scheduled HD sessions on admission Days 1, 3, and 5 of the 5-day study period. HD-related variables (ultrafiltration, rate, and total fluid removed) were abstracted from the postdialysis electronic medical record on dialysis days.
Body fluid composition was measured with BIS (ImpediMed Imp_ SFB7, ImpediMed Limited, Queensland, Australia). Measurements included TBW, ECF, and ICF in percentage (%). Three consecutive measurements were taken on the surface of the skin by placing electrodes on the right hand and right foot before and after dialysis. The average of these measurements was used for analysis.
Descriptive statistics were computed for all study variables, which were then compared in terms of their distributions across sodium intake groups using chi-square (χ2) tests for categorical variables and one-way analysis of variance for continuous variables. We conducted a set of three regression analyses with TBW, ICF, and ECF as outcomes with the following set of covariates: Day 1 TBW, intracellular fluid, or extracellular fluid; demographics (age, gender, race); health (tobacco use, alcohol use, perceived general health, years on HD); and sodium intake group membership (unrestricted, 1,500 mg, 2,400 mg). Regression models were run with and without sodium intake group membership, and change in R2 was tested with an F test. Analyses were conducted using IBM SPSS Statistics Version 23.
We approached 50 potential study participants; all but one met inclusion criteria. The first 42 signed informed consent and were enrolled into the study. Two patients withdrew from the study due to illness unrelated to the study during the intervention phase, for a total sample of 40 participants. Our overall retention rate was 95%. After controlling for other variables, given a small-to-moderate effect of sodium intake group membership on our outcomes (R2 change = .02–.13 [Cohen’s f2 = .02–.15]), a sample of 42 participants (14 per group), and an alpha of .05, we were able to achieve power in the 15%–69% range.
Our sample was overwhelmingly African American (85%), reported hypertension as the etiology of ESRD (45%), was balanced in terms of gender (male, 52.5%), and had a mean age of 56 years (SD = 11.7; see Table 1). Most of the participants perceived their health to be at least “good” (55%). Seventy-five percent reported never using tobacco, and 65% reported never consuming alcohol. For the total sample on Day 1, (a) mean dietary sodium intake was 1,987.93 mg Na (SD = 1,316.5); (b) baseline (before HD on Day 1) interdialytic weight gain was 2.62 kg (SD = 1.5); (c) symptom/problem scale score was 79.63 (SD = 12.5); (d) ultrafiltration rate was 1.41 (SD = 0.20); (e) ICF was 54.89 (SD = 10.2); (f) ECF was 42.53 (SD = 8.4); and (g) TBW was 49.89 (SD = 12.8). On average, 2.80 L were removed at the baseline HD session (SD = 1.1).
Results of regression models showed there were no statistically significant differences in TBW (Table 2), ICF (Table 3), or ECF (Table 4) among sodium intake groups. Except for etiology of HD, no predictor variables were significantly associated with any of the outcomes. Adjusting for all other predictors and regardless of sodium intake group membership, those with a diabetes etiology had significantly higher levels of ICF and ECF as compared to those with a hypertension etiology (both ps < .02). Hypertensive individuals had, on average, a 15-unit (p = .013) decrease in the intracellular fluid volume variance and an 11-unit (p = .021) decrease in the extracellular fluid volume variance compared to diabetics.
We determined that a study of this magnitude—using these particular methods—is feasible. The principal investigator and the research staff developed a rapport with the dialysis centers’ staff, and as a result, patient referral and self-referral were accomplished easily. Prior to the study, we were concerned that we would not be able to retain patients for the entire hospital stay or provide prepared meals that the participants would find palatable. Our recruitment, enrollment, and retention statistics allayed these concerns. We conducted a satisfaction survey in preparation for reproducing this study and found that 79% of patients would participate in a similar study in the future if contacted. Anecdotally, patients recounted the intervention phase of the study as a “vacation” or a “break” from their day-to-day lives.
We found no statistically significant differences in fluid volume body composition in our sample of HD patients regardless of group assignment. Although we did not have sufficient power to reach statistical significance, the control group had higher ECF and lower ICF volumes than the two sodium restriction groups.
In our study sample, participants who identified diabetes as the cause of their ESRD had significantly higher ICF and ECF than those with ESRD caused by hypertension. This is a potentially interesting phenomenon within this subsample of patients with respect to fluid compartments. A multicenter trial demonstrated that diabetic HD patients had greater fluid volume between dialysis sessions than nondiabetics (Davenport, Cox, & Thuraisingham, 2008). Similarly, diabetic control or lack thereof was linked to greater fluid volume in another sample of HD patients (Ramdeen, Tzamaloukas, Malhotra, Leger, & Murata, 1998). McCafferty, Fan, and Davenport (2014) found that patients in their sample with diabetes also had higher hydration status at baseline and at follow-up. From the analyses we conducted to address our study aims, we were unable to extrapolate any conclusions other than those already posited by previous investigations into the phenomenon.
