Skip Navigation


NDT Advance Access originally published online on November 22, 2006
Nephrology Dialysis Transplantation 2007 22(3):870-879; doi:10.1093/ndt/gfl654
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
22/3/870    most recent
gfl654v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (7)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Selby, N. M.
Right arrow Articles by McIntyre, C. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Selby, N. M.
Right arrow Articles by McIntyre, C. W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author [2006]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

The haemodynamic and metabolic effects of hypertonic-glucose and amino-acid-based peritoneal dialysis fluids

Nicholas M. Selby1, Jana Fialova1, James O. Burton1 and Christopher W. McIntyre1,2

1Department of Renal Medicine, Derby City Hospital, Derby and 2Centre for Integrated Systems Biology and Medicine, University of Nottingham, UK

Correspondence and offprint requests to: Dr C. McIntyre, Department of Renal Medicine, Derby City Hospital, Uttoxeter Road, Derby, DE22 3NE, UK. Email: Chris.McIntyre{at}derbyhospitals.nhs.uk



   Abstract
Top
 Abstract
Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
Background. Continuous ambulatory peritoneal dialysis (CAPD) may exert significant effects on systemic haemodynamics. We have previously demonstrated that hypertonic glucose solutions are associated with higher blood pressure (BP) due to a rise in stroke volume (SV) and cardiac output (CO). However, the mechanisms underlying these changes have not been established.

Methods. Ten non-diabetic CAPD patients entered a randomized crossover study (eight completed) to compare conventional glucose-based fluid, biocompatible pH-neutral glucose-based fluid and 1.1% amino acid solution (lactate-buffered but completely free of glucose degradation products). BP and haemodynamic variables were measured using continuous arterial pulse wave analysis, and serial plasma glucose and insulin concentrations were collected. Left ventricular (LV) diameters were measured at the start and end of each dwell period using M-mode echocardiography.

Results. BP was similar during 1.36% glucose and 1.1% amino acid dwells, but was significantly higher during 3.86% glucose dwells with both conventional and biocompatible fluids (P < 0.001). This was associated with a significantly higher SV and CO (P < 0.001), although the haemodynamic response differed between conventional and biocompatible 3.86% solutions. Plasma glucose and insulin levels did not differ from baseline during 1.36% and amino acid dwells, but increased significantly during 3.86% glucose dwells. Despite a significantly higher ultrafiltration volume with 3.86% glucose, LV diameters were similar throughout.

Conclusions. In conclusion, we have confirmed our previous findings demonstrating higher BP and adverse haemodynamics during 3.86% glucose dwells. These changes are associated with hyperglycaemia and hyperinsulinaemia, but are not related to differences in cardiac filling.

Keywords: amino acids; blood pressure; glucose; haemodynamics; insulin; pulse wave analysis



   Introduction
Top
 Abstract
 Introduction
Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
It is becoming increasingly recognized that continuous ambulatory peritoneal dialysis (CAPD) exerts significant effects on blood pressure (BP) and systemic haemodynamics [1, 2]. Previously, we have demonstrated that BP rose during 3.86% glucose dwells, and this effect was abrogated by using a non-glucose, high ultrafiltration (UF) alternative (7.5% Icodextrin) [2]. The higher BP was due to higher heart rate (HR), stroke volume (SV) and cardiac output (CO) during 3.86% glucose dwells and occurred despite higher UF volumes. However, in that initial study, it was not possible to determine the mechanisms underlying these changes. One possible explanation was that the changes were due to the effects of greater systemic absorption of glucose from the 3.86% glucose solution, leading to hyperglycaemia and hyperinsulinaemia. Certainly, both hyperglycaemia and hyperinsulinaemia are recognized to elevate BP and exert independent effects on systemic haemodynamics [3, 4]. However, insulin levels were not measured and some of the included subjects were diabetic. Alternatively, the greater UF and therefore intraperitoneal volumes with 3.86% glucose may have increased venous return and therefore cardiac filling. In addition, the fluids differed not only in glucose content but also in buffer type and in the amount of glucose degradation products (GDPs).

Conventional CAPD fluids are bioincompatible due to high glucose concentration, presence of GDPs and the low pH of a lactate buffer. High glucose and GDP levels have both been linked to adverse effects on the peritoneal membrane, which may ultimately lead to UF failure [5]. In addition, there is evidence that GDPs disappear from the peritoneal cavity and enter the circulation, raising the possibility that conventional PD fluids may have adverse systemic effects, beyond those on the peritoneal membrane [6]. Registry data has suggested that conventional solutions may even have a negative impact on patient survival, although this must be interpreted within the limitations of a retrospective observational study [7]. These findings have driven the development of more physiologic fluids. These include bicarbonate/lactate pH neutral fluids that are low in GDPs, and are therefore more biocompatible despite being glucose based. There is also a commercially available solution containing 1.1% amino acids as the osmotic agent (Nutrineal®, Baxter Healthcare, Norfolk, UK) and is therefore entirely free of glucose and GDPs. However, this latter solution contains lactate as the buffer and so has an acidic pH.

