Search
Close
Search
Advanced Search
Search Menu
AI Discovery Assistant

ABSTRACT

Background

Chronic kidney disease is an important contributor to morbidity and mortality. 3-methylhistidine (3-MH) is the by-product of actin and myosin degradation reflecting skeletal muscle turnover. Markedly elevated 3-MH levels have been documented in uraemic patients, but the interpretation of high 3-MH concentration in maintenance haemodialysis (MHD) patients remains unclear. Indeed, it is not known whether elevated serum 3-MH levels are a marker of excessive muscle catabolism or a better lean tissue mass. Here, we evaluated the association between serum 3-MH levels and clinical outcomes in these patients.

Methods

Serum 3-MH concentration was measured by reverse-phase liquid chromatography/tandem mass spectrometry in a cohort of MHD patients. We analysed the relationships between various clinical/laboratory indices, lean tissue mass measured by bioimpedance spectroscopy, mortality and cardiovascular (CV) events.

Results

Serum 3-MH concentration was positively correlated with serum albumin, normalized protein catabolic rate (nPCR), simplified creatinine index (SCI) and lean tissue mass. Of 291 MHD patients, during a mean follow-up of 847 days, 91 patients died and 101 patients experienced a CV event. Survival was significantly better in patients with high 3-MH concentrations (P = .002). A higher level of 3-MH was also associated with a lower CV mortality and lower incidence of CV events (P = .015 and P < .001, respectively). Low serum 3-MH levels remained significantly associated with CV events but not with mortality after adjustment for demographic, metabolic and CV risk factors.

Conclusion

Elevated serum 3-MH concentration appears to be a marker of better lean tissue mass and nutritional status. Low serum 3-MH is a robust and independent predictor of CV events in the MHD population.

Graphical Abstract
Graphical Abstract
Open in new tabDownload slide
3-methylhistidine, cardiovascular events, lean tissue mass, maintenance haemodialysis, mortality
KEY LEARNING POINTS

What is already known about this subject?

  • Patients on maintenance haemodialysis have a high cardiovascular (CV) risk and mortality rate.

  • 3-methylhistidine (3-MH) appears as an interesting marker of frailty, nutritional status and lean tissue mass.

  • Scant data are available about 3-MH in dialysis patients, so we evaluated 3-MH performance to predict poor outcomes in this specific population.

What this study adds?

  • We highlighted that 3-MH is a robust marker of nutritional status and lean tissue mass in dialysis patients and we demonstrated that 3-MH is a good predictor of CV events in this population.

  • We want to share with readers that a low lean tissue mass estimated by 3-MH is significantly correlated with poor clinical outcomes in haemodialysis patients.

What impact this may have on practice or policy?

  • Our results suggest that 3-MH may be a new marker to monitor lean tissue mass in dialysis patients and a potentially relevant marker to evaluate CV risk in these patients.

INTRODUCTION

Patients on maintenance haemodialysis (MHD) experience a significantly increased risk of death and cardiovascular (CV) events [1]. Besides CV traditional risk factors, MHD patients are exposed to non-traditional risk factors such as protein energy wasting (PEW) and sarcopaenia [2]. Instrumental methods, including bioimpedance spectroscopy (BIS) or dual-emission X-ray absorptiometry (DXA), are of great value to measure lean body mass, yet they are expensive, limited to clinical research and not ubiquitously used in clinical practice. Serum creatinine and the serum creatinine index (SCI) have been used to evaluate muscle mass and were validated in several studies [3, 4]. Nevertheless, several parameters may influence creatinine levels, including protein intake and physical activity. Furthermore, creatinine might be partially eliminated through the gastrointestinal tract to compensate for renal dysfunction [5]. Because of the high prevalence of sarcopaenia and its involvement in morbidity and mortality, there is an unmet need to have an accurate and simple marker to evaluate and monitor muscle mass in MHD patients.

3-methylhistidine (3-MH) is a histidine derivative produced by the degradation of several tissues, especially skeletal muscle [6]. 3-MH is the by-product of actin and myosin degradation in muscle [7–9]. In humans, ∼75% of 3-MH is issued from skeletal muscle turnover, while 10% is produced in the intestine and 15% in the skin [10]. A high plasma level of 3-MH seemed to be a valuable marker of skeletal muscle catabolism in different physiological and pathological conditions, both in young and elderly subjects [11–13]. As a result, the ability to evaluate increased protein turnover through the monitoring of muscle 3-MH plasma release under clinical and pathological conditions is very attractive to evidence individuals with increased risk of deteriorating muscle mass.

In chronic kidney disease (CKD) patients, particularly in those undergoing dialysis, there is a lack of robust data about 3-MH metabolism, especially in adult patients [14–17]. After its synthesis, 3-MH is not metabolically reused and is only removed [18]. 3-MH is cleared by the kidneys (filtered, not secreted and poorly reabsorbed) [19]. Serum 3-MH concentration is correlated with glomerular filtration rate (GFR) [14, 15] and 3-MH seems partially removed during dialysis sessions [17, 20]. Conflicting results were reported in patients with CKD. On the one hand, 3-MH was poorly positively correlated with creatinine production and muscle mass in MHD patients [20]. On the other hand, a low serum 3-MH level is an independent predictor of aortic arterial stiffness in MHD patients [21] and is related to the first hospitalization after kidney transplantation [22]. One explanation for these discordant results could be that the 3-MH level is influenced by meat consumption and a low level of 3-MH may further represent poor nutritional status (e.g. malnutrition) [23–25]. Therefore it is unclear whether 3-MH is a marker of muscle mass or an index of excessive muscle catabolism and if it is a robust biomarker to predict clinical outcomes in MHD patients.

Because we thought that serum 3-MH levels might be an interesting marker of lean tissue in MHD patients, we measured the serum concentration of 3-MH in a large prospective cohort of 291 MHD patients. We sought to examine the association between serum 3-MH levels, mortality and CV events in a large prospective cohort of MHD patients undergoing measurement of clinical, laboratory and body composition and anthropometry characteristics. Also, we tested if serum 3-MH concentrations were correlated with these different parameters.

MATERIALS AND METHODS

Study design and population

The study protocol was approved by the local ethics committee (DC-2009-1066) and was conducted in accordance with the ethical standards and principles of the second Declaration of Helsinki. All participants involved in the study signed written informed consent forms before enrolment. Prevalent MHD adult patients followed between 1 January 2014 and 31 December 2016 at a dialysis centre [Association pour l'Utilisation du Rein Artificiel dans la Région Lyonnaise (AURAL)] in Lyon (France) were asked to participate. Inclusion criteria were regular haemodialysis (HD) sessions at least three times a week for at least 3 months without current hospitalization. All patients received uniform HD treatments via high-flux membranes with reuse and standard water purification and processing techniques. Pregnant women were excluded. Demographic factors, relevant medical history and the usual drug treatment were collected at enrolment. Comorbidity severity was assessed for all patients by the Charlson comorbidity index (CCI) [26]. Dialysis vintage corresponded to the period between the date of HD initiation and the date of inclusion. Residual renal function (RRF) was defined as zero in patients without the use of a diuretic drug. The diuretic prescription was assessed every 3 months in this centre from the results of quarterly biological evaluations and 24-h diuresis. The CV history included a history of myocardial infarction, stroke, heart failure, angina pectoris or surgical procedures for angina or coronary/peripheral artery disease (including percutaneous transluminal angioplasty). The history of cancer was also collected.

Laboratory measurements

Fasted blood samples were obtained at the beginning of the mid-week dialysis session. Standard methods in the routine clinical laboratory were used to measure biological parameters. Albumin was measured by immunonephelometry. Creatinine was assayed via an enzymatic method (Roche, Meylan, France) with calibrators assigned by isotope-dilution mass spectrometry. 3-MH was measured by reverse-phase liquid chromatography/tandem mass spectrometry according to the method described by Piraud et al. [27], with an API3200 triple quadrupole apparatus (AB Sciex, Framingham, MA, USA) coupled to a 1200 Series HPLC system (Agilent Technologies, Les Ulis, France) equipped with a YMC-Pack ODS-AQ Narrowbore C18 column. Serum samples were deproteinized with methanol and injected without derivatization. Amino acid standards (Merck, Saint Quentin-Fallavier, France) were used for calibration and d3-Histidine (Cluzeau Info Labo, Sainte-Foy-La-Grande, France) was used as an internal standard.

Follow-up

During the follow-up period, until the endpoint date of 1 January 2019, clinical events, including all-cause mortality and CV events, were identified in all patients included. Thus the patients were followed up for at least 24 months. All-cause mortality was the main outcome. We performed secondary analysis for CV mortality and CV events. CV events were defined as CV death or any event related to the CV system or any procedures to cure coronary or peripheral arterial disease. CV death corresponded to death attributed to cardiac arrest, myocardial infarction, cardiogenic shock, peripheral ischaemia or stroke. When death occurred outside the hospital, if no other cause was specified, it was considered as cardiac arrest.

Nutritional status

Dietary protein intakes were assessed by normalized protein catabolic rate (nPCR) calculated with the Garred formula [28]. Body mass index (BMI) was defined as the post-dialysis weight (in kilograms) divided by height (in meters squared). The simplified creatinine index (SCI) was calculated according to Canaud et al. [29]:
$$\begin{eqnarray*} {\rm{SCI}}\left( {{\rm{mg}}/{\rm{kg}}/{\rm{day}}} \right) &=& 16.21 + 1.12 \times {\rm{Gender}} - 0.06\\ && \times\, \rm {age} -\, 0.08\times {\rm{spKt}}/{\rm{V}} + 0.009\\ && \times\, {\rm{creatinine}}\, \rm{(}{\mu}\rm{mol/L)}, \end{eqnarray*}$$
with Gender = 1 for males and Gender = 0 for females, age in years, spKt/V being single-pool Kt/V.

Whole-body bioimpedance spectroscopy (BIS) measurement was performed before a midweek dialysis session in a sub-group of 161 patients (55.3%) using a body composition monitor (BCM; Fresenius Medical Care, Hamburg, Germany) as previously reported [30]. The BCM measures the electrical responses at 50 different frequencies between 5 and 1000 kHz. Input variables include the patient's height, weight, age and sex. Electrodes were attached to the hand and foot on the non-dominant side of the body after the patient had been in a recumbent position for at least 10 min. This method expresses body composition as a three-compartment model, providing overhydration (OH), lean tissue index (LTI) and fat tissue index (FTI), whereby LTI and FTI are the respective tissue mass [lean tissue mass (LTM) and adipose tissue mass (ATM)] normalized to the height squared (expressed in kg/m²). Extracellular water (ECW), intracellular water (ICW) and total body water (TBW) were determined from the measured impedance data, calculated using the equations of Moissl et al. [31] and expressed in litres (L).

