PEDIATRICS Vol. 108 No. 5 November 2001, pp. 1187-1192

Glucose Monitoring With Long-Term Subcutaneous Microdialysis in Neonates

Friedrich A. M. Baumeister, MD^*, Boris Rolinski, MD, Raymonde Busch, MS^§, and Peter Emmrich, MD^*

From the ^* Children's Hospital of the Technical University Munich, Children's Clinic, Munich, Germany;  Dr. v. Hauner Children's Hospital, Children's Hospital of the Ludwig Maximilian University Munich, Munich, Germany; and ^§ Department for Medical Statistics and Epidemiology of the Technical University Munich, Munich, Germany.

	ABSTRACT

Top Abstract Methods Results Discussion References

Background.  Microdialysis is a new approach for continuous monitoring of small molecules in the extracellular space, and hypoglycemia is a common problem in neonatal intensive care. The objective of this study was to evaluate subcutaneous microdialysis for long-term glucose monitoring in neonatal intensive care. We determined the relative recovery of the microdialysis system in vitro and in vivo, the stability of the relative recovery in vivo during long-term microdialysis, and the correlation between blood and dialysate concentrations of glucose and urea. Furthermore, we evaluated the sensitivity and specificy of subcutaneous microdialysis for the diagnosis of hypoglycemia.

Patient and Methods.  Thirteen infants (10 neonates) with gestational ages of 30.2 to 45.6 weeks were investigated by microdialysis of subcutaneous adipose tissue and blood sampling. Subcutaneous microdialysis was performed for a median (range) duration of 9 (4-16) days.

Results.  The application was safe, even in extremely low birth weight infants (<1000 g) with scanty subcutaneous adipose tissue. The mean ± standard deviation of the relative recovery in vitro was 101 ± 3% for glucose and 100 ± 2% for urea. Using urea as the internal standard, the mean relative recovery in vivo was 96.4 ± 12.7% at the beginning and remained constant up to 16 days. The correlation between microdialysate and blood was significant for glucose (r = 0.88) and urea (r = 0.98). Subcutaneous microdialysis allowed the detection of asymptomatic hypoglycemias. The diagnostic sensitivity of a dialysate glucose 2.9 mM to predict a blood glucose level 2.8 mM was 92.3%, with 88.1% specificy. The positive predictive value with a 13.4% prevalence of a blood glucose 2.8 mM was 54.5%, with a negative predictive value of 98.7% and an accuracy of 88.7%.

Conclusions.  Subcutaneous microdialysis is a safe method, well suited for long-term glucose monitoring in neonates during intensive care. Subcutaneous microdialysis can be used to reduce blood loss and painful stress resulting from diagnostic blood sampling in high-risk neonates.  Key words:  microdialysis, monitoring, glucose, recovery, neonates.

Blood loss from diagnostic sampling is the most common cause of anemia in hospitalized infants.¹ In ill neonates, especially very low birth weight infants, the blood sampled during the first weeks may exceed the infant's total blood volume.¹ The amount of diagnostic blood loss is the main determinant for the transfused blood volume.^1,2 The risks of blood transfusions are well known, especially transmission of viral infections, graft versus host reaction, volume overload, antigenic sensitization, and increased risk of retrolental fibroplasia. However, the minimization of the diagnostic blood loss by restriction of blood testing and the use of ultramicroassays are limited.

Microdialysis is a promising approach to reduce diagnostic blood loss and painful stress associated with blood sampling. Microdialysis is a sampling method that permits continuous analysis of a patient's extracellular tissue chemistry without consuming blood.^3,4

The clinical application of microdialysis became feasible in recent years with the introduction of commercial available microdialysis catheters, certified for human tissues. Microdialysis is based on diffusion about a semipermeable membrane. The microdialysis catheter mimics a blood capillary. The tubular dialysis membrane is continuously perfused by a liquid that equilibrates with the surrounding interstitial fluid (Fig 1). After insertion of the microdialysis catheter into the tissue, dialysate samples can continuously be collected for several days.^3,4

View larger version (12K): [in this window] [in a new window]   Fig. 1.   The tip of a microdialysis catheter (length: 10 mm; diameter: 0.6 mm). The perfusion fluid is guided through the double-lumen shaft (a) to the proximal end of the tip where it enters the space (b) between the inner outlet tube and the surrounding tubular outer dialysis membrane and while flowing to the distal end of the tip dialysis takes place.

