TO: NDA 21-321, 7.5% Icodextrin Peritoneal Dialysis Solution

From: Stephen Fredd, M.D., HFD-110

Subject: Medical Review












Executive Summary                                                                                p.2

I. Introduction and Background                                                            p. 3

II. Clinically Relevant Information re Chemistry

      And Non-Clinical Pharmacology and Toxicology                          p.3-5

III. Human PK and PD                                                                           p.5

IV. Description of Clinical Data                                                             p.6

V. Clinical Review                     

     1. RD-97-CA-130                                                                                p.7-16

     2. RD-97-CA-131                                                                                p.17-33

     3. MIDAS                                                                                            p.34-37

     4. MIDAS-2                                                                                         p.38-39

     5. PRO-RENAL                                                                                  p.40-47

     6. DIANA                                                                                            p.48-51

     7. DELIA                                                                                             p.52

     8. RD-99-CA-060 and MIDAS Substudy                                         p.52

     9. IDEAL                                                                                             p.53-54

VI. Integrated Review of Efficacy                                                         p.55

VII. Integrated Review of Safety                                                           p.55-68

VIII. Dosing, Regimen, and Administration                                        p.69

IX. Use in Special Populations                                                               p.69

X. Conclusions and Recommendations                                                 p.69

XI. Labeling                                                                                             p.70-85































Chronic ambulatory peritoneal dialysis (CAPD) and automated peritoneal dialysis (APD) are currently performed using Dextrose in various concentrations as the osmotic agent to remove excess fluid and waste products from the system of End Stage Renal Disease (ESRD) patients. Because the osmotic gradient across the peritoneum decreases with Dextrose over the course of a long-dwell dialysis, Baxter has developed a different drug, Icodextrin, designed to maintain the gradient over the long-dwell period of peritoneal dialysis, and therefore increase the efficiency of dialysis.

Icodextrin is a high molecular weight glucose polymer derived from maltodextrin, and is administered with electrolytes. The dialysis solution contains 7.5 g  per 100 ml, and 2.0 or 2.5 L are used for the long-dwell period of dialysis. Approximately 30-40% of Icodextrin is absorbed from a single exchange depending on dwell time (between 8-16 hours), and is systemically hydolyzed to smaller oligosaccharides.

The clinical program compared the dialysis efficacy and safety to Dextrose for the long-dwell period in both CAPD and APD. No placebo controlled efficacy studies were performed.

Clinical studies 130, MIDAS and Pro-Renal demonstrated superiority of Icodextrin versus 1.5% or 2.5% Dextrose for net ultrafiltration and creatinine and urea clearance in  long-dwell CAPD or APD. Superiority of Icodextrin versus 4.5% Dextrose was not demonstrated. Data to demonstrate that the increased ultrafiltration and creatinine and urea clearances benefited the patients clinically were not convincing, but it was clear that Icodextrin was an effective dialysis drug.

A serious safety concern was raised by study 131, a 52 week safety study with mortality as the primary endpoint, in which the mortality data were adverse for Icodextrin compared to Dextrose. In the 13 month post-enrollment follow-up results, there were 20 Icodextrin deaths(n=175,11.4%) and 9 Dextrose deaths (n=112, 8%). Review of each case and exploratory subgroup analyses did not provide an explanation for the numerically adverse result. A pooling of all known deaths from all controlled trials did not replicate the adverse finding in study 131. Nevertheless, that result remains a concern.

Other adverse findings associated with Icodextrin were rash, decreased serum sodium and chloride, elevated alkaline phosphatase and AST (SGOT). A decline in serum amylase due to assay interference, and a slight decline in serum cholesterol was noted. No difference in effect on serum glucose, insulin requirements or HgA1C was found between treatments.

Considering the safety and efficacy data, a recommendation for approval for those patients inadequately responding to CAPD or APD with Dextrose for the long-dwell period is made. A post-marketing, long term, active-controlled, randomized mortality study should be considered.




On 12/20/2000, Baxter Healthcare Corporation submitted an NDA for Extraneal (7.5% Icodextrin with electrolytes) peritoneal dialysis solution for the treatment of chronic renal failure. The drug is a designated orphan drug. The total submission consists of 155 volumes, SAS data sets for the Phase III clinical trials, and pdf files for the case report forms. Certifications re financial interests and arrangements with clinical investigators, and patent information covering the formulation, composition and/or method of use are included.                                                                                                                                                        The medical portion of the submission includes volumes 1.1, 1.23-1.76 and 1.39A, and amendments dated 3/20/2001, 4/5/2001, 4/16/01, 4/18/01 and 5/3/01. On October 19,2000 a closed meeting of the CardioRenal Advisory Committee was held to discuss development of peritoneal dialysis solutions, at which meeting some of the data contained in this application was discussed. As a result of that discussion some additional data was gathered, analyzed and included in the application i.e. follow-up status of patients who participated in study RD 97-CA-131, a one year randomized safety study.

The drug is a new molecular entity, and has been approved in 17 European countries including the UK as well as in Canada. It is a designated orphan drug.





1. Chemistry

Icodextrin is a soluble glucose polymer derived from maltodextrin that in turn was derived by partial hydolysis of starch.  It has an average molecular weight of 12000-20000 Daltons, and its molecular structure is represented as follows:




















It is formulated as a 7.5% aqueous solution with electrolytes, and is manufactured by ML laboratories, PLC of Liverpool, England. The proposed fill volumes for various containers is 1.5L, 2.0L, and 2.5L. For the US, the composition of the electrolyte solution would be the same as for the currently available Dianeal  PD-2 . The formulation proposed for marketing is:


For further information, see chemistry review.


2. Non-Clinical Pharmacology and Toxicology

See Pharmacology review. Most studies were carried out using the ip route for 28 days, and blood levels were not detected in rats and low in dogs. A chart comparing those levels to man was provided as follows:



No carcinogenicity tests were performed. Ames test, CHO test and mouse micronucleus test were performed and no genotoxicity was observed. A reproductive study by ML Laboratories was not included in the NDA. The sponsor notes that maltodextrin is classied as GRAS as a food ingredient.

In metabolic animal studies it was shown that the route of elimination was renal, and icodextrin was hydrolyzed, probably by alpha amylase, to oligosaccharides including maltose and maltotriose.






One single-dose and several multiple-dose studies evaluated the PK of Icodextrin. See the Pharmacology review for details of those studies.

Since Icodextrin is directly instilled into the abdominal cavity, bioavailability is assured. In the abdominal cavity the drug works as a colloid osmotic agent to effect ultrafiltration in peritoneal dialysis. The osmotic pressure created by Icodextrin is thought to be relatively constant with little loss of osmotic gradient during long-dwells.

As noted in the preclinical section, Icodextrin is hydrolyzed by alpha amylase and smaller oligosaccharides such as maltose,  maltotriose and maltotetraose have been quantified in the plasma. Maltase further metabolizes the oligosaccharides to glucose.

During daily Icodextrin administration in single, long-dwell exchanges, plasma levels of 4-6.5 g/L of Icodextrin were found within one week and remained constant. Steady-state plasma levels of maltose ranged from 0.81 to 1.35 g/L. Steady state plasma levels of maltotriose were similar to maltose levels, and only small increases in plasma levels of larger metabolites were found. From a single dose of 150 g of Icodextrin, approximately 30-40% was absorbed, depending on dwell time. After discontinuation the plasma levels of Icodextrin and metabolites return to baseline in one to two weeks. Absorption from the peritoneal cavity into the blood follows zero order kinetics, and the drug is renally excreted, depending on residual renal function.




Clinical Trials


a)The sponsor identified 4 pivotal clinical trials which are outlined in the following chart.








Prospective, DB, Randomized comparison of Icodextrin and 2.5% Dextrose




CAPD patients

4 weeks

Net UF



Vol. 1.57-1.63

Open, Randomized comparison of Icodextrin and

1.5%,2.5%,4.5% Dextrose




CAPD patients

6 months

Net UF


Vol. 1.54-1.56

Open, Randomized comparison of Icodextrin and 2.5% Dextrose




APD patients

16 weeks

UF, Creatinine and Urea clearances


Vol. 1.38-1.53

Prospective, Randomized, DB

Comparison of Icodextrin and 2.5% Dextrose




CAPD and APD patients

52 weeks


Quality of Life


b) Supportive controlled clinical studies provided were:

ML/1B/004 (MIDAS-2): an open label long-term extension of MIDAS.

ML/1B/020 (DELIA): an open two-way crossover study comparing Icodextrin to a dry day.

ML/1B/011 (DIANA): an open, randomized comparison of Icodextrin to Dextrose in 38 APD patients for 2 years. 13 patients completed the 2 years.

RD-99-CA-060: an open single dose PK study of Icodextrin in a single exchange.

ML/1B/014: an open uncontrolled study of serum concentrations of drug and metabolites at steady state, after treatment cessation and after restarting.

ML/1B/002: an open randomized cross-over study of adding insulin to CAPD solutions in diabetics comparing Icodextrin and 1.5% glucose.

c) Cancelled studies due to slow enrollment were:

ML/1B/009 (IDEAL): an open, uncontrolled study that was to include 100 patients, but enrolled only 16 over more than a year.

PRO-RENAL-REF-037A: an open, uncontrolled study that was to include 80 patients but cancelled after 27 patients were enrolled. According to the sponsor analyses of the study are ongoing, and data were not included in the submission.


This review will consider the 4 pivotal studies in detail, and the others briefly.


Each of the clinical studies provided by the sponsor is summarized in the following clinical review.


1.  RD-97-CA-130:  This randomized, double-blind study of 7.5% Icodextrin peritoneal dialysis solution compared to 2.5% Dextrose peritoneal dialysis solution was initiated on April 1, 1998 and completed on December 29, 1998, and conducted in the US and Canada. Dianeal PD-2 was used in the US, and Dianeal PD-4 was used in Canada each with 2.5% Dextrose (2.27% glucose). The composition of the Icodextrin solution was:


That of the PD-2 and PD-4 solutions were:

The solutions were provided in Ultrabag, Twinbag or single bag configurations and the fill volume for each long-dwell dialysis was 2.0 or 2.5 liters.

                                                                                                                                                                        The study was designed as a non-inferiority trial which in the August 13, 1998 protocol amendment was defined as established if the difference between groups was within 30% of the mean ultrafiltration (UF) in the long-dwell exchange for the control group.  In the original December 5, 1997 protocol the non-inferiority definition used a 95% one-sided confidence interval with a lower bound greater than 150 ml. Secondary variables were peritoneal urea nitrogen exchange and peritoneal creatinine clearance.

