| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Icodextrin |
Department of Clinical Nephrology and Dialysis,1 CHPC Hôpital Louis Pasteur, Cherbourg; Peritoneal Dialysis Unit,2 La Pitié-Salpétrière University Hospital, Paris, France; Department of Nephrology and Dialysis,3 Brugmann University Hospital, Brussels, Belgium; Department of Nephrology and Dialysis,4 CHR Clemenceau, Caen, France; Statistics, Epidemiology and Surveillance,5 Baxter Healthcare Corporation, Round Lake; Renal Division,6 Baxter Healthcare Corporation, McGaw Park, Illinois, USA; Divisions of Baxter Novum and Renal Medicine,7 Department of Clinical Science, Intervention and Technology, Karolinska Institutet, Stockholm, Sweden; Baxter Renal Division Latin America,8 Mexico City, Mexico
Correspondence to: P. Freida, Department of Clinical Nephrology and Dialysis, CHPC Hôpital Louis Pasteur, rue Trottebec, Cherbourg, 50102 France. p.freida{at}ch-cherbourg.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT PATIENTS AND METHODSRESULTSDISCUSSIONDISCLOSUREREFERENCES |
|---|
Background: Evidence is accumulating that the
continuous exposure to high glucose concentrations during peritoneal dialysis
(PD) is an important cause of ultrafiltration (UF) failure. The cornerstone of
prevention and treatment of UF failure is reduction of glucose exposure, which
will also alleviate the systemic impact of significant free glucose
absorption. The challenge for the future is to discover new therapeutic
strategies to enhance fluid and sodium removal while diminishing glucose load
and exposure using combinations of available osmotic agents. Objectives: To investigate in patients on automated PD
(APD) with a fast transport pattern whether there is a glucose-sparing
advantage to replacing 7.5% icodextrin (ICO) during the long dwell with a
mixed crystalloid and colloid PD fluid (bimodal UF) in an attempt to promote
daytime UF and sodium removal while diminishing the glucose strength of the
dialysate at night. Design: A 2 parallel arm, 4 month, prospective
nonrandomized study. Setting: PD units or university hospitals in 4 French
and Belgian districts. Results: During the 4-month intervention period, net UF
and peritoneal sodium removal during the long dwell when treated by bimodal UF
was about 2-fold higher than baseline (with ICO). The estimated percent change
(95% confidence interval) from baseline in net daytime UF for the bimodal
solution was 150% (106% – 193%), versus 18% (–7% – 43%) for
ICO (p < 0.001). The estimated percent change from baseline in
peritoneal sodium removal for the bimodal solution was 147% (112% –
183%), versus 23% (–2% – 48%) for ICO (p < 0.001). The
estimated percent change from baseline in UF efficiency (24-hour net UF
divided by the amount of glucose absorbed) was significantly higher
(p < 0.001) when using the bimodal solution was 71%, versus
–5% for ICO. Conclusion: Prescription of bimodal UF during the day
in APD patients offers the opportunity to optimize the long dwell exchange in
a complete 24-hour APD cycle. The current study demonstrated that a bimodal
solution based on the mixing of glucose (2.6%) and icodextrin (6.8%) achieved
the double target of significantly improving UF and peritoneal sodium removal
by exploring a new concept of glucose-sparing PD therapy.
KEY WORDS: Bimodal ultrafiltration; fluid and sodium removal; glucose exposure; glucose sparing.
Although glucose-based peritoneal dialysis (PD) solutions have been used for more than two decades as an effective treatment for patients with end-stage renal disease, it is widely assumed that exposure of the peritoneal membrane to high concentrations of glucose contributes to long-term structural and functional changes in the peritoneal membrane (1). In addition, concern over the total amount of glucose that is absorbed systemically has become an increasing focus of attention in recent years, indicating that both diabetic and nondiabetic patients may benefit from a glucose-sparing approach to PD solution prescription (2).