In neonates, subcutaneous microdialysis has been used for metabolic monitoring up to 4 days after birth⁵ and after surgery.^6,7 Hypoglycemia is a common problem in neonatal intensive care and glucose is one of the most frequent blood tests.

The purpose of this study was to evaluate the feasibility of subcutaneous microdialysis long-term glucose monitoring in neonates receiving intensive care, the determination of the relative recovery of the microdialysis system in vitro and in vivo, the stability of the relative recovery in vivo during long-term microdialysis, and the correlation between the interstitial and blood concentrations of glucose and urea. Furthermore, we evaluated the sensitivity and specificy of subcutaneous microdialysis for the diagnosis of hypoglycemia.

	METHODS

Top Abstract Methods Results Discussion References

The study was approved by the local ethics committee, and written informed consent was obtained from the parents.

Thirteen infants (10 neonates) receiving intensive care were enrolled in the study and a total of 15 microdialysis catheters were inserted. When subcutaneous microdialysis was started the median (range) age was 0.7 (0.1-8.4) weeks, the gestational age was 38.7 (30.2-45.6) weeks, and the weight was 2.8 (0.8-7.0) kg. Two were extremely low birth weight (ELBW) infants with a body weight of 0.8 and 0.9 kg.

The indications for enrollment in this subcutaneous microdialysis study were repeated hypoglycemias (glucose <2.2 mM; N = 3), SGA (N = 4), metabolic diseases with lactate acidosis (N = 2), asphyxia (N = 2), and sepsis (N = 2).

The principle of the microdialysis technique used has been described in detail previously.^48-10 The insertion of the microdialysis catheter was performed under sterile conditions after transdermal local anesthesia (EMLA ointment, Wedel, Germany) into the subcutaneous adipose tissue of the lateral thigh. For the insertion, we used a steel introducer in a surrounding plastic tube. Only the tip of the microdialysis catheter (10 mm) was inserted subcutaneously. After implantation, the microdialysis catheter was fixed on the skin with a transparent sterile plastic film.

The microdialysis catheter (CMA 70; CMA/Microdialysis AB, Solna, Sweden) is a double-lumen plastic cannula, with a shaft (length 60 mm, diameter 0.9 mm) and a tip for dialysis (length 10 mm, diameter 0.6 mm; Fig 1). The dialysis membrane is a semipermeable polyamide membrane with a molecular cutoff of approximately 20 kDa.

The microdialysis catheter was continuously perfused with a sterile Ringer's solution. The low flow rate of 0.3 uL/min was achieved by a battery driven pump (CMA 106 microdialysis pump, CMA/Microdialysis AB, Solna, Sweden) using disposable precision syringes. The dialysate samples were collected in microvials (CMA/Microdialysis AB, Solna, Sweden) placed in a vial holder fixed at the end of the outlet tube of the microdialysis catheter (Fig 2). For analysis of the samples, the microvials were changed and transferred manually to the analyzer. The concentrations of glucose and urea in the dialysate were determined enzymatic spectrophotometric on an automated analyzer, which needs a sample volume of 0.5 uL per analysis. (CMA 600 Microdialysis Analyser, CMA/Microdialysis AB, Solna, Sweden).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Microdialysis is a continuous sampling method. The tip of the microdialysis catheter (B) where dialysis takes place is inserted into the subcutaneous tissue. The perfusion fluid is continuously pumped (0.3 uL/min) from the microdialysis pump (A) through the microdialysis catheter (B) into the microvial (C). The microvial is fixed in a vial holder at the end of the outlet tube. The microvials with the samples were transferred manually to the analyzer.

