A Case Report of Pediatric Fulminant Hepatitis TreatedWith Plasma Diafiltration

Hajime Nakae,1 Toshiko Igarashi,1 Kimitaka Tajimi,1 Atsuko Noguchi,2 Ikuko Takahashi,2

Satoko Tsuchida,2 Tsutomu Takahashi,2 and Yoshihiro Asanuma3

Departments of 1Integrated Medicine, Emergency and Critical Care Medicine, and 2Reproductive andDevelopmental Medicine, Division of Pediatrics, and 3School of Health Sciences, Akita University School of

Medicine, Akita, Japan

Abstract: Plasma diafiltration (PDF) is blood purificationtherapy in which simple plasma exchange is performedwith a membrane plasma separator while dialysate flowsoutside the hollow fibers. A 14-year-old boy with fulminanthepatitis underwent two sessions of PDF and one session ofhemodiafiltration.We infused filtered replacement fluid forartificial kidneys at a dialysate flow rate of 600 mL/h and areplacement flow rate of 450 mL/h.We infused fresh frozenplasma (1200 mL) and 25% albumin solution (50 mL)

intravenously over 8 h. Each PDF session lasted 8 h. Thepatient’s total bilirubin, interleukin-18, and cystatin Clevels decreased with treatment, and he recovered fromhepatic failure. PDF may be an extremely useful bloodpurification therapy for pediatric fulminant hepatitisin terms of both medical economics and cytokineremoval. Key Words: Fulminant hepatitis, Pediatric,Plasma diafiltration, Plasma exchange.

Plasma diafiltration (PDF) is blood purificationtherapy in which simple plasma exchange (PE) isperformed with a membrane plasma separator whiledialysate flows outside the hollow fibers. PDF, requir-ing about one-third of the amount of fresh frozenplasma (FFP) required by conventional PE, mayyield results similar to those of conventional PE (1,2).We have previously reported the effectiveness ofPDF in adults with acute liver failure (2); here, wereport the first case of a pediatric patient who under-went PDF for the treatment of fulminant hepatitis.

CASE REPORT

Plasma diafiltration was conducted with theinformed consent of the patient and his family and

the approval of the Ethical Committee of Akita Uni-versity School of Medicine.

A 14-year-old boy suffered a sudden loss ofconsciousness during an automobile trip. He lostconsciousness again at the end of the trip and wasadmitted to Akita University Hospital. Upon admis-sion, Grade II hepatic encephalopathy was present.An electroencephalographic study showed back-ground alpha waves (10–11 Hz); however, slow waves(5–6 Hz), qwaves (6–7 Hz), and a triphasic wave werealso evident. Computed tomography showed no defi-nite abnormality. Laboratory values on admissionare shown in Table 1. The hepaplastin test value was29.0%, which was remarkably low. Serum aspartateaminotransferase (1140 IU/L), alanine aminotrans-ferase (2004 IU/L), and total bile acid (270.9 mmol/L)levels were indicative of severe liver dysfunction. Ful-minant hepatitis was diagnosed and we decided toperform PDF to remove protein-bound substancesand nephrotoxic substances simultaneously.

We used an Evacure EC-2 A plasma separator(Kuraray, Tokyo, Japan). The PDF session lasted 8 hand the blood flow rate was 100 mL/min. Filteredreplacement fluid for artificial kidneys (Na+

Received February 2008.Address correspondence and reprint requests to Dr Hajime

Nakae, Department of Integrated Medicine, Division of Emer-gency and Critical Care Medicine, Akita University School ofMedicine, 1-1-1 Hondo, Akita 010-8543, Japan. Email: nakaeh@doc.med.akita-u.ac.jp

Presented in part at the 27th Annual Meeting of the JapaneseSociety for Apheresis held 2–4 March 2007 in Yokohama, Japan.