Icodextrin substituted for glucose as the osmotic agent in dialysate for the long dwell exchange provides sustained ultrafiltration (UF) through colloid osmosis and offers the double advantage of reduced exposure to and absorption of glucose, along with a consistent reduction in extracellular fluid volume without the expected fall in urine output (3,4). In spite of its noticeable contribution, the osmotic effect of icodextrin in automated PD (APD)/continuous ambulatory PD (CAPD) patients has not been sufficient to achieve adequate fluid balance in many patients without resorting to hypertonic glucose during the daytime exchanges in CAPD or the overnight APD session. Indeed, without reducing fluid and sodium intake, a long dwell using 7.5% icodextrin may not achieve the required fluid and sodium balance in a number of anuric or fluid overloaded patients (5). Recently, two practical strategies designed to tackle the challenge of glucose exposure and UF management in APD were proposed: the first focused on reduction of peritoneal glucose load using commercially available amino acid or icodextrin PD solutions in combination with glucose-based PD solutions during overnight APD (6,7); the second was based on the previously investigated concept of a mixed crystalloid and colloid osmosis based on a bimodal solution obtained by mixing a small molecular component (glucose or amino acids) with a glucose polymer, represented by icodextrin, at formulations ranging from 2% to 7.5% during a long dwell exchange (8–11).
The recent work by Freida et al. (10) has extended further the potential of bimodal solutions using a combination of glucose and icodextrin as osmotic agent. These investigators added 200 mL 30% glucose to 2 L 7.5% icodextrin to prepare a bimodal solution containing 2.61% glucose and 6.8% icodextrin. Note that more glucose was added to these solutions than in the previous studies by the Sheffield group (8,11); this bimodal solution also had a reduced concentration of sodium (121 mEq/L). Ultrafiltration and sodium removal were compared in 7 APD patients during a 15-hour dwell using 3.86% glucose, 7.5% icodextrin, or the bimodal solutions. Only a single dwell of each solution was compared in each patient; all the patients had high transport status. The results showed an increase in mean net UF: from 462 mL using icodextrin to 990 mL using the bimodal solution. The mean increase in peritoneal sodium removal was from 49 mEq using icodextrin to 158 mEq using the bimodal solution. Thus, this bimodal solution was shown to approximately double net UF and triple sodium removal compared to that obtained using icodextrin solutions during a 15-hour dwell in APD patients (10).
The current protocol was designed to demonstrate the capability of this same bimodal solution to achieve a long-standing 4-month duration increase in UF and sodium removal while reducing the trade-offs of the therapy in terms of membrane protection and diminution of systemic consequences, specifically, its effect on systemic absorption of glucose in high and high-average transport APD patients. With respect to the osmotic composition of this novel PD solution, the innovative modality of UF that was investigated throughout the current study will be referred to as "bimodal UF."
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONDISCLOSUREREFERENCES |
|---|
|
Baseline prescriptions (Table 2) were not significantly different between the control and intervention groups with respect to session duration, total cycler therapy volume, fill volume per exchange in continuous cycling PD or initial fill volume and tidal volume in tidal PD, and last fill volume (7.5% icodextrin), with the exception of average nocturnal cycler dextrose concentration, which differed significantly between the two groups likely due to the different hypertensive profiles of the two groups. Blood pressure control required fewer antihypertensive medications in control patients than in the intervention group, which likely accounts for the prescription of more hypertonic glucose-based dialysate (higher glucose exposure) in the latter. Peritoneal membrane transport characteristics obtained from a peritoneal equilibration test (PET) performed with 2 L 3.86% glucose-based solution allowed the screening of 23 patients with above average peritoneal transport rates, resulting in 21 being eligible. Twenty-four-hour urine collection was requested during the screening period in an attempt to compare the involvement of diuresis and renal sodium elimination between the two groups throughout the study period. All 21 enrolled patients fulfilled the primary criteria of evaluation (intention-to-treat analysis); 14 patients completed the whole protocol (per protocol analysis).