175 patients were randomized: 90 to Icodextrin and 85 to control. Patients 18 years of age or older who had been on CAPD for at least 90 days, and who were treated by a long-dwell night exchange time of 12±4 hours with a fill volume of at least 2.0L but not more than 2.5L of 2.5% Dextrose were eligible. The randomization for each assignment was stratified for either 2.0 or 2.5L fill volumes.

Eligible patients also needed to be requiring a minimum of 4 peritoneal dialysis exchanges per 24 hour period, one of which was a night exchange. Allergy to starch-based polymers, liver disease, and women who were pregnant, lactating or not using acceptable birth control methods were among the exclusion criteria.


Patients continued the same formulation of Dextrose during the other dialysis periods. If a patient was taking Dianeal PD-4, which contains less calcium chloride than PD-2, for the other exchanges, he or she would, if randomized to Icodextrin, get the PD-2 composition of electrolytes for the long-dwell.


Net ultrafiltration was determined by subtracting the inflow amount from the total weight of the long-dwell collection.

The sponsor provided a flowchart of procedures as follows:












The term PET in this context means peritoneal equilibrium test, and the baseline QoL test was added in the August 1998 amendment after the study had begun for comparison with a follow-up QoL test to be administered to some patients participating in the long term 52 week safety study RD-97-CA-131.



The sponsor provided some baseline characteristics of the 175 randomized patients:

The groups were also well balanced at baseline for long-dwell glucose concentration and fill volume being used, as well as serum calcium.

Slightly more past episodes of peritonitis were reported for the Icodextrin group (1.2% versus 5.6%,p=0.076),  and more months had elapsed from the last exit site infection

for the Icodextrin group (mean 14.7 versus 9.1, p=0.062).

59 patients were diabetic (31 Icodextrin patients or 34.4% of the full cohort, and 28of the Dextrose patients or 32.9%), mostly type II.

Patient disposition was provided by the sponsor as follows:




Details of the reasons for withdrawal were provided:


Results were provided for the evaluable population with additional analyses to demonstrate that the ITT results did not materially differ. Compliance was estimated by the number of days the bag was used and was over 80% in both groups. Changes in daytime dialysis prescriptions were similar for the groups. Time of the long-dwell averaged 10.44 hours for control and 10.63 hours for Icodextrin at week 4. Fill volume at that time was 2.2L for each treatment, and drain volumes were 2.6 L and 2.8 L for control and Icodextrin respectively.





Primary Endpoint-Net UF


At baseline, 2 and 4 week data on dwell start and stop times, volume infused and volume drained were collected on each patient’s case report form. For weeks 2 and 4, data from 166 and 163 patients respectively were analyzed by the sponsor to provide the following net UF results:


Mean dwell time was 10.36 hours and mean volume was 2222.55 ml for all patients with a slightly longer dwell time recorded for the Icodextrin patients but no difference in fill volumes between groups.

One patient with an extreme value at week 2 was excluded in an alternative analysis with no change in results. Negative UF values occurred in 13.4% of the control patients versus 0% in the Icodextrin treated patients at 4 weeks (p<0.001).

Nondiabetics had slightly less net UF at 2 weeks versus diabetics, but similar results at 4 weeks. Icodextrin was significantly more effective than control at 2 and 4 weeks in both diabetics and nondiabetics with a somewhat larger net UF benefit for diabetics versus nondiabetics.






Negative UF

Analyses of negative and nonnegative ultrafiltration were carried out at baseline, 2 and 4 week timepoints with the following results:


Secondary Endpoints-Peritoneal Urea Nitrogen and Creatinine Clearance


Blood urea nitrogen and creatinine and dialysate urea nitrogen and creatinine data were collected at baseline, 2 and 4 weeks. Results of peritoneal urea and creatinine clearances (ml/min) were provided as follows:


These results support the sponsor’s contention that Icodextrin is an effective peritoneal dialysis solution, and provides more net UF and peritoneal urea and creatinine clearance than 2.5% Dextrose for long-dwell dialysis.





175 patients were exposed to either Icodextrin or 2.5% Dextrose. 84.4% of the Icodextrin and 89.4% of the Dextrose patients were treated for more than 27 days. No deaths occurred.



The reasons for withdrawal were previously presented but is repeated below:

Serious Adverse Reactions

18 patients (9-Icodextrin, 9-Dextrose) reported serious adverse events with 7 Icodextrin and 8 Dextrose patients being hospitalized. The following list provides a brief primary problem description for each case.



                       2.5% Dextrose

Confusion, hypercalcemia, Rocalcetrol dc’d

Chest pain, anterior MI

Unresponsive, seizures, bleeding

Abd. pain,nausea and vomiting, Kleb. in dialysate

Nausea and vomiting, bleeding

Flank pain, actinobacter in effluent

Elective renal transplant


Chest pain


Tremor, jerky movements, confusion, had Reglan


Chest pain, CAD



Syncope, overuse of 2.5% Dextrose to lose weight

Blocked catheter



Peritonitis seems to be more frequent as a cause of serious adverse events in the control group, but as noted for all adverse events whether serious or not, 13 or 15% was reported for the Dextrose group compared to 10 or 11% for the Icodextrin group.


Other Adverse Reactions

135 patients reported adverse reactions; 77 Icodextrin (85.6%) versus 58 control (68.2%), p=0.006.

Headache, rash and exfoliative dermatitis were more frequently reported in the Icodextrin group. The sponsor notes that rash and exfoliative dermatitis have been ascribed to the use of Icodextrin in the literature.

No difference between cohorts in the incidence of edema during treatment was noted.




Laboratory Findings

Significant changes from baseline values in each group and between groups were noted in the sponsor’s chart:


The sponsor claims that the decrease in sodium and chloride is dilutional due to the osmotic effect of Icodextrin and metabolites in the blood that drew water from the intracellular to intravascular compartment.

While the slight reduction in cholesterol for the Icodextrin treated patients had been previously reported, it was not thought to be of clinical significance nor is a mechanism for this effect suggested.

The slight decline in AST had not been previously reported had not been previously seen, but was not thought to be of clinical significance and the AST values were all within the normal range.

The decline in serum amylase was ascribed to assay interference, and was previously reported as due to competition by Icodextrin for the substrate used in the assay.

The increase in serum alkaline phosphatase had been previously reported in the PRO-RENAL study, was not associated with changes in other liver enzymes per the sponsor, and was not explained by assay interference. No explanation was proposed by the sponsor.

Another significant shift from baseline to either higher or lower levels was found in the platelet counts for the Icodextrin group. The shifts however were modest and not to levels of clinical concern. 

There were changes from baseline glucose in both cohorts mainly from high to normal with no significant difference between cohorts. HgA1C remained normal for both cohorts.


Finally it should be noted that some patients completing the study entered the long-term safety study, RD-97-CA-131.





2. RD-97-CA-131: This was a 52 week randomized, double-blind prospective safety study in 287 ESRD patients undergoing CAPD or APD. The study began on 4/1/98 and ended on 3/17/00. The original protocol was amended twice after study initiation. On 8/13/98 the protocol was amended to increase enrollment to allow inclusion of 60 patients from a European study that was never initiated, and  on 1/29/99 to increase enrollment by 75 patients to include patients from study RD-97-CA-130 on the same assignment as designated in that study. The products involved were the same as described above for study RD-97-CA-130.                                                                                                                                             The primary endpoints were safety endpoints including mortality rates, changes in membrane transport characteristics, adverse reactions, laboratory abnormalities, clinical signs such as edema. The protocol specified reasons for removal of patients from therapy or assessment. These included withdrawal due to adverse event, protocol deviation, transplantation, transfer to hemodialysis and death. For patients terminating prematurely follow-up evaluation was to be completed no more than two weeks following the last dose administered.

No efficacy data was collected, but QoL questionaires (KDQoL and SF-35) at baseline and at 13, 26, 39 and 52 week timepoints were added and evaluated for those who completed these at baseline and week 52.

The sizing of the study was based on mortality estimates for the two groups with the hypothesis being that the mortality rates would be comparable. Mortality rates were to be calculated by determining number of deaths of any patient during the treatment or follow-up periods of the study, and comparing the rates for each group. In the 1/29/99 amendment a secondary analysis of mortality was added which was to do a survival analysis of time to death using a logrank test.

The schedule of study procedures was:



158 new patients were randomized and enrolled. 129 patients from other studies as described above were included in the study but not rerandomized. The disposition of subjects in the study was:




Baseline demographics for the 287 patients entered into the study were:


Primary renal diagnoses were balanced for the two cohorts:



As per the sponsor, the number of patients remaining in the study at the scheduled visits were:



Exposure to study drug calculated from first dose to discharge or time in study was:









The protocol called for a 30 day follow-up post-withdrawal or completion,

In the original submission, it was stated that 5 patients in the control group and 13 in the Icodextrin group had died. This count was based on deaths occurring during the study or 30 days following the study or withdrawal. The sponsor’s initial listing and brief synopsis of each patient who died follows:



10 of the 13 deaths on Icodextrin died before 6 months, while 2 of the 5 control patients died in that timeframe.

Of the Icodextrin patients who died 8 (62%) were diabetics compared to 1 (20%) in the Dextrose group.

Alkaline phosphatase elevations were associated with transaminase elevations in 3 of the Icodextrin patients who died. History of hypertension and cardiovascular disease was frequently present in both cohorts. 5 Icodextrin patients who died had an episode of hypotension documented compared to 1 in the Dextrose group, however visits for evaluation in this study were infrequent.















The results for this initial mortality result were:


90% Confidence Intervals are presented as specified in the protocol to estimate whether the two drugs had similar mortality risk associated with their use. Based on the initial results, it could not be concluded that the risk was similar, although the numerical difference was not statistically significant.

At the October 19, 2000 CRDAC meeting it was suggested that for the mortality analysis follow-up of all randomized patients should be done for the 52 week duration of the trial plus 30 days.

The sponsor therefore amended the protocol to provide a 13 month follow-up for all patients.

In the March 20, 2001 submission the sponsor provided final mortality results which included follow-up results on all but 3 of the randomized subjects. In this new tally it was noted that 16 patients had died (9-Icodextrin, 7-Control) in addition to the 18 already reported. A brief narrative for each of the new patients reported dead follows. A star following the patient number indicates that the death occurred later than 13 months post-enrollment of follow-up.





