Gut permeability in paediatric cardiac surgery

I. Malagon1,*, W. Onkenhout2, G. Klok2, P. F. H. van der Poel2, J. G. Bovill1 and M. G. Hazekamp3

1 Department of Anaesthesia, 2 Department of Paediatrics and 3 Department of Paediatric Cardiac Surgery, Leiden University Medical Centre, 2300 RC Leiden, The Netherlands

* Corresponding author: Department of Anaesthesia, Leiden University Medical Centre, Albinusdreef 2, PO Box 9600, 2300 RC Leiden, The Netherlands. E-mail: jmalagon{at}lumc.nl

Accepted for publication September 27, 2004.


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Background. Intestinal mucosal ischaemia can occur in infants and children during and after cardiac surgery. Severe decreases in mucosal perfusion may cause complications such as necrotizing enterocolitis and postoperative mortality. We investigated gut permeability in paediatric patients undergoing cardiac surgery using the dual sugar permeability test and absorption of two other saccharides.

Methods. Thirty-four patients undergoing palliative or corrective surgical procedures with and without cardiopulmonary bypass were investigated. Intestinal permeability was measured using 3-O-methyl-D-glucose, D-xylose, L-rhamnose and lactulose, given orally after induction of anaesthesia and 12 and 24 h later.

Results. Lactulose/rhamnose ratios were raised from the outset [median 0.39 (confidence interval 0.07–1.8 for patients undergoing operations without cardiopulmonary bypass and 0.30 (0.02–2.6) with cardiopulmonary bypass]. The highest lactulose/rhamnose ratios were recorded 12 h after surgery 0.32 (0.07–6.9), when cardiopulmonary bypass was used. This is approximately seven times the value expected in healthy children. There was an improvement in patients not undergoing cardiopulmonary bypass: 0.22 (0.03–0.85) 12 h and 0.11 (0–0.48) 24 h after induction of anaesthesia. Patients undergoing repair of aortic coarctation showed the fastest recovery: 0.09 (0.03–0.31) 12 h and 0.07 (0.04–0.35) 24 h after induction of anaesthesia.

Conclusions. Patients with congenital heart defects have abnormal gut permeability when compared with healthy children of similar age. Cardiopulmonary bypass seems to affect the intestinal barrier morphologically (lactulose and rhamnose absorption) and functionally (3-O-methyl-D-glucose and D-xylose absorption).

Keywords: gastrointestinal tract, gut permeability ; surgery, cardiovascular ; surgery, paediatric


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Intestinal mucosal ischaemia, although transient, can occur in infants and children during and after cardiopulmonary bypass (CPB).1 Severe decreases in mucosal perfusion may cause complications such as necrotizing enterocolitis (NEC) and postoperative mortality. NEC can have devastating consequences, especially in patients undergoing repair of hypoplastic left heart syndrome2 (mortality more than 90%). Neonates with aortic arch anomalies and infants subjected to CPB-induced profound hypothermia may be at particular risk of developing splanchnic ischaemia in the perioperative period.3 Patients with coarctation of the aorta may, on the other hand, be exposed to reperfusion injuries after the surgical repair, manifesting as the postcoarctectomy syndrome.4 The previous studies have used indirect indicators of intestinal mucosal perfusion (e.g. laser Doppler probe or gastric tonometry).2 3

Intestinal permeability can be evaluated non-invasively by measuring urinary excretion of orally administered water-soluble, non-degradable test molecules.5 6 This barrier function test is based on the comparison of intestinal permeation of larger molecules with that of smaller molecules by measuring the ratio of their urinary excretion. These two types of molecules follow different routes of intestinal permeation: the larger molecules are assumed to permeate paracellularly and the smaller molecules transcellularly. Preabsorption factors, such as gastric emptying, dilution by secretion and intestinal transit time, and post-absorption factors, such as systemic distribution and renal clearance, are assumed to affect both molecules equally. Four saccharides, 3-O-methyl-D-glucose (molecular weight 194 Da), D-xylose (molecular weight 150 Da), L-rhamnose (R, molecular weight 164 Da) and lactulose (L, molecular weight 342 Da) are used to assess active carrier-mediated, passive carrier-mediated, transcellular, and paracellular transport respectively in the small intestine. Intestinal permeability is considered to be normal if the lactulose (% recovery)/rhamnose (% recovery) (L/R) ratio is below 0.05.5 Intestinal absorptive capacity for saccharides is considered to be normal when the recoveries of D-xylose (passive carrier-mediated transport) and 3-O-methyl-D-glucose (active carrier–mediated transport) are around 10 and 30% respectively.7 The aim of our study was to observe the changes in gut permeability in paediatric patients undergoing cardiac surgery using the dual sugar permeability test (DSPT).


