Balance About balANZ

The most significant publications in peritoneal dialysis (PD) in 2012 are undoubtedly those from the balANZ study. This randomized controlled trial conducted in Australia, New Zealand, and Singapore looked at the effect of Balance—the neutral-pH, double-pouch PD solution low in glucose degradation products (GDPs) from Fresenius Medical Care (Bad Homburg, Germany)—on important clinical outcomes in 185 incident PD patients. Results from this important study have appeared in three publications. The first, published in the Journal of the American Society of Nephrology, looked at both the primary outcome of residual renal function (RRF) and the secondary outcome of peritonitis (1). The second, published in Nephrology Dialysis Transplantation, looked at peritoneal membrane function (2). The third, which appears in this issue of Peritoneal Dialysis International (PDI), looks in more detail at the peritonitis outcome (3).

The background to all this work is that low-GDP, normal-pH, double-pouch solutions have been around for about 15 years (4). The underlying rationale is that these solutions are more “biocompatible” than standard solutions, which would justify their extra cost. The notion is that greater biocompatibility will lead to better long-term preservation of membrane function, with consequent stable ultrafiltration (UF). Based on early data, other suggested advantages are better preservation of RRF and improved host defenses, leading to less peritonitis. To complicate matters further, a variety of such solutions, with differences in the buffer used and, reportedly, in the relative concentrations of various GDPs, have been brought to market (1,5). The two best known are Physioneal solution from Baxter Healthcare Corporation (Deerfield, IL, USA), which is buffered using a combination of bicarbonate and lactate, and the Balance product, which is lactate-buffered and reported to have lower levels of particular GDPs (1,5).

Do these solutions offer clinically important benefits to patients? Before this year, at least ten randomized trials had been performed in adults, including four each on Physioneal and Balance (4,6-15). Unfortunately, those trials yielded unsatisfactory, often conflicting, results, with no consistent benefits and even a suggestion of disadvantage in the form of decreased UF. The result, not surprisingly, has been very uneven uptake of these solutions. They have been widely used in some western European countries (such as the Netherlands and the United Kingdom) and in Japan and Korea. They have been much less prescribed elsewhere, particularly in Canada and Hong Kong, where the need to show cost-effectiveness has prevented their widespread use. Most notably, they have not even been available in the United States until very recently (16).

The balANZ study is a potential “game changer,” however. Although modest in size and unblinded, it is well done. Larger than almost all the previous trials, it appears to demonstrate important clinical benefits for the biocompatible solution. Specifically, it shows a longer time to anuria even though, overall, no significant difference in the rate of decline of RRF (the primary outcome) was observed (1). More surprisingly and most impressively, balANZ shows a significantly lower rate of peritonitis for patients on Balance solution: about 1 episode every 3 years compared with about 1 every 2 years in the control group (1,3). With regard to membrane function, the Balance group experienced an initial disadvantage, with faster creatinine transport and less UF in the first 12 months; however, that difference went away in the second year of follow-up (1,2). Disappointingly, no difference was observed in the key outcome of technique survival—and not surprisingly given the modest size of the study, in patient survival either (1).

How should these results be interpreted? Is there enough here to justify a recommendation for the routine use of these solutions? Indeed, is there evidence to go further and say that Balance solution specifically is superior to the competing Physioneal solution?

Before attempting to answer those questions, it is important to congratulate the investigators, David Johnson and colleagues, for carrying out this difficult study. As is invariably the case in such trials, recruitment was very challenging, taking initiative and persistence to ensure that the study was successfully completed. Initial hopes for recruitment were not realized, but the authors still managed to enroll enough patients to identify statistically significant differences in outcomes (1). They and the sponsor, Fresenius Medical Care, deserve congratulations from the PD community for completing the study and for attempting to answer questions that are important for PD patients. Efforts like these are how improvements to patient care in dialysis come about.

What about the RRF findings? Even if no difference in the rate of RRF decline was observed, is the longer time to anuria a convincing benefit for Balance? Unfortunately, there is a large caveat here. During the first year, the Balance patients consistently showed significantly less UF and a trend toward greater weight gain, strongly suggesting that the lower rate of anuria may have been related to less-effective fluid removal and consequent volume expansion (1). Without comparative body composition studies, such an interpretation cannot be considered proved, but the data are suggestive, and this time would not be the first that such an observation has been made. Some previous trials with both Balance (9,12) and other biocompatible solutions (8) have shown this same trend, whereby improvements in urine output have been associated with decreases in UF. It is far from clear that improving urine output at the expense of volume expansion is a good tradeoff (17). Experienced PD practitioners will be aware of the constant need to juggle RRF and fluid removal in PD patients.