We performed a study to determine whether the acute effects on BP and haemodynamics observed with hypertonic glucose dialysate were related to systemic absorption of glucose and subsequent hyperinsulinaemia, differences in cardiac filling or differences in buffer type or biocompatibility.



   Subjects and methods
Top
 Abstract
 Introduction
 Subjects and methods
Results
 Discussion
 Acknowledgements
 References
 
Patients
We recruited 10 non-diabetic patients on CAPD for a prospective crossover study. Baseline characteristics and CAPD prescriptions are shown in Table 1. All patients had been on CAPD for more than 6 months (mean 26 ± 26 months, range 6–89).


View this table:
[in this window]
[in a new window]

  		Table 1. Baseline characteristics of study patients

 
Patients were eligible only if their BP had been stable (BP <140/85 mmHg with no changes in anti-hypertensive medications) over the 4 weeks prior to recruitment, and if <50% of their PD regime was made up of 3.86% glucose solution or Icodextrin. Patients were excluded if they had diabetes mellitus, severe peripheral vascular disease, or if they had an arterio venous fistula or renal transplant in situ.

All patients underwent clinical examination prior to commencing the study to ensure that they were at their optimal weight, and had standard peritoneal equilibration testing (PET) and assessment of dialysis adequacy (Adequest 2.0® program, Baxter Healthcare, Norfolk, UK).

Study protocol
Study design is summarized in Figure 1. Patients attended for three study days (A, B and C), the order of which was randomly determined. For each investigatory session, patients were admitted to a clinical investigations unit where CAPD was performed. All fluids were manufactured by Baxter Healthcare (Norfolk, UK) and were warmed to 37°C before instillation. Non-invasive haemodynamic monitoring was undertaken using a Finometer, which was fitted for the entirety of each session. To obtain baseline values, monitoring commenced 30 min prior to draining the night-time dwell.


Figure 1
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. Schematic representation of the study design. PDFCONV, conventional glucose-based peritoneal dialysis solution; PDFBIO, biocompatible glucose-based peritoneal dialysis solution; PDFAA, 1.1% amino-acid-based peritoneal dialysis solution.

 
On day A, conventional bioincompatible fluids (PDFCONV) were used (Dianeal®, lactate buffered, acidic pH). Patients underwent CAPD with 2.5 l of 1.36% glucose solution followed by fluid containing 3.86% glucose. On day B, patients received dialysis with pH-neutral bicarbonate/lactate-buffered fluid low in GDPs (Physioneal®, PDFBIO); patients again received 1.36% glucose followed by 3.86% glucose-based solutions. On day C, patients had PDFCONV with 1.36% glucose followed by 1.1% amino acid solution (PDFAA—Nutrineal®, lactate buffer, acidic pH, no GDPs). Dwell times were 120 min and each drain/dwell cycle was planned to last ~2.5 h, although this was not absolute due to variable draining times of different patients. There was at least a week's washout period between the study days. All study days commenced at 8 a.m. and patients were fasted from the midnight before and throughout the study period.

M-mode echocardiography was performed using commercially available equipment at the start and end of each dwell period to measure left ventricular (LV) dimensions (1.5–3.6 MHz 3S probe, Vivid 3®, GE medical systems, Sonigen, Germany). The volume of the peritoneal waste fluid from each dwell was also recorded. Blood samples were collected before and after each session in lithium heparin and EDTA tubes, and biochemical analysis performed on a multichannel autoanalyser. In addition, blood samples were collected at baseline and then at 10, 30 and 60 min of each dwell phase for analysis of plasma glucose (fluoride oxidase tubes) and at 10 and 60 min for measurement of insulin (EDTA tubes). Samples for insulin analysis were centrifuged at 685 g for 15 min to separate the plasma, which was immediately frozen at –80°C. Insulin was subsequently measured using a commercially available enzyme-linked immunosorbent assay kit (Insulin ELISA, DRG diagnostics, Marburg, Germany). Insulin resistance was assessed by calculating the homeostasis model of assessment index (HOMA-IR) using the following equation [8]:


Formula

in which a value >3.8 indicates insulin resistance. This measure has been validated in renal patients [9].

Primary endpoints were percentage change in BP, SV, CO and total peripheral resistance (TPR) in relation to plasma glucose and insulin levels and cardiac dimensions.