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
After approval from the local ethics committee and informed consent from the parents or legal guardians, 34 patients were enrolled in this prospective, non-randomized observational study. Patients received a premedication consisting of oral atropine 0.02 mg kg–1 and midazolam 0.5 mg kg–1 30 min before induction of anaesthesia. Anaesthesia was induced with sevoflurane followed by a bolus of sufentanil 1 µg kg–1 and pancuronium 0.2 mg kg–1. Anaesthesia was maintained using continuous infusions of midazolam 0.2 mg kg–1 h–1 and sufentanil 2 µg kg–1 h–1. The lungs of the patients were mechanically ventilated with a mixture of oxygen and air. Mechanical ventilation was maintained until CPB commenced. After heparin administration (3 mg kg–1) and aorta cannulation, CPB was instituted with a Dideco hollow-fibre oxygenator with blood flow between 200 and 300 ml kg–1 min–1. The prime volume, between 325 and 750 ml according to the patient's weight, contained lactate-free Ringer's solution, albumin, mannitol, blood and heparin. Patients underwent modified ultrafiltration at the end of the surgical procedure. The effect of heparin was reversed with protamine sulphate at a ratio of 1 mg protamine to 1 mg heparin.

After induction of anaesthesia, the sugar solution 2 ml kg–1 was administered through a nasogastric tube. Urine was subsequently collected through a urinary catheter for 3 h, the total volume was recorded and samples were stored at –20°C until analysis. This process was repeated 12 and 24 h after induction of anaesthesia. The sugar solution, prepared by the hospital pharmacy, contained 3-O-methyl-D-glucose 2 g litre–1, D-xylose 5 g litre–1, L-rhamnose 10 g litre–1 and lactulose 50 g litre–1. The osmolarity of the solution is approximately 240 mosm litre–1.

Sugar concentrations in urine were determined by gas chromatography following a slight modification8 of the procedure described by Jansen and colleagues.9 Briefly, to an aliquot of urine corresponding to creatinine 0.5 µmol we added ribitol 30 µg and trehalose 10 µg (Sigma-Aldrich, St Louis, MO, USA) as internal standards (ribitol for the determination of 3-O-methyl-D-glucose, D-xylose and L-rhamnose and trehalose for lactulose) for the determination of 3-O-methyl-D-glucose, D-xylose, L-rhamnose and lactulose. The sample was dried, derivatized with 300 µl Tri-Sil TBT (Pierce, Rockford, IL, USA) at 100°C and partly hydrolysed with water. Subsequently, the intact sugar trimethylsilyl derivatives were extracted with hexane. After concentrating, gas chromatographic analysis was performed on a 30 m capillary fused silica HP-1 column (Agilent, Palo Alto, CA, USA) using split injection. Quantification was performed after the construction of standard addition calibration curves.

The types of vasoactive drugs and their amount were recorded after admission to the intensive care unit and 24 h later. To quantify inotropic support, inotrope scores were calculated as the sum of all inotrope doses corrected for potency (dopamine, dobutamine=1; milrinone=15; epinephrine=100).10 11 Fluid intake (including crystalloids, colloids and blood products), output (urine, blood and serous fluid loss) and balance were recorded over a 36 h period after admission to the intensive care unit.

