For Peer Review Only

BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

The effect of increased water intake on plasma copeptin in patients with chronic kidney disease

ForJournal: peerBMJ Open review only Manuscript ID: bmjopen-2015-008634

Article Type: Research

Date Submitted by the Author: 30-Apr-2015

Complete List of Authors: Sontrop, Jessica; London Health Sciences Centre, Nephrology Huang, Shih-Han; Western University, Medicine; London Health Sciences Centre - Victoria Hospital, Medicine - Nephrology Garg, Amit; University of Western Ontario Moist, Louise; Western University, Medicine; London Health Sciecnes Centre - Victoria Hospital, Medicine - Nephrology House, Andrew; London Health Sciences Centre - University Hospital, Medicine - Nephrology Gallo, Kerri; London Health Sciences Centre - Victoria Hospital, Medicine - Nephrology Clark, William; University of Western Ontario, Medicine

Primary Subject Renal medicine Heading:

Secondary Subject Heading: Renal medicine http://bmjopen.bmj.com/

coeptin, water intake, glomerular filtration rate, 24-hour urine volume, Keywords: chronic kidney disease

on September 28, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 21 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 The effect of increased water intake on plasma copeptin in patients with chronic kidney 4 5 disease: results from a pilot randomized controlled trial 6 1,2 1 1,2 7 Jessica M. Sontrop PhD, ShiHan Huang MSc MD, Amit X. Garg MD PhD, Louise Moist 1,2 1 1 1 8 MSc MD, Andrew A. House MD, Kerri Gallo RN, William F. Clark MD * 9 10 1. Division of Nephrology, Department of Medicine, London Health Sciences Centre, London 11 ON, Canada N6A 4G5 12 13 2. Department of Epidemiology & Biostatistics, Western University, London ON, Canada 14 N6A 5C1 15 For peer review only 16 *Corresponding Author 17 Dr. William F. Clark 18 Victoria Hospital 19 20 800 Commissioner’s Road East, A2343 21 London ON, Canada N6A 5W9 22 Phone: 5196858361 23 Fax: 5196858047 24 Email: [email protected] 25 26 27 28 29 30 Word count: main body: 2293 words; abstract: 250 words 31 32 33 Key words: 24hour urine volume, chronic kidney disease, copeptin, randomized controlled

34 trial, vasopressin http://bmjopen.bmj.com/ 35 36 37 38 DISCLOSURES 39 40 Dr. William Clark and Dr. Louise Moist have received consulting fees and/or honorarium and 41 42 support to travel to meetings with Danone Research. No other authors have relevant conflicts of on September 28, 2021 by guest. Protected copyright. 43 interest to declare. 44 45 This study was funded by Danone Research and the Program of Experimental Medicine, Western 46 University, Canada. The study sponsors had no direct role in the data collection, statistical 47 analysis, interpretation of the results or drafting of the manuscript. Dr. Garg is supported by the 48 Dr. Adam Linton Chair in Kidney Health Analytics. 49 50 51 An abstract of this work was presented at the World Congress of Nephrology, March 1317, 52 2015, Cape Town, South Africa. 53 54 55 56 57 58 59 60 1

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 21 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 ABSTRACT 4 5 6 7 Objectives: Increased water intake may have a beneficial effect on the kidney through 8 9 suppression of plasma vasopressin. We examined the effect of increased oral water intake on 10 11 plasma copeptin (a marker of vasopressin) over six weeks in patients with stage 3 chronic kidney 12 13 14 disease. 15 For peer review only 16 17 Design: Secondary analysis of a randomized controlled parallelgroup pilot trial. 18 19 20 Setting: Canada, 20122013. 21 22 23 Participants: Twentynine patients with stage 3 Chronic Kidney Disease randomized (2:1) to a 24 25 26 hydration (n=18) or control group (n=11). 27 28 29 Intervention: The hydration group was coached to increase water intake by up to 1.5 L/day for 6 30 31 weeks (in addition to usual intake, depending on sex and weight). The control group was asked 32 33 to maintain regular water intake. 34 http://bmjopen.bmj.com/ 35 36 37 Measures and Outcomes: Participants provided blood and 24hr urine samples at baseline and 6 38 39 weeks after randomization. Change in plasma copeptin was compared within and between study 40 41 42 groups. on September 28, 2021 by guest. Protected copyright. 43 44 45 Results: Participants (n=28) were 63% male with a mean age of 61 years and had an estimated 46 47 glomerular filtration rate (eGFR) of 40 ml/min/1.73 m². Between baseline and 6 48 49 50 weeks, 24hr unrine volume increased by 0.7 L/d in the hydration group, rising from 2.3 to 3.0 51 52 L/d, while decreasing by 0.3 L/d among controls, from 2.0 to 1.7 L/d (p=0.002 for the between 53 54 group change). In the hydration group, median copeptin decreased by 3.6 pmol/L, from 15.0 to 55 56 57 58 59 60 2

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 21 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 10.8 pmol/L (p=0.005), while remaining stable among controls at 19 pmol/L (p=0.76) (p=0.19 4 5 6 for the betweengroup change). 7 8 9 Conclusions: Adults with stage 3 chronic kidney disease can be successfully randomized to 10 11 drink approximately 1 L more per day than controls. This increased water intake caused a 12 13 14 significant decrease in plasma copeptin concentration. Future trials should examine whether 15 For peer review only 16 increased water intake can slow renal decline in patients with chronic kidney disease. 17 18 19 20 21 22 23 24 25 26 Strengths and Limitations 27 28 29 • In this randomized controlled pilot trial, 28 patients with stage 3 chronic kidney disease were 30 31 successfully randomized to drink approximately 1 L more per day than controls for six 32 33

34 weeks, as verified by 24hour urine collections. This increased water intake caused a http://bmjopen.bmj.com/ 35 36 significant decrease in plasma copeptin concentration. 37 38 39 • Patients were followed for only six weeks. Whether the effect of increased water intake on 40 41 plasma copeptin concentration is clinically significant, beneficial, or sustainable over time is 42 on September 28, 2021 by guest. Protected copyright. 43 unknown. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 3

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 21 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 INTRODUCTION 4 5 6 Vasopressin is an essential antidiuretic hormone in mammals that regulates thirst and urine 7 8 water.1 However, chronically elevated levels of vasopressin may have adverse health effects. 9 10 Until recently, it was not possible to reliably measure vasopressin due to the limited sensitivity of 11 12 13 available assays; however, in 2006, Morgenthaler et al. developed an immunoassay for 14 15 copeptin—a glycosylatedFor peptide peer that is coreleased review with vasopressin only from the hypothalamus.2 16 17 Copeptin can be measured from blood samples and is demonstrated to be a reliable marker of 18 19 2,3 20 vasopressin. The availability of this assay has renewed scientific interest into the role of 21 22 vasopressin and copeptin in chronic illness. In longitudinal studies, a higher baseline 23 24 25 concentration of plasma copeptin predicts kidney function decline patients with autosomal 26 4,5 27 dominant polycystic kidney disease, and other studies have linked copeptin to the development 28 29 of diabetes,6 the metabolic syndrome,7 and heart failure.8,9 In cohort studies, elevated plasma 30 31 10,11 32 copeptin at baseline is strongly predictive of subsequent myocardial infarction and endstage 33 12 34 kidney disease. http://bmjopen.bmj.com/ 35 36 37 Many factors are known to activate vasopressin, including acute stress and illness.13 Other 38 39 40 stimuli include high plasma osmolality and dehydration—vasopressin is the first hormone 41 42 released during dehydration.1 In experimental studies of rats, increased water intake was shown on September 28, 2021 by guest. Protected copyright. 43 44 to suppress vasopressin, reduce proteinuria, and improve creatinine clearance.14,15 In large cohort 45 46 16–18 47 studies, renal decline was slower in those with greater water inake, but evidence from 48 49 clinical trials is lacking. Copeptin may be an intermediate variable in the relationship between 50 51 water intake and renal decline; however, it is not known whether copeptin can be adequately 52 53 54 suppressed by water intake in patients with Chronic Kidney Disease. In 2012, we launched the 55 56 pilot phase of a randomized controlled trial that will test whether increased water intake can slow 57 58 59 60 4

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 21 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 renal decline in patients with chronic kidney disease. The results of the pilot trial were published 4 5 19 6 in December, 2013; however, data on plasma copeptin did not become available until 2014. 7 8 Here, we examine the effect of increased water intake on plasma copeptin. We also examine the 9 10 11 relationships between copeptin and 24hr urine volume, osmolality, sodium, albuminuria, and 12 13 estimated glomerular filtration rate (eGFR). 14 15 For peer review only 16 METHODS 17 18 Study Population 19 20 21 We analyzed data from the pilot phase of the Water Intake Trial (WIT), a parallelgroup 22 23 randomized controlled trial conducted between October, 2012 and March, 2013 in London, 24 25 26 Ontario, Canada (registered at clinicaltrials.gov NCT01766687). A figure detailing participant 27 19 28 selection and followup is available in Clark et al. This trial was designed to assess the 29 30 feasibility and safety of asking adults with stage 3 chronic kidney disease to increase their water 31 32 33 intake by up to 1.5 L/day. All patients provided informed consent consistent with the Declaration

34 http://bmjopen.bmj.com/ 35 of Helsinki and ethics approval was obtained from the Western University Research Ethics 36 37 Board for Health Sciences. 38 39 40 2 41 Eligibility criteria included age 3080 years, an eGFR 3060 ml/min/1.73m , presence of 42 on September 28, 2021 by guest. Protected copyright. 43 albuminuria (albumin/creatinine >2.8 mg/mmol [if female] or >2.0 mg/mmol if male] from a 44 45 spot urine sample or trace protein [albustix]), and a 24hr urine volume <3L/day. We excluded 46 47 48 patients who met any of the following criteria: selfreported fluid intake ≥10 cups/day; had 49 50 received a dialysis treatment in the past month; kidney transplant recipient (or on waiting list); 51 52 under fluid restriction; pregnant or breastfeeding; symptomatic kidney stones in past 5 years; a 53 54 55 life expectancy less than two years; serum sodium ≤130 mmol/L; serum calcium >2.6 mmol/L; 56 57 currently taking lithium (a drug which affects thirst and urination) or high daily doses of the 58 59 60 5

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 21 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 following diuretics: hydrochlorothiazide >25 mg/day, indapamide >1.25 mg/day, 4 5 6 furosemide >40 mg/day, or metolazone >2.5 mg/day. 7 8 9 Intervention 10 11 12 We randomized 29 patients by computergenerated randomization in block sizes of 3 to a 13 14 hydration or control group (2:1), stratified by gender. This 2:1 randomization in the pilot phase 15 For peer review only 16 was chosen to provide experience delivering the hydration intervention to more patients within 17 18 19 an overall sample of 29 patients. The hydration group (n=18) was coached to increase their oral 20 21 water intake by 1.0 to 1.5 L/day depending on sex, weight, and 24hour urine osmolality (in 22 23 addition to usual consumed beverages) for 6 weeks. The control group (n=11) was asked to 24 25 26 continue with their usual water intake or to decrease water intake by 12 cups/day depending on 27 28 their baseline 24hour urine osmolality. 29 30 31 Outcomes, Measurements and Definitions 32 33 In this secondary analysis of the WIT pilot trial, the primary outcome was the betweengroup 34 http://bmjopen.bmj.com/ 35 36 change in plasma copeptin. All participants provided 24hour urine samples and blood samples at 37 38 baseline and 6weeks after randomization. Serum creatinine was measured using the isotope 39 40 41 dilution/mass spectroscopytraceable enzymatic method, and eGFR was calculated using the 42 on September 28, 2021 by guest. Protected copyright. 43 Chronic Kidney Disease Epidemiology Collaboration (CKDEPI) equation.20 44 45 46 We measured the concentration of sodium, osmolality, and urea in blood and 24hour urine 47 48 19 49 samples and the 24hour urine albumin to creatinine ratio using standard methods. Copeptin 50 51 concentration was measured in plasmaEDTA samples. Samples were stored at 20°C and 52 53 54 analyzed for copeptin in a single batch analysis to eliminate interassay variation. Copeptin 55 56 57 58 59 60 6