All patients gave informed consent prior to commencement, and ethical approval for the project was granted by Derbyshire Research Ethics Committee.

Finometer
The use of the Finometer (Finapres Medical Systems, Arnhem, The Netherlands) in dialysis patients has been described in detail elsewhere [2, 10]. Previously, we have shown good concordance between echocardiographic and Finometer-derived measurements of SV in dialysis patients [11]. The Finometer is accurate in tracking relative change (as opposed to absolute values), so data are presented as percentage change from baseline except for BP, which is calibrated against brachial readings using a return to flow method and for this absolute values are shown.

Statistical analysis
Results are expressed as mean ± SD unless otherwise stated. For BP and haemodynamic data, the mean refers to the complete dwell period. After demonstration of a normal distribution, all data were compared using one-way ANOVA with a design for repeated measures and Bonferroni's test to correct for multiple comparisons. An alpha error at P < 0.05 was judged to be significant.



   Results
Top
 Abstract
 Introduction
 Subjects and methods
 Results
Discussion
 Acknowledgements
 References
 
Two patients did not complete the study—one became ill after attending for only the first study day, and the second patient had an inaccurate Finometer trace (intermittent detection of pulse wave resulting in unacceptable data quality). These patients were excluded from the analysis. Analysis of baseline measurements showed that there were no differences in BP or in any of the haemodynamic variables between the three different study days.

Peritoneal transport characteristics and urea kinetic modelling
All patients had a weekly Kt/Vurea of over 2.0 (mean 2.57 ± 0.3, range 2.17–3.1). Mean weekly creatinine clearance was 88 ± 17l/week (59–106) and residual renal function was 4.6 ± 2 ml/min (0.2–7.8). Peritoneal transport characteristics for each patient are listed in Table 1. Transport characteristics did not appear to affect the haemodynamic response to the different fluid types, or the magnitude or rate of change in plasma glucose during the 3.86% glucose dwells.

Blood pressure, PDFCONV vs PDFAA
BP was higher during 3.86% PDFCONV dwells as compared with PDFAA. During 3.86% PDFCONV, mean systolic BP (SBP) for the entire dwell was 163.5 ± 5 mmHg, mean diastolic BP (DBP) was 92.8 ± 3 mmHg and mean arterial pressure (MAP) 118.9 ± 3 mmHg, as compared with a mean SBP of 158.9 ± 6 mmHg (P < 0.01), DBP of 87.1 ± 6 (P < 0.001) and MAP of 113.2 ± 5 mmHg (P < 0.001) during PDFAA dwells. The higher overall BP with 3.86% PDFCONV was due to differences in the first half of the dwell, and systolic readings were similar during the last third of the dwell. These data are summarized in Figure 2. No differences were observed when comparing BP during 1.36% PDFCONV and PDFAA dwells (BP data for 1.36% PDFCONV dwells are shown in Table 2.).


Figure 2
View larger version (22K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. BP during PDFCONV and PDFAA dwells. BP was significantly higher during 3.86% PDFCONV as compared with either 1.36% PDFCONV or PDFAA (P < 0.001 for all comparisons). The 1.36% PDFCONV data points are combined values for both days as values were extremely similar. Data are shown as mean ± SEM.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Blood pressure during 1.36% and 3.86% PDFCONV and PDFBIO dwells

 
Blood pressure, PDFCONV vs PDFBIO
BP was similar when comparing PDFCONV and PDFBIO during both 1.36% and 3.86% dwells. These data are shown in Table 2. However, when comparing 1.36% and 3.86% glucose dwells, BP was significantly higher with 3.86% glucose for both PDFCONV and PDFBIO (P < 0.001 for all comparisons). BP tended to decline throughout 3.86% PDFCONV dwells, which was in contrast to the pattern seen with all other fluids where BP rose progressively. These data are summarized in Figure 3.


Figure 3
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. BP during PDFCONV and PDFBIO dwells. There were no differences in mean BP when comparing the two fluid types, but mean BP was higher when comparing 3.86% glucose fluids with the corresponding 1.36% glucose fluid (P < 0.001 for both comparisons). Data are shown as mean ± SEM.

 
Haemodynamics, PDFCONV vs PDFAA
Mean HR for the entire dwell period was 69 ± 8 bpm with 3.86% PDFCONV, which reflected a rise of +2 ± 5% from baseline. This was not significantly different from a mean HR of 71 ± 13 bpm with PDFAA (+5 ± 6% from baseline, P = NS). However, SV and CO were both higher during PDFCONV dwells. SV for the entire dwell period was –9 ± 4% from baseline with 3.86% PDFCONV and was –20 ± 7% with PDFAA (P < 0.001). CO for the entire dwell was –7 ± 5% beneath baseline with 3.86% PDFCONV as compared with –15 ± 6% with PDFAA (P < 0.001). TPR also differed significantly between 3.86% PDFCONV and PDFAA. With the former, TPR rose to a mean of +12 ± 8% above baseline for the entire dwell period, and with PDFAA, mean TPR was +25 ± 11% above baseline (P < 0.001). These data are summarized in Figure 4.