Therapeutic Apheresis and Dialysis 12(4):329–332doi: 10.1111/j.1744-9987.2008.00595.x© 2008 The AuthorsJournal compilation © 2008 International Society for Apheresis

329

140.0 mEq/L, K+ 2.0 mEq/L, Ca2+ 3.5 mEq/L, Mg2+

1.0 mEq/L, Cl- 111.0 mEq/L, CH3COO- 3.5 mEq/L,HCO3

- 35.0 mEq/L, C6H12O6 100 mg/dL) (Sublood-BS; Fuso Pharmaceutical, Osaka, Japan) was infusedat a dialysate flow rate of 600 mL/h and a replace-ment flow rate of 450 mL/h. We infused 1200 mL ofFFP and 50 mL of 25% albumin solution intrave-nously over 8 h, and we used nafamostat mesilate(Futhan; Torii Pharmaceutical, Tokyo, Japan)(20 mg/h) as an anticoagulant (3,4). A schematic rep-resentation of the flow of PDF is shown in Figure 1.We performed PDF on hospital days 1 and 3, andhemodiafiltration (HDF) on hospital day 2. Afterthree sessions of blood purification, the patient’slevel of consciousness increased and biochemicalparameters improved. Laboratory values before andafter PDF are shown in Table 2. Total bilirubin, inter-leukin (IL)-18, and cystatin C decreased after PDF.IL-18 levels were determined by ELISA (MBL,Nagoya, Japan); the detection limit is 12.5 pg/mL(normal, 126 � 44.5 pg/mL). Cystatin C levelswere measured by ELISA (BioVendor, Brno,Czech Republic); the normal cystatin C level is1081.9 � 304.7 ng/mL.

Histopathological features of a liver biopsy spe-cimen obtained on hospital day 22 are shown inFigure 2. Fibrous expansion of the portal region withmoderate infiltration by inflammatory cells, mainlylymphocytes, was observed. Liver regeneration wasgood and there were masses of regenerating liver cellsthat were slightly nodular in almost all liver paren-chyma areas. Mild vacuolar degeneration of hepato-cytes was observed. Slight infiltration of lymphocytesand a small colonization of Kupffer cells was seen inthe liver parenchyma area. In short, the histopatho-

FIG. 1. Schematic representation ofthe flow of plasma diafiltration.

FFP1 , 200 mL +FFP1, 25% albumin (50 mL) Anticoaguant25% 150mL/hr (Nafamostat mesilate)

Blood flow rate100 mL/min

Dialysate600 mL/hr

PDFEvacureEC 2A-

Extract 450mL/hr

Replacement fluid

TABLE 1. Laboratory data upon admission

HematologyWhite blood count (WBC) 6500/mm3

Red blood count (RBC) 500 ¥ 104/mm3

Hemoglobin (Hb) 14.2 g/dLHematocrit (HCT) 42.4%Platelets (Plt) 17.0 ¥ 104/mm3

Coagulation systemHepaplastin (HPT) 29.0%Prothrombin time (PT) 40.4%International normalized ratio (INR) 2.48Fibrin degradation products (FDP) 7.30 mg/mLProtein C 40.4%a2-plasmin inhibitor complex (PIC) 1.1 mg/mLThrombin-antithrombin (TAT) 5.8 ng/mLPlasminogen (PLG) 44.3%

Others

Ketone body fractionationTotal ketone bodies 33.7 mmol/LAcetoacetic acid 13.5 mmol/L3-hydroxybutyric acid 20.2 mmol/L

BiochemistryAspartate aminotransferase (AST) 1140 IU/LAlanine aminotransferase (ALT) 2004 IU/LTotal protein (TP) 6.3 g/dLAlbumin (Alb) 3.7 g/dLBlood urea nitrogen (BUN) 6.9 mg/dLCreatinine (Cr) 0.49 mg/dLTotal bilirubin (T-Bil) 3.0 mg/dLDirect bilirubin (D-Bil) 2.2 mg/dLC reactive protein (CRP) 1.14 mg/dLNa+ 136 mEq/LK+ 4.8 mEq/LCl– 101 mEq/LNH3 52 mg/dLTotal bile acid 270.9 mmol/LHepatocyte growth factor (HGF) 0.66 ng/mL

H Nakae et al.330

© 2008 The AuthorsJournal compilation © 2008 International Society for ApheresisTher Apher Dial, Vol. 12, No. 4, 2008

logical findings were indicative of the healing stage offulminant hepatitis.The patient recovered completelyand was discharged on hospital day 31.