Statistical analysis

Data were analysed using GraphPad Prism 8.0 (GraphPad Software, La Jolla, CA, USA) and R software version 4.0.3 (https://www.r-project.org/). Data distributions were tested for normality using the D'Agostino–Pearson test. 3-MH appeared non-normally distributed (P < .001) and was therefore log-transformed prior to analyses. All other data also exhibited non-normal distribution. Data were expressed as median and minimum–maximum range. Simple comparisons were made using the Mann–Whitney U test. The chi-squared test was used to compare categorical variables. Multiple comparisons were performed using the Kruskal–Wallis test. Univariable analysis was performed using the Pearson correlation method. For the survival analysis, we stratified our population into tertiles according to the log-transformed 3-MH concentration [lowest tertile (n = 95): median log-transformed 3-MH concentration 1.336 µmol/L (range 1.076–1.384); middle tertile (n = 99): median log-transformed 3-MH concentration 1.447 µmol/L (range 1.386–1.508); highest tertile (n = 97): median log-transformed 3-MH concentration 1.573 µmol/L (range 1.509–1.794)]. We compared survival differences between tertiles of 3-MH for all-cause mortality, CV mortality and CV events using Kaplan–Meier analysis with the log-rank test. Data were censored at renal transplantation, loss to follow-up or the end of the study observation period. The diagnostic performance of 3-MH was assessed by analysing the receiver operating characteristics (ROC) curve and was compared to the areas under the ROC curves of usual mortality risk factors using non-parametric tests according to DeLong et al. [32]. A multivariable Cox analysis was performed for all-cause mortality, CV mortality and CV events. 3-MH was used as a continuous variable with four levels of adjustment: model 1, adjusted for demographic characteristics (age, sex, dialysis dose estimation by Kt/V and dialysis vintage); model 2, adjusted for phospho-calcic metabolism parameters [age, sex, phosphorus, parathyroid hormone (PTH) and 25-hydroxy vitamin D3]; model 3, adjusted for metabolic factors [age, sex, albumin, low-density lipoprotein (LDL), high-density lipoprotein (HDL), triglycerides, nPCR and BMI]; model 4, adjusted for CV risk factors (gender, smoking, presence of diabetes, dyslipidaemia and history of cardiovascular disease, RRF); and model 5, adjusted for malnutrition/inflammation factors [BMI, nPCR, prealbumin and C-reactive protein (CRP)]. The assumption of proportional hazards was tested for each of the models using the cos.zph() function available in the ‘Survival’ package (https://www.rdocumentation.org/packages/survival/versions/3.2–13/topics/cox.zph). The Schoenfeld residuals were calculated for each regression model and analysed both visually using a plot and analytically using the Cox–Snell test. Multicollinearity was evaluated through the calculation of the variance inflation factor (VIF) for each predictor. The VIF estimates by how much the variance of a coefficient is ‘increased’ due to a linear relationship with some other predictors (i.e. collinearity). In R, the function vif() from the ‘car’ package was used to calculate the VIF for each model. Collinearity was considered for a VIF >2.5. P-values <.05 were considered statistically significant in all analyses.

RESULTS

Baseline characteristics

A total of 297 patients were recruited. Six patients were excluded and 291 patients were ultimately included (Supplementary data, Figure S1). Baseline characteristics of the cohort are summarized in Table 1. The mean follow-up period was 847 ± 495 days. The median serum 3-MH concentration was 28 µmol/L (range 11.9–58.1), 40.6% (n = 118) of patients were female and the median age was 67.1 years (range 17.4–90). Patients in the first tertile of 3-MH concentration were older and exhibited a higher CCI (P = .006), lower albumin level (P < .001), lower creatinine level (P < .001), higher N-terminal pro-brain natriuretic peptide (NT-pro-BNP) level (P = .004) and lower triglycerides level (P < .001). A total of 181 patients (62.2%) were considered with RRF. The 3-MH concentration was not significantly different between patients with and without RRF [median 3-MH concentration 28.2 µmol/L (range 11.9–58.1) versus 27.95 µmol/L (range 14–48.8); P = .144] (Supplementary data, Figure S2). The 3-MH concentration was significantly lower in patients who presented CV events during the follow-up period [n = 101; median 3-MH concentration 25.6 µmol/L (range 11.9–51.1)] than in those who did not present any CV event [n = 190; median 3-MH concentration 29.1 µmol/L (range 13.9–58.1); P < 0.001] (Supplementary data, Figure S3).

Table 1.
Open in new tab

Baseline characteristics of haemodialysis patients according to 3-MH tertiles

CharacteristicsOverallTertile 1Tertile 2Tertile 3P-valuea
Patients, n291959997
Clinical
Age (years)67.1 (17.4–90)73.6 (26.4–90)b,c67.1 (17.4–86.9)c,d60.4 (17.4–87.8)b,d<.001
Female (%)40.655.8c42.4c23.7b,d<.001
BMI (kg/m2)25.6 (15.5–49.3)24.9 (15.5–48.4)26.2 (18.6–49.3)25.7 (17.2–40.8).181
SCI (mg/kg/day)19.14 (14.21–27.82)17.27 (14.21–23.11)b,c19.47 (14.37–25.47)c,d20.75 (15.28–27.82)b,d<.001
Relative lean mass (%)46.2 (9.3–86.8)41.1 (9.3–74.8)c46.1 (20.9–86.8)52.9 (21.8–84.4)d.016
Lean tissue mass (kg)32.7 (10.6–74.7)29.1 (10.6–47.4)c34.2 (14.5–62.1)c39.3 (18.1–74.7)b,d<.001
Lean tissue index (kg/m²)12.2 (4.5–23.1)10.6 (4.5–17.7)c11.9 (7.1–20.5)c13.1 (7.8–23.1)b,d<.001
Dialysis vintage (years)1.71 (0.2–21.7)2.7 (0.2–21.7)c3.4 (0.2–20.3)c1.0(0.2–15.0)b,d<.001
Predialysis systolic blood pressure (mmHg)131 (77–192)128 (91–190)133 (83–186)132 (7–192).874
Predialysis diastolic blood pressure (mmHg)68 (23–125)67 (27–96)70 (36–125)71 (23–106).089
nPCR (g/kg/day)1.12 (0.47–2.33)1.08 (0.47–2.33)b,c1.13 (0.55–1.97)d1.15 (0.52–1.76)d.040
Dialysis dose (Kt/V)1.72 (0.44–3.29)1.89 (0.76–3.29)b,c1.72 (0.57–2.63)c,d1.52 (0.44–2.42)b,d<.001
Residual renal function (%)62.26057.669.1.218
Diabetes (%)37.845.334.334.188
AVF/AVG (%)89.787.489.991.8.775
Smoking (%)43.329.5b,c47.5d52.6d.002
CV history (%)46.747.446.546.4.989
HTA (%)95.590.59996.9.012
Dyslipidaemia (%)60.862.154.666.248
Cancer history (%)16.81916.215.5.7936
CCI6 (2–15)7 (2–15)b,c6 (2–11)d6 (2–14)d.006
Causes of nephropathy, %
Diabetes15.514.713.118.6.559
Vascular18.921.116.219.6.669
Diabetes + vascular17.824.216.213.4.365
Glomerulopathy14.412.612.118.6.873
ADPKD8.67.49.19.3.796
Interstitial nephritis6.57.47.15.2.668
Urology9.37.4b11.1c,d9.3b.028
Unknown8.95.315.26.2.128