Figure 4
View larger version (20K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Haemodynamics during PDFCONV and PDFAA dwells. SV and CO were significantly higher during PDFCONV dwells and TPR was significantly lower, as compared with PDFAA. The 1.36% PDFCONV data points are combined values for both days as values were extremely similar. Data are shown as mean ± SEM.

 
Haemodynamics, PDFCONV vs PDFBIO
There were no differences in any of the haemodynamic variables when comparing PDFCONV and PDFBIO during 1.36% glucose dwells. These data are shown in Table 3. However, as compared with 3.86% PDFBIO, TPR was significantly lower with 3.86% PDFCONV, with a mean for the entire 3.86% PDFBIO dwell of +31 ± 18% from baseline (P < 0.001). As a result, SV and CO were therefore higher during 3.86% PDFCONV, and means during the PDFBIO dwell were –17 ± 8% (P < 0.05) and –13 ± 6% (P < 0.01) respectively. These data are summarized in Figure 5.


View this table:
[in this window]
[in a new window]

  		Table 3. Systemic haemodynamics during 1.36% PDFCONV and PDFBIO dwells

 

Figure 5
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. Haemodynamics during PDFCONV and PDFBIO dwells. Parameters were very similar during 1.36% glucose dwells. SV and CO were significantly higher during 3.86% PDFCONV dwells and TPR was significantly lower, as compared with 3.86% PDFBIO. Data are shown as mean ± SEM.

 
Mirroring the trend for BP to increase during dwell periods, TPR also increased throughout all dwell periods (although TPR did fall during the last 30min of PDFAA dwell). The exception was during 3.86% PDFCONV dwell, when TPR progressively fell, which resulted in a smaller decline in CO as compared with the other fluid types.

Plasma glucose and insulin levels
Mean fasting glucose at baseline was 97.2 ± 19 mg/dl (5.4 ± 1.1 mmol/l). Plasma glucose did not change significantly during 1.36% PDFCONV, 1.36% PDFBIO or PDFAA dwells. Highest mean glucose was 104.5 ± 13 mg/dl (5.8 ± 0.7 mmol/l) during 1.36% PDFCONV dwells, 101 ± 13 mg/dl (5.6 ± 0.7 mmol/l) during 1.36% PDFBIO dwells and 91.9 ± 11 mg/dl (5.1 ± 0.6 mmol/l) during PDFAA dwells (P = NS for all comparisons). However, during 3.86% glucose dwells, plasma glucose rose significantly. Glucose was significantly higher than baseline by 10 min during 3.86% PDFCONV and PDFBIO dwells, with mean levels of 135 ± 23 mg/dl (7.5 ± 1.3 mmol/l, P < 0.01 vs baseline) and 138.9 ± 19 mg/dl (7.7 ± 1.1 mmol/l, P < 0.001 vs baseline) respectively, By 60 min, mean plasma glucose had risen to 163.9 ± 38 mg/dl (9.1 ± 2.1 mmol/l) during 3.86% PDFCONV and 163.9 ± 34 mg/dl (9.1 ± 1.9 mmol/l) during 3.86% PDFBIO. These values were significantly higher than baseline (P < 0.001), and higher than the peak levels during PDFAA and 1.36% glucose dwells (P < 0.001 for all comparisons).

A similar pattern was seen with plasma insulin levels. Mean fasting plasma insulin was 16.9 ± 6.7 µIU/ml (117.4 ± 47 pmol/l), and did not change significantly during 1.36% glucose or PDFAA dwells. At 60 min, mean plasma insulin during 1.36% PDFCONV was 18.1 ± 6.3 µIU/ml (125.7 ± 44 pmol/l), 18.4 ± 6.7 µIU/ml (127.8 ± 47 pmol/l) during 1.36% PDFBIO and 16.7 ± 6.1 µIU/ml (116.0 ± 42 pmol/l) during PDFAA dwells. However, insulin did rise significantly in response to the hyperglycaemia of the 3.86% glucose dwells. Mean plasma insulin was 22.9 ± 8.8 µIU/ml (159.0 ± 61 pmol/l) at 10 min (P = NS vs baseline) and 31.6 ± 14.4 µIU/ml (219.5 ± 100 pmol/l) at 60 min during 3.86% PDFCONV dwells (P < 0.05 vs baseline). During 3.86% PDFBIO, mean plasma insulin was 29.0 ± 11.1 µIU/ml (201.4 ± 77pmol/l) at 10 min (P = NS vs baseline) and 35.6 ± 16.2 µIU/ml (247.2 ± 113 pmol/l) at 60 min (P < 0.01 vs baseline). For PDFCONV and PDFBIO, both 10 min and 60 min plasma insulin levels were significantly higher than the corresponding values for PDFAA (P < 0.05 for 10 min comparison, P < 0.001 for 60 min comparison PDFCONV vs PDFAA, P < 0.01 for 60 min comparison PDFBIO vs PDFCONV). Plasma glucose and insulin data are summarized in Figure 6.