Results of viral screening upon admission wereas follows: HBsAg (+), anti-HBs (-), anti-HBc (+),HBeAg (-), anti-HBe (+), IgM anti-HBc 38.8, HBVDNA (PCR) 7.2 log copy/mL; IgM anti-HAV (-);anti-HCV (-); IgM anti-CMV (-), and IgG anti-CMV (-). Results of viral screening on hospitalday 25 were as follows: HBsAg (+), anti-HBs (-), andIgM anti-HBc 28.0. The patient’s father is an HBVcarrier, and results of viral screening of his motherwere as follows: HBsAg (-), anti-HBs (+), anti-HBc (+). We therefore diagnosed fulminant hepatitistype B in this case.

DISCUSSION AND CONCLUSION

The Evacure EC-2 A plasma separator used for PEin our patient has a pore size of 0.008–0.01 mm,which is much smaller than that of the standardplasma separation membrane (0.2–0.4 mm). Thismembrane has a sieving coefficient of 0.3 for albuminand thus can selectively remove low- or intermediate-molecular-weight albumin-bound substances. Inaddition, coagulation factors are preserved becausethis membrane has a sieving coefficient of 0 forfibrinogen (1,2).

In this case, the levels of cystatin C and creatinineare within their normal ranges; nevertheless, bilirubinand cystatin C were removed during treatment of ourpatient (5). In addition, we found that IL-18 couldalso be removed. Endotoxin-induced liver injury orFas ligand-induced hepatitis is caused by endogenousIL-18 in mice. Moreover, patients with liver diseasessuch as fulminant hepatitis, liver cirrhosis due tohepatitis virus infection, and primary biliary cirrhosisshow an elevation of the IL-18 serum level thatcorrelates with the corresponding disease severity.Therefore, endogenous IL-18 plays a major role inthe induction of some types of liver injuries in miceand human (6). IL-18 was involved in the pathologyof liver failure in our previous reports (7–10). ThePDF technology may remove both water-soluble andalbumin-bound toxins. With PE, it is difficult to com-pletely control an increase in the citric acid concen-tration even if continuous hemodiafiltration (CHDF)is performed concomitantly with administration oflarge amounts of FFP. With PDF, however, theincrease in the citric acid concentration is inhibitedmore effectively because smaller amounts of FFP arerequired and because HDF is performed simulta-neously. In addition, the use of smaller amounts ofFFP reduces medical costs (7–10).

Acute liver failure in children leads to severeencephalopathy, coagulopathy, and subsequent

TABLE 2. Changes in total bilirubin, citrate, interleukin (IL)-18, and cystatin C

PDFTotal bilirubin

(mg/dL)Citrate

(mg/dL)IL-18

(pg/mL)Cystatin C(ng/mL)

1stBefore 3.5 2.8 650 887After 2.4 7.8 415 784

2ndBefore 2.7 3.3 640 918After 1.5 8.7 270 873

FIG. 2. Histopathological features of the liver biopsy specimen are indicative of the healing stage of fulminant hepatitis.

PDF in Pediatric Fulminant Hepatitis 331

© 2008 The AuthorsJournal compilation © 2008 International Society for Apheresis Ther Apher Dial, Vol. 12, No. 4, 2008

multiple organ failure, resulting in a high mortalityrate; liver transplantation is therefore considered oneof the best treatment options. PE, CHDF, or a com-bination of PE and CHDF is generally chosen fortreatment of pediatric liver failure (7,11,12). Ours isthe first report of the successful use of PDF for thetreatment of pediatric fulminant hepatitis. PDF maybe effective in the treatment of fulminant hepatitisnot only in adults, but also in children.