Biological analysis
3-MH (µmol/L)28.0 (11.9–58.1)21.7 (11.9–24.2)28.0 (24.3–32.2)37.4 (32.3–58.1)
Log-transformed 3-MH1.447 (1.076–1.764)1.336 (1.076–1.384)1.447 (1.386–1.508)1.573 (1.509–1.764)
Creatinine (µmol/L)704 (213–1399)565 (213–963)b,c755 (322–1264)c,d820 (376–1399)b,d<.001
Urea (mmol/L)20 (7.9–41.1)16.5 (7.9–34.3)b,c20.2 (9.8–31.2)c,d22.5 (8.5–41.1)b,d<.001
Bicarbonate (mmol/L)21 (14–30)22 (18–27.8)b,c21.0 (14–30)c,d20 (15–30)b,d<.001
Calcium (mmol/L)2.20 (1.48–2.76)2.20 (1.92–2.76)2.20 (1.48–2.58)2.21 (1.60–2.54).804
Phosphorus (mmol/L)1.55(0.37–3.35)1.35 (0.37–2.76)c1.48 (0.45–3.08)c1.75 (0.56–3.35)b,d<.001
PTH (ng/L)258 (2–1949)209 (15–1659)b,c299 (2–1949)d357 (2–1913)d.039
25-hyfroxyvitamin D3 (µg/L)33.3 (7.6–82.9)32.4 (10.4–77.7)33.2 (7.6–82.9)36.1 (9.2–67.7).053
Serum albumin (g/L)38.1 (25.5–48.3)37.1 (25.5–45.4)b,c38.5 (28.8–45.8)c,d39.5 (27.8–48.3)b,d<.001
Serum prealbumin (g/L)0.31 (0.04–0.65)0.28 (0.07–0.53)b,c0.30 (0.13–0.65)c,d0.35 (0.04–0.53)b,d<.001
Glycaemia (mmol/L)6.00 (2.94–30.67)6.39 (3.06–30.67)6.06 (3.33–17.39)5.78 (2.94–27.78).151
Haemoglobin A1c (%)5.7 (4.2–11.9)5.8 (4.3–11.8)5.6 (4.2–10.7)5.7 (4.2–11.9).524
Total cholesterol (g/L)1.62 (0.39–3.67)1.62 (0.69–3.01)1.54 (0.39–3.21)1.69 (0.74–3.67).411
HDL cholesterol (g/L)0.39 (0.15–1.09)0.44 (0.15–1.09)b,c0.37 (0.16–0.91)d0.36 (0.17–0.95)d.007
LDL cholesterol (g/L)0.90 (0.20–3.78)0.89 (0.22–3.29)0.86 (0.23–3.25)0.94 (0.20–3.78).329
Triglycerides (g/L)1.37 (0.32–10.80)1.16 (0.47–6.37)b,c1.52 (0.32–10.80)d1.47 (0.32–6.97)d.003
Haemoglobin (g/L)109 (71–141)109 (83–133)c109 (72–138)c112 (71–141)b,d.041
Platelet count (G/L)209 (73–495)213 (88–475)201 (73–422)220 (101–495).088
WBC count (G/L)5.8 (1.1–19.2)5.8 (2.9–15.5)5.3 (2.4–10.7)6.2 (1.1–19.2).050
CRP (mg/L)4.2 (0.2–94.4)5.1 (0.3–94.4)4.0 (0.2–88.3)3.6 (0.2–65.9).198
Ferritin (ng/mL)289 (7–2752)320 (7–1770)321 (15–2752)207 (12–1354).137
NT-proBNP (ng/L)2602 (122–70 000)3790 (167–44 166)b,c2431 (263–70 000)d2309 (122–58 546)d.004
TSHus (mUI/L)1.29 (0.01–88.40)1.24 (0.03–88.40)1.30 (0.19–6.81)1.33 (0.01–5.14).798
Treatment (%)
Anti-diabetic therapy40.643.237.441.2.704
Statin therapy53.652.654.653.6.965
RAS blockers4337.944.446.4.461
Beta-blockers52.946.353.558.8.222
AAP55.354.758.652.6.692
Corticotherapy10.37.412.111.3.509
PPIs51.663.2b,c45.5d46.4d.022
Thyroid hormone replacement therapy1116.88.18.3.085
Anti-thyroid medication0.341.100N/A
CharacteristicsOverallTertile 1Tertile 2Tertile 3P-valuea
Patients, n291959997
Clinical
Age (years)67.1 (17.4–90)73.6 (26.4–90)b,c67.1 (17.4–86.9)c,d60.4 (17.4–87.8)b,d<.001
Female (%)40.655.8c42.4c23.7b,d<.001
BMI (kg/m2)25.6 (15.5–49.3)24.9 (15.5–48.4)26.2 (18.6–49.3)25.7 (17.2–40.8).181
SCI (mg/kg/day)19.14 (14.21–27.82)17.27 (14.21–23.11)b,c19.47 (14.37–25.47)c,d20.75 (15.28–27.82)b,d<.001
Relative lean mass (%)46.2 (9.3–86.8)41.1 (9.3–74.8)c46.1 (20.9–86.8)52.9 (21.8–84.4)d.016
Lean tissue mass (kg)32.7 (10.6–74.7)29.1 (10.6–47.4)c34.2 (14.5–62.1)c39.3 (18.1–74.7)b,d<.001
Lean tissue index (kg/m²)12.2 (4.5–23.1)10.6 (4.5–17.7)c11.9 (7.1–20.5)c13.1 (7.8–23.1)b,d<.001
Dialysis vintage (years)1.71 (0.2–21.7)2.7 (0.2–21.7)c3.4 (0.2–20.3)c1.0(0.2–15.0)b,d<.001
Predialysis systolic blood pressure (mmHg)131 (77–192)128 (91–190)133 (83–186)132 (7–192).874
Predialysis diastolic blood pressure (mmHg)68 (23–125)67 (27–96)70 (36–125)71 (23–106).089
nPCR (g/kg/day)1.12 (0.47–2.33)1.08 (0.47–2.33)b,c1.13 (0.55–1.97)d1.15 (0.52–1.76)d.040
Dialysis dose (Kt/V)1.72 (0.44–3.29)1.89 (0.76–3.29)b,c1.72 (0.57–2.63)c,d1.52 (0.44–2.42)b,d<.001
Residual renal function (%)62.26057.669.1.218
Diabetes (%)37.845.334.334.188
AVF/AVG (%)89.787.489.991.8.775
Smoking (%)43.329.5b,c47.5d52.6d.002
CV history (%)46.747.446.546.4.989
HTA (%)95.590.59996.9.012
Dyslipidaemia (%)60.862.154.666.248
Cancer history (%)16.81916.215.5.7936
CCI6 (2–15)7 (2–15)b,c6 (2–11)d6 (2–14)d.006
Causes of nephropathy, %
Diabetes15.514.713.118.6.559
Vascular18.921.116.219.6.669
Diabetes + vascular17.824.216.213.4.365
Glomerulopathy14.412.612.118.6.873
ADPKD8.67.49.19.3.796
Interstitial nephritis6.57.47.15.2.668
Urology9.37.4b11.1c,d9.3b.028
Unknown8.95.315.26.2.128
Biological analysis
3-MH (µmol/L)28.0 (11.9–58.1)21.7 (11.9–24.2)28.0 (24.3–32.2)37.4 (32.3–58.1)
Log-transformed 3-MH1.447 (1.076–1.764)1.336 (1.076–1.384)1.447 (1.386–1.508)1.573 (1.509–1.764)
Creatinine (µmol/L)704 (213–1399)565 (213–963)b,c755 (322–1264)c,d820 (376–1399)b,d<.001
Urea (mmol/L)20 (7.9–41.1)16.5 (7.9–34.3)b,c20.2 (9.8–31.2)c,d22.5 (8.5–41.1)b,d<.001
Bicarbonate (mmol/L)21 (14–30)22 (18–27.8)b,c21.0 (14–30)c,d20 (15–30)b,d<.001
Calcium (mmol/L)2.20 (1.48–2.76)2.20 (1.92–2.76)2.20 (1.48–2.58)2.21 (1.60–2.54).804
Phosphorus (mmol/L)1.55(0.37–3.35)1.35 (0.37–2.76)c1.48 (0.45–3.08)c1.75 (0.56–3.35)b,d<.001
PTH (ng/L)258 (2–1949)209 (15–1659)b,c299 (2–1949)d357 (2–1913)d.039
25-hyfroxyvitamin D3 (µg/L)33.3 (7.6–82.9)32.4 (10.4–77.7)33.2 (7.6–82.9)36.1 (9.2–67.7).053
Serum albumin (g/L)38.1 (25.5–48.3)37.1 (25.5–45.4)b,c38.5 (28.8–45.8)c,d39.5 (27.8–48.3)b,d<.001
Serum prealbumin (g/L)0.31 (0.04–0.65)0.28 (0.07–0.53)b,c0.30 (0.13–0.65)c,d0.35 (0.04–0.53)b,d<.001
Glycaemia (mmol/L)6.00 (2.94–30.67)6.39 (3.06–30.67)6.06 (3.33–17.39)5.78 (2.94–27.78).151
Haemoglobin A1c (%)5.7 (4.2–11.9)5.8 (4.3–11.8)5.6 (4.2–10.7)5.7 (4.2–11.9).524
Total cholesterol (g/L)1.62 (0.39–3.67)1.62 (0.69–3.01)1.54 (0.39–3.21)1.69 (0.74–3.67).411
HDL cholesterol (g/L)0.39 (0.15–1.09)0.44 (0.15–1.09)b,c0.37 (0.16–0.91)d0.36 (0.17–0.95)d.007
LDL cholesterol (g/L)0.90 (0.20–3.78)0.89 (0.22–3.29)0.86 (0.23–3.25)0.94 (0.20–3.78).329
Triglycerides (g/L)1.37 (0.32–10.80)1.16 (0.47–6.37)b,c1.52 (0.32–10.80)d1.47 (0.32–6.97)d.003
Haemoglobin (g/L)109 (71–141)109 (83–133)c109 (72–138)c112 (71–141)b,d.041
Platelet count (G/L)209 (73–495)213 (88–475)201 (73–422)220 (101–495).088
WBC count (G/L)5.8 (1.1–19.2)5.8 (2.9–15.5)5.3 (2.4–10.7)6.2 (1.1–19.2).050
CRP (mg/L)4.2 (0.2–94.4)5.1 (0.3–94.4)4.0 (0.2–88.3)3.6 (0.2–65.9).198
Ferritin (ng/mL)289 (7–2752)320 (7–1770)321 (15–2752)207 (12–1354).137
NT-proBNP (ng/L)2602 (122–70 000)3790 (167–44 166)b,c2431 (263–70 000)d2309 (122–58 546)d.004
TSHus (mUI/L)1.29 (0.01–88.40)1.24 (0.03–88.40)1.30 (0.19–6.81)1.33 (0.01–5.14).798
Treatment (%)
Anti-diabetic therapy40.643.237.441.2.704
Statin therapy53.652.654.653.6.965
RAS blockers4337.944.446.4.461
Beta-blockers52.946.353.558.8.222
AAP55.354.758.652.6.692
Corticotherapy10.37.412.111.3.509
PPIs51.663.2b,c45.5d46.4d.022
Thyroid hormone replacement therapy1116.88.18.3.085
Anti-thyroid medication0.341.100N/A

