Hypoglycemia in Nondiabetic Patients Undergoing Albumin Dialysis by Molecular Adsorbent Recirculating System Ai-Leng Khoo,* Lai-San Tham,* Gek-Kee Lim,* and Kang-Hoe Lee†‡ It was observed that patients developed episodes of hypoglycemia during molecular adsorbent recycling system (MARS) treatment. The aim of this study is to assess the effect of MARS treatment on blood glucose concentration to formulate appropriate dextrose replacement guidelines during MARS dialysis. Five patients with liver failure each underwent a 6- to 8-hour MARS treatment. No patient had a history of diabetes or was administered insulin or oral antihyperglycemic agents throughout the period of albumin dialysis. There was no active intervention or restriction on glucose intake. Rather, a dextrose drip and boluses were allowed based on each patient’s condition and the clinical judgment of the attending physician. Blood glucose concentration was monitored hourly during the period of MARS treatment. Glucose loss in dialysate fluid was quantified hourly by measuring the total volume of dialysate fluid and assaying the glucose concentration in dialysate fluid. Mean glucose removal during a 6-hour MARS session was 37.19 ⴞ 5.58 g. Mean glucose removal rate was 6.20 ⴞ 0.93 g/h. In addition to a maintenance drip supporting the caloric requirement of patients, a dextrose replacement drip that paralleled the rate of glucose removal would prevent patients from experiencing episodes of hypoglycemia during MARS treatment. Dextrose replacement at a mean rate of 6 g/h (range, 5 to 7 g/h) in patients without diabetes undergoing albumin dialysis by MARS is recommended. (Liver Transpl 2003;9:949-953.)
T
he recently developed molecular adsorbent recirculating system (MARS) represents a new nonbiological liver support system that enables the selective removal of albumin-bound substances. Disease states treated with MARS included acute exacerbation of chronic hepatic failure, hepatorenal syndrome, acute hepatic failure, and graft dysfunction after liver transplantation.1 MARS uses a hollow-fiber dialysis module in which the patient’s blood is dialyzed across an albumin-impregnated membrane, with an albumin dialysate in a closed-loop dialysis circuit.1-4 The albumin molecule has free binding sites that compete with toxins bound to proteins in the perfused blood. Adsorbed toxins from the membrane return to the dialysate, which then is dialyzed against bicarbonate-buffered dialysate and subsequently regenerated by passage through an anion-exchange and activated-charcoal column. The
system thus detoxifies protein-bound toxins, for example, bilirubin, aromatic amino acids, and phenols that accumulate in liver failure. Water-soluble substances, such as ammonia and creatinine, also are removed by means of diffusion and hemofiltration. Removal of albumin-bound toxins has been shown to result in improvement in hepatic encephalopathy, hemodynamic parameters, and kidney and liver function.5-13 From the authors’ experience with this novel liver support device, it was observed through routine monitoring of plasma glucose concentrations that patients undergoing MARS dialysis experienced episodes of asymptomatic hypoglycemia (plasma glucose concentration ⬍ 4 mmol/L) during the second and third hour of MARS treatment and onward. Without dextrose replacement, plasma glucose concentrations continued to decrease throughout the course of dialysis. Changes in blood glucose concentrations have been studied extensively in patients undergoing hemodialysis and peritoneal dialysis.14-18 Hemodialysis-induced hypoglycemia has been observed in patients with and without diabetes using a glucose-free dialysate fluid. Similarly, MARS uses a glucose-free dialysate and removes water-soluble substances. We hypothesized that glucose, with a low molecular weight of 180 d, is removed by MARS, resulting in a decrease in plasma glucose concentration. Because hypoglycemia often has been associated with liver failure and MARS treatment could potentially aggravate glucose control, the imporFrom the Departments of *Pharmacy and †Medicine, National University Hospital; and the ‡Department of Medicine, National University of Singapore, Singapore. Supported in part by a grant from the National Healthcare Group Research Fund, Singapore. Address reprint requests to Ai-Leng Khoo, BSc(Pharm)(Hon), Department of Pharmacy, National University Hospital, 5 Lower Kent Ridge Rd, Singapore 119074. Telephone: 65-6772-5239; FAX: 65-6873-7121; E-mail:
[email protected] Copyright © 2003 by the American Association for the Study of Liver Diseases 1527-6465/03/0909-0008$30.00/0 doi:10.1053/jlts.2003.50178
Liver Transplantation, Vol 9, No 9 (September), 2003: pp 949-953
949
950
Khoo et al
tance of close blood glucose monitoring with appropriate therapy should be recognized. Therefore, the purpose of this study is to investigate the effect of MARS treatment on plasma glucose concentration and the significance of glucose removal by the current mode of dialysis. Knowledge of the rate of glucose loss in dialysate fluid would be of value in making an informed decision with regard to dextrose replacement during the course of MARS dialysis.