Patients assigned to Icodextrin:

02401 was a 48 year old female Caucasian with type I diabetes. She entered on 12/14/98 and was withdrawn on 2/3/99 for pericarditis. She died on 6/7/99 from a CVA including intracranial hemorrhage.

19503 was a 50 year old Black male with hypertension who entered on 2/16/99 (BP 90/70). The patient was withdrawn on 4/20/99 due to a rash and itching (BP118/60). The patient died on 11/21/99 of some unspecified cardiac problem.

24204* was a 69 year old male Caucasian with type I diabetes who entered on 9/17/98 with a BP 141/68. On 11/20/98 complaints of hypotension, dizziness and chest pain were noted. The patient withdrew on 4/9/99 due to peritonitis, and died on 1/19/2000 of cardiac arrest. Death occurred after the 13 month post-enrollment period being considered.

24502* was a 62 year old male Caucasian who entered on 8/17/98 and was withdrawn on 9/12/98 due to appendicitis. He died on 7/21/2000 from cardiac arrest.

32301 was a 37 year old female Caucasian with type I diabetes. She entered on 2/16/99, had a myocardial infarction on 4/7/99 and was noted to have a problem with diabetic control on 5/20/99. She withdrew on 6/3/99 for muscle aches, and died on 12/5/99 from diabetes, severe peripheral vascular disease and withdrawal from dialysis.

35301 was a 70 year old female Caucasian with type II diabetes and hypertension. She entered 10/16/98, and withdrew on 1/14/99 after her husband’s death. She died 3/20/99 of renal failure.

38102 was an 82 year old male Caucasian with type II diabetes. He entered on 10/27/98 (BP140/70), and was noted to have hypotension on 11/4/98. On 5/10/99 fluid overload was found, and the patient was withdrawn on 5/20/99 for membrane failure. The patient died of ESRD on 7/9/99.

38103 was a 74 year old male Hispanic with hypertension. He entered on 10/27/98 (BP130/80), and developed hypotension and dehydration on 6/22/99. He was withdrawn for a cardiac mass on 7/7/99, and died on 10/25/99 with peripheral vascular disease noted.

61603 was a 66 year old male Caucasian with hypertension. He entered 11/13/98 (BP170/80), and withdrew for joint aches on 1/12/99. He died on 6/20/99 of cardiac arrest.

Patients assigned to control:

01501 was a 50 year old Caucasian female with type I diabetes. She entered on 8/17/98, was withdrawn on 3/5/99 for hypoglcemia, and died on 4/13/99 from unknown cause.

15202 was a 42 year old Black female with type II diabetes. She entered on 7/24/98 and was withdrawn on 8/15/98 for peritonitis. She died on 12/17/98 from unknown cause.

24202* was a 50 year old Black male with hypertension. He entered on 7/29/98 (BP 162/78), and died 12/9/99 of cardiac arrest/arrhythmia.

24501* was a 60 year old Caucasian male with type I diabetes. He entered on 7/31/98, and withdrew on 9/23/98 for peritonitis. He died on 9/10/99 from cardiac arrest.

30601* was a 69 year old Caucasian male with diabetes who entered on 11/16/98 and withdrew on 12/9/98 for unresolved peritonitis. He died on 5/26/00, no cause given.

32401 was a 64 year old Black male who entered on 1/14/99, withdrew on 10/14/99 for peritonitis, and died 1/7/2000 from cardiac arrest.

43403 was a 60 year old male Caucasian with type II diabetes.  He entered 3/8/99, withdrew 7/23/99, and died 3/2/2000 from multisystem organ failure.

Although 16 new deaths were reported (7 on control and 9 on Icodextrin), only 11 (4 on control and 7 on Icodextrin) occurred 13 months post-enrollment. Therefore, the sponsor’s analysis included 29 deaths: 20 (11.4%) randomized to Icodextrin and 9 (8.0%) to control. With these additional cases, the number of diabetics was 12 (60%) in the Icodextrin mortality group and 4 (44%) in the Dextrose group.








The sponsor provided a variety of analyses.

Their survival analysis indicating days to death or censoring was:




Mortality rates per-month and per-year with 90% confidence intervals were:






Since there was some numerical difference in mortality rates suggesting a possible increased risk with Icodextrin, numerous subgroup analyses were done. These should be considered exploratory, and since the overall result was inconclusive, such further analyses should be considered with more scepticism than usual.                                                                                                                                                            There were 4 prespecified randomized strata: 1) APD/2L, APD/2.5L, CAPD/2L, and CAPD/2/5L. The results for each stratum with 90%confidence intervals follows.


77 patients underwent APD; 41 assigned to Icodextrin and 36 to control. There were 40 males, 37 females with a mean age of 53.5 years. 52 were Caucasian, 19 Black, 4 Asian and 1 Hispanic.  22 had diabetic nephropathy, 17 hypertensive nephropathy, 12 glomerulonephritis, 3 autoimmune disease and 23 other.

22 were Canadians. The demographics were well balanced between treatments.

Of these 77 patients, 50 completed the study, 5 withdrew for transplantation, 13 for an adverse experience, 1 for a protocol violation, 4 for other reasons, and 4 died in the per-protocol analysis.

In most respects the APD cohort behaved similarly to the CAPD cohort, though the sponsor notes that there was less evidence of any favorable trend in weight maintenance for Icodextrin. Also, comparing the Icodextrin CAPD group with the APD group, larger decreases in blood glucose levels in the APD Icodextrin group were noted. 

The mortality rates with 90%CIs were:


For the APD/2L stratum;

For the APD/2.5L stratum:



210 patients undewent CAPD. 76 were assigned to Dextrose for the long-dwell, 134 were assigned to Icodextrin. The mortality rates for this cohort using the per-advisory committee follow-up database were.


For the CAPD/2L stratum, the results were:

For the CAPD/2.5L stratum, the results were:



Some patients entered from study 130 They continued on the assignment they were randomized to in that study and had successfully completed the 4 week treatment period of that study. To explore the mortality results of that cohort versus the newly randomized patients who entered study 131, the following analyses were done.








The mortality rate for the control patients was 8% for each cohort.

For those taking Icodextrin the mortality rate was 10.5% for study 130 patients, and 12% for study 131 patients. The difference is not large enough to suggest that results would have been significantly different had all new patients entered study 131.


One requested analysis was to compare mortality in those taking PD-2 electrolytes versus PD-4 electrolytes in the other exchanges. As previously stated, PD-4 has slightly less calcium chloride than PD-2, and might be selected for a patient who had elevated serum calcium. Since Icodextrin is supplied with PD-2 electrolytes only, during the long-dwell that patient would get the slightly greater amount of calcium chloride in that formulation.

The number taking PD-2 or PD-4 at entry was:

The mortality rate for those taking PD-2 at baseline was:

For those taking PD-4 at baseline the mortality rates were:

For the Icodextrin patients taking PD-2 at baseline the mortality rate was 6.9% versus 13.7% in those taking PD-4.


Since Canadians took only PD-4 at baseline a tabulation of their mortality rate was requested:

Their mortality rate was 17.3% on Icodextrin. Control mortality rate here was 7.1%.


The US results were also calculated:














Since many of those who died were diabetic, a comparison of mortality rates for diabetes and no diabetes at baseline was requested. For diabetics these results were:


For nondiabetics results were:

The mortality rate for those with diabetes at baseline and assigned to Icodextrin was 14.7% compared to 9.3% in nondiabetics assigned to Icodextrin.

None of these subgroup results are significant, and any hypotheses that might be considered would be speculative, needing prospective testing, at best.




The kidney disease quality of life (KDQoL) and short form 36 (SF-36) questionaires were used. The KDQoL form contained a 35 sympton/problem list. SF-36 had 36 questions about the patient’s general health covering mental and physical health. The protocol did not specify how these results were to be interpreted.

The KDQol results for the 66 patients (41-Icodextrin and 25-Dextrose) who completed baseline and 52 week questionaires as well as for 138 patients (63-Icodextrin, 75-Dextrose) for whom there was some data.There were no significant differences in overall score between treatments for either the KDQoL or SF-36 instruments for either cohort. Nor were changes from baseline to week 52 for individual questions such as soreness of muscles, trouble breathing significantly different between treatments.


A technical report from the Ovation Research group of Highland Park, Illinois analyzed the data for clinically significant differences. They state “The determination of a clinically meaningful change score (also referred to as a minimally important difference) is a relatively recent pursuit by HRQOL scientists, and, as such, more research will be required before validated rules can be established for all HRQOL measures.”    


They note that guidelines from the SF-36 developer suggest that a 5-10 point change in any subscale is clinically meaningful, and using a ¾5 point difference between groups they provided the following results from the KDQoL data:







From the SF-36 data they provided the following:


Additionally, a health transition frequency summary based on responses to the question “compared to one year ago, how would you rate your health in general now?” was provided for those with baseline and week 52 responses.


It is not clear why 40 Icodextrin patients rather than 41are included in this analysis, but whatever nominal significance was claimed in this analysis was not present at weeks 13, 26 and 39.





Serious adverse reactions were noted in 86 (51.2%) Icodextrin and 57 (51.4%) Dextrose patients. Hospitalization was the reason for classifying these events as serious in over 80% of cases with lesser percentages due to death or the life-threatening nature of the event. Hospitalization rates were similar for the two groups (17.1% for Icodextrin patients and 21.4% for the Dextrose patients). Events such as peritonitis, nausea and vomiting, MI were frequently noted in these patients as they were for other adverse reactions reported in this study.

60 patients (18.9% Icodextrin patients; 24.1% Dextrose patients) discontinued the study for adverse events. Most of these withdrawals were due to peritonitis, infection, dehydration, 5 Icodextrin patients withdrew for rash compared to none in the Dextrose group. Rash was reported 13 times in the Dextrose group versus 37 times in the Icodextrin group.



The frequencies of adverse reactions in the treatment and follow-up periods of the study was:


While overall the percentages in each costart category were similar, for specific terms where there was at least a 5% difference in incidence between groups, there were numerical differences that in most cases favored Icodextrin.




Concerning edema, the sponsor claimed a significant difference in the “no edema” category favoring Icodextrin at weeks 26 and 39 but not at week 52. As can be seen from the sponsor’s table with these results, the number of patients reporting varies from period to period.

As noted above, more adverse events termed edema were noted in patients on control versus Icodextrin.