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 21 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 concentration was measured using a sandwich immunoassay (Thermo Fisher Scientific 4 5 6 B.R.A.H.M.S, Hennigsdorf/ Berlin, Germany). 7 8 9 Statistical Analyses 10 11 Normally distributed data were summarized using means and standard deviations (SD), and non 12 13 14 normally distributed data were summarized using medians and interquartile ranges (IQR). All 15 For peer review only 16 randomized participants were included in the analysis and analyzed according to group 17 18 assignment. Withingroup changes in the amount of urine volume and the concentration of 19 20 21 plasma copeptin were compared using the paired ttest and the relatedsamples Wilcoxon signed 22 23 rank test, respectively; the betweengroup change was compared using the independent ttest and 24 25 26 MannWhitney U, as appropriate. Correlations between copeptin and measures of kidney 27 28 function, urine volume, sodium, osmolality, and albumin were analyzed using the Spearman’s 29 30 rank correlation coefficient (rho) for nonnormally distributed data. Correlations are presented 31 32 33 for the overall sample at baseline and 6weeks follow up. Data were analyzed using SPSS

34 http://bmjopen.bmj.com/ 35 version 21. 36 37 38 RESULTS 39 40 41 Of 74 participants who met the initial eligibility criteria and were invited to participate, 33 were 42 on September 28, 2021 by guest. Protected copyright. 43 consented and 29 were randomized (4 patients withdrew before randomization). One participant 44 45 46 in the hydration group withdrew after randomization due to a flare up of Crohn’s disease. 47 48 49 Participants were 63% male and 81% Caucasian with an average age of 61 years (SD 14) and an 50 51 average eGFR of 40 ml/min/1.73m2 (SD 11 ml/min/1.73m2); 54% had diabetes and 86% had 52 53 54 hypertension. Median copeptin at baseline was 17 pmol/L (IQR 9, 30), and was higher in males 55 56 [18 (IQR 12, 31) vs. 12 (IQR 6, 30) in females]. Participant characteristics by treatment group 57 58 59 60 7

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 21 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 are shown in Table 1 (as per CONSORT guidelines,21 pvalues were not calculated). Although 4 5 6 randomization protects against baseline differences between groups, differences can arise in 7 8 smaller samples,22 for example, participants in the control group were older, had more 9 10 11 comorbidities, and had more diuretic use compared with those in the hydration group. As well, 12 13 there were three participants with polycystic kidney disease in the hydration group (and none in 14 15 the control group);For however, peer study results remainedreview unchanged whenonly these patients were 16 17 18 excluded in sensitivity analysis. 19 20 21 Change in copeptin between baseline and 6 weeks is shown in Table 2 and Figure 1. During this 22 23 time, 24hour urine volume increased by 0.7 L/d in the hydration group (p=0.01) and decreased 24 25 26 by 0.3 L/d in the control group (p=0.07); p=0.002 for the betweengroup difference in change. In 27 28 the hydration group, the concentration of plasma copeptin decreased significantly between 29 30 baseline and followup, from 15.0 pmol/L to 10.8 pmol/L (p=0.005), while remaining relatively 31 32 33 stable among controls, at 19 pmol/L (p=0.76); p=0.19 for the betweengroup difference in

34 http://bmjopen.bmj.com/ 35 change. 36 37 38 39 Correlations between the concentration of plasma copeptin and urine volume, eGFR, and other 40 41 urine and serum measures are shown in Table 3. At baseline, copeptin was inversely correlated 42 on September 28, 2021 by guest. Protected copyright. 43 with eGFR (rho=0.53; p=0.003) and positively correlated with serum osmolality (rho=0.58; 44 45 46 p=0.001) and serum urea (rho=0.51; p=0.001). At six weeks followup, copeptin was inversely 47 48 correlated with 24hr urine volume (rho= 0.48; p=0.01) and eGFR (rho=0.56; p=0.002), and 49 50 positively correlated with the albumin to creatinine ratio (rho= 0.44; p=0.02), serum urea (rho= 51 52 53 0.49; p=0.008) and urine and serum osmolality (rho=0.53; p=0.004 and rho=0.47; p=0.01, 54 55 respectively). No correlation was seen with urine sodium or serum sodium at either time point. 56 57 58 59 60 8

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 21 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 4 5 6 DISCUSSION 7 8 9 In this randomized controlled pilot trial of 28 patients with stage 3 chronic kidney disease, 10 11 patients were successfully randomized to drink approximately 1 L more per day than controls for 12 13 six weeks. In the hydration group, 24hr urine volume increased significantly and the between 14 15 For peer review only 16 group difference was 1.3 L/d at 6 weeks. This increased water intake caused a significant 17 18 decrease in plasma copeptin concentration among patients in the hydration group; however, the 19 20 21 betweengroup difference was not statistically significant. No adverse effects were reported nor 22 19 23 observed. 24 25 26 The median baseline copeptin concentration in our study was 17 pmol/L, which is approximately 27 28 four times higher than values reported in healthy volunteers.13 Other studies have shown that 29 30 3 6 31 copeptin is elevated in patients with chronic kidney disease, diabetes, myocardial 32 33 infarction,10,11 and heart failure.8,9 In kidney transplant recipients, higher levels of copeptin at

34 http://bmjopen.bmj.com/ 35 36 baseline predicted a significantly faster decline in kidney graft function over a median followup 37 23 38 of 3.6 years, independent of baseline eGFR, proteinuria and other risk factors. 39 40 41 In our study, copeptin was inversely correlated with both 24hr urine volume and eGFR, and 42 on September 28, 2021 by guest. Protected copyright. 43 44 positively correlated with urine protein, urine osmolality, serum osmolality, and serum urea. If 45 46 copeptin is cleared by the kidneys then it will necessarily increase as kidney function declines, 47 48 and this could explain the inverse relationship between copeptin and eGFR. However, in a study 49 50 51 by Zittema et al., copeptin was not associated with GFR in healthy living kidney donors—and 52 53 copeptin levels did not change after donation despite a significant drop in kidney function after 54 55 24 56 nephrectomy. These data suggest that GFR alone is not a principal determinant of copeptin. 57 58 59 60 9

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 21 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 What is the link between copeptin and kidney function? 4 5 6 An inverse relationship between copeptin and kidney function is consistently demonstrated in 7 8 experimental and observational studies, in diverse patient groups, and appears to be independent 9 10 3,23,25 11 of age, sex, blood pressure, and other risk factors. However, with the exception of 12 13 autosomal polycystic kidney disease (ADPKD), the mechanism linking copeptin and kidney 14 15 function remainsFor unresolved. peer In the unique review case of ADPKD, a riseonly in vasopressin (and copeptin) 16 17 18 stimulates the formation of cAMP (3’,5’cyclic adenosine monophosphate), which promotes cell 19 20 proliferation and cyst growth leading to kidney enlargement and a subsequent decline in kidney 21 22 function.26,27 Accordingly, in several animal studies, researchers showed that blocking the 23 24 25 cAMPmediated pathway with V2receptor antagonists resulted in significantly reduced cyst 26 27 growth,28,29 and, in a 3year doubleblind randomized placebocontrolled trial, patients treated 28 29 with tolvaptan (a V2receptor antagonist) experienced a significantly slower increase in total 30 31 30 32 kidney volume and a significantly slower decline in kidney function. Interestingly, increased 33

34 oral water intake in rats has been demonstrated to have a similar suppressing effect on http://bmjopen.bmj.com/ 35 36 28 37 vasopressin secretion. 38 39 40 Vasopressin has potent vasoconstrictive effects, and some hypothesize that the associations 41 42 observed between copeptin and renal/cardiovascular outcomes may be partly explained by on September 28, 2021 by guest. Protected copyright. 43 44 12,31,32 45 vasopressin’s effect on blood pressure; however, associations remain significant after 46 47 controlling for blood pressure or hypertension, and the relationship between copeptin and blood 48 49 pressure in human cohort studies is inconsistent and weak.25,32,33 In our study, correlations 50 51 52 between copeptin and blood pressure were below 0.1 (data not shown). Many other mechanisms 53 54 have been proposed to explain the link between vasopressin/copeptin, GFR, and albuminuria; 55 56 these include vasopressin’s effect on urinary concentrating activity,34,35 hyperosmolarity,36 57 58 59 60 10

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 21 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 activation of the reninangiotensin system,37 glomerular hyperfiltration and hypertrophy,1,23,27 4 5 12,38 6 high salt intake and V2receptordependent tubular effects on sodium reabsorption, As 7 8 described in Bolignano et al., these pathways are not mutually exclusive and may act together to 9 10 39 11 affect renal outcomes. 12 13 14 Copeptin is also a marker of the body’s endocrine stress response, which is mediated through the 15 For peer review only 16 hypothalamuspituitaryadrenal system, and is activated in acute illness.7,9 For example, copeptin 17 18 19 levels spike in concert with cortisol and corticotropinreleasing hormone within hours of acute 20 21 myocardial infarction onset.40 However, ample evidence suggests that copeptin is not simply a 22 23 marker of stress or illness, but that it plays a role in the pathophysiology of cardiovascular and 24 25 41 26 chronic kidney disease. In studies of animals and humans, stimulating vasopressin causes an 27 28 increase in proteinuria, while suppressing vasopressin through V2 antagonism or water intake 29 30 reduces proteinuria.14,15,37 Further, in studies that controlled for inflammatory biomarkers (C 31 32 33 reactive protein and Nterminal pro brain natriuretic peptide), copeptin remained significantly

34 http://bmjopen.bmj.com/ 35 associated with renal decline and proteinuria.23,25 These data have sparked interest in the use of 36 37 38 vasopressin receptor antagonists to improve renal and cardiovascular outcomes, and several are 39 27,42 40 currently being investigated for clinical use. If proved effective, copeptin may serve as a 41 42 40 useful biomarker to identify patients who may most benefit from treatment. on September 28, 2021 by guest. Protected copyright. 43 44 45 46 Our data are from a small sample of 29 patients who were followed for only six weeks. 47 48 Nonetheless, a modest increase in water intake was associated with a significant decrease in the 49 50 concentration of plasma copeptin. At six weeks, plasma concentrations of copeptin were 51 52 53 inversely correlated with 24hour urine volume and eGFR, and positively correlated with the 54 55 amount of urine protein. We encouraged participants in the hydration group to increase water 56 57 58 intake by up to 1.5 L/d, and, compared with controls, 24hr urine volume was significantly 59 60 11

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 21 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 higher in the hydration group; however, we do not know how much these participants increased 4 5 6 their intake of water vs. other fluids. This is important because in several studies, a potentially 7 8 beneficial effect of higher fluid intake is not observed when fluid intake comes from greater 9 10 3,17,43 11 intake of sweetened beverages. 12 13 14 Our study and others have clearly demonstrated that copeptin is sensitive to changes in water 15 For peer review only 16 intake and is associated with 24hour urine volume.25,44 Copeptin is also inversely related to 17 18 eGFR, is positively associated with urine protein, and is a prognostic biomarker for future events 19 20 6,8–11,24,25,45 21 of diabetes and heart disease. Does it follow that hydration is a unifying factor linking 22 23 copeptin to kidney function and other diseases, as some suggest?46 Possibly, however evidence 24 25 26 from a large randomized controlled trial is needed to determine if increased water intake has a 27 28 direct protective effect on kidney function, and if so, to what extent and by what mechanism. The 29 30 possibility that hydration may play a protective role in chronic kidney disease is biologically 31 32 33 plausible and is supported by a diverse and growing body of research; however, whether this