|
STUDY DESIGN
The design of the study (Figure
1) included a 1-week screening period followed by a 1-month
baseline period. The screening period was used to determine eligibility for
study participation, to obtain informed consent, and to perform the screening
PET with 2.5 % Physioneal solution (Baxter, Castlebar, Ireland). Patients with
a peritoneal transport pattern below average (low and low-average groups) were
not eligible. All patients that met the entrance criteria used 7.5% icodextrin
for the long dwell for at least 1 month before the baseline visit. At
baseline, patients were assigned to receive either the mixed crystalloid and
colloid PD solution (glucose + polyglucose) or a control solution of 7.5%
icodextrin (Extraneal, Baxter) for the long dwell. The study was originally
designed as a randomized controlled trial; however, randomization proved to be
difficult in practice. Daytime fill volume ranged from 1.5 to 2.5 L and did
not vary for any patient during the study. The APD regimen, nocturnal and
daytime session durations, number of cycles, and infused nocturnal and daytime
volumes were kept fixed from baseline until the washout period.
|
Patients assigned to the bimodal PD solution (intervention group) were requested to use only 1.36% glucose-based dialysate at night but investigators were permitted to prescribe hypertonic glucose-based solution according to 24-hour UF needs during follow-up. Alternatively, the baseline nocturnal cycler dextrose concentration was continued throughout the study period in patients that received 7.5% icodextrin (control group) for the daytime dwell, although the investigators were also permitted to change the night prescription according to 24-hour UF needs during follow-up. The dual osmotic solution was prepared by mixing 200 mL 30% glucose (MacoPharma, Tourcoing, France) with 2.1 L (bag overfill) 7.5% icodextrin, which resulted in a 121 mEq/L low sodium formulation with 2.61% glucose and 6.8% icodextrin concentrations; the calculated osmolarity of this solution is 412 mOsm/L (10). Patients in the study were trained to do a safe transfer of the 30% glucose solution into the icodextrin bag using a Y-set tubing (Bieffe Medital/Baxter) with Luer-lock connection according to a procedure described elsewhere (10) (Figure 2).
|
This parameter will be comparable to that originally defined, assuming that the amounts of carbohydrate as glucose polymer absorbed from icodextrin and bimodal solutions are equivalent. Thus, it was assumed that the amount of icodextrin metabolized during the day dwell in patients treated with icodextrin solution (control group) was identical to that in patients assigned to the bimodal solution (intervention group). Therefore, only the direct absorption of free glucose from the dialysate was used to compare UF efficiency.
Because 24-hour glucose absorption could be determined only in a subgroup of patients treated at one institution, comparisons of 24-hour glucose absorption and UF efficiency were based on a subgroup of 14 patients (7 in the bimodal solution group and 7 in the icodextrin solution group) instead of the entire 21-patient group (10 in the bimodal group and 11 in the icodextrin group).
Statistical analysis of the data and the rationale for the methods were as follows. Since the sample size of this study was relatively small and there was some evidence of imbalance of hypertension-related risk factors between treatment groups at baseline (Table 1), the study end points were transformed into percent change from baseline on an individual patient basis. Additional analysis adjusting for the number of antihypertensive medications was conducted; the results were not materially changed and hence not presented. Generalized estimating equations with heterogeneous variances were used to evaluate the treatment effect between the bimodal and icodextrin arms for each study parameter. An autocorrelation model of order one [AR(1)] was selected to account for the temporal correlation among the repeated measures. Using likelihood ratio tests, a constant treatment effect model was chosen to be the parsimonious model since the simplified model appeared to be sufficient in describing the underlying biological process. No adjustment was made for multiple comparisons. Parameters after ending the study intervention (washout period) were recorded and tabulated; however, these data were not used in the main statistical analysis. All analyses were conducted using SAS 9.2 (SAS Institute, Cary, NC, USA).