The sponsor stated that body weight increases were observed in the Dextrose group throughout the study which were significantly different from the Icodextrin group which showed a slight gain only at week 4. Their conclusion was based on the following data:


These, however, were the before drain data. While the after drain data are sparse, those results do not support the sponsor’s conclusion.







Laboratory Findings

The following chart provides the sponsor’s assessment of significant changes from baseline in laboratory values over time. The week 4 baseline values include the data of patients from RD-97-CA-130 who were rolled over into study RD-97-CA-131.






As adverse events hyonatremia or hypochloremia were reported for 6.9% of the Icodextrin patients and 4.5% of the Dextrose patients. Increased alkaline phosphatase was reported in 6.9% and 5.4% of the Icodextrin and Dextrose groups respectively. One Icodextrin patient had cholestatic jaundice associated with the elevated alkaline phosphatase. Two Icodextrin patients who died were reported to have had elevated alkaline phosphatase. No significant differences between treatments in platelet shifts were reported in this study, but slight reductions in cholesterol at several timepoints were noted for the Icodextrin group.

The sponsor postulated that the increases in plasma osmolality were due to low molecular weight metabolites of Icodextrin.

While there was no significant difference in serum calcium levels between treatments over time in either this study or study 130, one could not rule out a detrimental effect in certain patients such as those with hypercalcemia. A review of the case report forms for those assigned to Icodextrin who had been on the low calcium PD4 solution before admission did not reveal data to suggest a problem due to use of the PD 2 solution for the long-dwell.




Results for the Peritoneal Equilibrium Test (PET) and the Mass Transfer Area Coefficient (MTAC) were presented.

PET is a test of peritoneal membrane transport of solutes and water. Dialysis/Plasma (D/P) ratios of urea, creatinine and glucose were determined at weeks 0, 26 and 52. If peritonitis developed, a minimum of 30 days had to elapse between resolution of the peritonitis and the PET. The PET was used to calculate the MTAC that provides a measure of diffusive solute mass transport based on membrane permeability and surface area.

The PET D/P ratios were not significantly different for each timepoint between treatments.

For glucose at week 52 there was a suggestion of a difference in the MTAC results as follows:




This safety study raises a serious question because of a unfavorable numerical mortality result suggesting that Icodextrin might increase the risk of death in ESRD patients compared to 2.5% Dextrose. While the difference in mortality was not statistically significant, neither does it support the conclusion that the two drugs are similar in terms of mortality risk.

The data were provided to support a clinical benefit related to Icodextrin administration, i.e. Quality of Life, edema status, and weight were not convincing due to incomplete data, inconsistency over time, and selection of timepoints, endpoints and conditions post-hoc.



3. ML/1B/001 (MIDAS): This was an open, randomized study performed at 11 centers in the UK with a product called Dextrin 20 that was essentially the same drug product as the Icodextrin formulation used in studies 130 and 131. Eligible patients were those adults on CAPD for at least three months, using 3-4 exchanges per 24 hours and free of peritonitis and mechanical drainage complications for at least 1 month prior to entry. The primary endpoint was the comparison of the median volume of ultrafiltrate at weeks 4, 13 and 21 (called special weeks by the protocol) produced after the long-dwell 12 hour dialysis with Dextrin 20 or the Dextrose solution (concentrations of 1.5%, 2.5% or 4.5%) as used prior to randomization for the ITT population (last value carried forward). Other secondary analyses involved different timepoints (weeks 3, 12 and 20 using an 8 hour long-dwell time), bag sizes (1.5 or 2.0L), subsets of the ITT population,and efficacy evaluable populations using the Bonferroni correction. Other analyses were done of “weak” glucose concentration i.e. 1.36%, “ medium” i.e.2.3%, and “strong” i.e. 4.25% versus Dextrin 20.


209 patients were randomized, 103 to control and 106 to Dextrin 20. Some demographic features of the population were:


    Control   n=103

  Dextrin 20  n=106

    Total   n=209

Age, mean in years(SD)

















The type and duration of renal disease was provided as follows:




The plan for the study was:

The monitoring schedule for RD-97-CA-131, the long term US safety study described above, included scheduled visits for the newly randomized patients at baseline, week 13 and every 13 weeks to week 52. Those entering from study RD-97-CA-130 had data at 2 and 4 weeks post baseline from that earlier study. As can be seen from the chart above, the UK study design had a more frequent visit schedule, and at each visit medical events and CAPD symptoms were assessed which formed part of the safety database.


The disposition of patients can be assessed from the chart below:


The case report forms included a temporary withdrawal form. Since patients were seen frequently where not only were diaries checked and symptoms elicited , where clinical circumstances indicated one temporary withdrawal period of up to a month to stabilize the patient on his/her usual pretrial CAPD regimen before reentering. To illustrate, patient 0608 was temporarily withdrawn on 6/21/91 for fluid overload, swollen ankles, shortness of breath and hypertension causing migraine, stabilized and reentered 7/15/91.


Results of the median ultrafiltrate volumes for the “total population and last values”at weeks 4, 13 and 21, where a 12 hour long-dwell time was used, were:

At weeks 3, 12 and 20 where an 8 hour long-dwell time was used the overall results were similar.

This analysis subdivided by weak (1.36%) or medium or strong (2.27 or 3.84%) glucose concentrations suggested a larger benefit of Dextrin 20 versus those in the weak concentration group.

The report does not provide the results for “medium” and “strong” glucose concentrations separately, and the literature report (Mistry et al: Kidney International, vol.46, 1994. Pp.496-503) gives the results as “weak: versus “strong” concentrations.


Regarding safety, there were 14 serious adverse events including 3 deaths; 2 in the Dextrose group and 1 in the Dextrin 20 group as detailed in the following chart:


Other findings of interest were 28 skin medical events (9 in control, 19 in Dextrin), and slight but statistically significant decline in serum sodium and chloride.



4. ML/1B/004 MIDAS-2 was a long-term open, uncontrolled treatment extension of 48 Dextrin patients of the 67 Dextrin assigned completers from MIDAS. The report of MIDAS-2 covered 10/91 to 3/95 at which point 3 patients were continuing.

Of the 48 patients enrolled 36 were male and 12 were female. Their average age was 57 years and 8 were diabetics.

Attrition over time was portrayed as follows:




25 serious adverse events were reported including 12 deaths (three occurring after discontinuing Dextrin in patients 0266, 0961, 1202), and 5 withdrawals. The sponsor’s listing of these cases was:

Patient 719 who died of a GI bleed should be included in the above listing.

A substudy of net ultrafiltration in 12 patients gave results as follows: 1 month 424ml±221sd (n=11); 3 months 418±195 (n=12); 6 months 493±197 (n=12); and 24 months 480±280 (n=12).

While uncontrolled, given the low mortality rate on Dextrin in the 6 months of the Midas study, this follow-up of  72% of the Dextrin completers from the MIDAS study showed no early increase in deaths that might suggest a reason for the low mortality found in MIDAS.








5. PRO-RENAL: This was an open, randomized study of Icodextrin versus 2.27% glucose all utilizing one 2 liter bag for the long-dwell day exchange in 39 chronic stable APD adult peritoneal dialysis patients. Patients who had been hospitalized, were pregnant or lactating, had chronic exit site infections, HIV positive as well as other reasons were excluded. The duration of the study was 16 weeks including a 2 week baseline period, a 12 week treatment period and a 2 week follow-up period during which all patients used the control solution for the long-dwell exchange.  The Icodextrin was provided as a single 2 L bag with the following composition:



The composition of the control solution was:

                                                                                                                                                                    The study began on 1/21/97 and ended on 1/12/98. Eight European centers including Germany, France, the Netherlands and Belgium participated. The primary efficacy measure was net ultrafiltration for the long-dwell exchange (14±2hours) with peritoneal clearance of creatinine and urea as secondary variables. The ITT population was defined as all randomized patients and at least 1 long-dwell dialysis with the assigned solution. The evaluable population completed the 2 week baseline period and at least the first 6 weeks of the treatment period. Change from baseline was assessed at weeks 1, 6 and 12 with between treatment results analyzed.

Safety was assessed during the study period and any patient experiencing a serious adverse event was followed-up for 3 months. In addition to the usual laboratory tests, the protocol included assessments of carbohydrate absorption, changes in insulin requirements for diabetic subjects and was amended to include determination of the sodium content of the dialysate during the long-dwell exchanges. This was added to assess whether the decrease in serum sodium with Icodextrin noted in other studies was due at least in part to greater loss of sodium during the treatment dialyses.


The flowchart of procedures was:



41 patients were screened, and 39 patients entered: 19 assigned to Dextrose and 20 to Icodextrin. Patient disposition was noted as follows:


The baseline characteristics of those randomized were:


As noted under renal diagnosis, 4 diabetics entered.  2 were assigned to Dextrose and 2 to Icodextrin.




The results for the primary efficacy variable of net UF were:


The means of the long-dwell time and of the infused volumes was presented in the following two charts.


For the secondary variables of peritoneal creatinine and urea clearances respectively, results were:



As with other studies, this study confirmed that for the long-dwell dialysis Icodextrin removes more fluid, creatinine and urea than 2.27% glucose. The sponsor noted that the result exceeded the 250 ml difference proposed in the protocol as the smallest meaningful clinical difference, however data were provided that a clinical benefit was associated with this physiological change.



Of the 39 randomized patients, 6 withdrew for adverse reactions for the following reasons:



While 1 death was noted during the trial, there were 2 other deaths in the Icodextrin group shortly after the treatment period. No deaths occurred in the Dextrose group. A brief narrative of those deaths follows.

Patient 0304 was a 29 year old Caucasian male who had a history of hypertension for 2 years prior to entry on 4/8/97. On 5/27/97 BP was 150/80. Developed acute cardiac failure on 5/29/97 and died. Post mortem showed marked LVH and circulatory failure was listed as the cause of death. No acute MI was found.   

Patient 0801 was a 53 year old Caucasian female who entered the trial on 10/21/97, developed hypertension on 2/2/98, was switched to the control solution and became normotensive on 2/9/98. No date or cause of death given.

Patient 0212 was a 59 year old Caucasian diabetic male who entered the trial on 11/6/97, was noted to have hypertension on 2/4/98 and had a stroke on 2/15/98. Presumably that was the cause of death on 2/21/98. 6 additional patients with serious adverse reactions were identified; 5 taking Dextrose and 1 on Icodexrtin. The reactions were extraosseous calcification (history of hyperparathyroidism), inguinal hernia, hyperhydration, hernia, stomach perforation in the Dextrose group, and peritonitis in the Icodextrin group. These reactions were classified as serious because of the need for hospitalization.