34 http://bmjopen.bmj.com/ 35 relationship is causal, clinically significant, and sustained over time is unknown. We anticipate 36 37 that the results from larger trials (including our own 12month trial, currently in progress), with 38 39 40 repeated measures of copeptin, 24hour urine volume, osmolality, and eGFR over time will 41 42 provide key insight into the causality, direction, and magnitude of the relationships between on September 28, 2021 by guest. Protected copyright. 43 44 water intake, copeptin, and kidney function. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 12

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 21 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 References 4 5 6 1. Bouby N, Fernandes S. Mild dehydration, vasopressin and the kidney: animal and human 7 8 studies. Eur J Clin Nutr. 2003;57 Suppl 2:S3946. doi:10.1038/sj.ejcn.1601900. 9 10 2. Morgenthaler NG, Struck J, Alonso C, Bergmann A. Assay for the measurement of 11 copeptin, a stable peptide derived from the precursor of vasopressin. Clin Chem. 12 2006;52:112119. 13 14 15 3. Roussel R,For Fezeu L, Marrepeer M, et al. reviewComparison Between only Copeptin and Vasopressin in a 16 Population from the Community and in People with Chronic Kidney Disease. J Clin 17 Endocrinol Metab. 2014;99:465663. doi:10.1210/jc.20142295. 18 19 4. Boertien WE, Meijer E, Zittema D, et al. Copeptin, a surrogate marker for vasopressin, is 20 21 associated with kidney function decline in subjects with autosomal dominant polycystic 22 kidney disease. Nephrol Dial Transplant. 2012;27(11):41317. doi:10.1093/ndt/gfs070. 23 24 5. Boertien WE, Meijer E, Li J, et al. Relationship of copeptin, a surrogate marker for 25 arginine vasopressin, with change in total kidney volume and GFR decline in autosomal 26 27 dominant polycystic kidney disease: results from the CRISP cohort. Am J Kidney Dis. 28 2013;61(3):4209. doi:10.1053/j.ajkd.2012.08.038. 29 30 6. Enhorning S, Wang TJ, Nilsson PM, et al. Plasma copeptin and the risk of diabetes 31 mellitus. Circulation. 2010;121:21022108. 32 33 7. Saleem U, Khaleghi M, Morgenthaler NG, et al. Plasma carboxyterminal provasopressin 34 http://bmjopen.bmj.com/ 35 (copeptin): a novel marker of insulin resistance and metabolic syndrome. J Clin 36 Endocrinol Metab. 2009;94(7):255864. doi:10.1210/jc.20082278. 37 38 8. Stoiser B, Mörtl D, Hülsmann M, et al. Copeptin, a fragment of the vasopressin precursor, 39 40 as a novel predictor of outcome in heart failure. Eur J Clin Invest. 2006;36(11):7718. 41 doi:10.1111/j.13652362.2006.01724.x. 42 on September 28, 2021 by guest. Protected copyright. 43 9. Yalta K, Yalta T, Sivri N, Yetkin E. Copeptin and cardiovascular disease: a review of a 44 novel neurohormone. Int J Cardiol. 2013;167(5):17509. 45 46 doi:10.1016/j.ijcard.2012.12.039. 47 48 10. Khan SQ, Dhillon OS, O’Brien RJ, et al. CTerminal Provasopressin (Copeptin) as a 49 Novel and Prognostic Marker in Acute Myocardial Infarction. Circulation. 50 2007;115(16):21032110. 51 52 53 11. Keller T, Tzikas S, Zeller T, et al. Copeptin Improves Early Diagnosis of Acute 54 Myocardial Infarction. J Am Coll Cardiol. 2010;55(19):20962106. 55 56 57 58 59 60 13

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 21 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 12. Velho G, Bouby N, Hadjadj S, et al. Plasma copeptin and renal outcomes in patients with 4 5 type 2 diabetes and albuminuria. Diabetes Care. 2013;36(11):363945. doi:10.2337/dc13 6 0683. 7 8 13. Morgenthaler NG, Struck J, Jochberger S, Dünser MW. Copeptin: clinical use of a new 9 biomarker. Trends Endocrinol Metab. 2008;19(2):439. doi:10.1016/j.tem.2007.11.001. 10 11 12 14. Bouby N, Bachmann S, Bichet D, Bankir L. Effect of water intake on the progression of 13 chronic renal failure in the 5/6 nephrectomized rat. Am J Physiol Ren Physiol. 14 1990;258(4):F973. 15 For peer review only 16 15. Sugiura T, Yamauchi A, Kitamura H, et al. High water intake ameliorates 17 18 tubulointerstitial injury in rats with subtotal nephrectomy: Possible role of TGFβ. Kidney 19 Int. 1999;55(5):18001810. 20 21 16. Clark WF, Sontrop JM, Macnab JJ, et al. Urine Volume and Change in Estimated GFR in 22 a CommunityBased Cohort Study. Clin J Am Soc Nephrol. 2011;6(11):26342641. 23 24 25 17. Sontrop JM, Dixon SN, Garg AX, et al. Association between water intake, chronic kidney 26 disease, and cardiovascular disease: A crosssectional analysis of NHANES data. Am J 27 Nephrol. 2013;37:434442. 28 29 30 18. Strippoli GF, Craig JC, Rochtchina E, Flood VM, Wang JJ, Mitchell P. Fluid and nutrient 31 intake and risk of chronic kidney disease. Nephrol. 2011;16:326334. 32 33 19. Clark WF, Sontrop JM, Huang SH, et al. The chronic kidney disease Water Intake Trial

34 (WIT): results from the pilot randomised controlled trial. BMJ Open. 2013;3(12):e003666. http://bmjopen.bmj.com/ 35 doi:10.1136/bmjopen2013003666. 36 37 38 20. Levey AS, Stevens LA, Schmid CH, et al. A New Equation to Estimate Glomerular 39 Filtration Rate. Ann Intern Med. 2009;150(9):604612. 40 41 21. Moher D, Hopewell S, Schulz KF, et al. CONSORT 2010 explanation and elaboration: 42 on September 28, 2021 by guest. Protected copyright. 43 updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c869. 44 45 22. Sedgwick P. Randomised controlled trials: balance in baseline characteristics. BMJ. 46 2014;349:g5721. doi:10.1136/bmj.g5721. 47 48 49 23. Meijer E, Bakker SJ, de Jong PE, et al. Copeptin, a surrogate marker of vasopressin, is 50 associated with accelerated renal function decline in renal transplant recipients. 51 Transplantation. 2009;88:561567. 52 53 24. Zittema D, van den Berg E, Meijer E, et al. Kidney function and plasma copeptin levels in 54 healthy kidney donors and autosomal dominant polycystic kidney disease patients. Clin J 55 56 Am Soc Nephrol. 2014;9(9):155362. doi:10.2215/CJN.08690813. 57 58 59 60 14

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 21 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 25. Meijer E, Bakker SJ, Halbesma N, de Jong PE, Struck J, Gansevoort RT. Copeptin, a 4 5 surrogate marker of vasopressin, is associated with microalbuminuria in a large population 6 cohort. Kidney Int. 2010;77:2936. 7 8 26. Meijer E, Casteleijn NF. Riding the waves: evidence for a beneficial effect of increased 9 water intake in autosomal dominant polycystic kidney disease patients? Nephrol Dial 10 11 Transplant. 2014;29(9):16157. doi:10.1093/ndt/gfu054. 12 13 27. Bankir L, Bouby N, Ritz E. Vasopressin: a novel target for the prevention and retardation 14 of kidney disease? Nat Rev Nephrol. 2013;9(4):22339. doi:10.1038/nrneph.2013.22. 15 For peer review only 16 28. Nagao S, Nishii K, Katsuyama M, et al. Increased Water Intake Decreases Progression of 17 18 Polycystic Kidney Disease in the PCK Rat. J Am Soc Nephrol. 2006;17(8):22202227. 19 20 29. Wang X, Gattone V, Harris PC, Torres VE. Effectiveness of vasopressin V2 receptor 21 antagonists OPC31260 and OPC41061 on polycystic kidney disease development in the 22 PCK rat. J Am Soc Nephrol. 2005;16(4):84651. doi:10.1681/ASN.2004121090. 23 24 25 30. Torres VE, Chapman AB, Devuyst O, et al. Tolvaptan in patients with autosomal 26 dominant polycystic kidney disease. N Engl J Med. 2012;367(25):24072418. 27 28 31. Bankir L, Bichet DG, Bouby N. Vasopressin V2 receptors, ENaC, and sodium 29 30 reabsorption: a risk factor for hypertension? Am J Physiol Ren Physiol. 2010;299:F917 31 F928. doi:10.1152/ajprenal.00413.2010. 32 33 32. Enhorning S, Bankir L, Bouby N, et al. Copeptin, a marker of vasopressin, in abdominal

34 obesity, diabetes and microalbuminuria: the prospective Malmo Diet and Cancer Study http://bmjopen.bmj.com/ 35 cardiovascular cohort. Int J Obes(Lond). 2012;37:598603. doi:10.1038/ijo.2012.88. 36 37 38 33. Enhorning S, Leosdottir M, Wallstrom P, et al. Relation between human vasopressin 1a 39 gene variance, fat intake, and diabetes. Am J Clin Nutr. 2009;89:400406. 40 41 34. Plischke M, Kohl M, Bankir L, et al. Urine osmolarity and risk of dialysis initiation in a 42 on September 28, 2021 by guest. Protected copyright. 43 chronic kidney disease cohorta possible titration target? PLoS One. 2014;9(3):e93226. 44 doi:10.1371/journal.pone.0093226. 45 46 35. Bankir L, Bichet DG. Polycystic kidney disease: An early ureaselective urine 47 concentrating defect in ADPKD. Nat Rev Nephrol. 2012;8(8):4379. 48 49 doi:10.1038/nrneph.2012.139. 50 51 36. Johnson RJ, RodriguezIturbe B, RoncalJimenez C, et al. Hyperosmolarity drives 52 hypertension and CKDwater and salt revisited. Nat Rev Nephrol. 2014;10(7):41520. 53 doi:10.1038/nrneph.2014.76. 54 55 56 57 58 59 60 15

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 21 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 37. Bardoux P, Bichet DG, Martin H, et al. Vasopressin increases urinary albumin excretion 4 5 in rats and humans: involvement of V2 receptors and the reninangiotensin system. 6 Nephrol Dial Transplant. 2003;18:497506. 7 8 38. Cirillo M. Determinants of kidney dysfunction : is vasopressin a new player in the arena ? 9 Epidemiology of atherosclerotic renal artery stenosis as a mirror. Kidney Int. 10 11 2010;77(1):56. doi:10.1038/ki.2009.408. 12 13 39. Bolignano D, Zoccali C. Vasopressin beyond water: implications for renal diseases. Curr 14 Opin Nephrol Hypertens. 2010;19:499504. doi:10.1097/MNH.0b013e32833d35cf. 15 For peer review only 16 40. Morgenthaler NG. Copeptin: a biomarker of cardiovascular and renal function. Congest 17 18 Hear Fail. 2010;16 Suppl 1:S37S44. 19 20 41. Fenske W, Wanner C, Allolio B, et al. Copeptin levels associate with cardiovascular 21 events in patients with ESRD and type 2 diabetes mellitus. J Am Soc Nephrol. 22 2011;22(4):78290. doi:10.1681/ASN.2010070691. 23 24 25 42. Decaux G, Soupart A, Vassart G. Nonpeptide argininevasopressin antagonists: the 26 vaptans. Lancet. 2008;371(9624):162432. doi:10.1016/S01406736(08)606959. 27 28 43. Palmer SC, Wong G, Iff S, et al. Fluid intake and allcause mortality, cardiovascular 29 30 mortality, and kidney function: a populationbased longitudinal cohort study. Nephrol 31 Dial Transplant. 2014. 32 33 44. Szinnai G, Morgenthaler NG, Berneis K, et al. Changes in Plasma Copeptin, the C