Patients and Methods Five patients without diabetes undergoing MARS treatment were studied. Our institutional criteria for the selection of patients undergoing albumin dialysis included grade III and higher hepatic encephalopathy and/or a total bilirubin level greater than 300 mol/L. All patients were receiving standard intensive care management. There was no active intervention or restriction on glucose intake. Dextrose drip and boluses were allowed based on the individual patient’s condition. During each course of 6- or 8-hour albumin dialysis, we monitored the patient’s plasma glucose concentration, as well as amount of glucose loss in the dialysate fluid hourly. The amount of dialysate fluid exhausted by the process was quantified, and the glucose concentration in the dialysate fluid was assayed every hour. Capillary blood glucose concentration was monitored using the Advantage blood glucose meter (Roche Diagnostics, Mannheim, Germany). Glucose concentration in the dialysate fluid was assayed by means of photocolorimetry using the Vitros 950 analyzer (Johnson and Johnson, Rochester, NY). From the amount of dialysate fluid and glucose concentration, the glucose removal rate was quantified. During MARS treatment, the extracoporeal blood circuit was driven by AK 100 Ultra (Gambro, Copenhagen, Denmark) continuous venovenous hemodiafiltration equipment. The closed-loop albumin circuit was controlled by the MARS monitor (Teraklin AG, Rostock, Germany). Components of the MARS treatment kit (Teraklin AG) included a MARS dialyzer (MARSFlux 1S), low-flux dialyser (diaFlux 1S),
adsorber cartridges filled with anionic exchanger resin (diaMARS AC 250), adsorber cartridges filled with activated charcoal (diaMARS IE 250), and tube set (AS-01). For each dialysis cycle, 600 mL of 20% human albumin (Albumin Human 20%; Octapharma, Vienna, Austria) of albumin dialysate used. Albumin dialysate was recirculated at a flow rate of 250 mL/min through the dialysate compartment of the dialyzer. Blood flow rates through the dialyzer ranged from 200 to 250 mL/min. Vascular access was obtained by insertion of a dialysis double-lumen catheter. The ultrafiltration rate was 2 L/h. Heparin was used as an anticoagulant to prevent thrombosis in the extracorporeal circulation.
Results Five patients without diabetes (four men, one woman; mean age, 51 years; range, 41 to 61 years) with liver impairment requiring albumin dialysis by MARS were included in this study. No patient was administered insulin or oral antihyperglycemic agents throughout the course of this study. Patient clinical and biochemical data are listed in Table 1. Indications for albumin dialysis were hyperbilirubinemia and recurrent grade III hepatic encephalopathy after previous MARS treatments in patients 1 and 2 and fulminant liver failure in patient 5. Causes of drug-induced fulminant hepatic failure requiring MARS treatment likely were associated with bupropion and methotrexate in patients 3 and 4, respectively. During the course of a 6-hour MARS treatment, mean amount of glucose removed was 37.19 ⫾ 5.58 g (n ⫽ 5). Amount of glucose removed per hour ranged from 4.82 to 7.47 g/h (Table 2). Based on the five patients, mean glucose removal rate was 6.20 ⫾ 0.93 g/h. During the second cycle of MARS treatment, patient 1 was administered 10% dextrose replacement
Table 1. Clinical Status of Patients Before MARS Treatment Patient No.
Gender
Age (yrs)
Cause of Liver Failure
Total Bilirubin (mol/L)
Ammonia (mol/L)
Prothrombin Time (sec)
Creatinine (mol/L)
1 2 3 4 5
M M M F M
61 55 41 43 50
HCC secondary to HCV infection HBV infection Drug-induced (bupropion) Drug-induced (methotrexate) HBV infection
406 359 1,437 155 214
22 33 40 184 184
31.4 45.4 24.9 ⬎100 59.8
90 66 398 56 192
Abbreviations: HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HBV, hepatitis B virus.
951
Hypoglycemia During MARS Treatment
Table 2. Loss of Glucose in Dialysate Fluid During a 6-Hour MARS Treatment Cycle 1
Cycle 2
Patient No.