The indication for blood sampling was not influenced by the microdialysis study. When the clinical management required the determination of glucose, lactate, or urea in the blood, a microdialysis sample was collected for 15 minutes, starting when the blood sample was drawn. Fifteen minutes is the lag phase for the sample to move from the tip of the microdialysis catheter to the collecting test tube. Microdialysis was performed until <2 venous blood samples were expected in 1 week.

Glucose was measured in plasma by an enzymatic-amperometric method based on glucose oxidase reaction on an automated analyzer (ESAT 6660, Eppendorf, Hamburg, Germany). Urea was determined in the blood by an enzymatic spectrophotometric method on a dry chemistry automated analyzer (Vitros 250, Ortho-Clinical Diagnostics, Neckargemünd, Germany).

For statistical analysis, the computer program Statistical Package for Social Science version 10.0 (SPSS Inc, Chicago, IL) was used. Data distribution was analyzed with the Kolmogorov-Smirnov test with lilliefors adaptation. Values were expressed as mean ± standard deviation or as median and range. When it was necessary, patient's data were logarithmically transformed to achieve a normal distribution for multiple linear regression analysis. Wilcoxon test was used to compare dialysate and blood values.

	RESULTS

Top Abstract Methods Results Discussion References

Microdialysis Technique

The implantation of the microdialysis catheter into the subcutaneous tissue can be compared with the insertion of an intravenous cannula. Care was taken to insert the microdialysis catheter without or only with minimal bleeding. Four times an accidental venous puncture occurred and the introduction was repeated.

Fifteen microdialysis catheters were inserted in 13 infants. Two were ELBW infants with a body weight of 800 g and 900 g. The median (range) time of microdialysis was 9 (4-16) days. The microdialysis catheters were well tolerated and less interfering with movements. In 2 agile infants, the catheter was accidentally withdrawn. Microdialysis was stopped in 1 child after 12 days because the perfusion flow was disturbed, and in another at day 4 of microdialysis because of death.

Antibiotic treatment was not influenced by microdialysis. Median (range) time of microdialysis with antibiotic treatment was 7 (0-12) days. Three infants received microdialysis for 8 days without any antibiotics. No systemic inflammatory reaction attributable to microdialysis or sign of infection at the site of implantation of the microdialysis catheter was observed in any patient. In 1 child there was a minor bleeding around the microdialysis catheter during microdialysis. Microdialysis was not interfering with nursing. By changing the collecting tube at the end of the perfusion line, the dialysate samples were obtained even during sleep without any disturbances of the patients. When the microdialysis catheters were withdrawn, all were complete and none of the infants developed a scar in the following weeks.

Relative Recovery of the Microdialysis Catheter
In Vitro and In Vivo

Samples collected by microdialysis are not derived directly from the extracellular space, so the concentrations in the dialysate only partially reflects the true concentrations in the extracellular fluid. For the determination of the true extracellular concentrations, it is important to know the relative recovery of the used microdialysis system. The relative recovery is the concentration of a particular substance in the dialysate when it leaves the microdialysis catheter expressed as percentage of the surrounding concentration.

The relative recovery in vitro was determined for 11 microdialysis catheters after withdrawal from the patients. The microdialysis catheters were immersed in a test solution (glucose 5.6 mM, urea 13.3 mM) and perfused with Ringer's solution (0.3 uL/min). From each catheter, 5 samples collected for 15 minutes were analyzed. The mean relative recovery ± standard deviation was 101% ± 3% for glucose and 100% ± 2% for urea.

For the determination of the relative recovery in vivo, urea was used as the internal standard. Urea is not metabolized in the adipose tissue and is in equilibrium between blood and extracellular space. Urea serum concentration can therefore be used as an accurate measure of its interstitial level, and the ratio between the dialysate and serum concentration of urea represents the relative recovery in vivo.¹¹ The relative recovery in vivo remained stable during the microdialysis period. In 9 patients, pairs of dialysate/blood samples were obtained on day 1 or 2 and also on day 8 or 9 during microdialysis. The mean (± standard deviation) relative recovery in vivo was 96.4% (±12.7%) at the beginning and 91.1% (±10.5%) after 1 week of microdialysis. There was no significant (P = .260) difference (Wilcoxon test). In 4 infants, multiple pairs of dialysate/blood samples were available and the relative recovery remained stable up to 16 days.

