alpha-1-antitrypsin deficiency in children: liver disease is not ...

December 7, 2005EUROPEAN JOURNAL OF MEDICAL RESEARCH 509Eur J Med Res (2005) 10: 509-514 © I. Holzapfel Publishers 2005ALPHA-1-ANTITRYPSIN DEFICIENCY IN CHILDREN: LIVER DISEASE IS NOTREFLECTED BY LOW SERUM LEVELS OF ALPHA-1-ANTITRYPSIN –A STUDY ON 48 PEDIATRIC PATIENTST. Lang 1 , M. Mühlbauer 1 , M. Strobelt 1 , S. Weidinger 2 , H. B. Hadorn 11 Children´s Hospital, Dr. v. Haunersches Kinderspital,University of Munich, Munich, Germany2 Bavarian Red Cross Transfusion Service, Regensburg, GermanyAbstractBackground: Alpha-1-antitrypsin (α1-AT) is an importantprotease inhibitor. The phenotypes are characterizedby a low total serum α1-AT or by an abnormalprotein accumulating in the hepatocytes. The aim ofour study was to examine a correlation of total serumα1-AT, phenotype, and liver involvement in pediatricpatients.Methods: 48 patients, deficient for α1-AT were included.The phenotypes for α1-AT were determined byisoelectric focusing. Liver disease was defined either aselevated transaminases or/and as elevated conjugatedbilirubin and γGT. Patients were reexamined after amean interval of 2 years.Results: Homozygous α1-AD was found in 12 patients,heterozygous in 24 patients. In 12 children rarevariants of α1-AD were diagnosed. Serum α1-AT levelsless than 60% of normal were found in all patientswith homozygous, in 37% of patients with heterozygousα1-antitrypsin deficiency (α1-AD), and in patientswith the homozygous variant PiM palermo . Liverdisease was found in 8/12 patients with the phenotypePiZZ and in 15/24 patients with heterozygous α1-AD.Three of 4 patients with the phenotype PiMQ0 had severeliver disease despite normal serum levels for α1-AT. In 11 patients with heterozygous α1-AD liver diseasewas apparent despite normal serum α1-AT levels.In two patients with the variant type Mpalermo serumlevels were as low as 11% of normal without any signsof liver disease.Conclusions: Our data clearly show that in the diagnosticworkup of neonatal cholestasis measurement of totalserum α1-AT does not exclude liver disease due toabnormal α1-AT variants. We suggest analysis of α1-AT-phenotype by isoelectric focussing in patients withunknown liver disease. Heterozygous or rare varianttypes might remain undiagnosed by measuring totalα1-AT only.Key words: α1-antitrypsin deficiency, children, liverINTRODUCTIONAlpha-1-antitrypsin or protease inhibitor (Pi) is a singlechain 52 kD glycoprotein consisting of 394 aminoacids The protein is encoded by a gene located on thedistal long arm of chromosome 14 (14q31-32.2) [17].Structural variants of α1-AT in humans are classifiedaccording to the protease inhibitor phenotype systemas defined by agarose gel electrophoresis or isoelectricfocusing of plasma [15]. To date more than 75 differentvariants are known, some of them leading to severeliver disease in early childhood or early pulmonaryemphysema [16, 21]. Several variants of α1-AT are associated with a reduction in serum α1-ATconcentrations, called deficiency variants. The geneticdeficiency is inherited in an autosomal co-dominantfashion. Homozygous PiZ α1-antitrypsin deficiencyhas been associated with liver disease in childhood,leading to endstage liver disease in a subset of patients[10]. This variant is characterized by a distinctivehistopathological feature of PAS positive inclusionbodies in the endoplasmatic reticulum of the hepatocytes[23]. In a prospective screening study of 200.000newborns in Sweden, 127 children were identified asbeing deficient for α1-AT due to homozygous α1-antitrypsin deficiency PiZ [19]. A subset of 14 patientspresented with neonatal cholestasis, 3 of themdeveloped severe liver disease. However α1-antitrypsindeficiency is still the most frequent metabolicliver disease leading to end stage liver disease. About10% of patients with the homozygous variant PiZ requireliver transplantation during childhood [6]. Anincreased prevalence of neonatal liver disease is reportedin patients with heterozygous PiMZ α1-antitrypsindeficiency [1]. However only a very few adultpatients with the heterozygous form are reported requiringliver transplantation [9]. The mechanismfor liver cell damage in patients with α1-antitrypsindeficiency is still unclear. Accumulation of α1-antitrypsinin the liver cell is thought to be directly relatedto liver cell damage [14]. Experiments in transgenicmice carrying the mutant Z allele of the human α1-antitrypsin gene showed histological approved livercell damage in these animals despite normal serumconcentrations of α1-antitrypsin. It is, therefore, likelythat liver injury cannot be caused by low serum concentrationsof the protein but by a intracellular accumulationof the abnormal α1-antitrypsin gene product[4]. The clinical consequence of these pathophysiologicalmodels is that total serum concentrations ofα1-antitrypsin cannot reveal enough information