56 other adverse reactions were noted; 25 in the control group and 31 in the Icodextrin group, most thought unrelated to drug treatment.

Laboratory abnormalities were found for serum sodium, chloride, alkaline phosphatase, serum amylase, and serum AST (SGOT) in direction and degree consistent with the findings of other studies.



The explorations of both serum and peritoneal sodium were more extensive with the following graphs depicting the patterns of change over time.



The change from baseline for carbohydrate absorption was somewhat greater in the Icodextrin arm compared to the Dextrose arm (+8gms/long-dwell versus +0.3gms/long-dwell, p=0.003).

Of the 4 diabetic patients, 1 in the Dextrose arm required an increase in insulin.

These safety data are not reassuring, since there was a numerically greater number of deaths in the Icodextrin group compared to control in this study where reasonably frequent clinical observations were made.























6. ML/1B/011 (DIANA): This was an open, randomized parallel study of Icodextrin 7.5% versus Dextrose (1.36% or 2.27% or 3.86%) for long-dwell dialysis in 38 adult ESRD adult APD patients. The electrolytic composition of the Control in mmol/liter was Sodium 132, Chloride 102, Calcium 1.75, Lactate 35, and Magnesium 0.75. For Icodextrin it was Sodium 133, Chloride 97, Calcium 1.75, Lactate 40, and Magnesium 0.25.



The duration of the study was two years, and it was conducted in the Netherlands at two hospitals (Rotterdam and Haarlem).  The primary purposes of this study were to:

A.      Evaluate the safety, efficacy and biocompatability of Icodextrin compared to Dextrose.

B.       To evaluate whether there was less damage to host resistance (macrophage function, peritonitis episodes) and to the peritoneal membrane with Icodextrin.

C.       To assess “whether the glycation of peritoneal membrane” was less with Icodextrin.


The following evaluations were to be made at approximately 3 month intervals:


No formal estimation of sample size was done, and corrections for multiple endpoints were to be made appropriately.



The demographics of the 38 randomized patients were:










The number of patients assessed at each timepoint was:











N on Icodex.










N on Control











ITT and completer analyses were done, but insufficient data was collected for the biocompatability endpoints specified in the protocol, such as IL-6 or TNF-", and where sufficient data were available, such as with macrophage function, no differences between groups were noted.

The only efficacy endpoint that did show a significant difference was net UF. At baseline both groups showed negative net ultrafiltration volumes. While this negative direction continued in those receiving Dextrose for the log-dwell dialysis, the UF volumes became positive for those assigned to Icodextrin.


Of greatest interest in this study are the safety results.

5 patients died during the study and 1 died two and one-half months after withdrawal from the study. All were in the Dextrose group. A brief cause of death for each patient follows.

A006/0006 66 year old Caucasian male-infection and sepsis after toe amputation.

A017/0017 66 year old Caucasian male-CVA.

A019/0019 68 year old Asian female-dehydration, hypotension, transferred to nursing home and died 4 months later.

A025/0025 57 year old diabetic Caucasian male-peritonitis due to bowel ischemia.

A027/0027 67 year old Caucasian male-myocardial infarction.

B010/0110 55 year old Caucasian male-acute necrotic pancreatitis.


While the Pro-Renal study results included 3 deaths all on Icodextrin, this study result shows an opposite numerical direction.


In addition to the deaths, 7 patients withdrew from the Dextrose group and 12 from the Icodextrin group. The major reason given was transplantation.






7. ML/1B 020 (DELIA): This was an open, single center randomized crossover study in 11 adult ESRD undergoing APD which compared Icodextrin to a dry dwell. The design of the study was as follows:


Of the 11 randomized patients, all were Caucasian, 3 were male and mean age was 51.9 years±13.6. 7 patients completed both study periods, and diabetes was the most frequent cause of the renal disease.

For the completers, there was no significant difference in 24 hour total ultrafiltration volume for the Icodextrin arm versus the control arm. There was a significant increase in creatinine clearance during the Icodextrin treatment compared to control (47.4l/week±12.0 versus 29.5±8.7,p<0.01).

Concerning safety, there were no deaths reported, there were 7 serious adverse reactions reported (4 in the Icodextrin period). 6 patients withdrew for peritonitis or diarrhea, 4 during the Icodextrin period.



8. RD-99-CA-060 and ML/1B/014 (MIDAS Substudy) were two pharmacokinetic studies; the first of a single Icodextrin exchange, and the second of Icodextrin levels at steady state, after stopping and after restarting Icodextrin. These will be reviewed in the Biopharmaceutics review, as will study ML1B/002 that evaluated insulin absorption when administered intraperitoneally during CAPD with Icodextrin or glucose.




9. ML/1B 009 (IDEAL): This was an open, noncomparator study of Icodextrin in CAPD patients who had lost ultrafiltration across the peritoneum as defined by a PET study. Although the study planned to enroll 100 patients at 10 European centers and treat patients for 6 months, it was stopped after 16 patients enrolled in over a year. No efficacy data are presented, but safety data were reported.

8 males and 8 females entered at two centers; one in London, the other in Brussels. They ranged from 19 to 78 years of age. No other demographic data were presented.

Patient disposition was detailed as follows:



5 patients died. Those were:

patient 305-50 year old diabetic Caucasian female died after a myocardial infarction.

patient 307-77 year old Caucasian female had peritonitis and was not responding to dialysis.

patient 308-78 year old Caucasian male died after a myocardial infarction.

patient 1104-75 year old Caucasian male died in his sleep.

patient 1105-45 year old Caucasian male died after a cardiac arrest.

Each of these patients had a history of cardiovascular disease.


Other serious adverse reactions were reported by 5 patients. These included hypertension, CVA, overhydration, diabetic management problem, exit site infection, and peritonitis.

Without a randomized control group, it is difficult to assess the significance of these safety results.




The three controlled efficacy studies, 130, MIDAS, Pro-Renal, demonstrate That Icodextrin is an effective peritoneal dialysis drug, and is superior to 1.5% and 2.5% Dextrose for ultrafiltraton amounts and creatinine and urea peritoneal clearance during the long-dwell period for CAPD and APD. Icodextrin long-dwell dialysis would be integrated into a daily treatment regimen which would still employ Dextrose for the other dialyses.

None of the patients entered were doing  poorly on their regimen which consisted of Dextrose for all dialyses, but greater volumes of fluid and waste products were removed when Icodextrin was substituted for the long-dwell. The sponsor does not make a convincing case that this represents a clinical benefit were everyone to be treated with the new drug. In some cases excess fluid removal could lead to dehydration, hypotension, electrolyte imbalance. What attempts were made to show a clinical benefit, e.g QoL results, edema status, were not convincing because of incomplete cohort results, post-hoc selection of time points and scales, and inadequate statistical consideration of non-preplanned endpoints and multiple comparisons.

That is not to say that the sponsor needs to prove that fluid and waste removal in ESRD is beneficial. Compared to historical expectations it is clear that Icodextrin is an effective dialysis drug, but compared to a currently used and well-tolerated drug there is no convincing data to demonstrate clinical superiority.




840 patients were included in the sponsor’s integrated summary of safety; 493 assigned to Icodextrin and 347 to control. The breakdown by study was as follows:


The duration of exposure was:













Some demographic characteristics were:


Diabetic, hypertensive and hypertensive nephropathy subpopulations were represented as follows:

















Disposition of patients was:


The mortality comparisons did not include all patients who died during and following the study. This will be presented later.


For adverse events occurring in 5% or more of patients by treatment group and all patients were presented as follows:



Some events of interest were peritonitis, hyperglycemia, edema and rash.

Episodes of peritonitis were similar between groups as was hyperglycemia. Edema was more frequently noted in the control group, and has been discussed in the results of individual studies.

Rash was approximately twice as frequent in the Icodextrin treated patients and deserves further comment.






A breakdown by study of skin adverse events leading to discontinuation was:


All were treated with Icodextrin although patient 307 had been assigned to Control in MIDAS-1 and Icodextrin in MIDAS-2.

For skin events judged related to study drug, 14 patients on control (4.0%) reported such events versus 49 patients on Icodextrin (9.9%; p<0.001).

By gender the following skin events were noted:


Exfoliative dermatitis and rash appear to be more frequent complaints in females.


Adverse events leading to discontinuation of the drug assigned were:


Proportions were similar between groups, though rash was a more frequent reason in Icodextrin treated patients.




By gender and treatment group the comparative incidence of selected adverse events was:


Hypertension seemed more frequent in the Icodextrin treated females versus control females, but the hypotension result was in the opposite direction. Nausea was also somewhat more frequent in Icodextrin treated females compared to control treated females, but was similar to males treated with Icodextrin. Rash was most frequently reported in Icodextrin treated females.


In the geriatric population versus all patients, the results were:



Most results between groups were of comparable frequency.

Concerning race comparative results for Caucasian and Black were provided:


Exit site infections seemed least frequent in Blacks on Icodextrin, while headache was most frequent in this group. Rather than put credence in these findings, one should be very cautious in any of the many numerical differences found in these exhaustive comparisons.





For diabetics versus all patients the results for selected adverse events were:


Since Icodextrin was developed with one expectation that it would provide less glucose load to diabetics and therefore be better tolerated, it is interesting that hyperglycemia was slightly more frequently reported in diabetics taking Icodextrin. Rash was also more frequent in this group.


Metabolic events were also compared in diabetics versus all patients:



For many comparisons diabetics had events more frequently than all patients (and nondiabetics by subtraction). Hypercalcemia and hyperglycemia was somewhat more frequent in diabetics treated with Icodextrin. Hypokalemia was more frequently reported in diabetics on the control drug.


The sponsor provided summary information on the following laboratory parameters: serum sodium, serum chloride, alkaline phosphatase, serum amylase, and osmolality which results are presented below.


Serum Sodium


Serum Chloride


Alkaline Phosphatase



Serum Amylase



Plasma Osmolality


In studies 130 and 131 decreases in serum sodium and chloride as well as serum cholesterol, amylase and AST (SGOT), and increases in alkaline phosphatase and plasma osmolality were statistically significant. The decreases in serum sodium and chloride are more likely due to increased loss in the dialysate during Icodextrin treatment per the results in the Pro-Renal study.













The mortality result of study 131 has been discussed in that section above.