Urea

For urea, 67 pairs of dialysate/serum samples were obtained from 13 patients (Fig 3). The concentrations of serum urea were in the range 1.3 mM to 12 mM. The distribution of the serum values and the distribution of the dialysate values were not significantly different from a normal distribution when tested by the Kolmogorov-Smirnov test with lilliefors adaptation. Multiple linear regression analysis with dummy-variables, characterizing the affiliation of the values to the various patients, demonstrated a high correlation (r = 0.984; P < .001) between the serum and the dialysate concentration of urea. Between the median concentration of urea in the serum (6.3 mM) and in the dialysate (6.0 mM), there was a minor but significant (P = .001) difference (0.3 mM).

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Relationship between serum-urea concentration and the dialysate-urea concentration (r = 0.984; P < .001; serum-urea [mM] = 0.76 + 0.92 x dialysate-urea [mM]). The 67 pairs of dialysate/serum samples were from 13 patients. Each patient is shown by a different symbol.

Glucose

For glucose, 194 pairs of dialysate/blood samples were obtained from 13 patients (Fig 4). Blood glucose ranged from 1.5 mM to 20.8 mM. Between the median concentration of glucose in the blood (4.3 mM) and in the dialysate (3.7 mM) there was a significant (P < .001) difference (0.6 mM). Values of blood glucose and dialysate glucose possessed an approximately log-normal distribution. Multiple linear regression analysis with dummy-variables, characterizing the affiliation of the values to the various patients, demonstrated a significant correlation (r = 0.882; P < .001) between the logarithmic values of blood- and dialysate glucose.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Relationship between the logarithmic blood-glucose concentration and the logarithmic dialysate-glucose concentration (r = 0.882; P < .001; blood-glucose [mM log] = 0.18 + 0.81 × dialysate-glucose [mM log]). The 194 pairs of dialysate/blood samples were from 13 patients. Each patient is shown by a different symbol.

Increase and decrease of the blood glucose concentration could be detected by subcutaneous microdialysis (Fig 5).

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Subcutaneous interstitial glucose during hyperglycemia treated with insulin. At the age of 5 weeks, an infant of consanguine parents, born at term, developed an acute liver failure. Despite intensive diagnostic, the cause remained unclear and the child died at the age of 7 weeks. Three days before death, the infant developed hyperglycemia and was treated with insulin (0.1-0.3 U/kg/hour, shaded area). This shows the subcutaneous interstitial glucose measured by microdialysis (O) and the blood glucose ().

In 26 of the 194 pairs of dialysate/blood samples, the blood glucose was below 2.8 mM and in 9 it was below 2.2 mM. Figure 6 shows the receiver-operating characteristics (ROC) for the detection of an optimal cutoff for a blood glucose below 2.8 mM. All values of blood glucose were used as cutoff for calculation of sensitivity and specificity and to draw the corresponding ROC-curve. The area under the curve was 0.930 (P < .001). For the diagnosis of a blood glucose below 2.8 mM, a dialysate glucose of 2.9 mM possessed a sensitivity of 92.3% and a specificity of 88.1%, a negative predictive value of 98.7%, a positive predictive value of 54.5% (prevalence of a blood glucose 2.8 mM was 13.4%), and an accuracy of 88.7%. The area under the curve of the ROC curve for a blood glucose below 2.2 mM was 0.933 (P < .001). For a blood glucose below 2.2 mM, a dialysate glucose of 2.4 mM possessed a sensitivity of 88.9%, a specificity of 87.0%, a negative predictive value of 99.4%, a positive predictive value of 25% (prevalence of a blood glucose 2.2 mM was 4.6%), and an accuracy of 87.1%.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   ROCs for the diagnosis of a blood glucose below 2.8 mM. The area under the curve is 0.930 (P < .001). For the diagnosis of a blood glucose below 2.8 mM, a dialysate glucose of 2.9 mM possessed a sensitivity of 92.3% and a specificity of 88.1%, a negative predictive value of 98.7%, a positive predictive value of 54.5% (prevalence of a blood glucose 2.8 mM was 13.4%) and an accuracy of 88.7%.