All data appear as non-normally distributed and are expressed as median (minimum–maximum) unless stated otherwise.

All analyses were performed using Kruskal–Wallis, Mann–Whitney U test or chi-squared test as appropriate.

aValues in bold are statistically significant.

bSignificantly different from the 2nd tertile.

cSignificantly different from the 3rd tertile.

dSignificantly different from the 1st tertile.

AAP, antiagregative treatment; AVF, arteriovenous fistula; AVG, arteriovenous graft; ADPKD, autosomal dominant polycystic kidney disease; PPIs, proton pump inhibitors; TSHus, ultrasensitive thyroid stimulating hormone; WBC, white blood cell; N/A, not applicable.

Table 1.
Open in new tab

Baseline characteristics of haemodialysis patients according to 3-MH tertiles

CharacteristicsOverallTertile 1Tertile 2Tertile 3P-valuea
Patients, n291959997
Clinical
Age (years)67.1 (17.4–90)73.6 (26.4–90)b,c67.1 (17.4–86.9)c,d60.4 (17.4–87.8)b,d<.001
Female (%)40.655.8c42.4c23.7b,d<.001
BMI (kg/m2)25.6 (15.5–49.3)24.9 (15.5–48.4)26.2 (18.6–49.3)25.7 (17.2–40.8).181
SCI (mg/kg/day)19.14 (14.21–27.82)17.27 (14.21–23.11)b,c19.47 (14.37–25.47)c,d20.75 (15.28–27.82)b,d<.001
Relative lean mass (%)46.2 (9.3–86.8)41.1 (9.3–74.8)c46.1 (20.9–86.8)52.9 (21.8–84.4)d.016
Lean tissue mass (kg)32.7 (10.6–74.7)29.1 (10.6–47.4)c34.2 (14.5–62.1)c39.3 (18.1–74.7)b,d<.001
Lean tissue index (kg/m²)12.2 (4.5–23.1)10.6 (4.5–17.7)c11.9 (7.1–20.5)c13.1 (7.8–23.1)b,d<.001
Dialysis vintage (years)1.71 (0.2–21.7)2.7 (0.2–21.7)c3.4 (0.2–20.3)c1.0(0.2–15.0)b,d<.001
Predialysis systolic blood pressure (mmHg)131 (77–192)128 (91–190)133 (83–186)132 (7–192).874
Predialysis diastolic blood pressure (mmHg)68 (23–125)67 (27–96)70 (36–125)71 (23–106).089
nPCR (g/kg/day)1.12 (0.47–2.33)1.08 (0.47–2.33)b,c1.13 (0.55–1.97)d1.15 (0.52–1.76)d.040
Dialysis dose (Kt/V)1.72 (0.44–3.29)1.89 (0.76–3.29)b,c1.72 (0.57–2.63)c,d1.52 (0.44–2.42)b,d<.001
Residual renal function (%)62.26057.669.1.218
Diabetes (%)37.845.334.334.188
AVF/AVG (%)89.787.489.991.8.775
Smoking (%)43.329.5b,c47.5d52.6d.002
CV history (%)46.747.446.546.4.989
HTA (%)95.590.59996.9.012
Dyslipidaemia (%)60.862.154.666.248
Cancer history (%)16.81916.215.5.7936
CCI6 (2–15)7 (2–15)b,c6 (2–11)d6 (2–14)d.006
Causes of nephropathy, %
Diabetes15.514.713.118.6.559
Vascular18.921.116.219.6.669
Diabetes + vascular17.824.216.213.4.365
Glomerulopathy14.412.612.118.6.873
ADPKD8.67.49.19.3.796
Interstitial nephritis6.57.47.15.2.668
Urology9.37.4b11.1c,d9.3b.028
Unknown8.95.315.26.2.128
Biological analysis
3-MH (µmol/L)28.0 (11.9–58.1)21.7 (11.9–24.2)28.0 (24.3–32.2)37.4 (32.3–58.1)
Log-transformed 3-MH1.447 (1.076–1.764)1.336 (1.076–1.384)1.447 (1.386–1.508)1.573 (1.509–1.764)
Creatinine (µmol/L)704 (213–1399)565 (213–963)b,c755 (322–1264)c,d820 (376–1399)b,d<.001
Urea (mmol/L)20 (7.9–41.1)16.5 (7.9–34.3)b,c20.2 (9.8–31.2)c,d22.5 (8.5–41.1)b,d<.001
Bicarbonate (mmol/L)21 (14–30)22 (18–27.8)b,c21.0 (14–30)c,d20 (15–30)b,d<.001
Calcium (mmol/L)2.20 (1.48–2.76)2.20 (1.92–2.76)2.20 (1.48–2.58)2.21 (1.60–2.54).804
Phosphorus (mmol/L)1.55(0.37–3.35)1.35 (0.37–2.76)c1.48 (0.45–3.08)c1.75 (0.56–3.35)b,d<.001
PTH (ng/L)258 (2–1949)209 (15–1659)b,c299 (2–1949)d357 (2–1913)d.039
25-hyfroxyvitamin D3 (µg/L)33.3 (7.6–82.9)32.4 (10.4–77.7)33.2 (7.6–82.9)36.1 (9.2–67.7).053
Serum albumin (g/L)38.1 (25.5–48.3)37.1 (25.5–45.4)b,c38.5 (28.8–45.8)c,d39.5 (27.8–48.3)b,d<.001
Serum prealbumin (g/L)0.31 (0.04–0.65)0.28 (0.07–0.53)b,c0.30 (0.13–0.65)c,d0.35 (0.04–0.53)b,d<.001
Glycaemia (mmol/L)6.00 (2.94–30.67)6.39 (3.06–30.67)6.06 (3.33–17.39)5.78 (2.94–27.78).151
Haemoglobin A1c (%)5.7 (4.2–11.9)5.8 (4.3–11.8)5.6 (4.2–10.7)5.7 (4.2–11.9).524
Total cholesterol (g/L)1.62 (0.39–3.67)1.62 (0.69–3.01)1.54 (0.39–3.21)1.69 (0.74–3.67).411
HDL cholesterol (g/L)0.39 (0.15–1.09)0.44 (0.15–1.09)b,c0.37 (0.16–0.91)d0.36 (0.17–0.95)d.007
LDL cholesterol (g/L)0.90 (0.20–3.78)0.89 (0.22–3.29)0.86 (0.23–3.25)0.94 (0.20–3.78).329
Triglycerides (g/L)1.37 (0.32–10.80)1.16 (0.47–6.37)b,c1.52 (0.32–10.80)d1.47 (0.32–6.97)d.003
Haemoglobin (g/L)109 (71–141)109 (83–133)c109 (72–138)c112 (71–141)b,d.041
Platelet count (G/L)209 (73–495)213 (88–475)201 (73–422)220 (101–495).088
WBC count (G/L)5.8 (1.1–19.2)5.8 (2.9–15.5)5.3 (2.4–10.7)6.2 (1.1–19.2).050
CRP (mg/L)4.2 (0.2–94.4)5.1 (0.3–94.4)4.0 (0.2–88.3)3.6 (0.2–65.9).198
Ferritin (ng/mL)289 (7–2752)320 (7–1770)321 (15–2752)207 (12–1354).137
NT-proBNP (ng/L)2602 (122–70 000)3790 (167–44 166)b,c2431 (263–70 000)d2309 (122–58 546)d.004
TSHus (mUI/L)1.29 (0.01–88.40)1.24 (0.03–88.40)1.30 (0.19–6.81)1.33 (0.01–5.14).798
Treatment (%)
Anti-diabetic therapy40.643.237.441.2.704
Statin therapy53.652.654.653.6.965
RAS blockers4337.944.446.4.461
Beta-blockers52.946.353.558.8.222
AAP55.354.758.652.6.692
Corticotherapy10.37.412.111.3.509
PPIs51.663.2b,c45.5d46.4d.022
Thyroid hormone replacement therapy1116.88.18.3.085
Anti-thyroid medication0.341.100N/A
CharacteristicsOverallTertile 1Tertile 2Tertile 3P-valuea
Patients, n291959997
Clinical
Age (years)67.1 (17.4–90)73.6 (26.4–90)b,c67.1 (17.4–86.9)c,d60.4 (17.4–87.8)b,d<.001
Female (%)40.655.8c42.4c23.7b,d<.001
BMI (kg/m2)25.6 (15.5–49.3)24.9 (15.5–48.4)26.2 (18.6–49.3)25.7 (17.2–40.8).181
SCI (mg/kg/day)19.14 (14.21–27.82)17.27 (14.21–23.11)b,c19.47 (14.37–25.47)c,d20.75 (15.28–27.82)b,d<.001
Relative lean mass (%)46.2 (9.3–86.8)41.1 (9.3–74.8)c46.1 (20.9–86.8)52.9 (21.8–84.4)d.016
Lean tissue mass (kg)32.7 (10.6–74.7)29.1 (10.6–47.4)c34.2 (14.5–62.1)c39.3 (18.1–74.7)b,d<.001
Lean tissue index (kg/m²)12.2 (4.5–23.1)10.6 (4.5–17.7)c11.9 (7.1–20.5)c13.1 (7.8–23.1)b,d<.001
Dialysis vintage (years)1.71 (0.2–21.7)2.7 (0.2–21.7)c3.4 (0.2–20.3)c1.0(0.2–15.0)b,d<.001
Predialysis systolic blood pressure (mmHg)131 (77–192)128 (91–190)133 (83–186)132 (7–192).874
Predialysis diastolic blood pressure (mmHg)68 (23–125)67 (27–96)70 (36–125)71 (23–106).089
nPCR (g/kg/day)1.12 (0.47–2.33)1.08 (0.47–2.33)b,c1.13 (0.55–1.97)d1.15 (0.52–1.76)d.040
Dialysis dose (Kt/V)1.72 (0.44–3.29)1.89 (0.76–3.29)b,c1.72 (0.57–2.63)c,d1.52 (0.44–2.42)b,d<.001
Residual renal function (%)62.26057.669.1.218
Diabetes (%)37.845.334.334.188
AVF/AVG (%)89.787.489.991.8.775
Smoking (%)43.329.5b,c47.5d52.6d.002
CV history (%)46.747.446.546.4.989
HTA (%)95.590.59996.9.012
Dyslipidaemia (%)60.862.154.666.248
Cancer history (%)16.81916.215.5.7936
CCI6 (2–15)7 (2–15)b,c6 (2–11)d6 (2–14)d.006
Causes of nephropathy, %
Diabetes15.514.713.118.6.559
Vascular18.921.116.219.6.669
Diabetes + vascular17.824.216.213.4.365
Glomerulopathy14.412.612.118.6.873
ADPKD8.67.49.19.3.796
Interstitial nephritis6.57.47.15.2.668
Urology9.37.4b11.1c,d9.3b.028
Unknown8.95.315.26.2.128
Biological analysis
3-MH (µmol/L)28.0 (11.9–58.1)21.7 (11.9–24.2)28.0 (24.3–32.2)37.4 (32.3–58.1)
Log-transformed 3-MH1.447 (1.076–1.764)1.336 (1.076–1.384)1.447 (1.386–1.508)1.573 (1.509–1.764)
Creatinine (µmol/L)704 (213–1399)565 (213–963)b,c755 (322–1264)c,d820 (376–1399)b,d<.001
Urea (mmol/L)20 (7.9–41.1)16.5 (7.9–34.3)b,c20.2 (9.8–31.2)c,d22.5 (8.5–41.1)b,d<.001
Bicarbonate (mmol/L)21 (14–30)22 (18–27.8)b,c21.0 (14–30)c,d20 (15–30)b,d<.001
Calcium (mmol/L)2.20 (1.48–2.76)2.20 (1.92–2.76)2.20 (1.48–2.58)2.21 (1.60–2.54).804
Phosphorus (mmol/L)1.55(0.37–3.35)1.35 (0.37–2.76)c1.48 (0.45–3.08)c1.75 (0.56–3.35)b,d<.001
PTH (ng/L)258 (2–1949)209 (15–1659)b,c299 (2–1949)d357 (2–1913)d.039
25-hyfroxyvitamin D3 (µg/L)33.3 (7.6–82.9)32.4 (10.4–77.7)33.2 (7.6–82.9)36.1 (9.2–67.7).053
Serum albumin (g/L)38.1 (25.5–48.3)37.1 (25.5–45.4)b,c38.5 (28.8–45.8)c,d39.5 (27.8–48.3)b,d<.001
Serum prealbumin (g/L)0.31 (0.04–0.65)0.28 (0.07–0.53)b,c0.30 (0.13–0.65)c,d0.35 (0.04–0.53)b,d<.001
Glycaemia (mmol/L)6.00 (2.94–30.67)6.39 (3.06–30.67)6.06 (3.33–17.39)5.78 (2.94–27.78).151
Haemoglobin A1c (%)5.7 (4.2–11.9)5.8 (4.3–11.8)5.6 (4.2–10.7)5.7 (4.2–11.9).524
Total cholesterol (g/L)1.62 (0.39–3.67)1.62 (0.69–3.01)1.54 (0.39–3.21)1.69 (0.74–3.67).411
HDL cholesterol (g/L)0.39 (0.15–1.09)0.44 (0.15–1.09)b,c0.37 (0.16–0.91)d0.36 (0.17–0.95)d.007
LDL cholesterol (g/L)0.90 (0.20–3.78)0.89 (0.22–3.29)0.86 (0.23–3.25)0.94 (0.20–3.78).329
Triglycerides (g/L)1.37 (0.32–10.80)1.16 (0.47–6.37)b,c1.52 (0.32–10.80)d1.47 (0.32–6.97)d.003
Haemoglobin (g/L)109 (71–141)109 (83–133)c109 (72–138)c112 (71–141)b,d.041
Platelet count (G/L)209 (73–495)213 (88–475)201 (73–422)220 (101–495).088
WBC count (G/L)5.8 (1.1–19.2)5.8 (2.9–15.5)5.3 (2.4–10.7)6.2 (1.1–19.2).050
CRP (mg/L)4.2 (0.2–94.4)5.1 (0.3–94.4)4.0 (0.2–88.3)3.6 (0.2–65.9).198
Ferritin (ng/mL)289 (7–2752)320 (7–1770)321 (15–2752)207 (12–1354).137
NT-proBNP (ng/L)2602 (122–70 000)3790 (167–44 166)b,c2431 (263–70 000)d2309 (122–58 546)d.004
TSHus (mUI/L)1.29 (0.01–88.40)1.24 (0.03–88.40)1.30 (0.19–6.81)1.33 (0.01–5.14).798
Treatment (%)
Anti-diabetic therapy40.643.237.441.2.704
Statin therapy53.652.654.653.6.965
RAS blockers4337.944.446.4.461
Beta-blockers52.946.353.558.8.222
AAP55.354.758.652.6.692
Corticotherapy10.37.412.111.3.509
PPIs51.663.2b,c45.5d46.4d.022
Thyroid hormone replacement therapy1116.88.18.3.085
Anti-thyroid medication0.341.100N/A

All data appear as non-normally distributed and are expressed as median (minimum–maximum) unless stated otherwise.

All analyses were performed using Kruskal–Wallis, Mann–Whitney U test or chi-squared test as appropriate.

aValues in bold are statistically significant.

bSignificantly different from the 2nd tertile.

cSignificantly different from the 3rd tertile.

dSignificantly different from the 1st tertile.

AAP, antiagregative treatment; AVF, arteriovenous fistula; AVG, arteriovenous graft; ADPKD, autosomal dominant polycystic kidney disease; PPIs, proton pump inhibitors; TSHus, ultrasensitive thyroid stimulating hormone; WBC, white blood cell; N/A, not applicable.

Correlation of serum 3-MH levels with nutritional parameters and LTM

The Pearson correlation coefficient as well as different clinical, biochemical and nutritional parameters are presented in Table 2. 3-MH levels were positively correlated with nutritional parameters such as albumin, pre-albumin and nPCR (P < .001, P < .001 and P = .004, respectively) but not with BMI (P = .987). Protein metabolism markers were also significantly positively correlated with 3-MH levels (phosphorus, bicarbonate and urea; P < .001). The 3-MH concentration was negatively correlated with CRP but not with ferritin level. 3-MH levels were negatively correlated with Kt/V (P < .001). We further observed a significant positive correlation of 3-MH levels with LTM markers such as serum creatinine and SCI (P < .001). Measured using BIS, serum 3-MH concentration was positively associated with LTI and LTM in univariate analysis (both P < .001) (Table 3). We noted a statistically negative correlation between 3-MH levels and the FTI (P = .033) but not with adipose tissue mass. 3-MH levels were also positively correlated with TBW content. In good agreement, men with higher lean mass have a higher level of 3-MH [n = 173; median 3-MH concentration 30.3 µmol/L (range 13.9–55.5)] compared with women [n = 118; median 3-MH concentration 25.2 µmol/L (range 11.9–58.1)] (P < .001).

Table 2.
Open in new tab

Unadjusted Pearson correlation coefficients (r) of 3-MH and other relevant covariates in HD patients