Figure 6
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6. Plasma glucose (A) and insulin (B) levels during PDFCONV, PDFBIO and PDFAA dwells. Plasma glucose and insulin rose significantly during 3.86% glucose dwells as compared with both baseline and the corresponding values during PDFAA dwells. The two data points for 1.36% PDFCONV represent values for PDFCONV and PDFAA study days. Data are shown as mean ± SD. To convert glucose in mmol/l into mg/dl, multiply by 18.018; insulin in µIU/l into pmol/l, multiply by 6.945. *P < 0.001 vs baseline, corresponding measurements during 1.36% glucose and PDFAA by ANOVA, **P < 0.05 vs PDFAA, ***P < 0.05 vs baseline, P < 0.001 PDFCONV vs PDFAA and P < 0.01 PDFBIO vs PDFAA.

 
At baseline, there were only two fasting glucose readings of >126 mg/dl (7.0 mmol/l), and in both these patients, readings were <126 mg/dl on the other two study days. Mean HOMA-IR was 4.6 ± 2.7 and four patients had values suggestive of insulin resistance (>3.8).

Ultrafiltration volumes
UF volumes were similar during PDFAA, 1.36% PDFCONV and 1.36% PDFBIO dwells, with means of 177 ± 167 ml, 116 ± 129 ml and 134 ± 138 ml, respectively (P = NS for all comparisons). UF volumes however were significantly higher with 3.86% glucose dwells, with means of 736 ± 180 ml with PDFCONV and 649 ± 232 with PDFBIO (P < 0.001 for all comparisons vs 1.36% glucose and P < 0.05 for all comparisons vs PDFAA). There was no difference when comparing UF volumes between 3.86% glucose PDFCONV and PDFBIO (P = NS). These data are shown in Figure 7.


Figure 7
View larger version (24K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7. Ultrafiltration volumes during PDFCONV, PDFBIO and PDFAA dwells. The 1.36% PDFCONV data represent the combined value for both days as values were extremely similar. Data are shown as mean ± SD. *P < 0.001 vs 1.36% glucose, P < 0.05 vs PDFAA by ANOVA.

 
We also performed a secondary analysis to look for correlations between the degree of hyperglycaemia and hyperinsulinaemia and change in SBP. There was a trend for higher plasma glucose levels and, to a lesser extent, plasma insulin levels to be associated with larger change in BP; however, neither of these trends reached statistical significance (for glucose and delta SBP: r = 0.4, P = 0.1; for insulin and delta SBP: r = 0.26, P = 0.4).

Echocardiographic measurements
At the outset of the study, only two patients had LV hypertrophy (defined as LVMI >51 g/m2.7 or interventricular septal thickness >1.1 cm) and all had normal ejection fractions. Despite the differences in UF volume (and therefore in intraperitoneal volume during the dwell periods), there were no significant differences in LV dimensions throughout the study period. Equally, ejection fraction remained constant throughout all dwell periods. These data are shown in Table 4.


View this table:
[in this window]
[in a new window]

  		Table 4. Echocardiographic measurements of LV diameter

 
Biochemical parameters
Biochemical data from the start and end of each study session are shown in Table 5. There were no differences in any of the variables. In particular, mean post-PDFCONV bicarbonate was 28.3 ± 6 mEq/l and did not differ from either post-PDFBIO bicarbonate (29.8 ± 5 mEq/l) or post-PDFAA bicarbonate (mean of 27.4 ± 5 mEq/l, P = NS for both comparisons).


View this table:
[in this window]
[in a new window]

  		Table 5. Biochemical parameters for each of the study days

 


   Discussion
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
Acknowledgements
 References
 
This study shows that hypertonic glucose dialysate results in an acute elevation of BP as compared with low osmolar glucose or amino-acid-containing solutions. The elevation in BP was associated with hyperglycaemia and hyperinsulinaemia (although this association is not necessarily causal) but was not due to differences in cardiac filling. We also observed differences between the haemodynamic response to conventional and biocompatible hypertonic glucose solutions.