	DISCUSSION

Top Abstract Methods Results Discussion References

We demonstrated that continuous long-term microdialysis of subcutaneous adipose tissue can be used for glucose monitoring in neonates and infants during intensive care. The used microdialysis technique was safe and well tolerated by the neonates and infants.

The insertion of the microdialysis catheter into the subcutaneous tissue with an introducer is simple and can be compared with the placement of a venous cannula. The length of the used dialysis membrane (10 mm) was shorter than in the other studies concerning infants and children (30 mm).^5-7,12 The shorter membrane allowed the safe application even in ELBW infants (<1000 g) with scanty subcutaneous adipose tissue.

Subcutaneous microdialysis has previously been performed up to 4 days in neonates during intensive care^5,6 and for 3 and 7 days in adult insulin-dependent diabetic patients.^13,14 Subcutaneous microdialysis continued in this study for 4 to 16 days (median: 9 days). Long-term microdialysis for 7 to 16 days has also been used for neurochemical monitoring of the brain in adult patients without complications.^15-17 Microdialysis seems to be a safe technique for long-term monitoring of patients.

When microdialysis is used for long-term monitoring, it is important to know the reliability of the relative recovery during the course. Probably the tissue response to the microdialysis catheter may influence the diffusion characteristics because a scar may form a barrier altering the relative recovery in vivo.¹⁰ After minimal traumatic implantation, the microdialysis catheters used in this study showed a constant very high relative recovery in vivo of 90% during long-term microdialysis and a relative recovery in vitro of 100% after removal. The high relative recovery in vitro and in vivo is a prerequisite for clinical investigations if approximation of "true" values is mandatory.⁹

In the long-term, subcutaneous microdialysis study presented here a significant correlation was found between the subcutaneous dialysate and the blood levels of urea (r = 0.98). This indicates that for the monitoring of small molecules that are not metabolized in the subcutaneous tissue, microdialysate, which represents the interstitial fluid, can be used as equivalent for blood. Microdialysis has been applied in humans for pharmacokinetic studies.^18,19

From diabetic patients, it has been reported that the interstitial adipose tissue concentration of glucose closely mirrors the serum glucose.^14,20 In neonates, a close correlation (r = 0.97) has been described between blood and interstitial glucose in the range of 1.9 to 5.9 mM. This observation was based on 14 pairs of samples obtained by microdialysis from 7 neonates.⁵ Based on 194 pairs of samples in the range of 1.5 to 20.8 mM from 13 patients, we found also a significant correlation (r = 0.882) between blood glucose and the dialysate glucose, which represents the interstitial glucose.

Hypoglycemia is a common problem in neonatal intensive care. Despite the uncertainty concerning neonatal hypoglycemia with regard to definition and outcome, aggressive diagnosis and treatment is recommended to prevent threat to the brain.^21-23 No uniform standard is accepted for defining a blood glucose concentration diagnostic for hypoglycemia.^21,22,24 In neonates at risk for hypoglycemia, a serum glucose greater than 2.8 mM should be maintained²³ and values below 2.2 mM call for treatment.²²

Dependent on the prevalence of hypoglycemias, aggressive diagnostic by blood sampling induces numerous negative tests while the risk of missing asymptomatic hypoglycemias during the test interval persists. Furthermore, there is the uncertainty concerning the duration once the hypoglycemia is detected.

