The purpose of this study is to determine the prevalence of protein-energy wasting and obesity among Danish dialysis patients.

Undernutrition among dialysis patients is associated with increased morbidity and mortality.

Conversely, in hemodialysis (HD) patients a high BMI in the range of obesity has long been known protective against death and cardiovascular disease, a phenomenon called reverse epidemiology. Hence, what is good for the background population might differ from what is good for HD patients.

In peritoneal dialysis (PD) patients this association is less clear. The majority of studies find reverse epidemiology, however some studies find no survival advantage of obesity or even indicate a higher risk of death in obese patients. Some authors claim that reverse epidemiology is a misconception and that although in the short run a high BMI is protective a follow up period of more than five years show increased mortality. Considering this Mutsert et al. claim that HD patients have the same mortality risk patterns associated with BMI as the general population.

Moreover, it has been shown that the protective effect of a high BMI or weight gain is limited to those with normal or high muscle mass. Patients with high BMI and malnutrition have a high risk of death. Concordantly, recent studies show that abdominal obesity is associated with mortality among dialysis patients .

The prevalence of under-nutrition, often known as protein energy wasting (PEW), among dialysis patients varies from 31% to 51% and the prevalence of obesity varies from 10-56% presumably depending on the diagnostic criteria and the method used. These inconsistent results make comparison of the prevalence of undernutrition and obesity among dialysis populations somewhat problematic.

In order to standardize the diagnosis of PEW in dialysis patients International Society of Renal Nutrition and Metabolism (ISRNM) suggest diagnostic criteria. As goes for obesity, there are no standardized methods for diagnosing this among dialysis patients. However, reviews recommend using multiple different methods concurrently including BMI, fat percentage (FM%), and measures of abdominal fat.

Data on the nutritional status among dialysis patients in Northern Europe are primary derived from a single research center (Karolinska institute, Sweden) and there is no recent published data on the prevalence of PEW among dialysis patients. A single study examines the prevalence of obesity.

Few studies have compared the nutritional status of HD and PD patients. Hardly any studies examine the prevalence of undernutrition and obesity in the same population.

Knowing the nutritional status of a dialysis population and identifying already malnourished patients, can contribute the quality of the care provided.

Hence, the aim of this study is, for the first time in Northern Europe to assess the prevalence of both PEW and obesity in the same population using the ISRNM criteria as well as different obesity markers. Additionally, we compare the nutritional status of HD and PD patients.

Design The study was designed as a cross-sectional study, investigating the nutritional status of hemodialysis (HD) patients and peritoneal dialysis (PD) patients attending the dialysis centre at Roskilde Hospital, Denmark in February to June 2014.

Subjects All PD, HD and home-HD patients (n = 105) were invited to participate. Patients who refused to participate were invited to complete a partial nutritional examination. The participating eligible patients all gave their written consent. Patient data (age, gender, dialysis vintage, primary kidney disease and co morbid conditions were taken from the patient records. The local ethical committee approved the study protocol.

Anthropometric measurements Anthropometric measurements were performed immediately after a dialysis session (HD) as recommended by the HEMO study or after the monthly ambulatory control (PD, home-HD). All measurements were done on the non-access side of the body for HD-patients and on the right side for PD-patients. Skin folds (SF) were measured with a Harpenden caliper and all circumferences were done using a non-stretchable fiberglass insertion tape (seca 201). All measurements were made by to well-trained research assistants.

Height (h) and bodyweight (BW) was measured in light clothing using standard instrument in the dialysis centre. For PD the weighing was preferably done with empty abdominal cavity, for those (n=8) who refused to empty the abdominal cavity the bodyweight was calculated as BW with dialysate minus the volume of dialysate that was last drained into the abdomen. For patients with lower leg amputations (n=2) 6.2% of total bodyweight per amputated leg was added to the measured weight

Fat mass was assessed according to a formula described by Durnin and Womersley based on four skinfold thicknesses (biceps, triceps, subscapular and suprailiac). Mid-arm circumference (MAC) was measured midway between acromion and olecranon. All measurements were done in duplicate. If the two measurements were more than four mm apart, two additional measurements were done and the mean of all four was calculated. This approach is recommended by "CLINICAL PRACTICE GUIDELINES For Nutrition of Chronic Renal Failure". Triceps skinfold (TSF) and MAC was used to calculate corrected mid-arm muscle area (cMAMA) as described by Heymsfield SB et al.

Waist circumference (WC) was measured as recommended by WHO. Patients unable to stand had their waist and hip circumference measurement done in supine position. The waist and hip measurements were skipped if empty abdominal cavity was not feasible for PD patients. All measurements were done in duplicate if the difference between the two measurements exceeded 1 cm, two more measurements were done, and the mean of all four calculated.

Lean body mass index (LBMI) and fat mass index (FMI) was calculated as lean body mass (LBM)/h2 and fat mass (FM)/h2 respectively as described by Kyle et al.

Blood sampling and biochemical measurements For assessment of s-transferrin and albumin blood samples were from all patients was analyzed at the laboratory at Roskilde Sygehus, Denmark. Kt/V was measured as described by F.A.Gotch(Gotch 1998). For HD-patients normalized protein catabolic rate (nPCR) was determined by measuring the interdialytic rise of blood urea nitrogen after a midweek dialysis session. For patients with residual urine output > 300ml/day urinary nitrogen excretion was also taken into account. To account for day-to-day variations we calculated a mean of three nPCR measurements, which were made over a 2 months period after the anthropometric measurements, we only made one urine sample. For PD-patients nPCR was measured using the Bergstroms' formula. Only one nPCR measurement was done for PD patients

Inclusion Criteria:

• all dialysis patients receiving dialysis therapy at Roskilde Hospital, Denmark

Exclusion Criteria:

• < 3 months dialysis
• fever
• antibiotic treatment
• major surgery within 2 weeks
• disseminated cancer
• age below 18
• psychosis
• pregnancy
• physical or mental disability where participation was not feasible,
• language barriers

Patients receiving regular dialysis therapy at Roskilde Hospital, Denmark in February to June 2014.