Statistical analysis
Data were analysed with the statistical package SPSS 10. Data are presented as median (95% confidence intervals). The data were not normally distributed and we used the Mann–Whitney U test for unpaired data and the Friedman test for sequential data. Values of P<0.05 were considered significant. Based on a previous study6 that found a mean (SD) L/R ratio of 0.047 (0.018), a prospective analysis showed that we needed a sample size of 34 to detect a difference in the L/R ratio of 0.02, with {alpha}=0.05 and a power of 80%.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Table 1 shows the patients' characteristics. The groups were comparable with respect to age, sex and weight. The types of operations performed in each group are shown in Table 2. Table 3 shows the L/R ratios and percentage recovery of the four sugars throughout the study period. Figure 1 is a graphic representation of the L/R changes in both groups. Patients undergoing repair of aortic coarctation showed the fastest improvement in L/R ratios: 0.52 (0.21–1.01) at T0, 0.09 (0.03–0.31) at T12 and 0.07 (0.04–0.35) at T24 (P<0.01). Patients receiving a Blalock–Taussig shunt or banding of the pulmonary artery had the following L/R ratios: 0.37 (0.06–1.81) at T0, 0.27 (0.12–0.85) at T12 and 0.14 (0–0.48) at T24 (P<0.04). Inotropic scores (median 95% confidence interval) on admission to the intensive care unit [5 (0–59.7)] were not significantly different from those at 24 h after admission [5 (0–54)] in the CPB group. In the group without CPB the inotropic scores were 0 (0–10) after admission and 0 (0–15) at 24 h.


View this table:
[in this window]
[in a new window]
 
Table 1 Patient characteristics. Values are median (95% CI)

 

View this table:
[in this window]
[in a new window]
 
Table 2 Types of operation performed in each group

 

View this table:
[in this window]
[in a new window]
 
Table 3 Lactulose/rhamnose ratios (L/R) and percentage recovery for lactulose, rhamnose, 3-O-methyl-glucose (3OMG) and xylose without and with cardiopulmonary bypass (CPB). Values are expressed as median (95% CI).

 


View larger version (12K):
[in this window]
[in a new window]
 
Fig 1 Changes in L/R ratios in both groups at induction of anaesthesia (T0) and 12 h (T12) and 24 h (T24) after induction of anaesthesia. Values are expressed as median, 25th to 75th interquartile ranges (bars) and 2.5th and 97.5th percentiles (error bars). *Statistically significant difference between groups; {dagger}statistically significant difference within groups.

 
Patients operated without CPB had a fluid intake of 79 ml kg–1 (39–251) and output of 70 ml kg–1 (22–176), with an overall fluid balance of 18 ml kg–1 (–97 to 92). For patients undergoing operations with CPB, the fluid intake was 86 ml kg–1 (41–147) and output 58 ml kg–1 (23–110), with an overall balance of 31 ml kg–1 (–41 to 121). Differences between the two groups were not significantly different.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
To our knowledge this is the first report of changes in gut permeability in paediatric patients with congenital heart defects undergoing cardiac surgery. Our study shows that from the outset L/R ratios were well above the normal values expected in patients of similar age without cardiac defects. Only patients undergoing repair of coarctation of the aorta had near-normal L/R ratios 24 h after the surgical procedure. The DSPT has been used to assess intestinal function in healthy neonates, in whom the L/R ratios were around 0.05.5 Paediatric patients without intestinal pathology have a similar ratio: range 0.023–0.074, mean (SD) 0.047 (0.018).6

Several studies have investigated the effect of CPB on intestinal permeability in the adult population.1214 In all of them the investigators demonstrated an increase in gut permeability that reverted to normal during the postoperative period. In animals, exposure to CPB induces a transient mesenteric endothelial dysfunction with an increased contractile response to an {alpha}1-adrenergic agonist.15

Cyanotic patients may be at higher risk of developing intestinal mucosal ischaemia. This may also be true for patients with coarctation of the aorta because of reduced blood flow through the descending aorta. Infants undergoing cardiac surgery often have chronic low arterial oxygen concentrations as a result of intra- or extracardiac shunting. Neonates with hypoplastic left heart syndrome have at times preferential blood flow to the pulmonary circulation at the expense of the systemic circulation. Patients operated without CPB had either total correction of their disease (coarctation of the aorta) or an improvement in systemic oxygen delivery via a Blalock–Taussig shunt or pulmonary banding. It is therefore logical in these patients to expect an improvement in the L/R ratios and recovery of single markers.