34 Terminal Portion of Arginine Vasopressin during Water Deprivation and Excess in http://bmjopen.bmj.com/ 35 Healthy Subjects. J Clin Endocrinoll Metab. 2007;92:39733978. 36 37 38 45. Reichlin T, Hochholzer W, Stelzig C, et al. Incremental value of copeptin for rapid rule 39 out of acute myocardial infarction. J Am Coll Cardiol. 2009;54(1):6068. 40 41 46. Yun AJ, Lee PY, Bazar KA. Clinical benefits of hydration and volume expansion in a 42 on September 28, 2021 by guest. Protected copyright. 43 wide range of illnesses may be attributable to reduction of sympathovagal ratio. Med 44 Hypotheses. 2005;64(3):646650. doi:10.1016/j.mehy.2004.07.014. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 16

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 21 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 Table 1. Baseline characteristics by treatment assignment 4 5 6 Treatment Group 7 Control Hydration 8 n=11 n=18 9 Mean age, years (SD) 67 (11) 59 (14) 10 Males, n (%) 7 (64%) 12 (67%) 11 Caucasian, n (%) 10 (91%) 14 (78%) 12 2 13 Body Mass Index, kg/m (SD) 30 (6) 31 (6) 14 Waist circumference, cm (SD) 110 (11) 101 (18) 15 Smoking status, nFor (%) peer review only 16 Current 0 1 (6%) 17 Former 8 (73%) 9 (50%) 18 Cause of chronic kidney disease, n (%) 19 Diabetes 5 (46%) 3 (17%) 20 21 Hypertension 3 (27%) 3 (17%) 22 Polycystic kidney disease 0 3 (17%) 23 Unknown/other 4 (36%) 9 (50%) 24 Comorbidities, n (%) 25 Hypertension 11 (100%) 13 (72%) 26 Hyperlipidemia 8 (73%) 9 (50%) 27 Diabetes 7 (64%) 8 (44%) 28 29 Peripheral vascular disease 3 (27%) 1 (6%) 30 Gastric bleeding 2 (18%) 0 31 Malignancy 0 2 (11%) 32 Cerebrovascular/TIA 1 (9%) 1 (6%) 33 Coronary artery disease 1 (9%) 1 (6%)

34 COPD 1 (9%) 1 (6%) http://bmjopen.bmj.com/ 35 Mean blood pressure, mm Hg (SD) 36 37 Systolic 143 (17) 140 (22) 38 Diastolic 73 (11) 80 (11) 2 39 eGFR, ml/min/1.73m (SD) 39 (11) 41 (10) 40 Hematocrit, L/L (SD) 0.39 (0.05) 0.40 (0.06) 41 HbA1c, % (SD) 0.07 (0.02) 0.07 (0.01) 42 Medications, n (%) on September 28, 2021 by guest. Protected copyright. 43 ACE/ARB inhibitors 7 (64%) 12 (67%) 44 Statin 7 (64%) 8 (44%) 45 46 Diuretics 9 (82%) 5 (28%) 47 Calcium channel blockers 5 (46%) 4 (22%) 48 Aspirin 5 (46%) 3 (17%) 49 Angiotensin II receptor blockers 5 (46%) 3 (17%) 50 Beta blockers 3 (27%) 3 (17%) 51 Vasopressor 0 1 (6%) 52 First degree relative with 53 5 (46%) 11 (61%) hypertension or kidney failure, n (%) 54 55 Abbreviations: COPD, chronic obstructive pulmonary disorder; eGFR, estimated glomerular filtration rate; SD, standard 56 deviation. 57 58 59 60 17

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from Page 18 of 21 18 18

b p p 0.19 0.19 0.002 0.002 value

pvalue 19.4 (14, 33) 33) (14, 19.4 (p=0.76) 1.1 a 6-weeks 6-weeks

rho 0.01 0.01 0.94 0.13 0.13 0.50 0.44* 0.44* 0.02 0.47* 0.47* 0.01 0.48* 0.48* 0.01 0.49** 0.49** 0.008 0.53** 0.53** 0.004 BMJ Open 0.56** 0.56** 0.002 http://bmjopen.bmj.com/

Baseline Baseline weeks 6 Change pvalue a Baseline Baseline on September 28, 2021 by guest. Protected copyright. rho 0.15 0.15 0.44 0.04 0.04 0.83 0.17 0.17 0.38 0.51* 0.51* 0.006 0.53** 0.53** 0.003

Hydration (n=17) (n=17) Hydration 29) (8, 15.0 26) (6, 10.8 (p=0.005) 3.6 Hydration (n=17) (n=17) Hydration (n=11) Control (0.6) 2.3 36) (12, 19.3 (1.2) 3.0 (p=0.01) 0.7 Control (n=11) (n=11) Control (0.7) 2.0 (0.6) 1.7 (p=0.07) 0.2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only

2 Correlations with copeptin in 28 patients with stage 3 chronic kidney disease kidney 3 chronic stage with patients in28 copeptin with Correlations . Effect of increased water intake on the plasma concentration of copeptin of plasma concentration the intake on water increased of Effect . able 3. 3. able The betweengroup difference change baselinein from to 6week compared the was using ttestindependent (urine volume) MannWhitney and the U test Spearman’s correlationSpearman’s coefficient. Followup – pvalue baseline; for change calculatedwithingroup using the pairedsamples (urinettest volume) and the relatedsamples Wilcoxon Signed Rank Median copeptin, pmol/L (IQR) (IQR) pmol/L copeptin, Median *p<0.05; **p<0.01. a Serum sodium, mmol/L mmol/L sodium, Serum Urine sodium, mmol/d mmol/d sodium, Urine Albumin/creatinine, mg/mmol mg/mmol Albumin/creatinine, 0.35 0.07 Serum urea, mmol/L urea, mmol/L Serum Serum osmolality, mOsm/Kg mOsm/Kg osmolality, Serum 0.58** 0.001 Urine osmolality, mOsm/Kg mOsm/Kg osmolality, Urine 0.26 0.17 eGFR, ml/min/1.73m eGFR,

Urine volume, L/d L/d Urinevolume,

T Abbreviations: IQR,Abbreviations: interquartile SD, range; standard deviation. a b Mean urine volume, L/d (SD) (SD) L/d volume, urine Mean Table2 (copeptin).

Test (copeptin). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from 19 19

Group meanGroup 6 6 weeks Hydration (n=17) group Hydration Baseline 0 70 60 50 40 30 20 10 BMJ Open Group Group mean http://bmjopen.bmj.com/ 6 6 weeks on September 28, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Control group (n=11) group Control For peer review onlyBaseline 0 70 60 50 40 30 20 10 . Intraindividual change in copeptin between baseline and sixweeks after randomization after sixweeks and baseline between incopeptin . change Intraindividual (pmol/L) Copeptin Copeptin Figure 1 Figure

Page 19 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from Page 1 1 Page Page 20 of 21 * * Reported Reported on page No page on n\a n\a 5 5

5 5 6 n/a n/a 4 4 4-5 6 6 6 6 6 5 5 1 1 6 6 n\a n\a 5 5 2-3 (for specific guidance see CONSORT for abstracts) for see CONSORT guidance specific (for

BMJ Open http://bmjopen.bmj.com/ on September 28, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only actually administered administered actually Checklist item Checklist interventions interventions describing any steps taken to conceal the sequence until interventions were assigned assigned were interventions until sequence the conceal to stepstaken any describing were assessed assessed were 9 9 containers), numbered sequentially as (such sequence randomallocation the implement to used Mechanism 1a 1a titlethe in trial randomised as a Identification 6a 6a they when and how including measures, outcome secondary and primary pre-specified defined Completely 7b 7b guidelines stopping and analyses interimof any explanation applicable, When 6b 6b reasons with commenced, trial the after outcomes trial to changes Any 4b 4b collected were data the where locations and Settings 3b 3b reasons criteria), with aseligibility (such commencement trial after methods to changes Important 2a 2a of rationale explanation and background Scientific 2b 2b hypotheses or objectives Specific 8a 8a sequence allocation random the generate to used Method 8b size) block and blocking as (such restriction ofany details of randomisation; Type 1b 1b conclusions and results, methods, design, of trial summary Structured No No 11a 11a those providers, care participants, example, (for interventions to after assignment blinded was who Ifdone, n/a Item Item CONSORT CONSORT 2010 checklist of to include wheninformation randomised reporting a trial generation generation concealment concealment mechanism Implementation Implementation 10 to participants assigned and who participants, enrolled who sequence, random allocation the Whogenerated Sequence Sequence Allocation Allocation Blinding Blinding

Randomisation: Randomisation: Sample size size Sample 7a determined was size sample How Outcomes Outcomes Interventions Interventions 5 were they andwhen how including allowreplication, to details sufficient with group foreach Theinterventions Participants Participants 4a for criteria participants Eligibility Trial design design Trial 3a ratio allocation including factorial) parallel, as (such design oftrial Description Methods Methods Background and Background objectives Introduction Introduction

Title and abstract abstract and Title Section/Topic Section/Topic CONSORT 2010 checklist checklist 2010 CONSORT 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from Page 2 2 Page 9 9

1 Table 7, 5 5 7 7 8 5 5 6 6 5 5 5 5 11-12 11-12 8 7 1 1 n/a n/a 7 7 Table 1 1 Table 7 7

. www.consort-statement.org (for specific guidance see CONSORT for harms) for CONSORT see guidance specific (for BMJ Open http://bmjopen.bmj.com/ on September 28, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only by original assigned groups assignedgroups original by pre-specified from exploratory fromexploratory pre-specified assessing outcomes) and how how and outcomes) assessing precision (such as 95% confidence interval) interval) confidence 95% as (such precision were analysed for the primary outcome primary for analysed the were 25 25 funders of role ofdrugs), supply as (such support other and funding of Sources 23 23 registry trial ofname and number Registration 24 24 ifavailable accessed, be can protocol trial full the Where 19 19 group each in effects unintended or harms important All 17b 17b sizesrecommended is effect relative and absolute of both presentation outcomes, binary For 14b 14b stopped was or ended trial the Why 12b 12b analyses adjusted and analyses subgroup as such analyses, foradditional Methods 11b 11b ofinterventions similarity ofthe description Ifrelevant, 17a 17a andits size effect estimated the and group, eachfor results outcome, andsecondary primary each For 13a 13a and treatment, intended received assigned, randomly were who participants of numbers the group, each For 13b 13b reasons with together randomisation, after exclusions and losses group, each For

recommend reading recommend extensionsCONSORT for cluster trials,randomised non-inferiority trials, and equivalence non-pharmacological treatments, herbal and interventions, pragmatic trials. extensions Additional are forthcoming: for thosefor date references and up to this relevant checklist, to see recommendstrongly *We this reading conjunction statement in the with 2010 CONSORT Explanationand forElaboration important clarifications items.allthe If on relevant, also we Funding Registration Other information information Other Interpretation Interpretation 22 evidence relevant other considering and harms, and benefits balancing results, with consistent Interpretation 12 Limitations Limitations Generalisability 21 20 findings trial of the applicability) validity, (external Generalisability ofanalyses multiplicity if relevant, and, imprecision, bias, potential sources of addressing limitations, Trial 11-12 Discussion Discussion Harms Harms Outcomes and and Outcomes estimation Numbers analysed analysed Numbers 16 was analysis the whether and analysis each in included (denominator) participants of number group, each For Baseline data data Baseline 15 group each for characteristics clinical and demographic baseline showing Atable Recruitment Recruitment 14a follow-up and recruitment periods of the defining Dates Participant flow (a (a flow Participant strongly is diagram recommended) Results Results Statistical methods methods Statistical 12a outcomes secondary and primary for compare groups to used methods Statistical CONSORT 2010 checklist checklist 2010 CONSORT Protocol Ancillary analyses analyses Ancillary 18 distinguishing analyses, adjusted and analyses subgroup including performed, analyses other any of Results Page 21 of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