In accordance with results from a previous study in which we demonstrated higher BP with 3.86% glucose as compared with Icodextrin [2], results from this study also show a rise in BP with hypertonic glucose solutions as compared with either 1.36% glucose or PDFAA, the latter being entirely glucose free. This rise in BP occurred with both hypertonic PDFBIO and PDFCONV and did not appear dependent on buffer type. The higher BP also occurred despite the largest UF volumes, but the absence of change in LV diameters indicates the effects on BP were not caused by increased cardiac filling due to the larger intraperitoneal volume. Equally, the stable LV dimensions argue against the study patients being volume overloaded at baseline, in which situation an improvement in cardiac function may be seen in response to UF. Therefore, the hyperglycaemia and hyperinsulinaemia demonstrated in our study during 3.86% glucose dwells seem the most likely cause of the rise in BP. This is supported by the trend for the patients with the highest plasma glucose levels to display the largest changes in SBP, although this secondary analysis must be interpreted with caution.

BP has been shown to rise in response to hyperglycaemia in several studies in both non-diabetic and diabetic patients, while glucose and insulin exert independent effects on systemic haemodynamics [3, 4]. Marfella et al. [4] demonstrated a rise in BP during hyperglycaemia, during both normal and elevated insulin levels. BP was returned to baseline with a glutathione infusion, suggesting the direct effect of hyperglycaemia on haemodynamics may be mediated by free radical production leading to reductions in nitric oxide. Equally, hyperinsulinaemia during euglycaemia has been shown to cause elevations in HR, SV and CO whilst also causing a fall in TPR [12]. These changes would be consistent with those observed when comparing the haemodynamic response to PDFAA and PDFCONV, with SV and CO significantly higher and TPR lower in the presence of hyperinsulinaemia during 3.86% PDFCONV.

Despite a similar rise in BP, the underlying haemodynamic response differed between PDFCONV and PDFBIO 3.86% solutions. These two fluids produced equal changes in plasma insulin and glucose, similar UF volumes and the fluids were warmed to exactly the same temperature before instillation. Although the fluids differ in buffer, there was no significant difference in plasma bicarbonate at the end of the study period. In addition, there were no differences between 1.36% PDFCONV and PDFBIO, so it seems unlikely that the buffer type alone exerted a significant effect on systemic haemodynamics. However, as the metabolism of glucose and lactate are linked in both skeletal muscle and hepatocytes [13], it is possible to speculate that the difference in buffer type resulted in differences in subsequent glucose metabolism. Alternatively, PDFBIO does contain significantly fewer GDPs as compared with PDFCONV. The amount of GDPs generated during heat sterilization depends on both dialysate pH and the amount of glucose present, and therefore 3.86% PDFCONV contains more GDPs than 1.36% PDFCONV [14]. This may explain why 1.36% PDFBIO and PDFCONV behaved in a similar fashion, but why 3.86% fluids differed. One report describes mesenteric artery vasodilatation in the rat in response to acidic-buffered, 4.25% glucose fluid with high GDP content, and abrogation of this effect with similar fluid low in GDPs [15]. These data would be consistent with our observation of lower TPR during 3.86% PDFCONV dwells. Although very difficult to isolate from biological fluids, indirect measurements suggest that systemic absorption of GDPs occurs during CAPD with conventional solutions [6]. However, there are no available data on whether or not GDPs can affect systemic haemodynamics.

In short, the haemodynamic response during CAPD is affected by multiple factors. In addition to the possible effects of hyperglycaemia and hyperinsulinaemia, cooling due to the dialysate fluid plus the effects of UF are likely explanations for the rise in TPR and BP throughout the dwell periods (seen with all fluids except 3.86% PDFCONV) [1, 2]. In addition, autonomic function and baroreflex sensitivity can be altered by all of the above factors. Therefore, it is only possible to speculate as to the exact cause of the haemodynamic changes that we observed, particularly when comparing PDFCONV and PDFBIO. Equally, it is not possible to determine which of the haemodynamic profiles is most favourable, although from these results and previous studies the progressive fall in TPR seems to be the feature that differentiates the response to PDFCONV from other fluids.

We did not observe any effect of differing UF volumes on cardiac filling. This suggests that transmission of pressure from the peritoneal capillaries may not be directly transmitted to the venous system and the right atrium. It is possible that increased intraperitoneal pressure could exert external pressure on venules, favouring a reduction in venous return. In support of this, there are studies demonstrating increased accumulation of interstitial fluid at higher intraperitoneal pressures [16]. Alternatively, it may be that the difference in intraperitoneal volumes did not alter pressure to a degree at which venous return was affected; an increase in intraperitoneal volume of 1l increases pressure by only 2.2 cm/H2O [17].