The accuracy of the DSPT relies on the complete collection of urine samples during the study period. This factor limits the applicability of the test in non-cooperative patients without a urinary catheter. A high percentage of patients are discharged to the ward the day after surgery, with the consequent removal of urinary catheters. For this reason we performed the test during the first 24 h after the operation.

There has been criticism concerning the interpretation and significance of the DSPT in the literature.16 17 In an animal model it was shown that fluid loading increased the L/R ratios independent of changes in intestinal permeability. Rats received, in an 8-h period, a fluid bolus equivalent to twice the daily fluid oral intake. Put into perspective, this means an infant of 10 kg would receive in an 8-h period approximately 2 litres of fluid intravenously. We carefully documented the fluid balance during the study period. On average, patients received less than the daily maintenance fluid expected for their age.

When lactulose and rhamnose are combined in the test solution at a fixed concentration ratio, the effects of preabsorption factors (gastric emptying, dilution by secretions, intestinal transient time) and postabsorption factors (systemic distribution and renal clearance) will apply equally to both. Therefore, the L/R ratio is influenced only by the difference in gut permeability for each molecule.18 Pre- and postabsorption factors may influence single markers, such as D-xylose and 3-O-methyl-D-glucose. D-Xylose is absorbed by passive carrier-mediated transport and 3-O-methyl-D-glucose by active carrier-mediated transport. These two single markers provide information about the functional state of the intestinal mucosa, whereas the L/R ratio is a reflection of mucosal integrity at the morphological level. Results of recovery of single markers should always be interpreted in conjunction with L/R ratios.

From the outset, patients operated without CPB had better percentage recovery for both single markers, although the differences were significant after induction of anaesthesia only. However, the L/R ratio was worse in this group than in the group requiring CPB in the same period. It is difficult to draw conclusions at this point in time. Nevertheless, the trend was towards faster improvement of single markers in the group without CPB, keeping in mind that the values were far from normal 24 h after the operation.

We assume that the deterioration or lack of improvement in the L/R ratios 12 h after induction of anaesthesia must have been related to the use of CPB. Interestingly, CPB did not prevent a progressive and significant increase in the percentage recovery of D-xylose and 3-O-methyl-D-glucose. It appears that CPB temporarily damages the intestinal mucosa at the morphological level. At the functional level, though, CPB may only delay a process that begins with the surgical correction and consequent improvement in arterial oxygen supply and/or systemic perfusion.

Inotropic support was necessary more often in the group undergoing CPB. It may be argued that this factor alone can explain the differences in L/R ratios between the two groups. However, inotropic support in the CPB group did not change between admission to intensive care and 24 h later, whereas the L/R ratios did improve, although not significantly.

Used on a regular basis, the DSPT may help us to identify the optimal time to reintroduce enteral feeding in the postoperative period. This indeed deserves further research. Novel surgical techniques or drugs aimed at protecting the splanchnic circulation can be tested against the DSPT as an end-point. We are at present studying how the use of dexamethasone before the start of CPB affects gut permeability in the postoperative period during paediatric cardiac surgery.

In conclusion, we have shown that paediatric patients undergoing cardiac surgery with CPB have median L/R ratios up to seven times the normal values expected in healthy children. Patients undergoing surgical repair of aortic coarctation show a swift return to near-normal values 24 h after the operation. From our results, we can also conclude that the intestinal barrier is affected at both the morphological and the functional level. Measurement of mesenteric blood and oxygen supply in the paediatric population remains a difficult task. Only by using non-invasive, non-toxic surrogate markers of intestinal perfusion can we investigate and, most importantly, try to improve oxygen supply to the gut in the perioperative period.