The initial results were provided for those who died during the study or within 30 days after completion or withdrawal. These initial results were:




With follow-up for 13 months post-enrollment of all randomized patients, the results were:




For all deaths reported in study 131, the results were:




In this summary section all deaths reported in the controlled studies submitted in the NDA were evaluated.                                 Overview tables follow:

Controlled Studies                                                              


N  in Study

Duration of Study


Deaths         %


Deaths            %



52 weeks

22              12.6

12                 10.7



6 months

  1                0.9

2                    1.9           



16 weeks

  3              15

0                    0



2 years

  0                0

6                    31.6






Uncontrolled Studies


N in Study

Duration of Study


N                      %



6 months

5                       31

Midas II


53 months

12                     25






For all controlled studies, the sponsor provided an analysis of all deaths as follows:





These pooled results do not support an increased mortality risk in patients treated with Icodextrin compared to control. None of the individual study or pooled mortality comparisons were statistically significant. However, no study was sized to demonstrate a significant difference, and the adverse numerical result in study 131 was something of a surprise. While the most likely explanation for that result is chance, what was set out to be demonstrated, i.e. that Icodextrin and control had similar mortality risks, was not demonstrated. Rather than dismiss the study 131 finding, an additional long-term mortality study should be considered with the objective to rule out some predetermined mortality risk increase, taking into account the size of the study needed to do that.



Each 100ml of the Icodextrin peritoneal dialysis solution contains 7.5g of Icodextrin. For each 2L long-dwell dialysis, 150 g of Icodextrin would be given. Of this 30-40% is absorbed depending on the duration of the long-dwell (12±2hours generally for CAPD). For 2.5L , 187.5g of Icodextrin would be given. Since efficacy was demonstrated for both 2L and 2.5L bag sizes and no dose-related toxicity was identified, the selection of what dose to give a particular patient can be based on clinical judgment.

Concerning efficacy relevant to the duration of the long-dwell, in MIDAS 8 hour dwells were used during weeks 3,12 and 20, while 12 hour dwells were used during weeks 4, 13 and 21. A 14±2 hours long-dwell time was used in Pro-Renal, and 12±4 hours was used for study 130.   The glucose concentrations used in MIDAS were “weak”, i.e. 1.36% glucose, or “medium” or “strong”, i.e. 2.27 or 3.86% glucose respectively.

The comparative results were provided as follows:

Interpolating between trials it can be inferred that duration of the long-dwell does not much affect the net ultrafiltration benefit of Icodextrin versus 1.5% or 2.5% Dextrose.


Since the proposed market image of Icodextrin will include only the PD-2 electrolyte solution with contains 25.7mg/100ml calcium chloride, those patients who are taking Dextrose with PD-4 solution containing 18.3mg/100ml calcium chloride will be given the slightly higher calcium dose for the long-dwell. No case has been identified where this was associated with an adverse reaction. However, since a particular patient may be affected by the higher calcium dose, physician’s should be informed that use of Icodextrin for the long-dwell will include the slightly higher calcium dose as well.



No children have been studied.                                                                                                             Analyses of the studies by age, Caucasian or Black race, gender, diabetic status, and hypertension have been carried out. All showed a similar direction of net ultrafiltration benefit compared to the total randomized population.  Concerning Asian or Hispanic patients, too few were included to draw meaningful conclusions about efficacy or safety.




Icodextrin is an effective peritoneal dialysis drug based on historical expectations of patient outcome without it. It is more effective than 1.5% and 2.5% Dextrose in net ultrafiltration during the log-dwell dialysis period, but it has not been shown in the studies provided to provide a clinical benefit. However, it would be useful to have an alternative dialysis drug available for patients not adequately responding to their current regimen.

From a safety perspective, the mortality results in study 131 remain a concern. While this might be due to chance, it would be advisable to repeat that study. In addition to the mortality results, other findings such as rash and the laboratory abnormalities associated with the drug should be noted in the labeling.

Therefore, approval is recommended for the treatment of ESRD patients undergoing CAPD or APD during the single daily long-dwell 8-16 hours periods for those not adequately responding to their current regimen. A phase 4 commitment to repeat a mortality study similar to study 131.




The sponsor’s draft labeling with additional comments on the medical sections follow.

Package Insert Sections



MedicalReviewer Comments


(7.5% Icodextrin) Peritoneal Dialysis Solution



EXTRANEALÔ (7.5% Icodextrin) Peritoneal Dialysis Solution is a peritoneal dialysis solution containing the colloid osmotic agent icodextrin.  Icodextrin is a starch derived, water soluble glucose polymer linked by alpha (1-4) and alpha (1-6) glucosidic bonds with a weight average molecular weight between 12,000 and 20,000 Daltons and a number average molecular weight between 5,000 and 6,500 Daltons.  The representative structural formula of icodextrin is:




Each 1 liter of Extraneal contains:     

Electrolyte content per 1 liter:



Icodextrin                          75.0 g

Sodium Chloride                   5.4 g

Sodium Lactate                    4.5 g

Calcium Chloride               257 mg

Magnesium Chloride          51 mg


                Sodium                   132 mEq/l

                Calcium                  3.5 mEq/l

                Magnesium           0.5 mEq/l

                Chloride                 96 mEq/l

                Lactate                   40 mEq/l



Water for Injection, USP     qs

HCl/NaOH may have been used to adjust pH

Extraneal contains no bacteriostatic or antimicrobial agents.

Theoretical osmolarity:  285-288 mOsm/L;  pH=5.2
































































Extraneal is available for intraperitoneal administration only as a sterile, nonpyrogenic, clear solution in 1.5 L, 2.0 L and 2.5 L Ambu-Flex IIIÔ and UltrabagÔ containers.  The container systems are composed of polyvinyl chloride.






Mechanism of Action

Extraneal is an isosmotic peritoneal dialysis solution containing glucose polymers (icodextrin) as the primary osmotic agent.  Icodextrin functions as a colloid osmotic agent to achieve sustained ultrafiltration during long peritoneal dialysis dwells.  Icodextrin acts in the peritoneal cavity by exerting osmotic pressure across small intercellular pores resulting in a steady rate of transcapillary ultrafiltration throughout the dwell.  Extraneal also contains electrolytes to help normalize electrolyte balance and lactate to help normalize acid-base status.















Pharmacokinetics of Icodextrin


Absorption of icodextrin from the peritoneal cavity follows zero-order kinetics consistent with

convective transport via peritoneal lymphatic pathways.  In a single-dose pharmacokinetic study

using Extraneal, a median of 40.1% (60.2 g) of the instilled icodextrin was absorbed from the

peritoneal solution during a 12-hour dwell.









Plasma levels of icodextrin rose during the dwell and declined after the dwell was drained, consistent with a one-compartment model with zero order absorption and first order elimination.  Peak plasma concentrations (median Cpeak = 2.23 g/L) were observed at the end of the long dwell exchange (median Tmax = 12.7 hours) with plasma levels returning to baseline values within 3 to 7 days following cessation of icodextrin administration.  Icodextrin had a plasma half-life of 14.7 hours and a median clearance rate of 1.08 L/hr.









The mean steady-state plasma levels of icodextrin predicted from the above parameters (5.26 g/L) corresponded very closely to the stable plasma icodextrin values observed during long-term administration.





In multidose studies, steady-state levels of icodextrin were achieved within one week and returned to baseline within one week after discontinuation of Extraneal use.







Icodextrin is metabolized by alpha-amylase into oligosaccharides with a lower degree of polymerization (DP), including maltose (DP2), maltotriose (DP3), maltotetraose (DP4), and higher molecular weight species. In a single dose study, DP2, DP3 and DP4 showed a progressive rise in plasma concentrations with a profile similar to that for total icodextrin, with peak values reached by the end of the dwell and declining thereafter. Only very small increases in blood levels of larger polymers were observed.








Icodextrin is not completely metabolized. Provide data on steady-state blood levels.




Steady-state plasma levels of icodextrin metabolites were achieved within one week and stable plasma levels were observed during long-term administration.




Some degree of metabolism of icodextrin occurs intraperitoneally with a progressive rise in the concentration of the smaller polymers in the dialysate during the 12-hour dwell.





Icodextrin undergoes renal elimination in direct proportion to the level of residual renal function (r=0.824 vs creatinine clearance, p<0.01).  In nine patients with residual renal function (mean creatinine clearance: 5.0 ± 1.5 ml/min), the average daily urinary excretion of icodextrin was 473 ± 77 mg per ml of creatinine clearance.  Diffusion of the smaller icodextrin metabolites from plasma into the peritoneal cavity is also possible after systemic absorption and metabolism of icodextrin.








Special Populations


In clinical studies of Extraneal in which plasma levels of icodextrin and its metabolites were measured, 95 patients were aged 65 and older.  No apparent differences in plasma levels were observed in patients aged 65 and older as compared to patients under age 65.














Gender and Race

Although no specific studies were conducted to evaluate the differences between gender and race within the clinical trial data for icodextrin, no known differences have been detected.







If no PK or PD studies to investigate gender or race differences were done, we cannot conclude that we know there are no differences.

Pharmacodynamics and Clinical Effects

Extraneal has demonstrated efficacy as a peritoneal dialysis solution in clinical trials of approximately 400 patients studied with end-stage renal disease (ESRD).





Ultrafiltration, Urea and Creatinine Clearance, Negative Net Ultrafiltration

In active controlled trials from one to six months in duration, Extraneal used once daily for the

long dwell in either continuous ambulatory peritoneal dialysis (CAPD) or ambulatory peritoneal

dialysis (APD) therapy resulted in higher net ultrafiltration and clearances when compared with

2.5% Dextrose solutions.









In 175 CAPD patients randomized to Extraneal (N=90) or 2.5% Dextrose solution (N=85) for the 8-15 hour overnight dwell for one month, mean net ultrafiltration for the overnight dwell was significantly greater for the Extraneal group compared to the 2.5% Dextrose group when evaluated at weeks 2 and 4 (Figure 1).


Figure 1 - Mean Net Ultrafiltration for the Overnight Dwell (RD-97-CA-130)



In 39 APD patients randomized to Extraneal or 2.5% Dextrose solution for the long, daytime dwell (10-17 hrs) for three months, the average net ultrafiltration reported during the treatment period was 278 ± 43 ml for the Extraneal group and –138 ± 81 ml for the Dextrose group (P<0.001).








































Mean creatinine and urea nitrogen clearances were significantly greater for Extraneal as compared with 2.5% Dextrose in CAPD patients at weeks 2 and 4 (Figure 2) and in APD patients at weeks 6 and 12 (P<0.001).


