| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||
COMMENTARIES |
Renal Division Baxter Healthcare McGaw Park, IL, USA
e-mail: cliff_holmes{at}baxter.com
Since the start of peritoneal dialysis (PD), there has been a growing appreciation of the many characteristics of a PD solution that can influence its ultrafiltration (UF) performance. Factors that are well described and employed in clinical practice include the type of osmotic agent used (whether it be crystalloid or colloid), the concentration of the osmotic agent, the dwell time, and the volume instilled. Over the past two decades, numerous attempts to improve the UF efficiency of glucose have been described but with varying degrees of success. The use of vasoactive agents, such as nitroprusside, to pharmacologically alter peritoneal solute clearance was demonstrated as early as 1976 (1), but the mechanism behind this phenomenon was unclear. Douma and colleagues have since delineated the mechanism as a nitric oxide-mediated increase in the effective surface area and intrinsic permeability of the peritoneal membrane (2). Approaches to improving UF using phospholipids, proteoglycans, or glycosaminoglycans added to the PD solution have either been unsuccessful or demonstrated a trend toward improved UF (3–6). More recently, some studies have reported that "biocompatible" glucose-based solutions have altered UF characteristics versus the more acidic, higher glucose degradation product (GDP) conventional solutions (7,8). Indeed, mechanistic studies exist that describe the interaction of pH, buffer, and GDPs on peritoneal vasoreactivity, thereby offering a possible explanation for such clinical reports (9). However, a discourse on mechanisms seems premature given the current controversy over whether true differences in UF actually occur in the clinical setting with "biocompatible" solutions. In this respect, I refer the reader to recent publications and correspondence on two such studies with biocompatible solutions that draw opposite conclusions on this topic (10,11).
In this issue of Peritoneal Dialysis International, Cavallini and colleagues report yet another interaction between glucose (the osmotic agent employed) and a new buffer system that partially substitutes lactate with citrate (12). This is an acute rat study that reports an increase in net UF of about 25% during a 4-hour dwell when 10 mmol/L lactate is substituted with citrate. This net increase in UF appears to be due primarily to an increase in osmotically driven UF, and some evidence is provided that this may be caused by enhanced retention of glucose in the peritoneal cavity. The rationale behind the use of citrate derives from earlier studies by this same group and those by Sjoland et al. that demonstrated that low molecular weight heparin could improve UF via reduced glucose uptake (increased D4/D0), a phenomenon associated with reduced intraperitoneal activation of both the coagulation and the complement pathways (13–15). Citrate, due to its calcium-chelating properties, is also well known to exhibit anticoagulant and anti-complement properties. Furthermore, citrate can be readily metabolized by the liver to bicarbonate and thus should maintain acid–base balance and not accumulate in the presence of renal failure. Consequently, citrate would appear to offer, at least on first consideration, a potentially useful way of increasing UF without increasing glucose concentration. It is noteworthy that the co-agulation and complement pathways were described as activated within the environment of the peritoneum several years ago but with little exploration of the biological consequences and clinical relevance of such phenomena until now.
What questions are raised by this study and what additional work is advisable before human exposure in undertaken? First, it should be noted that animal studies are limited in their ability to predict performance in the clinical setting. Rats do offer, however, a reasonable model for evaluating glucose-based osmotic agents because the variability in UF measurement is small enough that reproducible and reliable results are possible. As these researchers and others have pointed out, the rat peritoneum tends to behave as a very fast transporter and thus the improved UF seen at 2 hours in these studies likely equates to a 4- to 5-hour continuous ambulatory PD dwell in humans. Thus, the effect of citrate in automated PD short dwells remains open to speculation until further animal work is completed or until human exposure studies are completed. Success in animal studies has frequently been reported with new technologies only to be irreproducible in the clinical setting, so a good dose of cautious optimism is appropriate here. Second, despite the anticoagulation and complement mechanisms that initially formed the basis behind citrate selection, this study could not confirm any inhibition of intraperitoneal coagulation, at least in terms of thrombin formation. In addition, due to the recent discontinuation of the commercial kit, complement activation was not measured. The study investigators speculate that reduced glucose uptake was more likely related to a cellular mechanism, such as mast cell degranulation. No data are provided, however, to support this hypothesis and thus a more detailed study to determine the mechanism behind the observed altered glucose uptake and increased UF is warranted. As complement data were not measured and there appeared to be an imbalance between groups in the baseline thrombin values, a role for these mechanisms still cannot be ruled out at this time. Furthermore, the invocation of mast cell degranulation, even at a low level, or another cell-mediated change clearly may have long-term consequences; this needs to be better described before human exposure.
The kinetics of intraperitoneal calcium and citrate ion are also provided in this study and suggest that peritoneal tissue would be exposed to subphysiological levels of free calcium for significant parts of the dwell time. The long-term effects of transient, but repetitive subphysiologic calcium on peritoneal homeostasis are unknown. However, chronic rat infusion models now exist that can expose the peritoneum to PD solution daily for up to 20 weeks; these models would be useful to better understand the effect of an altered calcium ion and citrated chelated calcium environment.
It is instructive to recall the experience of acetate in the 1980s as an alternative buffer to lactate. Acetate, in both the short-term clinical setting and animal models, seemed for all intents and purposes equivalent to lactate. However, retrospective analyses confirmed a strong association between acetate use and loss of UF (16). Unfortunately, chronic exposure animal models were not available in that decade and no studies with acetate have been performed since to determine their ability to predict adverse outcomes. Nevertheless, it would seem prudent to perform such studies with citrate-buffered solutions given the absence of a good mechanistic understanding of the observed increase in UF and the observed alteration in intraperitoneal calcium kinetics.
In summary, the study by Cavallini et al. describes an ingenious method to increase net UF in an acute rat model by partially substituting lactate with citrate. As the mechanism behind this effect remains obscure, combined with the observation of altered peritoneal calcium kinetics, additional animal work to better understand the risks of such an approach seems warranted. Ultimately, as with any new drug, any risks identified will have to be balanced against the benefit of increasing the UF capacity of glucose-based solutions by 20% – 30%.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