Variabler95% CIP-valuea
Age (years)−0.2807−0.3833 to −0.1713<.001
CCI−0.1885−0.2971 to −0.0751.001
Dialysis vintage (years)−0.1985−0.3065 to −0.0855<.001
Dialysis dose (Kt/V)−0.3614−0.4576 to −0.2570<.001
SCI (mg/kg/day)0.55370.4685–0.6287<.001
Systolic blood pressure (mmHg)−0.0077−0.1242–0.1090.897
Diastolic blood pressure (mmHg)0.14160.0250–0.2544.018
BMI (kg/m²)0.0009−0.1141–0.1159.987
Albumin (g/L)0.31220.2046–0.4124<.001
Prealbumin (g/L)0.31510.2076–0.4150<.001
nPCR (Garred)0.16950.0554–0.2792.004
Glucose (g/L)−0.0948−0.2077–0.0206.107
Hb A1c (%)0.0004−0.1227–0.1235.995
Cholesterol total (g/L)0.0607−0.0551–0.1749.304
LDL (g/L)0.0189−0.0967–0.1340.749
HDL (g/L)−0.1192−0.2314 to −0.0039.043
Triyglycerides (g/L)0.0963−0.0193–0.2094.102
Creatinine (µmol/L)0.54630.4603–0.6222<.001
Urea (mmol/L)0.46900.3742–0.5541<.001
Bicarbonates (mmol/L)−0.2859−0.3881 to −0.1767<.001
Calcium (mmol/L)−0.0483−0.1624–0.0671.412
Phosphorus (mmol/L)0.32730.2207–0.4263<.001
PTH (ng/L)0.12060.0055–0.2325.040
25-hydroxyvitamin D3 (µg/L)0.1098−0.0057–0.2224.062
CRP (mg/L)−0.1573−0.2680 to −0.0425.008
Ferritin (ng/mL)−0.0655−0.1794–0.050.266
NT-proBNP (ng/L)−0.1133−0.2256–0.0019.054
TSHus (mUI/L)−0.1029−0.2390–0.0370.149
Variabler95% CIP-valuea
Age (years)−0.2807−0.3833 to −0.1713<.001
CCI−0.1885−0.2971 to −0.0751.001
Dialysis vintage (years)−0.1985−0.3065 to −0.0855<.001
Dialysis dose (Kt/V)−0.3614−0.4576 to −0.2570<.001
SCI (mg/kg/day)0.55370.4685–0.6287<.001
Systolic blood pressure (mmHg)−0.0077−0.1242–0.1090.897
Diastolic blood pressure (mmHg)0.14160.0250–0.2544.018
BMI (kg/m²)0.0009−0.1141–0.1159.987
Albumin (g/L)0.31220.2046–0.4124<.001
Prealbumin (g/L)0.31510.2076–0.4150<.001
nPCR (Garred)0.16950.0554–0.2792.004
Glucose (g/L)−0.0948−0.2077–0.0206.107
Hb A1c (%)0.0004−0.1227–0.1235.995
Cholesterol total (g/L)0.0607−0.0551–0.1749.304
LDL (g/L)0.0189−0.0967–0.1340.749
HDL (g/L)−0.1192−0.2314 to −0.0039.043
Triyglycerides (g/L)0.0963−0.0193–0.2094.102
Creatinine (µmol/L)0.54630.4603–0.6222<.001
Urea (mmol/L)0.46900.3742–0.5541<.001
Bicarbonates (mmol/L)−0.2859−0.3881 to −0.1767<.001
Calcium (mmol/L)−0.0483−0.1624–0.0671.412
Phosphorus (mmol/L)0.32730.2207–0.4263<.001
PTH (ng/L)0.12060.0055–0.2325.040
25-hydroxyvitamin D3 (µg/L)0.1098−0.0057–0.2224.062
CRP (mg/L)−0.1573−0.2680 to −0.0425.008
Ferritin (ng/mL)−0.0655−0.1794–0.050.266
NT-proBNP (ng/L)−0.1133−0.2256–0.0019.054
TSHus (mUI/L)−0.1029−0.2390–0.0370.149

aValues in bold are statistically significant.

TSHus, ultrasensitive thyroid stimulating hormone.

Table 2.
Open in new tab

Unadjusted Pearson correlation coefficients (r) of 3-MH and other relevant covariates in HD patients

Variabler95% CIP-valuea
Age (years)−0.2807−0.3833 to −0.1713<.001
CCI−0.1885−0.2971 to −0.0751.001
Dialysis vintage (years)−0.1985−0.3065 to −0.0855<.001
Dialysis dose (Kt/V)−0.3614−0.4576 to −0.2570<.001
SCI (mg/kg/day)0.55370.4685–0.6287<.001
Systolic blood pressure (mmHg)−0.0077−0.1242–0.1090.897
Diastolic blood pressure (mmHg)0.14160.0250–0.2544.018
BMI (kg/m²)0.0009−0.1141–0.1159.987
Albumin (g/L)0.31220.2046–0.4124<.001
Prealbumin (g/L)0.31510.2076–0.4150<.001
nPCR (Garred)0.16950.0554–0.2792.004
Glucose (g/L)−0.0948−0.2077–0.0206.107
Hb A1c (%)0.0004−0.1227–0.1235.995
Cholesterol total (g/L)0.0607−0.0551–0.1749.304
LDL (g/L)0.0189−0.0967–0.1340.749
HDL (g/L)−0.1192−0.2314 to −0.0039.043
Triyglycerides (g/L)0.0963−0.0193–0.2094.102
Creatinine (µmol/L)0.54630.4603–0.6222<.001
Urea (mmol/L)0.46900.3742–0.5541<.001
Bicarbonates (mmol/L)−0.2859−0.3881 to −0.1767<.001
Calcium (mmol/L)−0.0483−0.1624–0.0671.412
Phosphorus (mmol/L)0.32730.2207–0.4263<.001
PTH (ng/L)0.12060.0055–0.2325.040
25-hydroxyvitamin D3 (µg/L)0.1098−0.0057–0.2224.062
CRP (mg/L)−0.1573−0.2680 to −0.0425.008
Ferritin (ng/mL)−0.0655−0.1794–0.050.266
NT-proBNP (ng/L)−0.1133−0.2256–0.0019.054
TSHus (mUI/L)−0.1029−0.2390–0.0370.149
Variabler95% CIP-valuea
Age (years)−0.2807−0.3833 to −0.1713<.001
CCI−0.1885−0.2971 to −0.0751.001
Dialysis vintage (years)−0.1985−0.3065 to −0.0855<.001
Dialysis dose (Kt/V)−0.3614−0.4576 to −0.2570<.001
SCI (mg/kg/day)0.55370.4685–0.6287<.001
Systolic blood pressure (mmHg)−0.0077−0.1242–0.1090.897
Diastolic blood pressure (mmHg)0.14160.0250–0.2544.018
BMI (kg/m²)0.0009−0.1141–0.1159.987
Albumin (g/L)0.31220.2046–0.4124<.001
Prealbumin (g/L)0.31510.2076–0.4150<.001
nPCR (Garred)0.16950.0554–0.2792.004
Glucose (g/L)−0.0948−0.2077–0.0206.107
Hb A1c (%)0.0004−0.1227–0.1235.995
Cholesterol total (g/L)0.0607−0.0551–0.1749.304
LDL (g/L)0.0189−0.0967–0.1340.749
HDL (g/L)−0.1192−0.2314 to −0.0039.043
Triyglycerides (g/L)0.0963−0.0193–0.2094.102
Creatinine (µmol/L)0.54630.4603–0.6222<.001
Urea (mmol/L)0.46900.3742–0.5541<.001
Bicarbonates (mmol/L)−0.2859−0.3881 to −0.1767<.001
Calcium (mmol/L)−0.0483−0.1624–0.0671.412
Phosphorus (mmol/L)0.32730.2207–0.4263<.001
PTH (ng/L)0.12060.0055–0.2325.040
25-hydroxyvitamin D3 (µg/L)0.1098−0.0057–0.2224.062
CRP (mg/L)−0.1573−0.2680 to −0.0425.008
Ferritin (ng/mL)−0.0655−0.1794–0.050.266
NT-proBNP (ng/L)−0.1133−0.2256–0.0019.054
TSHus (mUI/L)−0.1029−0.2390–0.0370.149

aValues in bold are statistically significant.

TSHus, ultrasensitive thyroid stimulating hormone.

Table 3.
Open in new tab

Unadjusted Pearson correlation coefficients (r) of 3-MH and bioimpedance parameters (N = 161)

VariableMedianRangeR95% CIP-valuea
BMI (kg/m2)25.917.9–48.4−0.0488−0.2015–0.1062.538
LTM (kg)32.710.6–74.70.37140.2300–0.4975<.001
LTI (kg/m2)12.24.5–23.10.31980.1737–0.4521<.001
ATM (kg)36.35.8–101.8−0.0974−0.2483–0.0582.219
FTI (kg/m2)13.52.3–43.5−0.1687−0.3151 to −0.0144.033
Extracellular body water (L)168.7–25.60.145−0.009 435–0.2926.066
Intracellular body water (L)179–35.30.37070.2296–0.4965<.001
TBW (L)33.220–59.80.29410.1465–0.4288<.001
VariableMedianRangeR95% CIP-valuea
BMI (kg/m2)25.917.9–48.4−0.0488−0.2015–0.1062.538
LTM (kg)32.710.6–74.70.37140.2300–0.4975<.001
LTI (kg/m2)12.24.5–23.10.31980.1737–0.4521<.001
ATM (kg)36.35.8–101.8−0.0974−0.2483–0.0582.219
FTI (kg/m2)13.52.3–43.5−0.1687−0.3151 to −0.0144.033
Extracellular body water (L)168.7–25.60.145−0.009 435–0.2926.066
Intracellular body water (L)179–35.30.37070.2296–0.4965<.001
TBW (L)33.220–59.80.29410.1465–0.4288<.001

aValues in bold are statistically significant.

Table 3.
Open in new tab

Unadjusted Pearson correlation coefficients (r) of 3-MH and bioimpedance parameters (N = 161)

VariableMedianRangeR95% CIP-valuea
BMI (kg/m2)25.917.9–48.4−0.0488−0.2015–0.1062.538
LTM (kg)32.710.6–74.70.37140.2300–0.4975<.001
LTI (kg/m2)12.24.5–23.10.31980.1737–0.4521<.001
ATM (kg)36.35.8–101.8−0.0974−0.2483–0.0582.219
FTI (kg/m2)13.52.3–43.5−0.1687−0.3151 to −0.0144.033
Extracellular body water (L)168.7–25.60.145−0.009 435–0.2926.066
Intracellular body water (L)179–35.30.37070.2296–0.4965<.001
TBW (L)33.220–59.80.29410.1465–0.4288<.001
VariableMedianRangeR95% CIP-valuea
BMI (kg/m2)25.917.9–48.4−0.0488−0.2015–0.1062.538
LTM (kg)32.710.6–74.70.37140.2300–0.4975<.001
LTI (kg/m2)12.24.5–23.10.31980.1737–0.4521<.001
ATM (kg)36.35.8–101.8−0.0974−0.2483–0.0582.219
FTI (kg/m2)13.52.3–43.5−0.1687−0.3151 to −0.0144.033
Extracellular body water (L)168.7–25.60.145−0.009 435–0.2926.066
Intracellular body water (L)179–35.30.37070.2296–0.4965<.001
TBW (L)33.220–59.80.29410.1465–0.4288<.001

aValues in bold are statistically significant.

3-MH level is a predictor of MHD patient outcomes

The relationships between 3-MH concentration and all CV events, CV mortality and all-cause mortality were examined using time-to-event analyses. During the follow-up period there were a total of 91 deaths (Supplementary data, Table S1) with a majority of CV causes. We observed 67 first non-fatal CV events during the follow-up, the causes of which are described in Supplementary data, Table S2. The number of CV events was significantly higher in the first and second tertiles compared with the higher tertile (42 events in tertile 1, 38 in tertile 2 and 21 in tertile 3; log-rank test P < .001) (Figure 1A). Similarly, CV mortality was significantly different according to 3-MH concentration tertiles (log-rank test P = .015; Figure 1B). There were significantly more all-cause deaths in the first and the second tertile of 3-MH concentration compared with the higher tertile (38 deaths in tertile 1, 36 in tertile 2 and 17 in tertile 3; log-rank P = .002) (Figure 1C).

Kaplan–Meier analysis of (A) time to first CV event, (B) cumulative cardiovascular survival and (C) cumulative all-cause survival of all patients according to 3-MH tertiles. Tertile cut-offs (3-MH level): tertile 1, <24.3 µmol/L; tertile 2, 24.3–32.2 µmol/L; tertile 3, >32.2 µmol/L. The number of CV events was 42, 38 and 21 for tertile 1, tertile 2 and tertile 3, respectively. The number of events for CV mortality was 19, 23 and 8 for tertile 1, tertile 2 and tertile 3, respectively. The number of deaths was 38, 36 and 17 for tertile 1, tertile 2 and tertile 3, respectively.
FIGURE 1:

Kaplan–Meier analysis of (A) time to first CV event, (B) cumulative cardiovascular survival and (C) cumulative all-cause survival of all patients according to 3-MH tertiles. Tertile cut-offs (3-MH level): tertile 1, <24.3 µmol/L; tertile 2, 24.3–32.2 µmol/L; tertile 3, >32.2 µmol/L. The number of CV events was 42, 38 and 21 for tertile 1, tertile 2 and tertile 3, respectively. The number of events for CV mortality was 19, 23 and 8 for tertile 1, tertile 2 and tertile 3, respectively. The number of deaths was 38, 36 and 17 for tertile 1, tertile 2 and tertile 3, respectively.