The effect of increased water intake on plasma copeptin in patients with chronic kidney disease: results from a pilot randomized controlled trial

For peer review only Journal: BMJ Open

Manuscript ID bmjopen-2015-008634.R1

Article Type: Research

Date Submitted by the Author: 18-Sep-2015

Primary Subject Renal medicine

Heading: http://bmjopen.bmj.com/

Secondary Subject Heading: Renal medicine

coeptin, water intake, glomerular filtration rate, 24-hour urine volume, Keywords: chronic kidney disease

on September 28, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 The effect of increased water intake on plasma copeptin in patients with chronic kidney 4 5 disease: results from a pilot randomized controlled trial 6 1,2 1 1,2 7 Jessica M. Sontrop PhD, Shi-Han Huang MSc MD, Amit X. Garg MD PhD, Louise Moist 1,2 1 1 1 8 MSc MD, Andrew A. House MD, Kerri Gallo RN, William F. Clark MD * 9 10 1. Division of Nephrology, Department of Medicine, London Health Sciences Centre, London 11 ON, Canada N6A 4G5 12 13 2. Department of Epidemiology & Biostatistics, Western University, London ON, Canada 14 N6A 5C1 15 For peer review only 16 *Corresponding Author 17 Dr. William F. Clark 18 Victoria Hospital 19 20 800 Commissioner’s Road East, A2-343 21 London ON, Canada N6A 5W9 22 Phone: 519-685-8361 23 Fax: 519-685-8047 24 Email: [email protected] 25 26 27 28 29 30 Word count: main body: 2896 words; abstract: 300 words 31 32 33 Key words: 24-hour urine volume, chronic kidney disease, copeptin, randomized controlled

34 trial, vasopressin http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 28, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 1

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

34 http://bmjopen.bmj.com/ 35 Measures and Outcomes: Participants provided blood and 24-h urine samples at baseline and 6 36 37 weeks. Change in plasma copeptin was compared within and between study groups. 38 39 40 Results: Participants were 64% male with a mean age of 62 years and an estimated glomerular 41 42

2 on September 28, 2021 by guest. Protected copyright. 43 filtration rate of 40 ml/min/1.73m . Between baseline and 6 weeks, 24-h urine volume increased 44 45 by 0.7 L/d in the hydration group, rising from 2.3 to 3.0 L/d (p=0.01), while decreasing by 0.3 46 47 L/d among controls, from 2.0 to 1.7 L/d (p=0.07); between-group difference: 0.9 L/d (95% CI: 48 49 50 0.37 to 1.46; p=0.002). In the hydration group, median copeptin decreased by 3.6 pmol/L, from 51 52 15.0 to 10.8 pmol/L (p=0.005), while remaining stable among controls at 19 pmol/L (p=0.76) 53 54 (p=0.19 for the between-group difference in median change); the between-group difference in 55 56 57 mean change was 5.4 pmol/L (95% CI: -1.2 to 12.0; p=0.11). 58 59 60 2

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 Conclusions: Adults with stage 3 chronic kidney disease can be successfully randomized to 4 5 6 drink approximately 1 L more per day than controls. This increased water intake caused a 7 8 significant decrease in plasma copeptin concentration. Our larger 12-month trial will examine 9 10 11 whether increased water intake can slow renal decline in patients with chronic kidney disease 12 13 (clinicaltrials.gov: NCT01766687). 14 15 For peer review only 16 17 18 19 20 21 22 23 Strengths and Limitations 24 25 26 • In this randomized controlled pilot trial, 28 patients with stage 3 chronic kidney disease were 27 28 successfully randomized to drink approximately 1 L more per day than controls for six 29 30 31 weeks, as verified by 24-hour urine collections. This increased water intake caused a 32 33 significant decrease in plasma copeptin concentration.

34 http://bmjopen.bmj.com/ 35 • Patients were followed for only six weeks. Whether the effect of increased water intake on 36 37 38 plasma copeptin concentration is clinically significant, beneficial, or sustainable over time is 39 40 unknown. 41 42 on September 28, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 3

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 INTRODUCTION 4 5 6 Vasopressin is an essential antidiuretic hormone in mammals that regulates thirst and urine 7 8 water.1 However, chronically elevated levels of vasopressin may have adverse health effects. 9 10 Until recently, it was not possible to reliably measure vasopressin due to the limited sensitivity of 11 12 13 available assays; however, in 2006, Morgenthaler et al. developed an immunoassay for 14 15 copeptin—a glycosylatedFor peptide peer that is co-released review with vasopressin only from the hypothalamus.2 16 17 Copeptin can be measured from blood samples and is demonstrated to be a reliable marker of 18 19 2,3 20 vasopressin. The availability of this assay has renewed scientific interest into the role of 21 22 vasopressin and copeptin in chronic illness. In longitudinal studies, a higher baseline 23 24 25 concentration of plasma copeptin predicts kidney function decline patients with autosomal 26 4,5 27 dominant polycystic kidney disease, and other studies have linked copeptin to the development 28 29 of diabetes,6 the metabolic syndrome,7 and heart failure.8,9 In cohort studies, elevated plasma 30 31 10,11 32 copeptin at baseline is strongly predictive of subsequent myocardial infarction and end-stage 33 12 34 kidney disease. http://bmjopen.bmj.com/ 35 36 37 Many factors are known to stimulate vasopressin secretion, including acute stress and illness.13 38 39 40 Other stimuli include high plasma osmolality and dehydration—vasopressin is the first hormone 41 42 released during dehydration.1 In experimental studies of rats, increased water intake was shown on September 28, 2021 by guest. Protected copyright. 43 44 to suppress vasopressin, reduce proteinuria, and improve creatinine clearance.14,15 In large cohort 45 46 16–18 47 studies, renal decline was slower in those with greater water inake, but evidence from 48 49 clinical trials is lacking. It is not known whether copeptin can be adequately suppressed by 50 51 increased water intake in patients with Chronic Kidney Disease. In 2012, we launched the pilot 52 53 54 phase of a randomized controlled trial that will test whether increased water intake can slow 55 56 renal decline in patients with chronic kidney disease. The results of the pilot trial were published 57 58 59 60 4

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 in December, 2013;19 however, data on plasma copeptin did not become available until 2014. 4 5 6 Here, we examine the effect of increased water intake on plasma copeptin. We also examine the 7 8 relationships between copeptin and 24-hr urine volume, osmolality, sodium, albuminuria, and 9 10 11 estimated glomerular filtration rate (eGFR). 12 13 14 METHODS 15 For peer review only 16 Study Population 17 18 We analyzed data from the pilot phase of the Water Intake Trial (WIT), a parallel-group 19 20 21 randomized controlled trial conducted between October, 2012 and March, 2013 in London, 22 23 Ontario, Canada (registered at clinicaltrials.gov NCT01753466).19 A figure detailing participant 24 25 19 26 selection and follow-up is available in Clark et al. The primary aim was to assess the feasibility 27 28 and safety of asking adults with stage 3 chronic kidney disease to increase their water intake and 29 30 to test our planned procedures, recruitment, and operational strategies for use in our larger trial 31 32 33 (NCT01766687). We enrolled 29 patients as recommended by Moore et al and Julious for pilot

34 http://bmjopen.bmj.com/ 35 studies assessing feasibility.20,21 All patients provided informed consent consistent with the 36 37 Declaration of Helsinki and ethics approval was obtained from the Western University Research 38 39 40 Ethics Board for Health Sciences. 41 42 Eligibility criteria included age 30-80 years; chronic kidney disease (stage 3), defined as the on September 28, 2021 by guest. Protected copyright. 43 44 presence of reduced kidney function (an estimated glomerular filtration rate (eGFR) 30–60 45 46 2 47 mL/min/1.73 m ) determined from a blood sample taken from participants at baseline; 48 49 proteinuria (albumin/creatinine >2.8 mg/mmol [if female] or >2.0 mg/mmol [if male] from a spot 50 51 22 52 urine sample or trace protein [albustix]) ; and 24-hr urine volume <3L/day (all participants 53 54 provided a 24-hour urine sample at baseline). We excluded patients who met any of the 55 56 following criteria: self-reported fluid intake ≥10 cups/day; had received a dialysis treatment in 57 58 59 60 5

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 the past month; kidney transplant recipient (or on waiting list); under fluid restriction; pregnant 4 5 6 or breastfeeding; symptomatic kidney stones in past 5 years; a life expectancy less than two 7 8 years; serum sodium ≤130 mmol/L; serum calcium >2.6 mmol/L; currently taking lithium (a 9 10 11 drug which affects thirst and urination) or high daily doses of the following diuretics: 12 13 hydrochlorothiazide >25 mg/day, indapamide >1.25 mg/day, furosemide >40 mg/day, or 14 15 metolazone >2.5For mg/day. peer review only 16 17 18 19 Intervention 20 21 We randomized 29 patients by computer-generated randomization in block sizes of 3 to a 22 23 hydration or control group (2:1), stratified by gender. This 2:1 randomization in the pilot phase 24 25 26 was chosen to provide experience delivering the hydration intervention to more patients within 27 28 an overall sample of 29 patients. The hydration group (n=18) was coached to increase their oral 29 30 31 water intake by 1.0 to 1.5 L/day depending on sex, weight, and 24-hour urine osmolality (in 32 19 33 addition to usual consumed beverages) for 6 weeks (see Table 1 in Clark et al.). We advised a

34 http://bmjopen.bmj.com/ 35 gradual increase in water intake over two-weeks. During week one, we instructed participants to 36 37 38 consume one cup of water at breakfast, lunch and dinner, and during week two, the full amount 39 40 according to weight and sex (Table 1 in Clark et al.19). We used a variety of techniques to 41 42 encourage adherence to the fluid regimen. Participants were given reusable drinking containers, on September 28, 2021 by guest. Protected copyright. 43 44 45 and the study dietician provided individual consultations with all participants (in person or by 46 47 telephone). We also conducted informed hydration coaching (Table 2 in Clark et al.19) based on 48 49 urine colour charts and level of spot urine osmolality, which was measured every two weeks 50 51 52 after randomization. At these times the research coordinator also inquired about regimen 53 54 tolerance and adherence. The control group (n=11) was asked to continue with their usual water 55 56 57 58 59 60 6

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 intake or to decrease water intake by 1-2 cups/day depending on their baseline 24-hour urine 4 5 6 osmolality. 7 8 9 Outcomes, Measurements and Definitions 10 11 In this secondary analysis of the WIT pilot trial, the primary outcome was the between-group 12 13 14 change in plasma copeptin. All participants provided 24-hour urine samples and blood samples at 15 For peer review only 16 baseline and 6-weeks after randomization. Serum creatinine was measured using the isotope 17 18 dilution/mass spectroscopy-traceable enzymatic method, and eGFR was calculated using the 19 20 23 21 Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation. 22 23 24 We measured the concentration of sodium, osmolality, and urea in blood and 24-hour urine 25 26 samples and the 24-hour urine albumin to creatinine ratio using standard methods.19 Copeptin 27 28 29 concentration was measured in plasma-EDTA samples. Samples were stored at -20°C and 30 31 analyzed for copeptin in a single batch analysis to eliminate interassay variation. Copeptin 32 33 concentration was measured using a sandwich immunoassay (Thermo Fisher Scientific