Figure 2 – Mean Creatinine and Urea Clearance for the Overnight Dwell





What of comparisons to 1.5% and 4.5% Dextrose?

Extraneal resulted in a significant decrease in the percentage of patients with negative net UF during long peritoneal dialysis dwells (10-17 hrs).  When compared to 2.5% Dextrose solution, the percentage of patients who were unable to achieve positive or zero ultrafiltration was significantly lower for patients using Extraneal for the long dwell in both CAPD and APD.






Long-term (12 month) Use

A randomized 12-month safety study (N=287) evaluated a single daily exchange of Extraneal for the 8 to 16-hour dwell in ESRD patients using CAPD or APD.  One hundred seventy-five (175) patients were randomized to Extraneal and 112 patients to 2.5% Dextrose.


Body Weight: Long-term use (12 months) of Extraneal resulted in maintenance of stable body weight compared to a mean weight gain of 2.3 kg in the 2.5% Dextrose group. The lack of weight gain observed in the Extraneal group may be related to a reduction in the glucose load during long dwells.


Fluid Balance: Significantly fewer patients receiving Extraneal reported edema at Weeks 26 and 39 during the 12-month study when compared to patients on 2.5% Dextrose (20% vs 35%). Overall, 17.9% of patients in the control group reported peripheral edema as compared to 6.3 % in the Extraneal group.


Peritoneal Membrane Transport Characteristics: After one year of treatment with Extraneal during the long dwell exchange, there were no differences in membrane transport characteristics for urea and creatinine.  There was a slight increase in the mass transfer area coefficient (MTAC) for glucose at one year, but it was not different from the change in MTAC in patients receiving treatment with 2.5% Dextrose solution for the long dwell.


Quality of Life: Quality of life in the 12-month study was assessed by the Kidney Disease Quality of Life (KDQoL) evaluation.  When asked to evaluate their general health at study completion, versus their baseline assessment, a significantly greater percentage of patients in the Extraneal group (30%) responded that their health was “much better now than one year ago” compared to the Control group (4%) (p<0.03).



























The primary endpoint was mortality. Provide mortality results with the extended follow-up data.


Delete body weight, fkuid balance, and Quality of Life claims. Not established by the data as discussed in the review of study 131.


Extraneal is indicated for a single daily exchange for the long (8 – 16 hour) dwell during continuous ambulatory peritoneal dialysis (CAPD) or automated peritoneal dialysis (APD) for the management of chronic renal failure.


In clinical studies, Extraneal demonstrated enhanced ultrafiltration and creatinine and urea clearances when compared to 2.5% Dextrose solutions.  The percentage of patients with net negative ultrafiltration was significantly reduced with Extraneal compared to 2.5% Dextrose (See CLINICAL PHARMACOLOGY –Pharmacodynamics and Clinical Effects).





Add: in patients not adequately responding to their current dialysis regimen.

Delete second paragraph since clinical benefit was not established in the studies.



Extraneal is contraindicated in patients with a known allergy to cornstarch or icodextrin or in patients with glycogen storage disease.






Not for intravenous injection.







Peritoneal Dialysis Related

All peritoneal dialysis solutions, including Extraneal, should be used with caution in patients with a history of abdominal surgery within thirty days of commencement of therapy, abdominal fistulae, tumors, open wounds, hernia or other conditions which compromise the integrity of the abdominal wall, abdominal surface or intra-abdominal cavity.  Caution should also be used in patients with conditions that preclude normal nutrition, patients with impaired respiratory function, and patients with potassium deficiency.


Aseptic technique should be employed throughout the peritoneal dialysis procedure to reduce the possibility of infection.  If peritonitis occurs, the choice and dosage of antibiotics should be based upon the results of culture and sensitivity of the isolated organisms.  Prior to identification of involved organisms, broad-spectrum antibiotics may be indicated.




Patient’s volume status should be carefully monitored to avoid hyper- or hypovolemia and potentially severe consequences including congestive heart failure, volume depletion and hypovolemic shock.  An accurate fluid balance record must be kept and the patient’s body weight monitored.




Significant losses of protein, amino acids, and water-soluble vitamins may occur during peritoneal dialysis.  The patient’s nutritional status should be monitored and replacement therapy provided as necessary.


Extraneal solution should be inspected for clarity, absence of particulate matter and container integrity.  Solutions, which are cloudy, contain particulate matter, or evidence of leakage should not be used.


Treatment should be initiated and monitored under the supervision of a physician knowledgeable in the management of patients with renal failure.



Add:In  patients with hypercalcemia, particularly in those on low calcium peritoneal dialysis solutions, consideration should be given to the face that Icodextrin peritoneal dialysis solution is not provided with low calcium electrolyte solution.

Insulin dependent diabetes mellitus


Patients with insulin dependent diabetes may require modification of insulin dosage following

initiation of treatment with Extraneal.  Appropriate monitoring of blood glucose should be

performed and insulin dosage adjusted if needed (See Drug /Laboratory Test Interactions).




Information for Patients

Patients should be instructed to inspect each container of Extraneal solution for clarity, particulate matter, color and integrity of the container prior to use.  Solutions should not be used if they are cloudy, discolored, contain visible particulate matter or if they have evidence of leaking containers.


Aseptic technique should be employed throughout the procedure.


To reduce possible discomfort during administration, patients should be instructed that solutions may be warmed to 37°C (98°F) prior to use.  Only dry heat should be used. It is best to warm solutions within the overwrap. To avoid contamination, solutions should not be immersed in water for warming.  Do not use a microwave oven to warm Extraneal. Heating the solution above 40°C (104°F ) may be detrimental to the solution.  (See Directions for Use)


Additional information for patients is provided at the end of the labeling.




Laboratory Tests

Serum Electrolytes

Decreases in serum sodium and chloride have been observed in patients using Extraneal.  The declines in serum sodium and chloride may be related to dilution resulting from the presence of icodextrin metabolites in plasma.  Although these decreases have been regarded as clinically unimportant, monitoring of the patients’ serum electrolyte levels as part of routine blood chemistry testing is recommended.


Extraneal does not contain potassium.  Evaluation of serum potassium should be made prior to administering potassium chloride to the patient.











Alkaline Phosphatase

An increase in mean serum alkaline phosphatase has been observed in clinical studies of ESRD patients receiving Extraneal. No associated increases in liver function tests were observed.  Serum alkaline phosphatase levels did not show evidence of progressive increase over a 12-month study period.  Levels returned to normal approximately two weeks after discontinuation of Extraneal.







There have been individual cases where elevated alkaline phosphatase has been associated with elevated AST(SGOT), but neither elevation was thought t to be causally related to the drug.

Drug Interactions


No clinical drug interaction studies were performed.  No evaluation of Extraneal’s effects on the cytochrome P450 system was conducted.  As with other dialysis solutions, blood concentrations of dialyzable drugs may be reduced by dialysis.  Dosage adjustment of concomitant medications may be necessary.  In patients using cardiac glycosides, plasma levels of calcium, potassium and magnesium must be carefully monitored.





A clinical study in 6 insulin dependent diabetic patients demonstrated no effect of Extraneal on insulin absorption from the peritoneal cavity or on insulin’s ability to control blood glucose when insulin was administered intraperitoneally with Extraneal.  However, appropriate monitoring  (See Drug /Laboratory Test Interactions) of blood glucose should be performed when initiating Extraneal in diabetic patients and insulin dosage should be adjusted if needed (See Precautions).













No human drug interaction studies with heparin were conducted.  In vitro studies demonstrated no evidence of incompatibility of heparin with Extraneal.






No human drug interaction studies with antibiotics were conducted.  In vitro studies evaluating the minimum inhibitory concentration (MIC) of vancomycin, cefazolin, ampicillin, ampicillin/flucoxacillin, ceftazidime, gentamicin, and amphotericin demonstrated no evidence of incompatibility of these antibiotics with Extraneal. (See Dosage and Administration)












Drug/Laboratory Test Interactions

Blood Glucose

Blood glucose measurement must be done with a glucose specific method to prevent maltose interference with test results.  Glucose dehydrogenase pyrroloquinolinequinone (GDH PQQ) based methods should not be used.












Serum Amylase

An apparent decrease in serum amylase activity has been observed in patients administered Extraneal.  Preliminary investigations indicate that icodextrin and its metabolites interfere with enzymatic based amylase assays, resulting in inaccurately low values.  This should be taken into account when evaluating serum amylase levels for diagnosis or monitoring of pancreatitis in patients using Extraneal.








Carcinogenesis, Mutagenesis, Impairment of Fertility

Icodextrin did not demonstrate evidence of mutagenic potential in in vitro or in vivo studies performed.  Long-term animal studies to evaluate the carcinogenic potential of Extraneal or icodextrin have not been conducted.   Icodextrin is derived from maltodextrin, a common food ingredient that is generally regarded as safe.






A preliminary fertility study in rats revealed slightly low epididymal weights in parental males in the high dose group (1.5 g/kg/day), as compared to Control.  Toxicological significance of this finding was not evident as no other reproductive organs were affected and all males were of proven fertility.  Studies on the effects of icodextrin on male and female fertility have not been performed.







Pregnancy Category C

Complete animal reproduction studies have not been conducted with Extraneal or icodextrin.  Thus it is not known whether icodextrin or Extraneal solution can cause fetal harm when administered to a pregnant woman or affect reproductive capacity.

Extraneal should only be utilized in pregnant women when the need outweighs the potential risks.




A preliminary study of the effects of icodextrin on the fertility and pregnancy in rats demonstrated no effects of treatment with icodextrin on mating performance, fertility, litter response, embryo-fetal survival, or fetal growth and development.




Nursing Mothers

It is not known whether icodextrin or its metabolites are excreted in human milk.  Because many drugs are excreted in human milk, caution should be exercised when Extraneal is administered to a nursing woman.




Pediatric Use

Safety and effectiveness in pediatric patients have not been established.





Geriatric Use

No formal studies were specifically carried out in the geriatric population.  However, approximately 25% of the patients in clinical studies of Extraneal were age 65 or older, with ~ 4% of patients age 75 or older.  No overall differences in safety or effectiveness were observed between these patients and patients under age 65.  Although clinical experience has not identified differences in responses between the elderly and younger patients, greater sensitivity of some older individuals cannot be ruled out.