Open in new tabDownload slide

According to 3-MH tertiles, in the subgroup of men, a higher tertile of 3-MH was also associated with better survival, as observed in the overall population (Supplementary data, Figure S4A). In the female sample, we observed a similar association without reaching statistical significance (Supplementary data, Figure S4B).

Similarly, in patients with or without RRF, the higher tertile of 3-MH was associated with better survival (Supplementary data, Figures S5A and S5B).

Analysed as a continuous variable, 3-MH was significantly associated with CV events {hazard ratio [HR] 0.9493 [95% confidence interval (CI) 0.9239–0.9753] P < .001}, CV mortality [HR 0.9564 (95% CI 0.9221–0.9920); P = .017] and all-cause mortality [HR 0.9526 (95% CI 0.9265–0.9794); P < .001] (Table 4). The association between 3-MH and CV events remained significant in multivariate Cox analysis after adjustment for demographic parameters [model 1, HR 0.9391 (95% CI 0.9119–0.9671); P < .001], phospho-calcic metabolism parameters [model 2, HR 0.9377 (95% CI 0.9090–0.9673); P < .001], metabolic factors [model 3, HR 0.9527 (95% CI 0.9225–0.9838); P = .003], CV risk factors [model 4, HR 0.9482 (95% CI 0.9197–0.9776); P < .001] and malnutrition/inflammation [model 5, HR 0.9450 (95% CI 0.9181–0.9728); P < .001]. The association between 3-MH, CV mortality and overall mortality remained significant in multivariate Cox analysis after adjustment for demographic and mineral metabolism parameters. However, this association did not persist after adjustment for metabolic and nutritional parameters. 3-MH showed a borderline significance for the trend to be associated with a higher risk of all-cause mortality after adjustment for CV risk factors (Table 4).

Table 4.
Open in new tab

Multivariable Cox regression analysis of 3-MH as a predictor of all-cause mortality, cardiovascular mortality and cardiovascular events

EventsHR (95% CI)P-valuea
Cardiovascular events
Unadjusted0.9493 (0.9239–0.9753)<.001
Model 1b0.9391 (0.9119–0.9671)<.001
Model 2c0.9377 (0.9090–0.9673)<.001
Model 3d0.9527 (0.9225–0.9838).003
Model 4e0.9482 (0.9197–0.9776)<.001
Model 5f0.9450 (0.9181–0.9728)<.001
Cardiovascular mortality
Unadjusted0.9564 (0.9221–0.9920).017
Model 1b0.9557 (0.9192–0.9937).023
Model 2c0.9563 (0.9187–0.9955).029
Model 3d0.9789 (0.9393–1.0200).314
Model 4e0.9709 (0.9318–1.0120).159
Model 5f0.9593 (0.9224–0.9977).038
Overall mortality
Unadjusted0.9526 (0.9265–0.9794)<.001
Model 1b0.9566 (0.9285–0.9856).004
Model 2c0.9601 (0.9309–0.9901).009
Model 3d0.9809 (0.9507–1.0121).217
Model 4e0.9715 (0.9418–1.0020).067
Model 5f0.9521 (0.9243–0.9808).001
EventsHR (95% CI)P-valuea
Cardiovascular events
Unadjusted0.9493 (0.9239–0.9753)<.001
Model 1b0.9391 (0.9119–0.9671)<.001
Model 2c0.9377 (0.9090–0.9673)<.001
Model 3d0.9527 (0.9225–0.9838).003
Model 4e0.9482 (0.9197–0.9776)<.001
Model 5f0.9450 (0.9181–0.9728)<.001
Cardiovascular mortality
Unadjusted0.9564 (0.9221–0.9920).017
Model 1b0.9557 (0.9192–0.9937).023
Model 2c0.9563 (0.9187–0.9955).029
Model 3d0.9789 (0.9393–1.0200).314
Model 4e0.9709 (0.9318–1.0120).159
Model 5f0.9593 (0.9224–0.9977).038
Overall mortality
Unadjusted0.9526 (0.9265–0.9794)<.001
Model 1b0.9566 (0.9285–0.9856).004
Model 2c0.9601 (0.9309–0.9901).009
Model 3d0.9809 (0.9507–1.0121).217
Model 4e0.9715 (0.9418–1.0020).067
Model 5f0.9521 (0.9243–0.9808).001

aValues in bold are statistically significant.

b

Model 1 was adjusted for demographic parameters (age, sex, Kt/V and dialysis vintage).

c

Model 2 was adjusted for phospho-calcic metabolism (age, sex, phosphorus, PTH and 25-hydroxyvitamin D3).

d

Model 3 was adjusted for metabolic factors (age, sex, albumin, LDL, HDL, triglycerides, nPCR and BMI).

e

Model 4 was adjusted for cardiovascular risk factors (age, sex, smoking, presence of diabetes, dyslipidaemia and history of cardiovascular disease and residual renal function).

f

Model 5 was adjusted for malnutrition/inflammation (BMI, nPCR, pre-albumin and CRP).

Table 4.
Open in new tab

Multivariable Cox regression analysis of 3-MH as a predictor of all-cause mortality, cardiovascular mortality and cardiovascular events

EventsHR (95% CI)P-valuea
Cardiovascular events
Unadjusted0.9493 (0.9239–0.9753)<.001
Model 1b0.9391 (0.9119–0.9671)<.001
Model 2c0.9377 (0.9090–0.9673)<.001
Model 3d0.9527 (0.9225–0.9838).003
Model 4e0.9482 (0.9197–0.9776)<.001
Model 5f0.9450 (0.9181–0.9728)<.001
Cardiovascular mortality
Unadjusted0.9564 (0.9221–0.9920).017
Model 1b0.9557 (0.9192–0.9937).023
Model 2c0.9563 (0.9187–0.9955).029
Model 3d0.9789 (0.9393–1.0200).314
Model 4e0.9709 (0.9318–1.0120).159
Model 5f0.9593 (0.9224–0.9977).038
Overall mortality
Unadjusted0.9526 (0.9265–0.9794)<.001
Model 1b0.9566 (0.9285–0.9856).004
Model 2c0.9601 (0.9309–0.9901).009
Model 3d0.9809 (0.9507–1.0121).217
Model 4e0.9715 (0.9418–1.0020).067
Model 5f0.9521 (0.9243–0.9808).001
EventsHR (95% CI)P-valuea
Cardiovascular events
Unadjusted0.9493 (0.9239–0.9753)<.001
Model 1b0.9391 (0.9119–0.9671)<.001
Model 2c0.9377 (0.9090–0.9673)<.001
Model 3d0.9527 (0.9225–0.9838).003
Model 4e0.9482 (0.9197–0.9776)<.001
Model 5f0.9450 (0.9181–0.9728)<.001
Cardiovascular mortality
Unadjusted0.9564 (0.9221–0.9920).017
Model 1b0.9557 (0.9192–0.9937).023
Model 2c0.9563 (0.9187–0.9955).029
Model 3d0.9789 (0.9393–1.0200).314
Model 4e0.9709 (0.9318–1.0120).159
Model 5f0.9593 (0.9224–0.9977).038
Overall mortality
Unadjusted0.9526 (0.9265–0.9794)<.001
Model 1b0.9566 (0.9285–0.9856).004
Model 2c0.9601 (0.9309–0.9901).009
Model 3d0.9809 (0.9507–1.0121).217
Model 4e0.9715 (0.9418–1.0020).067
Model 5f0.9521 (0.9243–0.9808).001

aValues in bold are statistically significant.

b

Model 1 was adjusted for demographic parameters (age, sex, Kt/V and dialysis vintage).

c

Model 2 was adjusted for phospho-calcic metabolism (age, sex, phosphorus, PTH and 25-hydroxyvitamin D3).

d

Model 3 was adjusted for metabolic factors (age, sex, albumin, LDL, HDL, triglycerides, nPCR and BMI).

e

Model 4 was adjusted for cardiovascular risk factors (age, sex, smoking, presence of diabetes, dyslipidaemia and history of cardiovascular disease and residual renal function).

f

Model 5 was adjusted for malnutrition/inflammation (BMI, nPCR, pre-albumin and CRP).

The areas under the receiver operating characteristics (AUROCs) curves for 3-MH concentration and other usual mortality risk factors as predictors of CV events and overall mortality are described in Supplementary data, Tables S3 and S4, respectively. ROC curves for CV events and overall mortality are presented in Supplementary data, Figures S6 and S7, respectively. 3-MH concentration had a significant diagnostic value to predict CV events [AUROC 0.6324 (95% CI 0.5655–0.6964); P < .001] and all-cause mortality [AUROC 0.6295 (95% CI 0.5620–0.6969); P < .001]. 3-MH concentration had a similar predictive value compared with other variables for CV events.

DISCUSSION

In this study we evaluated for the first time the performance of 3-MH to predict clinical outcomes in MHD patients. First, we observed that a low level of 3-MH predicts all-cause mortality and especially CV events in the chronic HD population. The association between low 3-MH concentration and CV events remained statistically significant after adjusting for demographic, nutritional, metabolic and CV risk factors. When we considered model discrimination using ROC curves, 3-MH was found to be an interesting biomarker that predicts CV events in patients. Second, we found that 3-MH is positively associated with LTM estimated with BIS.

In a different clinical cohort, high plasma or urine levels of 3-MH was associated with an increase in muscle mass catabolism and poor outcomes [33–35]. Recently, Kochlik et al. [12] observed in a large cohort of elderly patients that higher plasma 3-MH levels were found in frail and pre-frail participants compared with robust participants, assuming an elevated muscle protein turnover in these subjects. After cardiac surgery, high urinary 3-MH concentration is associated with a decrease in LTM [36]. In a healthy population, a strong association between 3-MH and BMI was observed [37].

In contrast to what has already been observed in other populations, in MHD patients we noticed that a low level of 3-MH was associated with low LTM and poor clinical outcomes [11–13]. Although urinary 3-MH excretion has been reported to be correlated with LTM in normal individuals [38], the paradoxical high plasmatic 3-MH levels in CKD patients with better survival are not fully understood, but we can suggest several hypotheses.

First, 3-MH is higher in omnivores compared with vegetarians [23–25], suggesting that low 3-MH is a marker of low protein intake, which may explain why low serum 3-MH levels were associated with an increased risk of mortality. Indeed, in our cohort we observed a positive correlation between 3-MH, nPCR as well as other nutritional parameters. The interdependence of these two parameters could partially explain the absence of correlation between mortality and 3-MH after adjustment with nutritional parameters. The relationship between malnutrition, inflammation and mortality in end-stage kidney disease has been well established. However, the association between 3-MH and CV events in our study was attenuated but not mitigated after adjustment for surrogates of malnutrition or inflammation. We therefore believe that the malnutrition–inflammation complex may not fully account for the observed 3-MH–cardiovascular diseases association. Nevertheless, the relation between 3-MH and protein intake remains controversial. The elevated 3-MH concentrations significantly declined within 24 h after white meat intake [39], even if some other studies failed to confirm this observation [40]. Furthermore, cod and salmon intake did not affect serum concentrations of 3-MH [41]. It should be noted that the association between 3-MH and CV events remained significant after adjustment with nPCR and albumin. At present, protein intake was only estimated with nPCR and, unfortunately, we did not have an exhaustive dietician interview to determine the influence of the dietary origin of protein at the 3-MH level.

Second, the validity to use plasmatic or urine 3-MH levels as a proxy for the whole body and skeletal muscle protein breakdown is still under debate. Indeed, data on how accurately the 3-MH level reflects muscle metabolism remain ambiguous, as 3-MH can be formed from several other tissues [42]. For instance, when rats are fasted, the rate of protein breakdown in the smooth muscle of the gut increases rapidly, whereas the skeletal muscle protein breakdown rate remains low [43]. Plasmatic or urinary concentrations of 3-MH show that protein breakdown rate is important throughout this period, but gives no information on the tissue in which this phenomenon is taking place. In moderate fasting, the variation of serum 3-MH levels in uraemic rats was higher compared with sham rats, but not after 48 h. These observations highlighted the difficulty of interpreting 3-MH depending on the fasting state. In our cohort, all patients were supposed to be overnight fasted, but the exact duration of the fast can influence the results. The situation is similar to measuring whole-body protein synthesis, which gives no information on changes at the tissue level. It could explain why the whole-body protein synthesis rate in cachectic cancer patients was not shown to be that different from healthy controls. However, in this population, muscle protein synthesis was decreased by 80% but is offset by an increase in protein synthesis in non-muscle tissues [44]. Recently, some data have suggested that 3-MH concentration is more an index of global muscle protein turnover, reflecting good muscular health rather than just an index of myofibrillar breakdown [45].