34 http://bmjopen.bmj.com/ 35 36 B.R.A.H.M.S, Hennigsdorf/ Berlin, Germany). 37 38 39 Statistical Analyses 40 41 Normally distributed data were summarized using means and standard deviations (SD), and non- 42 on September 28, 2021 by guest. Protected copyright. 43 44 normally distributed data were summarized using medians and inter-quartile ranges (IQR). All 45 46 randomized participants were included in the analysis and analyzed according to group 47 48 49 assignment. Within-group changes in the amount of urine volume and the concentration of 50 51 plasma copeptin were compared using the paired t-test and the related-samples Wilcoxon signed- 52 53 rank test, respectively; the between-group change was compared using the independent t-test and 54 55 56 Mann-Whitney U, as appropriate. Correlations between copeptin and measures of kidney 57 58 59 60 7

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 function, urine volume, sodium, osmolality, and albumin were analyzed using the Spearman’s 4 5 6 rank correlation coefficient (rho) for non-normally distributed data. Correlations are presented 7 8 for the overall sample at baseline and 6-weeks follow up. Data were analyzed using SPSS 9 10 11 version 21. 12 13 14 RESULTS 15 For peer review only 16 Of 74 participants who met the initial eligibility criteria and were invited to participate, 33 were 17 18 19 consented and 29 were randomized (4 patients withdrew before randomization). One participant 20 21 in the hydration group withdrew after randomization due to a flare up of Crohn’s disease. 22 23 24 Participants were 64% male and 82% Caucasian with an average age of 62 years (SD 14) and an 25 26 2 2 27 average eGFR of 40 ml/min/1.73m (SD 11 ml/min/1.73m ); 50% had diabetes and 82% had 28 29 hypertension. Median copeptin at baseline was 17 pmol/L (IQR 9, 31), and was higher in males 30 31 32 [17 (IQR 11, 35) vs. 12 (IQR 6, 30) in females]. Participant characteristics by treatment group 33 24

34 are shown in Table 1 (as per CONSORT guidelines, p-values were not calculated). Although http://bmjopen.bmj.com/ 35 36 randomization protects against baseline differences between groups, differences can arise in 37 38 25 39 smaller samples, for example, participants in the control group were older, had more 40 41 comorbidities, and had more diuretic use compared with those in the hydration group. As well, 42 on September 28, 2021 by guest. Protected copyright. 43 there were three participants with polycystic kidney disease in the hydration group (and none in 44 45 46 the control group); however, study results remained unchanged when these patients were 47 48 excluded in sensitivity analysis (the copeptin levels of these three participants were 16.3, 12.0, 49 50 and 7.5; slightly below the group average). 51 52 53 54 Change in copeptin between baseline and 6 weeks is shown in Table 2 and Figure 1. During this 55 56 time, 24-hour urine volume increased by 0.7 L/d in the hydration group (p=0.01) and decreased 57 58 59 60 8

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 by 0.3 L/d in the control group (p=0.07); the between-group difference in change was 0.9 L/d 4 5 6 (95% CI: 0.37 to 1.46; p=0.002). In the hydration group, the median plasma copeptin 7 8 concentration decreased significantly between baseline and follow-up, from 15.0 pmol/L to 10.8 9 10 11 pmol/L (p=0.005), while remaining relatively stable among controls, at 19 pmol/L (p=0.76); 12 13 p=0.19 for the between-group difference in median change. The between-group difference in the 14 15 mean change wasFor 5.4 pmol/L peer (95% CI: -1.2 review to 12.0; p=0.11). only 16 17 18 19 Correlations between the concentration of plasma copeptin and urine volume, eGFR, and other 20 21 urine and serum measures are shown in Table 3 (scatter plots of these correlations are provided 22 23 in Supplementary Figures S1 to S8). At baseline, copeptin was inversely correlated with eGFR 24 25 26 (rho=-0.53; p=0.003) and positively correlated with serum osmolality (rho=0.58; p=0.001) and 27 28 serum urea (rho=0.51; p=0.001). At six weeks follow-up, copeptin was inversely correlated with 29 30 24-hr urine volume (rho= -0.48; p=0.01) and eGFR (rho=-0.56; p=0.002), and positively 31 32 33 correlated with the albumin to creatinine ratio (rho= 0.44; p=0.02), serum urea (rho= 0.49;

34 http://bmjopen.bmj.com/ 35 p=0.008) and urine and serum osmolality (rho=0.53; p=0.004 and rho=0.47; p=0.01, 36 37 38 respectively). No correlation was seen with urine sodium or serum sodium at either time point. 39 40 41 42 on September 28, 2021 by guest. Protected copyright. 43 44 DISCUSSION 45 46 In this randomized controlled pilot trial of 28 patients with stage 3 chronic kidney disease, 47 48 49 patients were successfully randomized to drink approximately 1 L more per day than controls for 50 51 six weeks. In the hydration group, 24-hr urine volume increased significantly and the between- 52 53 group difference was 1.3 L/d at 6 weeks. This increased water intake caused a significant 54 55 56 decrease in plasma copeptin concentration among patients in the hydration group, although the 57 58 59 60 9

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 between-group difference was not statistically significant. No adverse effects were reported nor 4 5 19 6 observed. 7 8 9 The median baseline copeptin concentration in our study was 17 pmol/L, which is approximately 10 11 four times higher than values reported in healthy volunteers.13 Other studies have shown that 12 13 3 6 14 copeptin is elevated in patients with chronic kidney disease, diabetes, myocardial 15 For peer review only 16 infarction,10,11 and heart failure.8,9 In kidney transplant recipients, higher levels of copeptin at 17 18 baseline predicted a significantly faster decline in kidney graft function over a median follow-up 19 20 26 21 of 3.6 years, independent of baseline eGFR, proteinuria and other risk factors. 22 23 24 In our study, copeptin was inversely correlated with both 24-hr urine volume and eGFR, and 25 26 27 positively correlated with urine protein, urine osmolality, serum osmolality, and serum urea. If 28 29 copeptin is cleared by the kidneys then it will necessarily increase as kidney function declines, 30 31 and this could explain the inverse relationship between copeptin and eGFR. However, in a study 32 33 by Zittema et al., copeptin was not associated with GFR in healthy living kidney donors—and 34 http://bmjopen.bmj.com/ 35 36 copeptin levels did not change after donation despite a significant drop in kidney function after 37 38 nephrectomy.27 These data suggest that GFR alone is not a principal determinant of copeptin. 39 40 41 42 What is the link between copeptin and kidney function? on September 28, 2021 by guest. Protected copyright. 43 44 An inverse relationship between copeptin and kidney function is consistently demonstrated in 45 46 47 experimental and observational studies, in diverse patient groups, and appears to be independent 48 3,26,28 49 of age, sex, blood pressure, and other risk factors. Nonetheless, it remains unknown whether 50 51 copeptin itself has a direct causal effect on kidney function. In the unique case of ADPKD, a rise 52 53 54 in vasopressin (and copeptin) stimulates the formation of cAMP (3’,5’-cyclic adenosine 55 56 monophosphate), which promotes cell proliferation and cyst growth leading to kidney 57 58 59 60 10

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 enlargement and a subsequent decline in kidney function.29,30 Accordingly, in several animal 4 5 6 studies, researchers showed that blocking the cAMP-mediated pathway with V2-receptor 7 8 antagonists resulted in significantly reduced cyst growth,31,32 and, in a 3-year double-blind 9 10 11 randomized placebo-controlled trial, patients treated with tolvaptan (a V2-receptor antagonist) 12 13 experienced a significantly slower increase in total kidney volume and a significantly slower 14 15 decline in kidneyFor function. 33peer Interestingly, reviewincreased oral water intakeonly in rats has been 16 17 31 18 demonstrated to have a similar suppressing effect on vasopressin secretion. 19 20 21 Vasopressin has potent vasoconstrictive effects, and some hypothesize that the associations 22 23 observed between copeptin and renal/cardiovascular outcomes may be partly explained by 24 25 12,34,35 26 vasopressin’s effect on blood pressure; however, there is no convincing evidence that 27 28 vasopressin increases blood pressure. In patients treated with selective V2 receptor antagonists, 29 30 the plasma level of vasopressin increases by about three times with no concomitant change in 31 32 36–38 33 blood pressure. In human cohort studies, the relationship between copeptin and blood

34 http://bmjopen.bmj.com/ 35 pressure is inconsistent and weak, and associations between copeptin and other outcomes remain 36 37 significant after controlling for blood pressure or hypertension.28,35,39 In our study, correlations 38 39 40 between copeptin and blood pressure were below 0.1 (data not shown). Many other mechanisms 41 42 have been proposed to explain the link between vasopressin/copeptin, GFR, and albuminuria; on September 28, 2021 by guest. Protected copyright. 43 44 these include vasopressin’s effect on urinary concentrating activity,40,41 hyperosmolarity,42 45 46 43 1,26,30 47 activation of the renin-angiotensin system, glomerular hyperfiltration and hypertrophy, 48 49 high salt intake and V2-receptor-dependent tubular effects on sodium reabsorption,12,44 As 50 51 52 described in Bolignano et al., these pathways are not mutually exclusive and may act together to 53 45 54 affect renal outcomes. 55 56 57 58 59 60 11

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 Copeptin is also a marker of the body’s endocrine stress response, which is mediated through the 4 5 7,9 6 hypothalamus-pituitary-adrenal system, and is activated in acute illness. For example, copeptin 7 8 levels spike in concert with cortisol and corticotropin-releasing hormone within hours of acute 9 10 46 11 myocardial infarction onset. It is unclear, however, whether copeptin is simply a marker of 12 13 stress or illness, but if it plays a direct causative role in the pathophysiology of cardiovascular 14 15 and chronic kidneyFor disease. 47peer In studies of reviewanimals and humans, only stimulating vasopressin causes 16 17 18 an increase in proteinuria, while suppressing vasopressin through V2 antagonism or water intake 19 20 reduces proteinuria.14,15,43 Further, in studies that controlled for inflammatory biomarkers (C- 21 22 reactive protein and N-terminal pro brain natriuretic peptide), copeptin remained significantly 23 24 26,28 25 associated with renal decline and proteinuria. These data have sparked interest in the use of 26 27 vasopressin receptor antagonists to improve renal and cardiovascular outcomes, and several are 28 29 currently being investigated for clinical use.30,48 If proved effective, copeptin may serve as a 30 31 46 32 useful biomarker to identify patients who may most benefit from treatment. 33

34 http://bmjopen.bmj.com/ 35 The main limitations of this study are its small sample size (28 patients) and short follow-up. As 36 37 38 the primary aim of this pilot study was to assess the feasibility and safety of increasing hydration 39 19 40 among adults with chronic kidney disease, our study was not powered to detect changes in 41 42 secondary outcome variables. However, few studies have examined short-term changes in on September 28, 2021 by guest. Protected copyright. 43 44 45 copeptin in the same participants. While we were not able to determine the precise time-lag 46 47 between change in water intake and change in plasma copeptin, copeptin decreased in nearly all 48 49 of the participants in the hydration group between baseline and six weeks while showing a more 50 51 52 variable pattern in the control group (Figure 1). Participants in the hydration group were 53 54 instructed to increase water intake by up to 1.5 L/d, and 24-hr urine volume was significantly 55 56 57 higher in the hydration group compared to controls at the end of six weeks. This increase in 58 59 60 12

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 water intake was associated with a significant decrease in the concentration of plasma copeptin. 4 5 6 However, it is not known if decreasing plasma copeptin actually leads to any health benefits for 7 8 patients, or if any such effects would be sustained over time. Another limitation of this study is 9 10 11 that we were not able to determine how much participants increased their intake of plain water 12 13 vs. other fluids (and if different types of fluid affect copeptin differently). In other studies that 14 15 examined intakeFor of plain water peer vs. other fluids review in relation to renal only outcomes, a beneficial effect 16 17 18 was seen with greater intakes of plain water, but not other beverages, with negative effects seen 19 20 for increased intake of sweetened beverages.3,17,49 Many of these limitations will be addressed in 21 22 our larger 12-month trial (expected completion date: December 2016; NCT01766687), which 23 24 25 will examine the effect of increased water intake and 24-hr urine volume on change in renal 26 27 function in patients with chronic kidney disease (primary outcome); the effect of increased water 28 29 intake on change in copeptin over 12 months will be examined as a secondary outcome. As well, 30 31 32 participants will complete a 3-day food-and-fluid intake record and a water survey at three times 33