If the RRF results are perhaps a disappointment, what about the peritonitis findings in the balANZ study? Those results do appear impressive, and the paper in this issue of PDI expands on them, giving detail that was not reported in the initial publication (1,3). We learn that the decline in peritonitis in the Balance group occurred at an almost equal rate for all causative organisms with the exception of Pseudomonas, for which no trend was observed. A statistically significant difference could be detected only for non-pseudomonal gram-negative peritonitis; however, in a relatively small study (186 patients), the issue may be one of statistical power—a so-called type B error. We learn that, although hospitalizations for peritonitis were no less common in the Balance patients, the duration of those hospitalizations and the severity of the peritonitis as assessed by the responsible clinician were both less with the biocompatible solution (3). We also know from the previous paper that an analogous decrease was seen in non-PD-related infections such as pneumonia. That general beneficial effect raises the possibility that Balance may be enhancing immune defenses, both systemically and locally (1). Less happily, we are told that neither technique failure in general nor technique failure attributable to peritonitis was any different between the groups. However, this issue may again be one of statistical power, given that, for example, only 5 patients in each group switched to hemodialysis because of peritonitis (1,3).

The main reason to be cautious about these very promising findings on peritonitis is that they have not been apparent in other randomized trials of biocompatible solutions. The failure to see even a trend toward a decrease in the peritonitis rate in the larger randomized trial conducted in 267 patients by Fan and Srivastava and their colleagues raises doubts, but could be explained by the fact that their trial studied mainly the Physioneal product and that the benefit may be specific to Balance (7,8). The hypothesis is plausible, but other randomized trials of Balance also failed to detect a benefit. Szeto et al. randomized 50 patients and Choi et al. randomized 104 to either Balance or standard solutions, and neither could show a difference after 12 months of follow-up (11,12). Kim et al. randomized 91 patients and, conversely, reported a higher peritonitis rate in the Balance group, a finding that almost reached statistical significance (13). It could be argued that the foregoing trials were too small or too short. A well-done meta-analysis of biocompatible solutions in general and of Balance solution in particular would potentially be very helpful, but there is currently no clear trend.

Nevertheless, it would be too cynical to dismiss the beneficial effect on peritonitis shown in balANZ. This trial is the best single randomized study conducted to date on any biocompatible fluid. Many negative randomized trials in nephrology have been conducted, and when a positive one appears, dismissing it would be wrong, even given the presence of valid concerns and reasons for caution. One possibility, for example, is that a beneficial effect of a biocompatible solution on peritonitis will be apparent only when the baseline rate with the control solution is relatively high—for example, once every 2 years in the balANZ study compared with the much lower rates in the studies by Szeto and Kim and their co-workers (3,11,13).

Are Balance and Physioneal different in terms of outcomes? The signal on peritonitis is not consistent for Balance, but it is not present at all in the Physioneal randomized trials. Both solutions have shown inconsistent and sometimes concerning effects on UF. The two have never been compared directly, and it is unlikely that they ever will be. Even the two standard solutions used as controls may not be equivalent, given that they are based on different connectology, although one randomized trial has compared them and shown no difference in peritonitis rates (18). No conclusion can be drawn that either of these biocompatible solutions is convincingly superior.

So, we are left with an impressive finding from the balANZ study, but also with doubt (might it be a random type I statistical error?) and an added concern (might there be an early deleterious effect of the solution on UF?). In this writer’s opinion, the evidence from balANZ is not strong enough to recommend a wholesale switch to biocompatible solutions; the potential benefit has to be weighed against the possible downsides of impaired fluid removal and extra cost. However, others may see peritonitis as the bigger problem for their PD patients and reach a different conclusion. It is impossible to be dogmatic either way. As usual, more studies and more data are needed.

However, before we ask for more, larger, and longer studies, two further points should be made. First, randomized trials are expensive and demanding. More may not be done. This evidence may be all we get. Second, it is possible that one or both major companies may eventually switch production entirely to their biocompatible products, with the whole issue becoming a moot point. Such a switch is essentially what happened in hemodialysis with the debate about the benefit of high- versus low-flux dialyzers. However, in the present economic environment, such a move would not be well received unless it were to be cost-neutral. Also, the concerns about UF would need to be more convincingly allayed. It would be regrettable if the promising findings of balANZ are not further investigated. There may be clues here to the pathogenesis of infectious complications in uremia generally and perhaps to an important systemic toxic role for GDPs.

In summary, no definitive conclusion can yet be drawn about the value to patients of biocompatible solutions in general or of Balance solution in particular, but the strong suggestion that Balance may have beneficial properties needs further evaluation. We should be grateful to the balANZ investigators for their work and contribution to our debate.

DISCLOSURES

The author has received recent research support from Baxter Healthcare Corporation.