Data presentation and statistical analyses in this repeated measures study design were complicated by the imbalance of factors between treatment groups at baseline. In order to be clinically relevant, the data are presented in absolute units as means and standard deviations at baseline and at each monthly follow-up visit. In order to be scientifically rigorous, however, statistical analysis of the percent change from baseline was used for the treatment comparison and the estimated treatment effect between the bimodal and icodextrin groups.
|
RESULTS |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONDISCLOSUREREFERENCES |
|---|
|
|
|
|
|
|
|
PRIMARY STUDY END POINTS
Long Dwell Exchange UF and Peritoneal Na Removal: Changes in the
primary study end points (net UF and peritoneal Na removal) during the long
dwell exchange differed between the control and bimodal groups. Both
parameters were about twofold higher than baseline during the 4-month
intervention period when treated by bimodal UF (Figures
3 and
4). The longitudinal changes in
these parameters for the control group were relatively small. Estimated mean
percent change from baseline in net UF for bimodal solution was 150%, versus
18% for icodextrin solutions. This difference was statistically significant
(p < 0.001). The estimated percent change from baseline in
peritoneal Na removal for bimodal solution was 147%, versus 23% for icodextrin
solutions. This difference was also statistically significant (p <
0.001; Table 3).
SECONDARY STUDY END POINTS
Twenty-Four-Hour UF and Peritoneal Na Removal: There was a
moderate change in 24-hour net UF in the icodextrin group during the study but
the use of bimodal solution resulted in a borderline significant increase in
24-hour net UF (Figure 5). The
estimated percent change from baseline in 24-hour net UF for bimodal solutions
was 48%, versus 0.2% for icodextrin solutions. The difference between
treatment groups was statistically significant (p = 0.03). Changes in
24-hour peritoneal Na removal during the study are shown in
Figure 6. The estimated percent
change from baseline in 24-hour peritoneal Na removal was 73% for bimodal
solutions, versus 47% for icodextrin solution. Although 24-hour peritoneal Na
removal was higher in the bimodal group than in the icodextrin group, this
difference was not statistically different (p = 0.54).
OTHER STUDY END POINTS
In a subgroup of 14 patients treated at one site, detailed observations
were recorded for 7 patients in the icodextrin group and 7 patients in the
bimodal group. Patients assigned to the bimodal group absorbed more glucose
during the initial baseline period than those in the icodextrin group
(Figure 7) because they were
using more hypertonic glucose solutions during the initial baseline period
(Table 2). No significant
difference was observed, however, in the percent changes in glucose absorption
from the initial baseline period between the bimodal and icodextrin groups
(Table 3). The estimated change
from baseline in 24-hour glucose absorption for bimodal solutions was
–5%, versus 5% for icodextrin solutions. This difference was not
statistically significant (p = 0.37). These observations of glucose
absorption are consistent with the study design. It was expected that patients
switching to bimodal solutions would absorb less glucose overnight because
nocturnal glucose exposure was decreased
(Table 2). In this patient
subgroup, therefore, the results suggest that the amount of glucose absorbed
from the long-dwell exchange bimodal solution during the study period was not
much different from the extra amount of glucose absorbed in the nighttime
exchanges during the initial baseline period.
Ultrafiltration efficiency was calculated by dividing 24-hour net UF by 24-hour glucose absorbed during the study, as described above; these calculated results are shown in Figure 8. The estimated percent change from baseline in UF efficiency for bimodal solutions was 71%, versus –5% for icodextrin solutions. This difference was statistically significant (p = 0.001). Ultrafiltration efficiency was higher during the study primarily because 24-hour net UF was higher in the bimodal group (Figure 5). Twenty-four-hour glucose absorption did not differ between the groups (Figure 7).
Of the 7 patients in the bimodal subgroup, 5 had residual urine volume, and of the 7 patients in the icodextrin subgroup, 6 had residual urine volume. Urine volumes were measured and are shown in Table 4. Residual urine volume was initially higher (p = 0.34) in the patients in the bimodal group and it decreased during months 1 – 4 from the initial baseline period in both groups. Indeed, when total fluid loss from the patient was considered (net UF plus residual urine volume), there were no observed differences between total fluid loss in the icodextrin and bimodal groups (detailed statistical analyses not shown).
|
Sodium loss in the urine was also determined in the patients from this single site; these data are also shown in Table 4. It appears there are clear differences between the icodextrin and bimodal groups. Sodium loss in the urine was initially higher in the patients in the bimodal group and it decreased during months 1 – 4 from the initial baseline period. Patients in the bimodal group were modestly heavier than those in the icodextrin group but substantial changes in body weight were not observed during the study.