34 during the trial, which will provide additional information on types of fluid consumed. http://bmjopen.bmj.com/ 35 36 37 38 Our study and others have clearly demonstrated that copeptin is sensitive to changes in water 39 28,50 40 intake and is inversely associated with 24-hour urine volume. Copeptin is also inversely 41 42 related to eGFR, is positively associated with urine protein, and is a prognostic biomarker for on September 28, 2021 by guest. Protected copyright. 43 44 6,8–11,27,28,51 45 future events of diabetes and heart disease. Does it follow that hydration is a unifying 46 47 factor linking copeptin to kidney function and other diseases, as some suggest?52 Possibly, 48 49 however evidence from a large randomized controlled trial is needed to determine if increased 50 51 52 water intake has a direct protective effect on kidney function, and if so, to what extent and by 53 54 what mechanism. The possibility that hydration may play a protective role in chronic kidney 55 56 53 57 disease is biologically plausible and is supported by a diverse and growing body of research; 58 59 60 13

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 however, whether this relationship is causal, clinically significant, and sustained over time is 4 5 6 unknown. We anticipate that the results from larger trials (including our own 12-month trial, 7 8 currently in progress), with repeated measures of copeptin, 24-hour urine volume, osmolality, 9 10 11 and eGFR over time will provide key insight into the causality, direction, and magnitude of the 12 13 relationships between water intake, copeptin, and kidney function. 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 28, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 14

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 AUTHOR CONTRIBUTIONS 4 5 6 Each author contributed to the conception and design of the study and interpretation of the data. 7 8 Dr. Sontrop analyzed the data, Drs Sontrop and Clark drafted the article, and all authors revised 9 10 11 it critically for important intellectual content. All authors had full access to all of the data in the 12 13 study (including statistical reports and tables) and can take responsibility for the integrity of the 14 15 data and the accuracyFor of the peer data analysis. review only 16 17 18 Specific contributions are as follows: 19 20 21 Study concept and design: WFC, JMS, SSH, AXG, LM, AH, KG 22 23 Acquisition of data: WFC, SSH, LM, AH, KG 24 25 26 Data Analysis: JMS 27 28 Interpretation of the data: WFC, JMS, SSH, AXG, LM, AH, KG 29 30 Drafting of the manuscript: JMS, WFC 31 32 33 Critical revision of the manuscript for important intellectual content: JMS, WFC, SSH, AXG,

34 http://bmjopen.bmj.com/ 35 LM, AH, KG 36 37 Guarantor: WFC 38 39 40 41 42 on September 28, 2021 by guest. Protected copyright. 43 Competing interests: Dr. William Clark and Dr. Louise Moist have received consulting fees 44 45 46 and/or honorarium and support to travel to meetings with Danone Research. Dr. Clark is 47 48 currently conducting a randomized controlled trial funded by Danone Research. No other authors 49 50 have relevant conflicts of interest to declare. 51 52 53 54 Funding: This work was supported by Danone Research and the Program of Experimental 55 56 Medicine, Western University, Canada. Dr. Garg is supported by the Dr. Adam Linton Chair in 57 58 59 60 15

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 Kidney Health Analytics. The study sponsors had no direct role in the data collection, statistical 4 5 6 analysis, interpretation of the results or drafting of the manuscript. 7 8 9 Data sharing statement: Dataset is available by emailing corresponding author at 10 11 12 [email protected] 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 28, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 16

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 References 4 5 6 1. Bouby N, Fernandes S. Mild dehydration, vasopressin and the kidney: animal and human 7 8 studies. Eur J Clin Nutr. 2003;57 Suppl 2:S39-S46. doi:10.1038/sj.ejcn.1601900. 9 10 2. Morgenthaler NG, Struck J, Alonso C, Bergmann A. Assay for the measurement of 11 copeptin, a stable peptide derived from the precursor of vasopressin. Clin Chem. 12 2006;52:112-119. 13 14 15 3. Roussel R,For Fezeu L, Marrepeer M, et al. reviewComparison Between only Copeptin and Vasopressin in a 16 Population from the Community and in People with Chronic Kidney Disease. J Clin 17 Endocrinol Metab. 2014;99:4656-4663. doi:10.1210/jc.2014-2295. 18 19 4. Boertien WE, Meijer E, Zittema D, et al. Copeptin, a surrogate marker for vasopressin, is 20 21 associated with kidney function decline in subjects with autosomal dominant polycystic 22 kidney disease. Nephrol Dial Transplant. 2012;27(11):4131-4137. 23 doi:10.1093/ndt/gfs070. 24 25 5. Boertien WE, Meijer E, Li J, et al. Relationship of copeptin, a surrogate marker for 26 27 arginine vasopressin, with change in total kidney volume and GFR decline in autosomal 28 dominant polycystic kidney disease: results from the CRISP cohort. Am J Kidney Dis. 29 2013;61(3):420-429. doi:10.1053/j.ajkd.2012.08.038. 30 31 6. Enhorning S, Wang TJ, Nilsson PM, et al. Plasma copeptin and the risk of diabetes 32 Circulation 33 mellitus. . 2010;121:2102-2108.

34 http://bmjopen.bmj.com/ 35 7. Saleem U, Khaleghi M, Morgenthaler NG, et al. Plasma carboxy-terminal provasopressin 36 (copeptin): a novel marker of insulin resistance and metabolic syndrome. J Clin 37 Endocrinol Metab. 2009;94(7):2558-2564. doi:10.1210/jc.2008-2278. 38 39 40 8. Stoiser B, Mörtl D, Hülsmann M, et al. Copeptin, a fragment of the vasopressin precursor, 41 as a novel predictor of outcome in heart failure. Eur J Clin Invest. 2006;36(11):771-778. 42

doi:10.1111/j.1365-2362.2006.01724.x. on September 28, 2021 by guest. Protected copyright. 43 44 9. Yalta K, Yalta T, Sivri N, Yetkin E. Copeptin and cardiovascular disease: a review of a 45 46 novel neurohormone. Int J Cardiol. 2013;167(5):1750-1759. 47 doi:10.1016/j.ijcard.2012.12.039. 48 49 10. Khan SQ, Dhillon OS, O’Brien RJ, et al. C-Terminal Provasopressin (Copeptin) as a 50 Novel and Prognostic Marker in Acute Myocardial Infarction. Circulation. 51 52 2007;115(16):2103-2110. 53 54 11. Keller T, Tzikas S, Zeller T, et al. Copeptin Improves Early Diagnosis of Acute 55 Myocardial Infarction. J Am Coll Cardiol. 2010;55(19):2096-2106. 56 57 58 59 60 17

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 12. Velho G, Bouby N, Hadjadj S, et al. Plasma copeptin and renal outcomes in patients with 4 5 type 2 diabetes and albuminuria. Diabetes Care. 2013;36(11):3639-3645. 6 doi:10.2337/dc13-0683. 7 8 13. Morgenthaler NG, Struck J, Jochberger S, Dünser MW. Copeptin: clinical use of a new 9 biomarker. Trends Endocrinol Metab. 2008;19(2):43-49. doi:10.1016/j.tem.2007.11.001. 10 11 12 14. Bouby N, Bachmann S, Bichet D, Bankir L. Effect of water intake on the progression of 13 chronic renal failure in the 5/6 nephrectomized rat. Am J Physiol Ren Physiol. 14 1990;258(4):F973. 15 For peer review only 16 15. Sugiura T, Yamauchi A, Kitamura H, et al. High water intake ameliorates 17 18 tubulointerstitial injury in rats with subtotal nephrectomy: Possible role of TGF-β. Kidney 19 Int. 1999;55(5):1800-1810. 20 21 16. Clark WF, Sontrop JM, Macnab JJ, et al. Urine Volume and Change in Estimated GFR in 22 a Community-Based Cohort Study. Clin J Am Soc Nephrol. 2011;6(11):2634-2641. 23 24 25 17. Sontrop JM, Dixon SN, Garg AX, et al. Association between water intake, chronic kidney 26 disease, and cardiovascular disease: A cross-sectional analysis of NHANES data. Am J 27 Nephrol. 2013;37:434-442. 28 29 30 18. Strippoli GF, Craig JC, Rochtchina E, Flood VM, Wang JJ, Mitchell P. Fluid and nutrient 31 intake and risk of chronic kidney disease. Nephrol. 2011;16:326-334. 32 33 19. Clark WF, Sontrop JM, Huang S-H, et al. The chronic kidney disease Water Intake Trial

34 (WIT): results from the pilot randomised controlled trial. BMJ Open. 2013;3(12):e003666. http://bmjopen.bmj.com/ 35 doi:10.1136/bmjopen-2013-003666. 36 37 38 20. Julious SA. Sample size of 12 per group rule of thumb for a pilot study. Pharm Stat. 39 4(4):287-291. 40 41 21. Moore CG, Carter RE, Nietert PJ, Stewart PW. Recommendations for planning pilot 42 on September 28, 2021 by guest. Protected copyright. 43 studies in clinical and translational research. Clin Transl Sci. 2011;4(5):332-337. 44 doi:10.1111/j.1752-8062.2011.00347.x. 45 46 22. National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney 47 disease: evaluation, classification, and stratification. Am J kidney Dis. 2002;39(2 Suppl 48 49 1):S1-S266. 50 51 23. Levey AS, Stevens LA, Schmid CH, et al. A New Equation to Estimate Glomerular 52 Filtration Rate. Ann Intern Med. 2009;150(9):604-612. 53 54 24. Moher D, Hopewell S, Schulz KF, et al. CONSORT 2010 explanation and elaboration: 55 56 updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c869. 57 58 59 60 18

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 25. Sedgwick P. Randomised controlled trials: balance in baseline characteristics. BMJ. 4 5 2014;349:g5721. doi:10.1136/bmj.g5721. 6 7 26. Meijer E, Bakker SJ, de Jong PE, et al. Copeptin, a surrogate marker of vasopressin, is 8 associated with accelerated renal function decline in renal transplant recipients. 9 Transplantation. 2009;88:561-567. 10 11 12 27. Zittema D, van den Berg E, Meijer E, et al. Kidney function and plasma copeptin levels in 13 healthy kidney donors and autosomal dominant polycystic kidney disease patients. Clin J 14 Am Soc Nephrol. 2014;9(9):1553-1562. doi:10.2215/CJN.08690813. 15 For peer review only 16 28. Meijer E, Bakker SJ, Halbesma N, de Jong PE, Struck J, Gansevoort RT. Copeptin, a 17 18 surrogate marker of vasopressin, is associated with microalbuminuria in a large population 19 cohort. Kidney Int. 2010;77:29-36. 20 21 29. Meijer E, Casteleijn NF. Riding the waves: evidence for a beneficial effect of increased 22 water intake in autosomal dominant polycystic kidney disease patients? Nephrol Dial 23 24 Transplant. 2014;29(9):1615-1617. doi:10.1093/ndt/gfu054. 25 26 30. Bankir L, Bouby N, Ritz E. Vasopressin: a novel target for the prevention and retardation 27 of kidney disease? Nat Rev Nephrol. 2013;9(4):223-239. doi:10.1038/nrneph.2013.22. 28 29 30 31. Nagao S, Nishii K, Katsuyama M, et al. Increased water intake decreases progression of 31 polycystic kidney disease in the PCK rat. J Am Soc Nephrol. 2006;17(8):2220-2227. 32 doi:10.1681/ASN.2006030251. 33

