End-stage kidney disease requiring dialysis has increased in prevalence by up to 2.5 times in the past 5 years (1). Currently, the life expectancy of dialysis patients is longer than in the past, but it is not comparable to that of the general population. This might, at least in part, be explained by the accumulation of middle-molecule uremic toxins that can result in uremic complications and lead to higher morbidity and mortality (2). Current advanced hemodialysis (HD) techniques that can effectively eliminate middle-molecule uremic toxins include high-volume online hemodiafiltration (HDF) and HD using super high-flux dialyzers (SHF), so-called expanded hemodialysis (HDx). Both modalities have been demonstrated to pose a comparable removal of middle-molecule uremic toxins such as β2-microglobulin (β2M, 11.8 kDa), α1-microglobulin (α1M, 31 kDa), and λ-free light chain (λ-FLC, 45 kDa). These removals are higher than those of conventional high-flux HD (3–7). Although online HD, which has been widely used for more than two decades, has numerous mechanistic studies and clinical trials demonstrating superior removal of middle uremic toxins and decreased mortality compared to conventional high-flux HD, it requires a specialized dialysis machine with additional programming and staff training, which are not available in many hemodialysis centers (8–12). HDx exploits novel SHF or medium cut-off (MCO) dialyzers, which have larger pore sizes, to effectively remove larger middle-molecule compounds with conventional and simpler HD machines (8). Some studies even demonstrated that HDx tends to remove middle-molecule uremic toxins more effectively than online HDF (6, 7, 13). This makes HDx more practical than online HDF for large-scale applications with comparable outcomes. However, the cost of SHF is still high and there is no manufacturer approval for dialyzer reuse, which are major obstacles to the widespread application of HDx, especially in countries with limited resources.

Dialyzer reuse is a protocol developed to permit repeated HD with the same dialyzer for up to 12–20 times with careful monitoring to prevent adverse effects in the patients. The purposes of dialyzer reuse include reducing the cost of new dialyzers for every HD session, preventing the shortage of new dialyzer supplies, and possibly reducing medical waste associated with the dialyzer discard process (14). These purposes are more important for the dialyzer with higher production costs and a more limited supply, such as the SHF. However, studies on dialyzer reuse to date have been restricted to conventional low- and high-flux dialyzers. Indeed, the reuse or reprocessing of these conventional dialyzers is widely adopted in Thailand because of economic benefit. Theoretically, the reprocessing of SHF with adequate surveillance of efficacy change is feasible and may endorse widespread implementation of HDx to improve patient outcomes while compromising an environmental burden, complying with the green nephrology policy (15). The safety issue is another concern for dialyzer reprocessing, both from inadequate dialysis if the solute clearance is significantly impaired and from the residual chemical used in the sterilizing process that could enter the patient’s circulation in subsequent dialysis. Therefore, we conducted this study to compare the efficacy and safety of hemodialysis with reused SHF.

Five patients with end-stage renal disease were enrolled. The mean age was 58.6 ± 7.1 years; three were male and two were female. The mean duration of hemodialysis was 4.0 ± 7.9 years and the mean Kt/V was 2.83 ± 0.8. Vascular access was arteriovenous fistulas in four patients and an arteriovenous graft in one patient. The underlying causes of end-stage renal disease were diabetes mellitus (n = 2), hypertension (n = 1), and unknown etiology (n = 2). Two patients had a history of previous kidney transplantation. The baseline clinical characteristics are presented in Table 2.

In this study, a total of 15 dialyzers were investigated. The number of dialyzers completed on the 1st, 2nd, 5th, 10th, and 15th used was 15, 15, 14, 14, and 11, respectively (Figure 1). Dialyzers that were discarded early had residual blood clots observed during the reprocessing. One dialyzer had a TCV of less than 80% after the 15th reuse, while the remaining TCV measurements in this study were more than 80% of the baseline.

Our study is the first to examine the effectiveness of middle-molecule uremic toxin removal by SHF after conventional reprocessing with peracetic acid (18). Although there was no impairment of the small solute clearance from SHF reuse as reflected by TCV higher than 80% until the last permitted reuse time, middle-molecule uremic toxins removal needs more attention for the high-flux dialyzer and SHF, since this is the focus of recent dialysis techniques.