Serum sodium levels and blood pressure were measured consistently throughout the study at only one study site. In this study subgroup, mean serum Na concentration was 135 or 136 mEq/L during each study period; no trends with time were noted. Further, systolic and diastolic blood pressure remained relatively unchanged throughout the entire study subgroup.
|
DISCUSSION |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONDISCLOSUREREFERENCES |
|---|
Of additional importance, this study demonstrated that a PD solution based on mixing glucose and icodextrin was capable of achieving the double target of significantly increasing UF and peritoneal Na removal and also explored a new concept of glucose-sparing PD therapy in a clinical setting for a period of 4 months. The concept of UF efficiency, which was originally defined as the amount of net UF obtained for every gram of carbohydrate absorbed from the dialysis solution, was more recently corroborated by the superiority of 7.5% icodextrin prescribed as the APD long-dwell solution over 4.25% dextrose in high and high-average transporters by Finkelstein et al. (15). During the current study, the demonstration of a 71% increase in UF efficiency with the mixed dialysis solution was the consequence of a 2.5-fold enhancement of daytime UF. Thus, all patients in the intervention group achieved a satisfactory 24-hour UF volume without needing prescription of hypertonic glucose during the nocturnal session. Therefore, the concept of bimodal UF potentially represents an important advance in attempts to minimize the systemic glucose absorption inherent with PD therapy. This minimization of glucose absorption was accompanied by decreased glucose exposure in many patients treated by bimodal solutions. For example, it can be estimated that glucose exposure will be less in patients treated using bimodal solutions if its use allows a change of more than 2 L 3.86% to 1.36% glucose solutions, or a change of more than 5 L 2.27% to 1.36% glucose solutions during nocturnal exchanges (both equivalent to a reduction of approximately 50 g of glucose exposure). There were reductions in both glucose absorption and glucose exposure in 6 of the 7 patients treated by bimodal solutions in the present study (where these data were simultaneously measured and could be directly compared). This decreased local exposure to and systemic absorption of glucose potentially reduces complications due to altered peritoneal membrane structure, as well as metabolic imbalances such as dyslipidemia, hyperinsulinemia, insulin resistance, oxidative stress, inflammation, and altered adipokine levels (2,16,17).
The mixed dialysis solution increased both UF and peritoneal Na removal by offering the opportunity to optimize the place of the long-dwell exchange in a complete 24-hour APD cycle. This is based not only on the dramatic efficacy of the combination of crystalloid and colloid osmosis per se during daytime dwells, but predominantly on its glucose-sparing effect allowing the prescription of a very limited, if any, volume of nonisotonic glucose dialysate at night. In 2000, Holmes and Shockley (18) described strategies to reduce glucose absorption during PD by combining non-glucose solutions with standard formulation. Subsequently, Holmes and Mujais proposed an approach to optimize dialysis prescription based on a quantification of glucose sparing (2). The authors emphasized that the impact of icodextrin on glucose sparing might go beyond the long dwell, as enhanced UF efficiency during the long dwell might lessen the burden of required UF during the short dwell and consequently lower glucose use in the short dwells. Furthermore, from a nutrition point of view, all carbohydrates are not created equal (19). For example, pure glucose is absorbed quickly, which can result in a rapid rise in both serum glucose and insulin. Others, such as complex carbohydrates, are absorbed much more slowly and produce only a modest blood sugar and insulin response. In addition, a modern approach to PD can optimize not only peritoneal UF but also Na removal with minimization of glucose use. In that respect, the concept of UF efficiency, which appears crucial to a glucose-sparing approach, must be extended to the amount of sodium removed. The central position of sodium ion as a determinant of extracellular volume legitimizes the necessity of elucidating the best strategy and the best solutions capable of achieving satisfactory control of sodium balance (5,20–22). This was highlighted in EAPOS, where the predominance of volume control over small solute elimination during APD in anuric patients was a determinant of patient outcomes (23). The enhancement of fluid and sodium transport along with the glucose-sparing advantage achieved throughout bimodal UF in the anuric PD patient population might potentially open a new era of peritoneal efficacy and tolerance.