Studies in adults and children demonstrated that subcutaneous microdialysis can detect the decrease of the interstitial glucose concentration during induced hypoglycemia.^12,1425-27 The detection of spontaneous hypoglycemias using subcutaneous microdialysis has been demonstrated in patients with insulin-dependent diabetes mellitus.^13,28,29

This study demonstrated that asymptomatic hypoglycemias can be detected in neonates by continuous subcutaneous microdialysis. Based on the ROCs an optimal cutoff value can yield a sensitivity and specificity of approximately 90% and a negative predictive value of 99%. Compared with blood sampling, subcutaneous microdialysis probably facilitates a more efficient glucose monitoring by the detection of asymptomatic hypoglycemias and by the reduction of negative blood tests, an unnecessary painful procedure and blood loss.

The present study demonstrates that microdialysis is well suited for monitoring the glucose metabolism in neonates during intensive care.

Additional studies should evaluate the significance of other metabolites and drugs in the subcutaneous tissue for diagnostic and therapy.

Subcutaneous long-term microdialysis is a promising approach for continuous therapy monitoring and the reduction of diagnostic blood loss during neonatal intensive care.

	FOOTNOTES

Received for publication Aug 30, 2000; accepted Jul 9, 2001.

Address correspondence to Friedrich A. M. Baumeister, MD, Kinderklinik und Poliklinik der Technischen Universität München, Kinderklinik Schwabing, Kölner Platz 1, 80 804 München. E-mail: u7r11cf{at}mail.lrz-muenchen.de