Third, renal function influences plasma 3-MH levels. Once released from the peptide during protein breakdown, 3-MH cannot be reincorporated into proteins and is excreted from the body in urine. According to the literature, the urinary 3-MH:creatinine ratio is commonly used to evaluate sarcopaenia by overcoming the renal function in the overall population that could not be performed in the MHD population. Because information for RRF was unavailable in our database, we compared the CV event predictabilities of serum 3-MH level among subgroups stratified by the use of diuretic as a surrogate of RRF. In our study, the presence or the absence of RRF did not influence the association between 3-MH level and CV events. The plasma 3-MH level could also be modified by the regimen of dialysis. 3-MH and creatinine clearance appeared similar [46]. Kaizu et al. [20] showed a similar production in dialysate of 3-MH and creatinine in HD. This may explain why Kt/V appeared negatively correlated with 3-MH level. However, in all observational studies including MHD patients [47], a higher Kt/V is associated with improved survival. Given that a higher Kt/V was found in the group with the highest mortality, we may conclude that this confounding factor has a limited impact. Supporting this view, in multivariate analysis after adjustment for Kt/V (model 1), 3-MH remained associated with all outcomes. Therefore Kt/V did not seem to be a significant confounding factor in our study.

Fourth, to date, the association between 3-MH and CV events has been poorly explored. One study observed a higher plasmatic concentration of 3-MH in type 2 diabetes patients with CV events [48]. However, in this study, patients with CV disease exhibited a lower estimated GFR than control patients, which makes the values of 3-MH uninterpretable in this cohort. In MHD, Chang et al. [21] demonstrated that increased aortic arterial stiffness was associated with reduced 3-MH levels. Arterial stiffness, reflecting vascular damage in CKD, is known to be associated with CKD progression and increased mortality [49]. Furthermore, there was a statistical trend of negative correlation between NT-proBNP and 3-MH levels. Unfortunately, we did not have any echocardiography data to distinguish a real cardiac dysfunction of inadequate dialysis with fluid overload to confirm such an association. Nevertheless, in univariate analysis, the TBW appeared positively correlated with the 3-MH level, suggesting that the trend to a negative correlation between NT-proBNP and 3-MH levels was explained by cardiac dysfunction rather than by fluid overload. 3-MH is a post-translationally methylated form of histidine found in myofibrillar proteins. The role of histidine methylation in the functioning of the protein and in cell physiology remains unclear. Some recent data suggest that histidine methylation is a key regulator of smooth muscle contractility [50]. This hypothesis might be the starting point for further research to explore vascular dysfunction regarding histidine methylation and 3-MH concentration in this specific population.

Here, 3-MH appeared as a relevant lean mass indicator in MHD patients in univariate analysis. In the present cohort, we estimated lean mass by using BIS. However, the most suitable method to estimate body composition in MHD patients is still debated. Isotope dilution methods, DXA, computed tomography and magnetic resonance imaging are the most common standard methods used to measure body composition. DXA has become recognized for its ability to accurately and precisely measure total body composition and is accepted as a gold standard. However, DXA is not a bedside technique, it requires patient transportation to the instrument and has high costs, thus hampering its use in routine practice. BIS appears to be a robust tool for measuring and monitoring body composition. Indeed, the three-compartment model of the BCM has been validated against standard reference methods for the assessment of fluid status and body composition in dialysis patients [51–55], even if some conflicting data exist [56]. Therefore we can conclude with high confidence that the circulating 3-MH level is correlated with muscle weight in MHD patients [2].

There are some limitations to our study. We conducted a monocentric study, with mostly Caucasian patients. It would be interesting to include more racial and ethnic diversity, especially because Black patients are known to have an increased lean tissue index compared with other ethnic groups [57]. We did not have longitudinal data, but our patients were in a steady state and well phenotyped. 3-MH is probably a better marker of acute sarcopaenia [36], but we lacked information about its potency to discriminate between the acute and chronic sarcopaenia phenomenon. Further studies to test its potential kinetic significance should be carried out. Muscle mass was only one aspect of muscle function and we did not have any evaluation of physical activity or muscle strength in our cohort. Since sarcopaenia is the association between a loss of muscle mass and a reduction of muscle function, further studies are needed to examine whether 3-MH is correlated with muscle strength.

In conclusion, 3-MH was associated with higher LTM and better nutritional parameters in MHD patients and a low serum 3-MH level was an independent and significant predictor of poor clinical outcomes in this population. Routine use of this marker does not seem feasible in relation to its cost and because we failed to demonstrate better performance than the simplified creatinine index [58] to predict muscle mass and outcomes in CKD. However, it could provide a more comprehensive assessment of the CV risk in clinical trials and monitor muscle status. In particular, the impact of physical activity and nutritional support on serum 3-MH concentration could be an interesting complementary biomarker to evaluate earlier and more-specific lifestyle interventions. These studies could help to determine whether increased serum 3-MH is causally related to CV events in MHD or simply indicates risk as an innocent bystander. Further studies are needed to validate this marker in a larger population and to elucidate underlying mechanisms.

ACKNOWLEDGEMENTS

This study was supported by the Hospices Civils de Lyon. E.B. was supported by a grant (Année Recherche) from Hospices Civils de Lyon and Agence Régionale de Santé. We thank all the patients for their participation in this study.

AUTHORS’ CONTRIBUTIONS

E.B. performed the study, researched data, analysed the results and wrote the manuscript. C.P., A.B., M.P., S.G., M.L. and D.F. researched data and analysed the results. C.S. and L.K. conceived the study, analysed the results and reviewed the manuscript. L.K. is the guarantor of this work and, as such, takes full responsibility for it. All authors read and approved the final manuscript.

CONFLICT OF INTEREST STATEMENT

The authors have no conflicts of interest to disclose. The results presented in this article have not been published previously in whole or part, except in abstract format.

REFERENCES

1.

Lees
JS
,
Welsh
CE
,
Celis-Morales
CA
et al.
Glomerular filtration rate by differing measures, albuminuria and prediction of cardiovascular disease, mortality and end-stage kidney disease
.
Nat Med
2019
;
25
:
1753
–
1760

Google Scholar

Crossref
Search ADS
PubMed

 
2.

Carrero
JJ
,
Johansen
KL
,
Lindholm
B
et al.
Screening for muscle wasting and dysfunction in patients with chronic kidney disease
.
Kidney Int
2016
;
90
:
53
–
66

Google Scholar

Crossref
Search ADS
PubMed

 
3.

Ajiro
J
,
Alchi
B
,
Narita
I
et al.
Mortality predictors after 10 years of dialysis: a prospective study of Japanese hemodialysis patients
.
Clin J Am Soc Nephrol
2007
;
2
:
653
–
660

Google Scholar

Crossref
Search ADS
PubMed

 
4.

Canaud
B
,
Ye
X
,
Usvyat
L
et al.
Clinical and predictive value of simplified creatinine index used as muscle mass surrogate in end-stage kidney disease haemodialysis patients—results from the international MONitoring Dialysis Outcome initiative
.
Nephrol Dial Transplant
2020
;
35
:
2161
–
2171

Google Scholar

Crossref
Search ADS
PubMed

 
5.

Yun
Y
,
Gao
T
,
Li
Y
et al.
Excretory function of intestinal tract enhanced in kidney impaired rats caused by adenine
.
Sci World J
2016
;
2016
:
2695718

Google Scholar

Crossref
Search ADS

 
6.

Mussini
E
,
Cornelio
F
,
Dworzak
F
et al.
Content of methylhistidines in normal and pathological human skeletal muscles
.
Muscle Nerve
1983
;
6
:
423
–
429

Google Scholar

Crossref
Search ADS
PubMed

 
7.

Young
VR
,
Munro
HN
.
Ntau-methylhistidine (3-methylhistidine) and muscle protein turnover: an overview
.
Fed Proc
1978
;
37
:
2291
–
2300

Google Scholar

PubMed
OpenURL Placeholder Text

 
8.

Bates
PC
,
Grimble
GK
,
Sparrow
MP
et al.
Myofibrillar protein turnover. Synthesis of protein-bound 3-methylhistidine, actin, myosin heavy chain and aldolase in rat skeletal muscle in the fed and starved states
.
Biochem J
1983
;
214
:
593
–
605

Google Scholar

Crossref
Search ADS
PubMed

 
9.

Nishizawa
N
,
Shimbo
M
,
Hareyama
S
et al.
Fractional catabolic rates of myosin and actin estimated by urinary excretion of NT-methylhistidine: the effect of dietary protein level on catabolic rates under conditions of restricted food intake
.
Br J Nutr
1977
;
37
:
345
–
353

Google Scholar

Crossref
Search ADS
PubMed

 
10.

Millward
DJ
,
Bates
PC
,
Grimble
GK
et al.
Quantitative importance of non-skeletal-muscle sources of N tau-methylhistidine in urine
.
Biochem J
1980
;
190
:
225
–
228

Google Scholar

Crossref
Search ADS
PubMed

 
11.

Trappe
T
,
Williams
R
,
Carrithers
J
et al.
Influence of age and resistance exercise on human skeletal muscle proteolysis: a microdialysis approach
.
J Physiol
2004
;
554
:
803
–
813

Google Scholar

Crossref
Search ADS
PubMed

 
12.

Kochlik
B
,
Stuetz
W
,
Pérès
K
et al.
Associations of plasma 3-methylhistidine with frailty status in French cohorts of the FRAILOMIC initiative
.
J Clin Med
2019
;
8
:
1010

Google Scholar

Crossref
Search ADS

 
13.

Koshikawa
M
,
Harada
M
,
Noyama
S
et al.
Association between inflammation and skeletal muscle proteolysis, skeletal mass and strength in elderly heart failure patients and their prognostic implications
.
BMC Cardiovasc Disord
2020
;
20
:
228

Google Scholar

Crossref
Search ADS
PubMed

 
14.

Broyer
M
,
Jean
G
,
Dartois
AM
et al.
Plasma and muscle free amino acids in children at the early stages of renal failure
.
Am J Clin Nutr
1980
;
33
:
1396
–
1401

Google Scholar

Crossref
Search ADS
PubMed

 
15.

Condon
JR
,
Asatoor
AM
.
Amino acid metabolism in uraemic patients
.
Clin Chim Acta
1971
;
32
:
333
–
337

Google Scholar

Crossref
Search ADS
PubMed

 
16.

Canepa
A
,
Divino Filho
JC
,
Forsberg
AM
et al.
Nutritional status and muscle amino acids in children with end-stage renal failure
.
Kidney Int
1992
;
41
:
1016
–
1022

Google Scholar

Crossref
Search ADS
PubMed

 
17.

Counahan
R
,
El-Bishti
M
,
Cox
BD
et al.
Plasma amino acids in children and adolescents on hemodialysis
.
Kidney Int
1976
;
10
:
471
–
477

Google Scholar

Crossref
Search ADS
PubMed

 
18.

Ward
LC
,
Buttery
PJ
.
NT-methylhistidine—an index of the true rate of myofibrillar degradation? An appraisal
.
Life Sci
1978
;
23
:
1103
–
1115

Google Scholar

Crossref
Search ADS
PubMed

 
19.

Whitehouse
S
,
Katz
N
,
Schaeffer
G
et al.
Histidines and renal function
.
Clin Nephrol
1975
;
3
:
24
–
27

Google Scholar

PubMed
OpenURL Placeholder Text

 
20.