In contrast to previous studies, we did not observe significant differences in HR between the fluid types during this current study [2]. Only one patient was taking a rate-limiting drug; (atenolol) so it may be that our patients, as with many dialysis patients, had impaired autonomic function. However, this was not formally assessed.

We observed a relatively high degree of insulin resistance in our patients at baseline. This, in combination with the large amount of glucose delivered during 3.86% dwells, explains the magnitude of hyperglycaemia seen in some of these non-diabetic patients. Patients were fasted throughout, yet plasma glucose exceeded 11 mmol/l in three patients. These data are consistent with other work that shows that patients on CAPD have a higher prevalence of insulin resistance and therefore display higher plasma glucose levels in response to a glucose load compared with normals [18, 19]. Furthermore, the hyperinsulinaemia seen in response to glucose may be abnormally prolonged in CAPD patients [19]. Although not assessed in our short-term study, the frequent use of hypertonic glucose solutions resulting in repeated excursions of glucose and insulin to outside of their normal ranges may well have the potential to exert negative long-term negative metabolic effects. Certainly, several large prospective studies in non-diabetics have shown that a hyperglycaemic response to a glucose load is a strong predictor of cardiovascular death [20]. This may be important in the context of the extremely high cardiovascular mortality rates in dialysis patients that are not explained by conventional risk factors alone. Crucially, this may be modifiable as preliminary data show that a PD regime employing Icodextrin to reduce glucose exposure can improve insulin resistance after a period of 9 months [18].

Our study does have some weaknesses. Patient numbers are relatively small and there were two dropouts. In part this is compensated by the high resolution of differences in measured variables provided by the Finometer, although it also has to be recognized that this Finometer derives haemodynamic variables as opposed to measuring them directly. Furthermore, although our study demonstrated an association between hyperglycaemia/hyperinsulinaemia and the rise in BP, the results do not prove a causal relationship. Study design dictated that 1.36% glucose was always given before the 3.86% glucose and PDFAA dwells, and this may make comparisons between 1.36% glucose and the other fluids less robust. However, this was intentional to avoid confounding effects from differing overnight dwells preceding the study days, and the main comparisons that were planned were between the 3.86% glucose and PDFAA dwells. Finally, we were not able to measure plasma GDP levels during the dwell periods.

In conclusion, we have demonstrated that the hyperglycaemia and hyperinsulinaemia observed during CAPD with hypertonic glucose dialysate is associated with an acute rise in BP. CAPD with 1.1% amino acid solution did not cause any such derangements. In addition, a differing haemodynamic response to conventional and newer biocompatible PD solutions was observed, the cause of which is at present unclear. We suggest that these adverse haemodynamic and metabolic effects may have the potential to negatively impact cardiovascular outcomes. As cardiovascular risk reduction is key to ensuring optimal outcome in CAPD patients, manipulation of these haemodynamic and metabolic consequences by using low-glucose, biocompatible fluids should be the subject of longer-term, outcome-based studies.



   Acknowledgements
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
References
 
The authors gratefully acknowledge the assistance of Jeanette Hunt, Maxine Ward and the Cardiothoracic Measurement Department at Derby City Hospital. We are also very grateful to Prof. Mike Rennie, Dr Kenny Smith and Dr Emily Wilkes for providing laboratory facilities and assistance.

Conflict of interest statement. Dr McIntyre has previously received an unrestricted educational grant from Baxter Healthcare plc. This study has been registered with UK National Research Register (study ID N0077161773).