An accumulation of β2M, a small middle-molecule uremic toxin, is associated with mortality and morbidity (19, 20). Indeed, augmented β2M clearance is an important characteristic of this novel dialyzers (21). HDx, an HD using SHF, has been shown to achieve a comparable β2M removal and a higher removal of larger uremic toxins, such as α1M and λ-FLC, compared to the high volume online HDF (8, 13). Although HDx is more feasible for most dialysis centers to apply compared to online HDF, the high cost of SHF could hamper widespread implementation. The plausibility of reusing SHF, which is originally designated as single use, without an unacceptable decrease in dialyzer performance could theoretically overcome this obstacle, since dialyzer reuse has been shown to reduce the cost of dialysis by 15%–30% and may pose an environmental benefit due to reducing medical waste from the dialyzer discard (22, 23). Unlike the reprocessing of high-flux dialyzer that has been shown to reduce β2M removal by up to 60% (24, 25), SHF reprocessing in our study does not significantly impair the RR of both β2M and α1M. Moreover, all mean β2M RR is still higher than 80% for all reuses, which is comparable to the RR achieved by high volume online HDF in previous studies that demonstrated the clinical benefit of such technique (7, 13). Therefore, we suggest this cut point of β2M RR as a checklist to discard SHF after reuse. Regarding α1M, another small middle-molecule uremic toxin, the optimal RR derived from a previous study of online HDF and HDx in our study is less rigid and is around 25%–30% (26).

Considering larger middle-molecule uremic toxins, the RR of λ-FLC significantly decreased since the second reuse and reached 40% since the 5th reuse. The assumption is that the size of λ-FLC is close to the pore size of the SHF. Therefore, after reprocessing the dialyzer with peracetic acid, the size of some pores decreases to be smaller than the size of λ-FLC due to the remaining protein on the dialyzer membrane. The RR of λ-FLC achieved by online HDF from previous studies is around 40% (5, 27, 28). This might limit the reuse time of SHF to not exceed 5 times in further practice, as suggested in Table 4.

Regarding protein-bound uremic toxins, such as indoxyl sulfate, there was a trend in decline of RR after reuse, although not statistically significant, in our study. This pattern was similar for both total and free indoxyl sulfate levels. Data from a previous study in online HDF suggested the optimal RR for free indoxyl sulfate of 56% (26). A major concern for HDx is that the loss of dialysate albumin is greater than that of the HDF, possibly resulting in transient hypoalbuminemia. According to previous research, the median loss of albumin per HDx session is approximately 3 g (29). In our study, albumin loss in the dialysate with SHF was 1.01 g initially, which was significantly reduced to almost none after reprocessing. This could be due to the smaller pore size of the membranes after exposure to peracetic acid. Ameliorating albumin loss via HDx might be the advantage of reusing SHF and correlated with the finding that serum albumin, along with other inflammatory and nutritional markers, in the patients who received HDx for 3 months was not significantly altered.

Reuse of dialyzers may be more cost-effective compared to a single-use strategy if the cost associated with dialyzer manufacturing is more expensive than the reprocessing or labor costs. In developing countries, evidence has shown that conventional dialyzer reuse can reduce the cost of hemodialysis by 30%–70% per session (23, 30). Therefore, dialyzer reuse may improve patient access to hemodialysis and increase the opportunity to use high-quality dialyzers, especially SHF, without compromising the economic and environmental burdens. According to our study, HDx with reused SHF is a plausible and safe novel technique. The limitations included the small sample size and the restriction to only short-term outcomes. The specific dialyzer type and reprocessing protocol used in our study, which were polyester sulfone and peracetic acid, respectively, require cautious generalizability to other reprocessing techniques or dialyzer membranes. Moreover, adverse events from the reuse process should be monitored upon long-term use in various centers. Further studies are required to explore the clinical benefits, environmental issues, economic burden, and cost-effectiveness of different dialysis techniques such as conventional HD, online HDF, HDx with single-use SHF, and HDx with reuse protocol.

Reprocessing of SHF for reuse up to 15 times did not significantly reduce the RR of β2M and α1M. λ-FLC RR was significantly reduced after the 2nd reuse and fell below the optimal level reported by previous studies since the 5th reuse. Albumin loss during dialysis was significantly reduced after reuse, along with the trend in the reduction of RR of protein-bound indoxyl sulfate. Nevertheless, serum albumin from the included patients remained stable. HDx with reused SHF could be an alternative to online HDF and HDx with single-use SHF to lower the cost, reducing the medical waste associated with dialyzer discard, while maintaining the modern dialysis efficacy.