Most papers in which urine volume or glomerular filtration rate were evaluated when prescribing icodextrin showed preservation of these parameters (1,3). A decrease in glomerular filtration rate was reported in one paper (24) but the authors later proposed underhydration as the reason for the decrease in glomerular filtration rate in that study (25). The patients in our study were fluid overloaded and hypertensive, and therefore in need of improved fluid status. Indeed, the data in Table 4 suggest that urine volume decreased more in the bimodal group than in the icodextrin group; however, a larger and longer study would be required to demonstrate that such differences were clinically significant. In the present study, 24-hour fluid and total Na removal did not increase in the bimodal group because of a reduction in residual renal function. At this time, one can envision that this mixed PD solution is well indicated for anuric patients being treated with PD and may be used with caution in patients with residual renal function. One potential negative feature of using bimodal solutions is the possibility that its use might lead to a more rapid decrease in urine volume and glomerular filtration rate in the first months of its use, whereas, in the long run, one could hypothesize that the reduction in glucose exposure and the correction of fluid status might have a positive impact on residual renal function and peritoneal UF.
Several limitations to this study should be emphasized. First, the randomization process was not successful due perhaps to the small number of patients studied in this trial. Second, more than half the patients in this trial were from a single center; full participation from additional centers was difficult because of the distance between the outlying centers and the main center in Cherbourg. Third, collection of all data from the follow-up visits was incomplete. It has been mentioned already that data detailing 24-hour glucose absorption from the peritoneal cavity was available only from the main center in Cherbourg; the same was true for additional data on glucose exposure, serum sodium concentration, and blood pressure during follow-up visits. These omissions are unfortunate but they do not detract from the conclusions derived from the parameters evaluated in this study.
Much research has emphasized the potential detrimental effect that glucose-based PD solutions may have on the peritoneal membrane while inducing systemic metabolic imbalance. Active volume control policy is now recognized as a first-rank intervention in patients treated with PD. The implementation of a combination of short-dwell APD with icodextrin for the long dwell already contributes to mitigation of the worse outcomes in high transport patients (13). At the present time, avoidance of hypertonic glucose solutions seems to be a more logical strategy to prevent the increase in peritoneal transport rate that will result in faster dissipation of crystalloid osmotic gradient — a direct consequence of increased membrane vascularity — and reduction in osmotic conductance, both of which impact UF capacity. So far, only a few small and short-term studies have been done showing the potential of this bimodal solution. This clinical study is the first time patients have been followed up for a period longer than 1 month using this bimodal solution and thereby bringing to daily clinical practice what, until now, was still on the bench. Moreover, the results are novel and even though the design is not perfect the improvements in UF and Na removal efficiency may suggest how a bimodal solution could benefit high and high-average transport patients on PD. These results should be reproduced in a randomized controlled trial where not only the local peritoneal potential benefits but also the systemic advantages of the bimodal solution can be evaluated. Due to its capacity to achieve most of the required fluid and sodium removal during the long dwell exchange, bimodal ultrafiltration might be confirmed as a powerful glucose-sparing strategy for volume control in high-transport APD patients.
|
DISCLOSURE |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONDISCLOSUREREFERENCES |
|---|
Received 24 February 2009; accepted 16 June 2009.
|
REFERENCES |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONDISCLOSUREREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. G. Blake NOVEL APPROACHES TO PRESCRIBING ICODEXTRIN Perit. Dial. Int., July 1, 2009; 29(4): 412 - 414. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