Although this preliminary work confirmed the feasibility of an RCT testing three levels of sodium intake in patients with ESRD on HD, our study was limited by a small sample size, which was not meant to provide an effect size calculation. Furthermore, we were unable to completely control the patient’s length of stay. The inpatient stay was not a five full days, as patients were admitted after and discharged before their dialysis sessions on Days 1 and 5, respectively. The mean length of stay was approximately 4.5 days and 4 nights. BIS measurements were conducted only by the study staff trained by the primary investigator. However, environmental and participant circumstances outside our control may have created error in the measurements. Furthermore, strict control of dietary sodium intake required 24-hour oversight by inpatient nursing and nutrition staff, dialysis center nursing staff, and study staff. It is possible that some patients received food not on their prescribed diets by trading snacks or elements of their food tray without our knowledge.
Our data suggest the need for additional research into the practicality of BIS to assess the effect of sodium restriction on body fluid composition but offer the first insight into how this effect can be assessed using BIS. Furthermore, we have developed a well-tolerated, methodologically sound investigation into body composition and sodium intake, despite the paucity of literature available in this regard. We will reproduce this study with a larger, more diverse sample in the future.
Agarwal R. (2010). Hypervolemia is associated with increased mortality among hemodialysis patients. Hypertension
, 56, 512–517.
Arslan S., & & Tuncalp F. B. (2017). The effects of diet and fluid education administered to patients of hemodialysis on some parameters. Journal of Human Sciences
, 14, 346–354. doi:10.14687/jhs.v14i1.4284
Cader R. A., Ibrahim O. A., Paul S., Gafor H. A., & & Mohd R. (2014). Left ventricular hypertrophy and chronic fluid overload in peritoneal dialysis patients. International Urology and Nephrology
, 46, 1209–1215. doi:10.1007/s11255-013-0615-8
Chazot C., Wabel P., Chamney P., Moissl U., Wieskotten S., & & Wizemann V. (2012). Importance of normohydration for the long-term survival of haemodialysis patients. Nephrology Dialysis Transplantion
, 27, 2404–2410. doi:10.1093/ndt/gfr678
Clark-Cutaia M. N., Ren D., Hoffman L. A., Burke L. E., & & Sevick M. A. (2014). Adherence to hemodialysis dietary sodium
recommendations: Influence of patient characteristics, self-efficacy, and perceived barriers. Journal of Renal Nutrition
, 24, 92–99. doi:10.1053/j.jrn.2013.11.007
Clark-Cutaia M. N., Ren D., Hoffman L. A., Snetslaar L., & & Sevick M. A. (2013). Psychometric validation of the self-efficacy for restricting dietary salt in hemodialysis scale. Topics in Clinical Nutrition
, 28(4), 384–391. doi:10.1097/01.TIN.0000437407.76867.65
Clark-Cutaia M. N., Sommers M. S., Anderson E., & & Townsend R. R. (2016). Design of a randomized controlled clinical trial assessing dietary sodium
restriction and hemodialysis-related symptom profiles. Contemporary Clinical Trials Communications
, 3, 70–73. doi:10.1016/j.conctc.2016.04.002
Davenport A. (2013). Dialysis: Bioimpedance spectroscopy for assessment of fluid overload. Nature Reviews Nephrology
, 9, 252–254. doi:10.1038/nrneph.2013.60
Davenport A., Cox C., & & Thuraisingham R. (2008). Blood pressure control and symptomatic intradialytic hypotension in diabetic haemodialysis patients: A cross-sectional survey. Nephron Clinical Practice
, 109, c65–c71. doi:10.1159/000139991
Davies S. J., & & Davenport A. (2014). The role of bioimpedance and biomarkers in helping to aid clinical decision-making of volume assessments in dialysis patients. Kidney International
, 86, 489–496. doi:10.1038/ki.2014.207
Hur E., Usta M., Toz H., Asci G., Wabel P., Kahvecioglu S., … … Ok E. (2013). Effect of fluid management guided by bioimpedance spectroscopy on cardiovascular parameters in hemodialysis patients: A randomized controlled trial. American Journal of Kidney Diseases