Adverse Reactions from Clinical Trials

Significance of Adverse Reaction Data Obtained from Clinical Trials

Because clinical trials are conducted under widely varying conditions, adverse reaction rates observed in clinical trials of a drug cannot be compared to rates in the clinical trials of another drug and may not reflect the rates observed in practice.  The adverse reaction information from clinical trials does, however, provide a basis for identifying the adverse events that appear to be related to drug use and for approximating rates.




Extraneal was studied in controlled clinical trials of 366 patients with end-stage renal disease, including 60 patients exposed for 6 months and 155 patients exposed for one year. The population was 18-93 years of age, 56% male and 44% female, 73% Caucasian, 18% Black, 4% Asian, 3% Hispanic and included patients with the following comorbid conditions:  26.8% diabetes, 49.3% hypertension and 23.1% hypertensive nephropathy.  All patients received a single daily exchange of Extraneal for the long dwell (8-16 hours).



N=493 patients in ISS. Update.

Rash was the most frequently occurring icodextrin-related adverse event (5.5%, Extraneal; 1.7% Control). A listing of adverse events reported in these same clinical studies, regardless of causality, occurring in > 5% of patients is presented in Table 1.





Give withdrawal for rash data including exfoliative dermatitis. Give results by gender as well.


Additional adverse reactions that were possibly, probably or definitely related to Extraneal with an incidence of less than 5% within each body system were as follows:

Body as a Whole - neck pain, PD catheter dysfunction, facial edema, bloody effluent;

Cardiovascul - postural hypertension, tachycardia, cardiovascular disease, syncope, cerebrovascular accident, palpitations; Hematologic and Lymphatic - leukocytosis, eosinophilia; Digestive - anorexia, abnormal liver function, constipation, gastrointestinal disorder, flatulence, gastritis, intestinal obstruction, stomach ulcer; Metabolic and Nutrition - dehydration, hypovolemia, hypochloremia, hypomagnesemia, weight increase, increase alkaline phosphatase, hyponatremia, hypoglycemia, increase SGOT, increase SGPT, decreased weight, decreased ultrafiltration, increase creatinine; Musculoskeletal - myalgia, cramps, leg cramping, bone pain; Nervous -  paresthesia, dry mouth, anxiety, hyperkinesia, nervousness, abnormal thinking; Respiratory - lung disorder, lung edema, hiccup; Skin - exfoliative dermatitis, nail disorder, psoriasis, macular-papular rash, eczema, furunculosis, bulbar vesicular rash, skin discoloration, dry skin, skin ulcer, urticaria; Special Senses - loss of taste; Urogenital - kidney pain.


Use data whether or not considered related.


Table 1 - Adverse Experiences in >5 % of Patients





N = 366





N (%)

N (%)


Body in General











130 (26.4)

88 (25.4)



Exit Site Infection

73 (14.8)

58 (16.7)




48 (9.7)

43 (12.4)




43 (8.7)

23 (6.6)



Pain Abdominal

39 (7.9)

20 (5.8)



Flu Syndrome

35 (7.1)

21 (6.1)



Injury Accidental

31 (6.3)

14 (4.0)




28 (5.7)

27 (7.8)



Lab Test Abnormal

25 (5.1)

12 (3.5)



Pain Chest

25 (5.1)

12 (3.5)



Pain Back

22 (4.5)

18 (5.2)




21 (4.3)

19 (5.5)








62 (12.6)

29 (8.4)




32 (6.5)

37 (10.7)










40 (8.1)

33 (9.5)




35 (7.1)

17 (4.9)



Nausea /vomiting

25 (5.1)

21 (6.1)




25 (5.1)

13 (3.7)




22 (4.5)

19 (5.5)


Hematologic & Lymphatic






55 (11.2)

39 (11.2)


Metabolic and Nutrition








34 (6.9)

37 (10.7)




34 (6.9)

32 (9.2)




28 (5.7)

20 (5.8)




28 (5.7)

17 (4.9)




25 (5.1)

26 (7.5)




25 (5.1)

12 (3.5)



Peripheral Edema

18 (3.7)

29 (8.4)








31 (6.3)

27 (7.8)








27 (5.5)

19 (5.5)








Upper Res Infection

74 (15.0)

46 (13.3)



Cough increase

35 (7.1)

13 (3.7)




26 (5.3)

24 (6.9)








50 (10.1)

16 (4.6)




27 (5.5)

23 (6.6)



Skin Disorder

11 (2.2)

18 (5.2)































Peritoneal Dialysis Related

Adverse events common to the treatment modality of peritoneal dialysis including peritonitis, infection around the catheter, fluid and electrolyte imbalance, and pain were observed at a similar frequency with Extraneal and Controls (See Precautions).





Changes in Alkaline Phosphatase and Serum Electrolytes

An increase in mean serum alkaline phosphatase has been observed in clinical studies of ESRD patients receiving Extraneal. No associated increases in liver function tests were observed.  Serum alkaline phosphatase levels did not show evidence of progressive increase over a 12-month study period.  Levels returned to normal approximately two weeks after discontinuation of Extraneal.










Decreases in serum sodium and chloride have been observed in patients using Extraneal.  The declines in serum sodium and chloride may be related to dilution resulting from the presence of icodextrin metabolites in plasma.  Although these decreases have been regarded as clinically unimportant, monitoring of the patients serum electrolyte levels as part of routine blood chemistry testing is recommended.






There has been no observed potential of drug abuse or dependence with Extraneal.





No data is available on experiences of overdosage with Extraneal.  Overdosage of Extraneal may result in higher levels of serum icodextrin and metabolites.  It is unknown what symptoms may be caused from exposure in excess of those observed in clinical trials.  In the event of overdosage with Extraneal, continued peritoneal dialysis with glucose-based solutions should be provided.








Extraneal is intended for intraperitoneal administration only.  It should be administered only as a

single daily exchange for the long dwell in continuous ambulatory peritoneal dialysis or

automated peritoneal dialysis.   The recommended dwell time is 8 to 16 hours.


Patients should be carefully monitored to avoid under or over hydration.  An accurate fluid

balance record must be kept and the patient’s body weight monitored to avoid over or under

hydration and potentially severe consequences including congestive heart failure, volume

depletion and hypovolemic shock.


Aseptic technique should be used throughout the peritoneal dialysis procedure.


To reduce possible discomfort during administration, patients should be instructed that solutions

may be warmed to 37°C (98°F) prior to use.  Only dry heat should be used.  To avoid

contamination, solutions should not be immersed in water for warming.  Do not use a microwave

oven to warm Extraneal.  Heating the solution above 40°C (104°F ) may be detrimental to the

solution.  (See Directions for Use)


Extraneal should be administered over a period of 10-20 minutes at a rate that is comfortable for

the patient.


Parenteral drug products, including Extraneal, should be visually inspected for particulate matter,

leakage and discoloration prior to use.  Should these be present, discard product; do not use.


Following use, the drained fluid should be inspected for the presence of fibrin or cloudiness,

which may indicate the presence of an infection.



































Addition of Insulin

Addition of insulin to Extraneal was evaluated in 6 insulin dependent diabetic patients undergoing CAPD for end stage renal disease. No interference of Extraneal on insulin absorption from the peritoneal cavity or on insulin’s ability to control on blood glucose was observed (See Drug /Laboratory Test Interactions).  Appropriate monitoring of blood glucose should be performed when initiating Extraneal in diabetic patients and insulin dosage adjusted if needed (See Precautions).












Addition of Heparin

No human drug interaction studies with heparin were conducted.  In vitro studies demonstrated no evidence of incompatibility of heparin with Extraneal.






Addition of Antibiotics

No formal clinical drug interaction studies have been performed.  In vitro compatibility studies with Extraneal and the following antibiotics have demonstrated no effects with regard to minimum inhibitory concentration (MIC): vancomycin, cefazolin, ampicillin/flucoxcillin, ceftazidime, gentamicin, and amphotericin. 

Patients undergoing peritoneal dialysis should be under careful supervision of a physician experienced in the treatment end-stage renal disease with peritoneal dialysis.  It is recommended that patients being placed on peritoneal dialysis should be appropriately trained in a program that is under supervision of a physician.  Training materials are available from Baxter Healthcare Corporation, Deerfield, IL 60015, USA.








Directions for Use

For complete CAPD and APD system preparation, see directions accompanying ancillary


Aseptic technique should be used.





For patient comfort, Extraneal can be warmed to 37°C (98°F).   Only dry heat should be used.  It is best to warm solutions within the overwrap.  Do not immerse Extraneal in water for warming. Do not use a microwave oven to warm Extraneal.  Heating above 40°C (104°F) may be detrimental to the solution.







To Open

To open, tear the over wrap down at the slit and remove the solution container.  Some opacity of the plastic, due to moisture absorption during the sterilization process, may be observed.  This does not affect the solution quality or safety and may often leave a slight amount of moisture within the overwrap.



Inspect for Container Integrity

Inspect the container for signs of leakage and check for minute leaks by squeezing the container firmly.




Adding Medications

Some drug additives may be incompatible with Extraneal.  See DOSAGE AND ADMINISTRATION section for additional information.  If the re-sealable rubber plug on the medication port is missing or partly removed, do not use the product if medication is to be added.

1.             Prepare medication port site.

2.             Using a syringe with a 1-inch long, 25 to 19-gauge needle, puncture the medication port and inject additive.

3.             Reposition container with container ports up and evacuate medication port by squeezing and tapping it.

4.             Mix container thoroughly.




Preparation for Administration

1.             Place Extraneal on flat surface or suspend from support (depending on ancillary equipment).

2.             Remove protector from outlet port on container.

3.             Attach solution transfer set.  Refer to complete instructions with ancillary equipment or transfer set.

4.             Discard any unused portion.




Extraneal (7.5% icodextrin) Peritoneal Dialysis Solution is available in the following containers and fill volumes:




Fill Volume




1.5 L

NDC 0941-0679-51



2.0 L

NDC 0941-0679-52



2.5 L

NDC 0941-0679-53



1.5 L

NDC 0941-0679-45



2.0 L

NDC 0941-0679-47



2.5 L

NDC 0941-0679-48



Each liter of Extraneal contains 75 grams of icodextrin in an electrolyte solution with 40 mEq/l lactate.


Extraneal should be stored at controlled room temperature 68–77°F (20–25°C).  Store in moisture barrier overwrap in carton until ready to use.


Avoid excessive heat (104°F/40°C) and protect from freezing.


Rx Only