Kaizu
Y
,
Ohkawa
S
,
Kumagai
H
.
Muscle mass index in haemodialysis patients: a comparison of indices obtained by routine clinical examinations
.
Nephrol Dial Transplant
2002
;
17
:
442
–
448

Google Scholar

Crossref
Search ADS
PubMed

 
21.

Chang
Y-C
,
Tang
C-C
,
Hsu
B-G
.
P0235 L serum 3-methyl histidine level is associated with aortic stiffness in hemodialysis patients
.
Nephrol Dial Transplant
2020
;
35
(Suppl 3):
gfaa142.P0235

Google Scholar

OpenURL Placeholder Text

 
22.

Lai
Y-H
,
Lee
M-C
,
Lin
T-J
et al.
Low serum 3-methylhistidine levels are associated with first hospitalization in kidney transplantation recipients
.
Transplant Proc
2020
;
52
:
3214
–
3220

Google Scholar

Crossref
Search ADS
PubMed

 
23.

Hansen
B
,
DiBartola
SP
,
Chew
DJ
et al.
Amino acid profiles in dogs with chronic renal failure fed two diets
.
Am J Vet Res
1992
;
53
:
335
–
341

Google Scholar

PubMed
OpenURL Placeholder Text

 
24.

Gutierrez
A
,
Alvestrand
A
,
Qureshi
GA
et al.
Influence of a meat-free diet on the urinary excretion of 3-methylhistidine and creatinine in chronic renal failure
.
J Intern Med
1992
;
232
:
129
–
132

Google Scholar

Crossref
Search ADS
PubMed

 
25.

Qureshi
GA
,
Gutierrez
A
,
Bergström
J
.
Determination of histidine, 1-methylhistidine and 3-methylhistidine in biological samples by high-performance liquid chromatography. Clinical application of urinary 3-methylhistidine in evaluating the muscle protein breakdown in uraemic patients
.
J Chromatogr
1986
;
374
:
363
–
369

Google Scholar

Crossref
Search ADS
PubMed

 
26.

Charlson
ME
,
Pompei
P
,
Ales
KL
et al.
A new method of classifying prognostic comorbidity in longitudinal studies: development and validation
.
J Chronic Dis
1987
;
40
:
373
–
383

Google Scholar

Crossref
Search ADS
PubMed

 
27.

Piraud
M
,
Ruet
S
,
Boyer
S
et al.
Amino acid profiling for the diagnosis of inborn errors of metabolism
.
Methods Mol Biol
2011
;
708
:
25
–
53

Google Scholar

Crossref
Search ADS
PubMed

 
28.

Garred
LJ
,
Barichello
DL
,
Canaud
BC
et al.
Simple equations for protein catabolic rate determination from pre dialysis and post dialysis blood urea nitrogen
.
J Am Soc Artif Intern Organs
1995
;
41
:
889
–
895

Google Scholar

Crossref
Search ADS

 
29.

Canaud
B
,
Vallée
AG
,
Molinari
N
et al.
Creatinine index as a surrogate of lean body mass derived from urea Kt/V, pre-dialysis serum levels and anthropometric characteristics of haemodialysis patients
.
PLoS One
2014
;
9
:
e93286

Google Scholar

Crossref
Search ADS
PubMed

 
30.

Hung
S-C
,
Kuo
K-L
,
Peng
C-H
et al.
Volume overload correlates with cardiovascular risk factors in patients with chronic kidney disease
.
Kidney Int
2014
;
85
:
703
–
709

Google Scholar

Crossref
Search ADS
PubMed

 
31.

Moissl
UM
,
Wabel
P
,
Chamney
PW
et al.
Body fluid volume determination via body composition spectroscopy in health and disease
.
Physiol Meas
2006
;
27
:
921
–
933

Google Scholar

Crossref
Search ADS
PubMed

 
32.

DeLong
ER
,
DeLong
DM
,
Clarke-Pearson
DL
.
Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach
.
Biometrics
1988
;
44
:
837
–
845

Google Scholar

Crossref
Search ADS
PubMed

 
33.

López Hellín
J
,
Baena-Fustegueras
JA
,
Sabín-Urkía
P
et al.
Nutritional modulation of protein metabolism after gastrointestinal surgery
.
Eur J Clin Nutr
2008
;
62
:
254
–
262

Google Scholar

Crossref
Search ADS
PubMed

 
34.

McKeran
RO
,
Halliday
D
,
Purkiss
P
et al.
3-Methylhistidine excretion as an index of myofibrillar protein catabolism in neuromuscular disease
.
J Neurol Neurosurg Psychiatry
1979
;
42
:
536
–
541

Google Scholar

Crossref
Search ADS
PubMed

 
35.

Mansoor
O
,
Breuillé
D
,
Béchereau
F
et al.
Effect of an enteral diet supplemented with a specific blend of amino acid on plasma and muscle protein synthesis in ICU patients
.
Clin Nutr
2007
;
26
:
30
–
40

Google Scholar

Crossref
Search ADS
PubMed

 
36.

Iida
Y
,
Yamazaki
T
,
Arima
H
et al.
Predictors of surgery-induced muscle proteolysis in patients undergoing cardiac surgery
.
J Cardiol
2016
;
68
:
536
–
541

Google Scholar

Crossref
Search ADS
PubMed

 
37.

Elliott
P
,
Posma
JM
,
Chan
Q
et al.
Urinary metabolic signatures of human adiposity
.
Sci Transl Med
2015
;
7
:
285ra62

Google Scholar

Crossref
Search ADS
PubMed

 
38.

Wassner
SJ
,
Li
JB
.
N tau-methylhistidine release: contributions of rat skeletal muscle, GI tract, and skin
.
Am J Physiol
1982
;
243
:
E293
–
E297

Google Scholar

PubMed
OpenURL Placeholder Text

 
39.

Kochlik
B
,
Gerbracht
C
,
Grune
T
et al.
The influence of dietary habits and meat consumption on plasma 3-methylhistidine-A potential marker for muscle protein turnover
.
Mol Nutr Food Res
2018
;
62
:
e1701062

Google Scholar

Crossref
Search ADS
PubMed

 
40.

Pechar
J
,
Málek
P
,
Doberský
P
et al.
Influence of protein intake and renal function on plasma amino acids in patients with renal impairment and after kidney transplantation
.
Nutr Metab
1978
;
22
:
278
–
287

Google Scholar

Crossref
Search ADS
PubMed

 
41.

Hagen
IV
,
Helland
A
,
Bratlie
M
et al.
TMAO, creatine and 1-methylhistidine in serum and urine are potential biomarkers of cod and salmon intake: a randomised clinical trial in adults with overweight or obesity
.
Eur J Nutr
2020
;
59
:
2249
–
2259

Google Scholar

Crossref
Search ADS
PubMed

 
42.

Sjölin
J
,
Stjernström
H
,
Henneberg
S
et al.
Splanchnic and peripheral release of 3-methylhistidine in relation to its urinary excretion in human infection
.
Metabolism
1989
;
38
:
23
–
29

Google Scholar

Crossref
Search ADS
PubMed

 
43.

Emery
PW
,
Cotellessa
L
,
Holness
M
et al.
Different patterns of protein turnover in skeletal and gastrointestinal smooth muscle and the production of n tau-methylhistidine during fasting in the rat
.
Biosci Rep
1986
;
6
:
143
–
153

Google Scholar

Crossref
Search ADS
PubMed

 
44.

Emery
PW
,
Edwards
RH
,
Rennie
MJ
et al.
Protein synthesis in muscle measured in vivo in cachectic patients with cancer
.
Br Med J (Clin Res Ed)
1984
;
289
:
584
–
586

Google Scholar

Crossref
Search ADS
PubMed

 
45.

Crossland
H
,
Smith
K
,
Atherton
PJ
et al.
A novel stable isotope tracer method to simultaneously quantify skeletal muscle protein synthesis and breakdown
.
Metab Open
2020
;
5
:
100022

Google Scholar

Crossref
Search ADS

 
46.

Davies
SP
,
Reaveley
DA
,
Brown
EA
et al.
Amino acid clearances and daily losses in patients with acute renal failure treated by continuous arteriovenous hemodialysis
.
Crit Care Med
1991
;
19
:
1510
–
1515

Google Scholar

Crossref
Search ADS
PubMed

 
47.

Lertdumrongluk
P
,
Streja
E
,
Rhee
CM
et al.
Dose of hemodialysis and survival: a marginal structural model analysis
.
Am J Nephrol
2014
;
39
:
383
–
391

Google Scholar

Crossref
Search ADS
PubMed

 
48.

Kume
S
,
Araki
S
,
Ono
N
et al.
Predictive properties of plasma amino acid profile for cardiovascular disease in patients with type 2 diabetes
.
PLoS One
2014
;
9
:
e101219

Google Scholar

Crossref
Search ADS
PubMed

 
49.

Zanoli
L
,
Lentini
P
,
Briet
M
et al.
Arterial stiffness in the heart disease of CKD
.
J Am Soc Nephrol
2019
;
30
:
918
–
928

Google Scholar

Crossref
Search ADS
PubMed

 
50.

Wilkinson
AW
,
Diep
J
,
Dai
S
et al.
SETD3 is an actin histidine methyltransferase that prevents primary dystocia
.
Nature
2019
;
565
:
372
–
376

Google Scholar

Crossref
Search ADS
PubMed

 
51.

Fürstenberg
A
,
Davenport
A
.
Comparison of multifrequency bioelectrical impedance analysis and dual-energy X-ray absorptiometry assessments in outpatient hemodialysis patients
.
Am J Kidney Dis
2011
;
57
:
123
–
129

Google Scholar

Crossref
Search ADS
PubMed

 
52.

Di Iorio
BR
,
Scalfi
L
,
Terracciano
V
et al.
A systematic evaluation of bioelectrical impedance measurement after hemodialysis session
.
Kidney Int
2004
;
65
:
2435
–
2440

Google Scholar

Crossref
Search ADS
PubMed

 
53.

Molfino
A
,
Don
BR
,
Kaysen
GA
.
Comparison of bioimpedance and dual-energy x-ray absorptiometry for measurement of fat mass in hemodialysis patients
.
Nephron Clin Pract
2012
;
122
:
127
–
133

Google Scholar

Crossref
Search ADS
PubMed

 
54.

Wabel
P
,
Chamney
P
,
Moissl
U
et al.
Importance of whole-body bioimpedance spectroscopy for the management of fluid balance
.
Blood Purif
2009
;
27
:
75
–
80

Google Scholar

Crossref
Search ADS
PubMed

 
55.

Lim
PS
,
Chen
CH
,
Zhu
F
et al.
Validating body fat assessment by bioelectric impedance spectroscopy in Taiwanese hemodialysis patients
.
J Ren Nutr
2017
;
27
:
37
–
44

Google Scholar

Crossref
Search ADS
PubMed

 
56.

Tewari
N
,
Awad
S
,
Macdonald
IA
et al.
A comparison of three methods to assess body composition
.
Nutrition
2018
;
47
:
1
–
5

Google Scholar

Crossref
Search ADS
PubMed

 
57.

Araujo
AB
,
Chiu
GR
,
Kupelian
V
et al.
Lean mass, muscle strength, and physical function in a diverse population of men: a population-based cross-sectional study
.
BMC Public Health
2010
;
10
:
508

Google Scholar

Crossref
Search ADS
PubMed

 
58.

Canaud
B
,
Ye
X
,
Usvyat
L
et al.
Clinical and predictive value of simplified creatinine index used as muscle mass surrogate in end-stage kidney disease haemodialysis patients-results from the international MONitoring Dialysis Outcome initiative
.
Nephrol Dial Transplant
2020
;
35
:
2161
–
2171

Google Scholar

Crossref
Search ADS
PubMed

 
© The Author(s) 2022. Published by Oxford University Press on behalf of the ERA.
This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://dbpia.nl.go.kr/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Topic:
Issue Section:
Original Article
Download all slides
  • Supplementary data

    • Supplementary data

    gfac050_Supplemental_Files - zip file

    Comments

    0 Comments
    Comments (0)
    Submit a comment
    You have entered an invalid code
    Thank you for submitting a comment on this article. Your comment will be reviewed and published at the journal's discretion. Please check for further notifications by email.