, 61, 957–965. doi:10.1053/j.ajkd.2012.12.017
Kayikcioglu M., Tumuklu M., Ozkahya M., Ozdofan O., Asci G., Duman S., … … Ok E. (2009). The benefit of salt restriction in the treatment of end-stage renal disease by haemodialysis. Nephrology Dialysis Transplant
, 24, 956–962. doi:10.1093/ndt/gfn599
Khalil S. F., Mohktar M. S., & & Ibrahim F. (2014). The theory and fundamentals of bioimpedance analysis in clinical status monitoring and diagnosis of diseases. Sensors
, 14, 10895–10928. doi:10.3390/s140610895
Kim Y. J., Jeon H. J., Kim Y. H., Jeon J., Ham Y. R., Chung S., … … Lee K. W. (2015). Overhydration measured by bioimpedance analysis and the survival of patients on maintenance hemodialysis: A single-center study. Kidney Research and Clinical Practice
, 34, 212–218. doi:10.1016/j.krcp.2015.10.006
Kotanko P., Levin N. W., & & Zhu F. (2008). Current state of bioimpedance technologies in dialysis. Nephrology Dialysis Transplantation
, 23, 808–812. doi:10.1093/ndt/gfm889
Lin Y. P., Yu W. C., Hsu T. L., Ding P. Y., Yang W.-C., & & Chen C.-H. (2003). The extracellular fluid-to-intracellular fluid volume ratio is associated with large-artery structure and function in hemodialysis patients. American Journal of Kidney Diseases
, 42, 990–999. doi:10.1016/j.ajkd.2003.07.002
Machek P., Jirka T., Moissl U., Chamney P., & & Wabel P. (2010). Guided optimization of fluid status in haemodialysis patients. Nephrology Dialysis Transplantation
, 25, 538–544. doi:10.1093/ndt/gfp487
Maduell F., & & Navarro V. (2000). Dietary salt intake and blood pressure in haemodialysis patients. Nephrology Dialysis Transplantation
, 15, 2063. doi:10.1093/ndt/15.12.2063-a
McCafferty K., Fan S., & & Davenport A. (2014). Extracellular volume expansion, measured by multifrequency bioimpedance, does not help preserve residual renal function in peritoneal dialysis patients. Kidney International
, 85, 151–157. doi:10.1038/ki.2013.273
National Kidney Foundation (2015). KDOQI clinical practice guideline for hemodialysis adequacy: 2015 update. American Journal of Kidney Diseases
, 66, 884–930. doi:10.1053/j.ajkd.2015.07.015
Onofriescu M., Hogas S., Voroneanu L., Apetrii M., Nistor I., Kanbay M., & & Covic A. C. (2014). Bioimpedance-guided fluid management in maintenance hemodialysis: A pilot randomized controlled trial. American Journal of Kidney Diseases
, 64, 111–118. doi:10.1053/j.ajkd.2014.01.420
Ramdeen G., Tzamaloukas A. H., Malhotra D., Leger A., & & Murata G. H. (1998). Estimates of interdialytic sodium
and water intake based on the balance principle: Differences between nondiabetic and diabetic subjects on hemodialysis. American Society for Artificial Internal Organs
, 44, 812–817.
Torino C., Gargani L., Sicari R., Letachowicz K., Ekart R., Fliser D., … … Zoccali C. (2016). The agreement between auscultation and lung ultrasound in hemodialysis patients: The LUST study. Clinical Journal of the American Society of Nephrology
, 11, 2005–2011. doi:10.2215/CJN.03890416
U.S. Renal Data System. (2016). USRDS annual data report: Epidemiology of kidney disease in the United States
. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases.
Wizemann V., Wabel P., Chamney P., Zaluska W., Moissl U., Rode C., … … Marcelli D. (2009). The mortality risk of overhydration in haemodialysis patients. Nephrology Dialysis Transplantation
, 24, 1579–1574. doi.org/10.1093/ndt/gfn707
Xu Y., Chen Y., Li D., Li J., Liu X., Cui C., & & Yu C. (2013). Hypertension, fluid overload and micro inflammation are associated with left ventricular hypertrophy in maintenance hemodialysis patients. Renal Failure
, 35, 1204–1209. doi:10.3109/0886022X.2013.819765
Zoccali C., Moissl U., Chazot C., Mallamaci F., Tripepi G., Arkossy O., … … Stuard S. (2017). Chronic fluid overload and mortality in ESRD. Journal of the American Society of Nephrology
, 28, 2491–2497. doi:10.1681/ASN.2016121341
Zoccali C., Torino C., Tripepi R., Tripepi G., D’Arrigo G., Postorino M., … … Mallamaci F. (2013). Pulmonary congestion predicts cardiac events and mortality in ESRD. Journal of the American Society Nephrology
, 24, 639–646. doi:10.1681/ASN.2012100990