	ABBREVIATIONS

ELBW, extremely low birth weight, ROC, receiver-operating characteristics.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Obladen M, Sachsenweger M, Stahnke M Blood sampling in very low birth weight infants receiving different levels of intensive care. Eur J Pediatr 1988; 147:399-404 [CrossRef][Medline]
2. Maier RF, Obladen M, Messinger D, Wardrop CAJ Factors related to transfusion in very low birthweight infants treated with erythropoietin. Arch Dis Child 1996; 74:F182-F186
3. Hamberger A, Jacobson I, Nyström B, Sandberg M Microdialysis sampling of the neuronal environment in basic and clinical research. J Intern Med 1991; 230:375-380 [Medline]
4. Ungerstedt U Microdialysis-principles and applications for studies in animal and man. J Intern Med 1991; 230:365-373 [Medline]
5. Horal M, Ungerstedt U, Persson B, Westgren M, Marcus C Metabolic adaptation in IUGR neonates determined with microdialysisa pilot study. Early Hum Dev 1995; 42:1-14 [Medline]
6. Hildingsson U, Selldén H, Ungerstedt U, Marcus C Microdialysis for metabolic monitoring in neonates after surgery. Acta Paediatr 1996; 85:589-594 [Medline]
7. Hildingsson U, Lönnqvist P-A, Selldén H, Eksborg S, Ungerstedt U, Marcus C Age-dependent variations in white adipose tissue glycerol and lactate production after surgery measured by microdialysis in neonates and children. Paediatr Anaesth 2000; 10:283-289 [CrossRef][Medline]
8. Arner P, Bolinder J Microdialysis of adipose tissue. J Intern Med 1991; 230:381-386 [Medline]
9. Landolt H, Langemann H Cerebral microdialysis as a diagnostic tool in acute brain injury. Eur J Anaesthesiol 1996; 13:269-278 [CrossRef][Medline]
10. Nakahara D, Ozaki N, Nagatsu T. In vivo microdialysis of neurotransmitters and their metabolites. In: Parvez SH, Naoi M, Nagatsu T, Parvez S, eds. Methods in Neurotransmitter and Neuropeptide Research. Paris, France: Elsevier Science Publishers; 1993:219-248
11. Eisenberg EJ, Eickhoff WM A method for estimation of extracellular concentration of compounds by microdialysis using urea as an endogenous recovery marker in vitro validation. J Pharmacol Toxicol Methods 1993; 30:27-31 [Medline]
12. Kamel A, Norgren S, Persson B, Marcus C Insulin induced hypoglycaemia: comparison of glucose and glycerol concentrations in plasma and microdialysate from subcutaneous adipose tissue. Arch Dis Child 1999; 80:1-3 [Free Full Text]
13. Bolinder J, Ungerstedt U, Arner P Long-term continuous glucose monitoring with microdialysis in ambulatory insulin-dependent diabetic patients. Lancet 1993; 342:1080-1085 [CrossRef][Medline]
14. Hashiguchi Y, Sakakida M, Nishida K, Uemura T, Kajiwara KI, Shichiri M Development of a miniaturized glucose monitoring system by combining a needle-type glucose sensor with microdialysis sampling method. Diabetes Care 1994; 17:387-396 [Abstract]
15. During MJ, Fried I, Leonne P, Katz A, Spencer DD Direct measurement of extracellular lactate in the human hippocampus during spontaneous seizures. J Neurochem 1994; 62:2356-2361 [Medline]
16. Hillered L, Valtysson J, Enblad P, Persson L Interstitial glycerol as a marker for membrane phospholipid degradation in the acutely injured human brain. J Neurol Neurosurg Psychiatry 1998; 64:486-491 [Abstract/Free Full Text]
17. Persson L, Valtysson J, Enbland P, Neurochemical monitoring using intracerebral microdialysis in patients with subarachnoid hemorrhage. J Neurosurg 1996; 84:606-616 [Medline]
18. Scheyer RD, During MJ, Hochholzer JM, Spencer DD, Cramer JA, Mattson RH Phenytoin concentrations in the human brain: an in vivo microdialysis study. Epilepsy Res 1997; 18:227-232
19. Stähle L, Alm C, Ekquist B, Lundquist B, Tomson T Monitoring free extracellular valproic acid by microdialysis in epileptic patients. Ther Drug Monit 1996; 18:14-18 [Medline]
20. Bolinder J, Ungerstedt U, Arner P Microdialysis measurement of the absolute glucose concentration in subcutaneous adipose tissue allowing glucose monitoring in diabetic patients. Diabetologia 1992; 35:1177-1180 [CrossRef][Medline]
21. Cowett RM. Hypoglycemia in the newborn. In: Lifshitz F, ed. Pediatric Endocrinology. New York, NY: Dekker; 1996:677-692
22. Polk DH. Disorders of carbohydrate metabolism. In: Taeusch HW, Ballard RA, eds. Avery's Diseases of the Newborn. Philadelphia, PA: Saunders; 1998:1235-1241
23. Schwartz RP Neonatal hypoglycemia: how low is too low? J Pediatr 1997; 131:171-173 [Medline]
24. Cornblath M, Hawdon JM, Williams AF, Controversies regarding definition of neonatal hypoglycemia: suggested operational thresholds. Pediatrics 2000; 105:1141-1145 [Abstract/Free Full Text]
25. Maggs DG, Jacob R, Rife F, Caprio S, Tamborlane WV, Sherwin RS Counterregulation in peripheral tissues: effect of systemic hypoglycemia on levels of substrates and catecholamines in human skeletal muscle and adipose tissue. Diabetes 1996; 46:70-76 [Abstract]
26. Moberg E, Hagström-Toft E, Arner P, Bolinder J Protracted glucose fall in subcutaneous adipose tissue and skeletal muscle compared with blood during insulin-induced hypoglycaemia. Diabetologia 1997; 40:1320-1326 [CrossRef][Medline]
27. Moberg E, Hagstrom-Toft E, Bolinder J Detection of hypoglycaemia by microdialysis measurements of glucose in subcutaneous adipose tissue. Horm Metab Res 1997; 29:440-443 [Medline]
28. Bolinder J, Hagström-Toft E, Ungerstedt U, Arner P Self-monitoring of blood glucose in type I diabetic patients: comparison with continuous microdialysis measurements of glucose in subcutaneous adipose tissue during ordinary life conditions. Diabetes Care 1997; 20:64-70 [Abstract]
29. Hagstrom-Toft E, Hellstrom L, Moberg E Lipolytic response during spontaneous hypoglycaemia in insulin-dependent diabetic subjects. Horm Metab Res 1998; 30:586-593 [Medline]