   References
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 

  1. Boon D, Bos WJ, van Montfrans GA, Krediet RT. (2001) Acute effects of peritoneal dialysis on hemodynamics. Perit Dial Int 21:166–171.[Abstract/Free Full Text]
  2. Selby NM, Fonseca S, Hulme L, Fluck RJ, Taal MW, McIntyre CW. (2005) Hypertonic glucose-based peritoneal dialysate is associated with higher blood pressure and adverse haemodynamics as compared with icodextrin. Nephrol Dial Transplant 20:1848–1853.[Abstract/Free Full Text]
  3. Jacobsen P, Rossing K, Hansen BV, Bie P, Vaag A, Parving HH. (2003) Effect of short-term hyperglycaemia on haemodynamics in type 1 diabetic patients. J Intern Med 254:464–471.[CrossRef][ISI][Medline]
  4. Marfella R, Verrazzo G, Acampora R, et al. (1995) Glutathione reverses systemic hemodynamic changes induced by acute hyperglycemia in healthy subjects. Am J Physiol 268:E1167–E1173.
  5. Davies SJ, Phillips L, Naish PF, Russell GI. (2001) Peritoneal glucose exposure and changes in membrane solute transport with time on peritoneal dialysis. J Am Soc Nephrol 12:1046–1051.[Abstract/Free Full Text]
  6. Zeier M, Schwenger V, Deppisch R, et al. (2003) Glucose degradation products in PD fluids: do they disappear from the peritoneal cavity and enter the systemic circulation? Kidney Int 63:298–305.[CrossRef][ISI][Medline]
  7. Lee HY, Park HC, Seo BJ, et al. (2005) Superior patient survival for continuous ambulatory peritoneal dialysis patients treated with a peritoneal dialysis fluid with neutral pH and low glucose degradation product concentration (Balance). Perit Dial Int 25:248–255.[Abstract/Free Full Text]
  8. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. (1985) Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28:412–419.[CrossRef][ISI][Medline]
  9. Shoji T, Emoto M, Nishizawa Y. (2001) HOMA index to assess insulin resistance in renal failure patients. Nephron 89:348–349.[CrossRef][ISI][Medline]
  10. Selby NM, Lambie SH, Camici PG, Baker CS, McIntyre CW. (2006) Occurrence of regional left ventricular dysfunction in patients undergoing standard and biofeedback dialysis. Am J Kidney Dis 47:830–841.[CrossRef][ISI][Medline]
  11. Selby NM, Fonseca S, Hulme L, Fluck RJ, Taal MW, McIntyre CW. (2006) Automated peritoneal dialysis has significant effects on systemic hemodynamics. Perit Dial Int 26:328–335.[Abstract/Free Full Text]
  12. Muscelli E, Emdin M, Natali A, et al. (1998) Autonomic and hemodynamic responses to insulin in lean and obese humans. J Clin Endocrinol Metab 83:2084–2090.[Abstract/Free Full Text]
  13. Miller BF, Fattor JA, Jacobs KA, et al. (2002) Lactate and glucose interactions during rest and exercise in men: effect of exogenous lactate infusion. J Physiol 544:963–975.[Abstract/Free Full Text]
  14. Van Biesen W, Veys N, Vanholder R, Lameire N. (2003) New concepts in peritoneal dialysis: new wine in old barrels? Artif Organs 27:398–405.[CrossRef][ISI][Medline]
  15. Mortier S, De Vriese AS, Van de Voorde J, Schaub TP, Passlick-Deetjen J, Lameire NH. (2002) Hemodynamic effects of peritoneal dialysis solutions on the rat peritoneal membrane: role of acidity, buffer choice, glucose concentration, and glucose degradation products. J Am Soc Nephrol 13:480–489.[Abstract/Free Full Text]
  16. Zakaria ER and Rippe B. (1995) Peritoneal fluid and tracer albumin kinetics in the rat. Effects of increases in intraperitoneal hydrostatic pressure. Perit Dial Int 15:118–128.[ISI][Medline]
  17. Durand PY, Chanliau J, Gamberoni J, Hestin D, Kessler M. (1996) Measurement of hydrostatic intraperitoneal pressure: a necessary routine test in peritoneal dialysis. Perit Dial Int 16:Suppl 1, S84–S87.[Abstract]
  18. Amici G, Orrasch M, Da Rin G, Bocci C. (2001) Hyperinsulinism reduction associated with icodextrin treatment in continuous ambulatory peritoneal dialysis patients. Adv Perit Dial 17:80–83.[Medline]
  19. Delarue J and Maingourd C. (2001) Acute metabolic effects of dialysis fluids during CAPD. Am J Kidney Dis 37:S103–S107.[ISI][Medline]
  20. Brunner EJ, Shipley MJ, Witte DR, Fuller JH, Marmot MG. (2006) Relation Between blood glucose and coronary mortality over 33 years in the Whitehall Study. Diabetes Care 29:26–31.[Abstract/Free Full Text]
Received for publication: 19. 7.06
Accepted in revised form: 13.10.06


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nephrol Dial TransplantHome page
A. Pletinck, F. Verbeke, L. Van Bortel, C. Dequidt, D. Vijt, W. V. Biesen, and R. Vanholder
Acute central haemodynamic effects induced by intraperitoneal glucose instillation
Nephrol. Dial. Transplant., July 7, 2008; (2008) gfn384v1.
[Abstract] [Full Text] [PDF]

Home page
 J. Am. Soc. Nephrol.Home page
B. B. McCormick and J. M. Bargman
Noninfectious Complications of Peritoneal Dialysis: Implications for Patient and Technique Survival
J. Am. Soc. Nephrol., December 1, 2007; 18(12): 3023 - 3025.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
22/3/870    most recent
gfl654v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (7)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Selby, N. M.
Right arrow Articles by McIntyre, C. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Selby, N. M.
Right arrow Articles by McIntyre, C. W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?