This is the first nationwide survey concerning the types of nutritional therapy and ET for patients on hemodialysis in Japan. In this study, we adopted age and serum albumin as the patients’ characteristics. The proportion of patients aged ≥ 65 years, which was 63.4% in this study, was almost identical to that of 61.8% found by JSDT for the entire Japanese dialysis population [1]. The prevalence of serum albumin < 3.5 g/dL was slightly greater at 37.5% in the present study than the 33.8% reported in a JSDT survey [22], a difference that may be due to the inclusion or exclusion of patients with serum albumin equal to 3.5 g/dL.
Approximately 40% of all facilities were administering ONS, predominantly the low residue diet that can be reimbursed by health insurance in Japan and has a well-balanced nutrient composition. About 50% of all facilities were offering IDPN. These results were independent of response probability. Although the typical composition of IDPN reported in the literature is a mixture of glucose, amino acids, and lipid emulsion [23], most facilities in the present study used only the amino acid formula for IDPN. This is possibly because the staff focused on the amino acids removed by the dialysis procedure [24]. Other reasons could be the higher costs of lipid emulsion compared with amino acid formula, as well as the time and labor required by medical staff to mix the components for IDPN preparations.
ONS administration may be associated with better survival [12]. Although IDPN did not have an additive effect to ONS in malnourished dialyzed patients [25], IDPN works to facilitate protein anabolism in patients receiving long-term hemodialysis in the acute phase [26]. This is evidenced by significant concomitant increases in lean body mass and albumin synthesis in the liver [27]. Therefore, both ONS and IDPN can potentially ameliorate malnutrition. In the present study, the proportion of patients treated by nutritional therapy was as small as 2%, while the proportion of patients with serum albumin ≤ 3.5 g/dL was as high as 37.5%. Therefore, it is possible that patients who should have been treated by nutritional therapy did not receive it. However, hypoalbuminemia is not the sole indication for nutritional therapy, and this criterion could overestimate the actual proportion of patients needing nutritional therapy. Further clarification of the indications for nutritional therapy is required in actual clinical practice. A positive point to note, however, is that logistic analysis showed nutritional therapy is being administered to patients with low nutrition in some facilities in Japan.
ET was enforced in about 20% of the facilities, and only 3% of all the patients were receiving ET. Patients on dialysis have extremely limited exercise capacity and poor physical functioning [28]. It has been reported that sedentary behavior in patients on hemodialysis is associated with an increased risk of death within 1 year even after adjusting for all covariates [29]. However, ET might increase aerobic capacity, muscle mass or strength, physical quality of life, and Kt/V [16, 30]. Additionally, ET during dialysis therapy is associated with few adverse events [16, 31, 32]. However, a study from the Dialysis Outcomes and Practice Patterns Study (DOPPS) showed that the percentage of units offering exercise programs was 14% in Japan from 2009 to 2011, while 27 and 37% of the facilities offered ET in Europe or Australia/New Zealand and in North America, respectively [33]. This difference between DOPPS and our study might indicate an increase in the facilities offering ET over the last few years, as the benefits of ET have become widely accepted.
ET performed on non-dialysis days is reportedly the most effective way of training, although difficulties in maintaining adherence have been demonstrated [34]. In the present study, a limited number of facilities (up to 20%) provided ET on hemodialysis days and only 9% provided ET on non-hemodialysis days. DOPPS showed that the proportion of facilities providing ET on dialysis and non-dialysis days was about 10% in japan from 2005 to 2006 [35]. After a decade, we found the facilities offering the exercise program on dialysis is increasing as discussed above, but the facilities offering ET on non-dialysis days have not increased. Recently it reported that two daily 10-min walking sessions during the off-dialysis days (every second day for patients on peritoneal dialysis) at a prescribed walking speed may improve physical performance and quality of life [36]. Our findings indicate that the awareness of ET needs to be further improved.
Each intervention type accounting for almost one-third of ET. Recently many studies have reported the evidences about exercise for dialysis patients. But the best way to offer ET in dialysis patients remains unclear. The current study demonstrated that the actual types of ET were diverse across the facility offering ET. Many investigations have employed ET as an intervention for at least 30 min, three times a week [16]. The duration of ET in Japanese facilities was found to be shorter in this study, possibly because the Japanese dialysis population investigated here was older than that in the reported clinical trials (aged 40–50 years) [16].
DOPPS reported an important finding: patients who exercise regularly showed better survival than those who do not (HR 0.73, 95% CI 0.69–0.78); moreover, patients who received treatment at a facility with a higher frequency of exercise also had better survival (HR per 10% increase in regularly exercising patients 0.92, 95% CI 0.89–0.94) [35]. These results demonstrate that the practice patterns implemented in a facility could affect the clinical outcomes of patients treated at that facility. So it is important for every dialysis facility to be aware of the usefulness and safety of ET in order to prevent PEW progression to further unfavorable outcomes. In fact, over half of the total facilities included in this survey intended to offer ET.
The present investigation into the background factors relating to the implementation of these therapies indicated that clinics tend to offer such therapies rather than hospitals. In general, clinics without beds manage larger numbers of chronic patients over longer terms, enabling them to provide these therapies more frequently than the other facility types. In contrast, we can consider that the patients in hospitals have different characteristics from those in clinics; they require the treatment for acute conditions because the hospitalized patients in other nephrology department tend to have many comorbidities.
The results obtained from the analyses where staff number was standardized by patient number confirmed that IDPN is being predominantly offered in clinics without beds.
Interestingly, our results indicate that the number of staff—key elements of each facility—was positively correlated with the implementation of nutritional therapy or ET. In particular, the results obtained from the analyses standardized by facility type and region revealed stronger associations between these factors.
Curiously, the proportions of the older patients or the patients with lower albumin levels were positively associated with the implementation of ONS or IDPN. On the other hand, ET did not have such associations with the proportions of the patients with such background. This fact shows the medical director of the facility might have different criteria to implement ET from those for ONS and IDPN.
Surprisingly, the provision of ONS or IDPN was independent of the presence of dietitians or pharmacists in the facility. This might indicate these therapies might be readily available irrespective of the professionals available, although the involvement of dietitians or pharmacists might improve the quality or effectiveness of these therapies.
There are several limitations to this study. The first is that this observational study did not examine the data of individual patients. Therefore, we could not investigate the detailed relationship between patient characteristics and the interventions or clinical outcomes. However, the aim of the study was to investigate the overall status of these therapies in Japan, which has not been investigated previously. Second, the response rate for the questionnaire was not high 26% of all facilities. Although detailed patient characteristics were not investigated and we cannot eliminate the effects of differences in patients characteristics between responded and non-responded facilities, the proportions of older patients or the patients with albumin of 3.5 g/dl or less were not virtually different from the entire dialysis population in Japan. On the other hand, the response rate did differ across the facility types. However, adjustment by affiliations of facility type or region yielded no significant differences in the proportion of facilities offering these interventions.