    Acknowledgments
 
We are grateful to our hospital pharmacy for the preparation and supply of the sugars solution.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
1 Booker PD, Prosser DP, Franks R. Effect of hypothermia on rectal mucosal perfusion in infants undergoing cardiopulmonary bypass. Br J Anaesth 1996; 77: 591–6[Abstract/Free Full Text]

2 Hebra A, Brown MF, Hirshl RB, et al. Mesenteric ischemia in hypoplastic left heart syndrome. J Pediatr Surg 1993; 28: 606–11[CrossRef][ISI][Medline]

3 Booker PD, Romer H, Franks R. Gut mucosal perfusion in neonates undergoing cardiopulmonary bypass. Br J Anaesth 1996; 77: 597–602[Abstract/Free Full Text]

4 Lerberg DB, Hardesty RL, Siewers RD, et al. Coarctation of the aorta in infants and children: 25 years of experience. Ann Thorac Surg 1982; 33: 159–70[Abstract]

5 Beach RC, Menzies IS, Clayden GS, et al. Gastrointestinal permeability changes in the preterm neonate. Arch Dis Child 1982; 57: 141–5[Abstract]

6 Miki K, Butler R, Moore D, et al. Rapid and simultaneous quantification of rhamnose, mannitol and lactulose by HPLC for estimating intestinal permeability in paediatric practice. Clin Chem 1996; 42: 71–5[Abstract/Free Full Text]

7 Van Elburg RM, Uil JJ, De Monchy, et al. Intestinal permeability in pediatric gastroenterology. Scand J Gastroenterol 1992; 27: 19–24[ISI]

8 Onkenhout W, Groener JEM, Verhoeven NM, et al. L-Arabinosuria: a new defect in human pentose metabolism. Mol Genet Metab 2002; 77: 80–5[CrossRef][ISI][Medline]

9 Jansen G, Muskiet FAJ, Schierbeek H, et al. Capillary gas chromatographic profiling of urinary, plasma and erythrocyte sugars and polyols as their trimethylsilyl derivatives, preceded by a simple and rapid prepurification method. Clin Chem Acta 1986; 157: 277–94[CrossRef][ISI][Medline]

10 Wernovsky G, Wypij D, Jonas RA, et al. Postoperative course and hemodynamic profile after the arterial switch operation in neonates and infants: a comparison of low-flow cardiopulmonary bypass and circulatory arrest. Circulation 1995; 92: 2226–35[Abstract/Free Full Text]

11 Shore S, Nelson DP, Pearl JM, et al. Usefulness of corticosteroid therapy in decreasing epinephrine requirements in critically ill infants with congenital heart disease. Am J Cardiol 2001; 88: 591–4[CrossRef][ISI][Medline]

12 Ohri SK, Bjarnason I, Pathi V, et al. Cardiopulmonary bypass impairs small intestinal transport and increases gut permeability. Ann Thorac Surg 1993; 55: 1080–6[Abstract]

13 Ohri SK, Somasundaram S, Koak Y, et al. The effect of intestinal hypoperfusion on intestinal absorption and permeability during cardiopulmonary bypass. Gastroenterology 1994; 106: 318–23[ISI][Medline]

14 Sinclair DG, Houldsworth PE, Keogh B, et al. Gastrointestinal permeability following cardiopulmonary bypass: a randomised study comparing the effects of dopamine and dopexamine. Intensive Care Med 1997; 23: 510–6[CrossRef][ISI][Medline]

15 Doguet F, Litzler PY, Tamion F, et al. Changes in mesenteric vascular reactivity and inflammatory response after cardiopulmonary bypass in a rat model. Ann Thorac Surg 2004; 77: 2130–7[Abstract/Free Full Text]

16 Hellemeesch MM, Lamers WH, Soeters PB, et al. Increased lactulose/rhamnose ratio during fluid load is caused by increased urinary lactulose excretion. Am J Physiol Gastrointest Liver Physiol 2000; 278: G83–8[Abstract/Free Full Text]

17 Fink MP. Interpreting dual-sugar absorption studies in critically ill patients: what are the implications of apparent increase in intestinal permeability to hydrophilic solutes? Intensive Care Med 1997; 23: 489–92[CrossRef][ISI][Medline]

18 Menzies IS. Transmucosal passage of inert molecules in health and disease. In: Skadhauge E, Heinze K, eds. Intestinal Absorption and Secretion. London: MTP Press, 1983; 527–43