34 32. Wang X, Gattone V, Harris PC, Torres VE. Effectiveness of vasopressin V2 receptor http://bmjopen.bmj.com/ 35 antagonists OPC-31260 and OPC-41061 on polycystic kidney disease development in the 36 37 PCK rat. J Am Soc Nephrol. 2005;16(4):846-851. doi:10.1681/ASN.2004121090. 38 39 33. Torres VE, Chapman AB, Devuyst O, et al. Tolvaptan in patients with autosomal 40 dominant polycystic kidney disease. N Engl J Med. 2012;367(25):2407-2418. 41 42 on September 28, 2021 by guest. Protected copyright. 43 34. Bankir L, Bichet DG, Bouby N. Vasopressin V2 receptors, ENaC, and sodium 44 reabsorption: a risk factor for hypertension? Am J Physiol Ren Physiol. 2010;299:F917- 45 F928. doi:10.1152/ajprenal.00413.2010. 46 47 35. Enhorning S, Bankir L, Bouby N, et al. Copeptin, a marker of vasopressin, in abdominal 48 49 obesity, diabetes and microalbuminuria: the prospective Malmo Diet and Cancer Study 50 cardiovascular cohort. Int J Obes(Lond). 2012;37:598-603. doi:10.1038/ijo.2012.88. 51 52 36. Goldsmith SR. Vasopressin as vasopressor. Am J Med. 1987;82(6):1213-1219. 53 doi:10.1016/0002-9343(87)90228-2. 54 55 56 57 58 59 60 19

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 37. Finley JJ, Konstam MA, Udelson JE. Arginine vasopressin antagonists for the treatment 4 5 of heart failure and hyponatremia. Circulation. 2008;118(4):410-421. 6 doi:10.1161/CIRCULATIONAHA.108.765289. 7 8 38. Wong L. Vasopressin V2 receptor antagonists. Cardiovasc Res. 2001;51(3):391-402. 9 doi:10.1016/S0008-6363(01)00315-7. 10 11 12 39. Enhorning S, Leosdottir M, Wallstrom P, et al. Relation between human vasopressin 1a 13 gene variance, fat intake, and diabetes. Am J Clin Nutr. 2009;89:400-406. 14 15 40. Plischke M,For Kohl M, peerBankir L, et al. review Urine osmolarity and only risk of dialysis initiation in a 16 chronic kidney disease cohort--a possible titration target? PLoS One. 2014;9(3):e93226. 17 18 doi:10.1371/journal.pone.0093226. 19 20 41. Bankir L, Bichet DG. Polycystic kidney disease: An early urea-selective urine- 21 concentrating defect in ADPKD. Nat Rev Nephrol. 2012;8(8):437-439. 22 doi:10.1038/nrneph.2012.139. 23 24 25 42. Johnson RJ, Rodriguez-Iturbe B, Roncal-Jimenez C, et al. Hyperosmolarity drives 26 hypertension and CKD--water and salt revisited. Nat Rev Nephrol. 2014;10(7):415-420. 27 doi:10.1038/nrneph.2014.76. 28 29 30 43. Bardoux P, Bichet DG, Martin H, et al. Vasopressin increases urinary albumin excretion 31 in rats and humans: involvement of V2 receptors and the renin-angiotensin system. 32 Nephrol Dial Transplant. 2003;18:497-506. 33

34 44. Cirillo M. Determinants of kidney dysfunction : is vasopressin a new player in the arena ? http://bmjopen.bmj.com/ 35 Epidemiology of atherosclerotic renal artery stenosis as a mirror. Kidney Int. 36 37 2010;77(1):5-6. doi:10.1038/ki.2009.408. 38 39 45. Bolignano D, Zoccali C. Vasopressin beyond water: implications for renal diseases. Curr 40 Opin Nephrol Hypertens. 2010;19:499-504. doi:10.1097/MNH.0b013e32833d35cf. 41 42 on September 28, 2021 by guest. Protected copyright. 43 46. Morgenthaler NG. Copeptin: a biomarker of cardiovascular and renal function. Congest 44 Hear Fail. 2010;16 Suppl 1:S37-S44. 45 46 47. Fenske W, Wanner C, Allolio B, et al. Copeptin levels associate with cardiovascular 47 events in patients with ESRD and type 2 diabetes mellitus. J Am Soc Nephrol. 48 49 2011;22(4):782-790. doi:10.1681/ASN.2010070691. 50 51 48. Decaux G, Soupart A, Vassart G. Non-peptide arginine-vasopressin antagonists: the 52 vaptans. Lancet. 2008;371(9624):1624-1632. doi:10.1016/S0140-6736(08)60695-9. 53 54 49. Palmer SC, Wong G, Iff S, et al. Fluid intake and all-cause mortality, cardiovascular 55 56 mortality, and kidney function: a population-based longitudinal cohort study. Nephrol 57 Dial Transplant. January 2014. 58 59 60 20

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 50. Szinnai G, Morgenthaler NG, Berneis K, et al. Changes in Plasma Copeptin, the C- 4 5 Terminal Portion of Arginine Vasopressin during Water Deprivation and Excess in 6 Healthy Subjects. J Clin Endocrinoll Metab. 2007;92:3973-3978. 7 8 51. Reichlin T, Hochholzer W, Stelzig C, et al. Incremental value of copeptin for rapid rule 9 out of acute myocardial infarction. J Am Coll Cardiol. 2009;54(1):60-68. 10 11 12 52. Yun AJ, Lee PY, Bazar KA. Clinical benefits of hydration and volume expansion in a 13 wide range of illnesses may be attributable to reduction of sympatho-vagal ratio. Med 14 Hypotheses. 2005;64(3):646-650. doi:10.1016/j.mehy.2004.07.014. 15 For peer review only 16 53. Wang CJ, Grantham JJ, Wetmore JB. The medicinal use of water in renal disease. Kidney 17 18 Int. 2013;84(1):45-53. doi:10.1038/ki.2013.23. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 28, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 21

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 35 BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 Figure Legends 4 5 6 7 8 Figure 1. Intra-individual change in copeptin between baseline and six-weeks after 9 randomization 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 28, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 22

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 Table 1. Baseline characteristics by treatment assignment 4 5 6 Treatment Group 7 Control Hydration 8 n=11 n=17 9 Mean age, years (SD) 67 (11) 60 (14) 10 Males, n (%) 7 (64%) 11 (65%) 11 Caucasian, n (%) 10 (91%) 13 (77%) 12 2 13 Body Mass Index, kg/m (SD) 30 (6) 31 (6) 14 Waist circumference, cm (SD) 110 (11) 101 (18) 15 Smoking status, nFor (%) peer review only 16 Current 0 1 (6%) 17 Former 8 (73%) 9 (53%) 18 Cause of chronic kidney disease, n (%) 19 Diabetes 5 (46%) 3 (18%) 20 21 Hypertension 3 (27%) 3 (18%) 22 Polycystic kidney disease 0 3 (18%) 23 Unknown/other 4 (36%) 8 (47%) 24 Comorbidities, n (%) 25 Hypertension 11 (100%) 12 (71%) 26 Hyperlipidemia 8 (73%) 8 (47%) 27 Diabetes 7 (64%) 7 (41%) 28 29 Peripheral vascular disease 3 (27%) 1 (6%) 30 Gastric bleeding 2 (18%) 0 31 Malignancy 0 2 (12%) 32 Cerebrovascular/TIA 1 (9%) 1 (6%) 33 Coronary artery disease 1 (9%) 1 (6%)

34 COPD 1 (9%) 1 (6%) http://bmjopen.bmj.com/ 35 Mean blood pressure, mm Hg (SD) 36 37 Systolic 143 (17) 139 (22) 38 Diastolic 73 (11) 79 (11) 2 39 eGFR, ml/min/1.73m (SD) 39 (11) 41 (10) 40 Hematocrit, L/L (SD) 0.39 (0.05) 0.39 (0.06) 41 HbA1c, % (SD) 0.07 (0.02) 0.07 (0.01) 42 Medications, n (%) on September 28, 2021 by guest. Protected copyright. 43 ACE/ARB inhibitors 7 (64%) 11 (65%) 44 Statin 7 (64%) 8 (47%) 45 46 Diuretics 9 (82%) 5 (29%) 47 Calcium channel blockers 5 (46%) 4 (24%) 48 Aspirin 5 (46%) 3 (18%) 49 Angiotensin II receptor blockers 5 (46%) 3 (18%) 50 Beta blockers 3 (27%) 3 (18%) 51 Vasopressor 0 1 (6%) 52 First degree relative with 53 5 (46%) 10 (59%) 54 hypertension or kidney failure, n (%) 55 Abbreviations: COPD, chronic obstructive pulmonary disorder; eGFR, estimated glomerular filtration rate; SD, standard 56 deviation. 57 58 59 60 23

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from Page 24 of 35 24 24

b p- p- 0.19 0.19 0.002 0.002 value

p-value 19.4 (14, 33) 33) (14, 19.4 (p=0.76) -1.1 a 6-weeks 6-weeks

rho 0.01 0.01 0.94 0.13 0.13 0.50 0.44* 0.44* 0.02 0.47* 0.47* 0.01 -0.48* -0.48* 0.01 0.49** 0.49** 0.008 0.53** 0.53** 0.004 BMJ Open -0.56** -0.56** 0.002 http://bmjopen.bmj.com/

Baseline Baseline weeks 6 Change p-value a Baseline Baseline on September 28, 2021 by guest. Protected copyright. rho 0.26 0.26 0.17 0.15 0.15 0.44 -0.04 -0.04 0.83 -0.17 -0.17 0.38 0.51* 0.51* 0.006 -0.53** -0.53** 0.003

Hydration (n=17) (n=17) Hydration 29) (8, 15.0 26) (6, 10.8 (p=0.005) -3.6 Hydration (n=17) (n=17) Hydration (n=11) Control (0.6) 2.3 36) (12, 19.3 (1.2) 3.0 (p=0.01) 0.7 Control (n=11) (n=11) Control (0.7) 2.0 (0.6) 1.7 (p=0.07) -0.2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only

2 . Effect of increased water intake on the plasma concentration of copeptin copeptin of plasma concentration the intake on water increased of Effect . Correlations with copeptin in 28 patients with stage 3 chronic kidney disease kidney 3 chronic stage with patients in28 copeptin with Correlations 3. able The between-group difference change baseline in from to 6week compared the was using t-testindependent (urine volume) Mann-Whitney and the U test Spearman’s correlationSpearman’s coefficient. Follow-up – p-value baseline; for change calculatedwithin-group using the paired-samples (urinet-test volume) and the related-samples Wilcoxon Signed Rank Median copeptin, pmol/L (IQR) (IQR) pmol/L copeptin, Median *p<0.05; **p<0.01. a Serum sodium, mmol/L mmol/L sodium, Serum Urine sodium, mmol/d mmol/d sodium, Urine Albumin/creatinine, mg/mmol mg/mmol Albumin/creatinine, 0.35 0.07 Serum urea, mmol/L urea, mmol/L Serum Serum osmolality, mOsm/kg mOsm/kg osmolality, Serum 0.58** 0.001 Urine osmolality, mOsm/kg mOsm/kg osmolality, Urine eGFR, ml/min/1.73m eGFR,

Urine volume, L/d L/d Urinevolume,

T Abbreviations: IQR,Abbreviations: inter-quartile SD, range; standard deviation. a b Mean urine volume, L/d (SD) (SD) L/d volume, urine Mean 2 Table (copeptin).

Test (copeptin). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 25 of 35 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Intra-individual change in copeptin between baseline and six-weeks after randomization 33 254x190mm (300 x 300 DPI)

34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 28, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2015-008634 on 24 November 2015. Downloaded from

BMJ Open Page 26 of 35