Total Glucose Removal in 6 H (g)
Mean Glucose Removal (g/h)
Total Glucose Removal in 6 H (g)
Mean Glucose Removal (g/h)
1 2 3 4 5
37.36 38.65 28.95 44.84 33.98
6.23 ⫾ 1.15 (4.22-7.25) 6.44 ⫾ 1.11 (4.36-7.44) 4.82 ⫾ 0.87 (3.78-6.28) 7.47 ⫾ 1.01 (6.28-8.94) 6.20 ⫾ 0.93 (4.36-6.53)
42.28 40.67 30.78 — —
7.05 ⫾ 0.84 (5.67-8.05) 6.78 ⫾ 1.07 (5.19-8.06) 5.13 ⫾ 1.37 (3.31-6.76) — —
Note. Values expressed as mean ⫾ SD (range).
(at 22 mL/h) throughout the cycle. This could account for a greater rate of removal (7.05 g/h) compared with the first cycle (6.23 g/h), in which 40 mL of 50% dextrose followed by 20% dextrose replacement (at 20 mL/h) was administered during the last 2 hours of dialysis. Patient 2 was administered a bolus dose of 50% dextrose for each cycle. Patient 3 was administered 5% dextrose at a rate of 22 mL/h throughout the first MARS treatment and 50% dextrose at 22 mL/h for the second MARS treatment. A continuous 50% dextrose drip (at 22 mL/h) was administered to patient 4 during the last 3 hours of his MARS treatment. We observed that all five patients experienced episodes of hypoglycemia during the course of MARS treatment (Figs.1 through 5), especially from the second hour onward.
Discussion Our results indicate a significant loss of glucose in dialysate fluid, which likely was a factor contributing to
Figure 1. Glucose removal and plasma glucose profile for patient 1.
hypoglycemic episodes observed in our patients. Loss of glucose could be attributed to its low molecular weight (180 d) and water-soluble nature, which enabled it to transverse the MARS membrane with ease. The MARS membrane is permeable to molecules not larger than 50 kilodaltons. Glucose is drawn from the albumin circuit into the secondary dialyzer and removed in dialysate fluid. The other contributing factor could be the high blood flow rate (250 mL/min) used in this dialysis system. Jackson et al15,16 reported glucose loss in dialysate fluid at a rate of 6 and 9.2 g/h in 21 patients without diabetes and 18 patients with diabetes dialyzed with glucose-free dialysate, respectively. Similarly, our patients underwent liver dialysis against a 20% albumin solution and glucose-free dialysate, and the rate of glucose loss ranged from 4.8 to 7.5 g/h. Hepatic glucose metabolism characterized by impaired insulin sensitivity, high levels of insulin and glucagon, and hypoglycemia has been observed in patients with liver diseases.19 Increased glucogenesis and impaired hepatic glycogenolysis have been reported
Figure 2. Glucose removal and plasma glucose profile for patient 2.
952
Khoo et al
Figure 3. Glucose removal and plasma glucose profile for patient 3.
Figure 5. Glucose removal and plasma glucose profile for patient 5.
in patients with liver cirrhosis.20 The resultant derangement in plasma glucose levels in such patients could have been compounded by its loss during MARS dialysis. We conclude that significant glucose loss in dialysate fluid during the course of MARS treatment possibly could lead to asymptomatic hypoglycemic episodes. It thus is essential to monitor and detect hypoglycemia during the course of albumin dialysis by MARS. To normalize glycemic control in such patients, the rate of dextrose replacement should essentially parallel the rate of glucose loss in dialysate. Therefore, in addition to the daily caloric requirement of 30 kcal/kg for a patient in the critical care unit or as calculated from the HarrisBenedict equation,19 we advocated additional dextrose replacement of 5 to 7 g/h throughout the course of liver dialysis. Alternatively, glucose-containing dialysate may be used, as commonly practiced in other countries.
Acknowledgment
Figure 4. Glucose removal and plasma glucose profile for patient 4.
The authors thank Linus Khoo, Margaret Lee, and the nursing team in the medical intensive care unit for their assistance throughout the study.