In several countries, albumin dialysis with themolecular adsorbent recirculating system (MARS) iscommonly performed in patients with acute liverfailure for the purpose of simultaneously removinghepatotoxic substances and nephrotoxic substances(13). MARS, however, costs more and involves morecomplicated preparation than PDF. The features ofPDF, PE with CHDF, and MARS are shown inTable 3. Our experience suggests that PDF may be anextremely useful blood purification therapy for pedi-atric fulminant hepatitis in terms of both medicaleconomics and cytokine removal. Whether thedecrease of IL-18 levels results in clinical improve-ment or PDF technology must be revealed by exam-ining its clearance. Larger studies are needed tofurther assess the value of this treatment.

REFERENCES

1. Mori T, Eguchi Y, Shimizu T et al. A case of acute hepaticinsufficiency treated with novel plasmapheresis plasma

diafiltration for bridge use until liver transplantation. TherApher 2002;6:463–6.

2. Nakae H, Igarashi T, Tajimi K et al. A case of hepatorenalsyndrome treated with plasma diafiltration. Ther Apher Dial2007;11:391–5.

3. Nakae H, Tajimi K. Pharmacokinetics of nafamostat mesilateduring continuous hemodiafiltration with a polyacrylonitrilemembrane. Ther Apher Dial 2003;7:483–5.

4. Nakae H, Igarashi T, Tajimi K. The dose of nafamostat mesi-late during plasma exchange with continuous hemodiafiltra-tion in the series-parallel circuit. Ther Apher Dial 2006;10:233–6.

5. Shimizu-Tokiwa A, Kobata M, Io H et al. Serum cystatin C isa more sensitive marker of glomerular function than serumcreatinine. Nephron 2002;92:224–6.

6. Tsutsui H, Adachi K, Seki E, Nakanishi K. Cytokine-inducedinflammatory liver injuries. Curr Mol Med 2003;3:545–59.

7. Nakae H, Asanuma Y, Tajimi K. Cytokine removal by plasmaexchange with continuous hemodiafiltration in critically illpatients. Ther Apher 2002;6:419–24.

8. Nakae H, Zheng YJ, Wada H, Tajimi K, Endo S. Involvementof IL-18 and soluble Fas in patients with postoperative hepaticfailure. Eur Surg Res 2003;35:61–6.

9. Nakae H, Yonekawa C, Moon S, Tajimi K. The series-parallelcircuit in the treatment of fulminant hepatitis. Ther Apher Dial2004;8:153–9.

10. Yonekawa C, Nakae H, Tajimi K, Asanuma Y. Effectiveness ofcombining plasma exchange and continuous hemodiafiltrationin patients with postoperative liver failure. Artif Organs 2005;29:324–8.

11. Lowrie LH. Renal replacement therapies in pediatric multi-organ dysfunction syndrome. Pediatr Nephrol 2000;14:6–12.

12. Singer AL, Olthoff KM, Kim H, Rand E, Zamir G, Shaked A.Role of plasmapheresis in the management of acute hepaticfailure in children. Ann Surg 2001;234:418–24.

13. Debray D, Yousef N, Durand P. New management options forend-stage chronic liver disease and acute liver failure: poten-tial for pediatric patients. Paediatr Drugs 2006;8:1–13.

TABLE 3. Features of PDF, PE + CHDF and MARS in pediatric liver failure

Method PDF PE + CHDF MARS

Circuit Simple Complicated More complicatedFFP (per day) 15 units (1.2 L) 15 units (1.2 L) Not usedCitrate increase Moderate Moderate UnknownBilirubin removal Moderate Good ModerateFibrinogen retention Good Poor GoodMaterial cost (per procedure [US$]) $1185 $1444 $2600

(excluding device costs: $33 600)RCT (-) (-) (-)

CHDF, continuous hemodiafiltration; FFP, fresh frozen plasma; MARS, molecular adsorbent recirculating system; PDF, plasmadiafiltration; PE, plasma exchange; RCT, randomized controlled trial.

H Nakae et al.332

© 2008 The AuthorsJournal compilation © 2008 International Society for ApheresisTher Apher Dial, Vol. 12, No. 4, 2008