References 1. MARS liver support therapy. Therapy information 2001:1. 2. Stange J, Mitzner S, Ramlow W, Gliesche T, Hickstein H, Schmidt R. A new procedure for the removal of protein bound drugs and toxins. ASAIO J 1993;39:M621-M625. 3. Stange J, Mitzner S. A carrier-mediated transport of toxins in a hybrid membrane. Safety barrier between a patient’s blood and a bioartificial liver. Int J Artif Organs 1996;19:677-691. 4. Stange J, Mitzner S, Risler T, Erley CM, Lauchart W, Goehl H, et al. Molecular adsorbent recycling system (MARS): Clinical results of a new membrane-based blood purification system for bioartificial liver support. Artif Organs 1999;23:319-330. 5. Mitzer S, Stange J, Klammt S, Peszynski P, Schmidt R, Schoburg GN. Extracorporeal detoxification using the molecular adsorbent recirculating system for critically ill patients with liver failure. J Am Soc Nephrol 2001;12(suppl):S75-S82. 6. Novelli G, Rossi M, Pretagostini R, Poli L, Peritore D, Berloco P, et al. Use of MARS in the treatment of acute liver failure: Preliminary monocentric experience. Transplant Proc 2001;33: 1942-1944. 7. Mitzner S, Stange J, Klammt S, Risler T, Erley CM, Bader BD, et al. Improvement of hepatorenal syndrome with extracorporeal albumin dialysis MARS: Results of a prospective, randomized, controlled clinical trial. Liver Transpl 2000;6:277-286. 8. Mitzner S, Stange J, Klammt S, Peszynski P, Risler T, Schmidt R. Albumin dialysis “MARS”: Clinical results in hepatorenal syndrome (HRS)—Treatment [abstract]. J Hepatol 1999;30(suppl 1):S83. 9. Stange J, Mitzner S, Klammt S, Freytag J, Klammt S, Peszynski P, et al. Extracorporeal liver support therapy for patients suffering from acute on chronic hepatic failure (AOCRF) results in an improvement of Child-Turcotte-Pugh (CTP) [abstract]. J Hepatol 1999;30(suppl 1):S79.
Hypoglycemia During MARS Treatment
10. Sorkine P, Abraham RB, Szold O, Biderman P, Kidron A, Merchav H, et al. Role of the molecular adsorbent recycling system (MARS) in the treatment of patients with acute exacerbation of chronic liver failure. Crit Care Med 2001;29:1332-1336. 11. Schmidt LE, Svendsen LB, Sorensen VR, Hansen BA, Larsen FS. Cerebral blood flow velocity increases during a single treatment with the molecular adsorbents recirculating system in patients with acute on chronic liver failure. Liver Transpl 2001;7:709712. 12. Schmidt LE, Sorensen VR, Svendsen LB, Hansen BA, Larsen FS. Hemodynamic changes during a single treatment with the molecular adsorbents recirculating system in patients with acuteon-chronic liver failure. Liver Transpl 2001;7:1034-1039. 13. Stange J, Mitzner S, Klammt S, Peszynski P, Freytag J, Hickstein H, et al. Inhibition of NO synthase by arginine removal by liver support (MARS) seems to be only a cofactor for reversal of hypotension in decompensated liver cirrhosis that seems to be linked more to removal of protein bound toxins [abstract]. Hepatology 2000;32:612A. 14. Tzamaloukas AH, Murata GH, Eisenberg B, Murphy G, Avasthi PS. Hypoglycemia in diabetics on dialysis with poor glycemic
15.
16.
17.
18.
19.
20.
953
control: Hemodialysis versus continuous ambulatory peritoneal dialysis. Int J Artif Organs 1992;15:390-392. Jackson MA, Holland MR, Nicholas J, Lodwick R, Forster D, MacDonald IA. Hemodialysis-induced hypoglycemia in diabetic patients. Clin Nephrol 2000;54:30-34. Jackson MA, Holland MR, Nicholas J, Talbot M, Spencer H, Lodwick R, et al. Occult hypoglycemia caused by hemodialysis. Clin Nephrol 1999;51:242-247. Catalano C, Bordin V, Fabbian F, Landro D. Glucose-free standard hemodialysis and occult hypoglycemia. Clin Nephrol 2000; 53:235-236. Passadakis P, Thodis E, Vargemezis V, Oreopoulos G. Recommendations for glucose control in diabetics on CAPD. Int J Artif Organs 1999;22:657-664. ASPEN Board of Directors. Guidelines for the use of parenteral and enteral nutrition in adult and pediatrics patients. JPEN 2002;26:1SA-138SA. Petersen KF, Krssak M, Navarro V, Chandramouli V, Hundal R, Schumann WC, et al. Contributions of net hepatic glycogenolysis and gluconeogenesis to glucose production in cirrhosis. Am J Physiol 1999;276:E529-E535.