510 EUROPEAN JOURNAL OF MEDICAL RESEARCHDecember 7, 2005about the risk and the outcome of liver disease for theindividual patient.On the basis of these observations we hypothesized:Determination of total serum α1-AT in the diagnosticworkup of neonatal cholestasis is an insufficienttool of detecting α1-antitrypsin deficiency andrare variants of the protein.Detection of rare variants of α1-AT coinciding withliver disease can only be detected by determination ofthe α1-AT phenotype using isoelectric focusing.We, therefore, reevaluated 48 children with differenttypes of α1-AT deficiency including total serum α1-AT, α1-AT phenotype, appearance of liver disease bymeasuring serum levels of AST, ALT, γGT, GLDH,bilirubin, and coagulation time at the time of diagnosisand after an observation period of 1-18 years (mean 2years).PATIENTS AND METHODSOur study included 48 pediatric patients, 27 girls and21 boys, aged between 1 week and 16 years at the timeof diagnosis of α1-antitrypsin deficiency. Diagnosis ofα1-antitrypsin deficiency was proven by isoelectric focusing(IEF) of the protein in all patients. Analysis ofα1-antitrypsin was performed in serum samples byusing the IEF technique followed by immunoprinting[15, 22]. Patients were grouped into three subsets:patients homozygous for α1-antitrypsin deficiency(PiZ, PiM Palermo ), patients heterozygous for α1-antitrypsindeficiency (PiMZ, PiMS, PiSZ, PiMP, PiVZ,PiMM procida , PiMQ0), and a patient with a normal rarevariant of the protein (PiMP donauwörth ).We retrospectively analyzed the prevalence of liverdisease in each group at the time of diagnosis. Liverdisease was defined as evidence of pathological liverfunction tests. Pathological elevation of transaminases(ALT > 40 U/l, AST > 45 U/l) and/or elevation ofγGT (newborn: γGT >120 U/l, infants and children:γGT > 45 U/l), conjugated bilirubin (> 1,5 mg/dl),and alkaline phosphatase (> 350 U/l) were regarded ashepatitis or cholestasis, respectively. Serum concentrationsof α1-antitrypsin were determined in each individualpatient. Liver biopsies were performed in patientswith persistent pathological liver function testsonly.All patients underwent a clinical and laboratoryreevaluation after a mean period of 2 years after beingdiagnosed as deficient for α1-antitrypsin. Appearanceof liver disease was again defined by pathological elevationof liver function tests as described above.Serum concentrations of α1-antitrypsin were measuredin all patients at the time of reevaluation.We compared the three subsets of patients regardingappearance of liver disease, total serum concentrationsof α1-antitrypsin, clinical course of the patients,especially development of cirrhosis.Differences between goups were analysed using theWilcoxon-Rank Sum test.RESULTSAbnormal variants of the α1-antitrypsin were detectedin all patients by isoelectric focusing. Analysis ofserum α1-antitrypsin and isoelectric focusing of theprotein was performed because of appearance of liverdisease in 50 %, clinical signs of airway obstruction in18 %, family history of α1-antitrypsin deficiency in10%. In 12% of the patients α1-antitrypsin deficiencywas detected during a routine follow up.Homozygous α1-antitrypsin deficiency was diagnosedin 12 patients (PiZZ), heterozygous α1-antitrypsindeficiency in 28 patients (PiMZ n = 17, PiMS n= 7, PiSZ n = 2, PiVZ n = 1, PiMP n = 1), rarevariants of the protein were found in 8 patients(PiM(Q0) n = 4, PiMP donauwörth n = 1, PiMM procida n =1, PiM palermo n = 2), respectively. Low serum concentrationsof α1-antitrypsin were detected in all patientswith homozygous deficiency, however 36% of thepatients with heterozygous variants showed normalserum levels of α1-antitrypsin. Serum levels of α1-antitrypsinranged between 78 and 100% in patients withthe rare variant M(Q0), between 9,5 and 12% in patientswith the variant PiM palermo , 10% in the phenotypePiMM procida , 100% in the patient with the phenotypePiMP donauwörth . Table 1 summarizes the results ofserum α1-antitrypsin according to the phenotypes.Table 1. Pi phenotypes and serum concentrations of α1-antitrypsin in 48 pediatric patients with (values of serum α1-antitrypsinare given in mg/dl (a) and in percent of normal values for age (b)*Phenotype Number of patients α1-AT in mg/dl α1-AT in %± SD of age related to normal valuehomozygous Z 12 34 24 ± 8M palermo 2 19 11 ± 4heterozygous MS 7 125 81 ± 11MZ 17 95 69 ± 9SZ 2 56 40 ± 7MP 1 115 88 ± 0VZ 1 60 46 ± 0MM procida 1 105 81 ± 0M(Q0) 4 190 94 ± 6variant MP donauwörth 1 200 100 ± 0* data are expressed as means.

December 7, 2005 EUROPEAN JOURNAL OF MEDICAL RESEARCH511At the time of diagnosis pathological elevation oftransaminases was found in 2 patients and significantcholestasis in 6 patients with the phenotype PiZZ,cholestasis was detected in one of the two patientswith the phenotype PiSZ. 8 of 17 patients with thephenotype PiMZ had biochemical evidence ofcholestasis. Transaminases were elevated in 4 patientswith PiMZ and in two patients with the phenotypePiMS. The patient with the rare phenotype PiMPshowed no clinical signs of cholestasis. All patientswith the rare variant type M(Q0) were severely affectedby liver disease (severe neonatal hepatitis in 2 patients,neonatal cholestasis in 2 patients) despite serumlevels of α1-antitrypsin of 80% of normal range accordingto patient’s age. One of them already had developedcirrhosis at the time of diagnosis. The patientwith the rare variant type MP donauwörth presenting witha normal serum level for α1-antitrypsin had histologicalevidence of cirrhosis. Two patients with the rarevariant M palermo in the homozygous state had no chemicalsigns of liver disease despite very low serum levelsof α1-antitrypsin (< 20% of normal range accordingto patient’s age). The patient with the variant phenotypeMM procida had normal serum levels of α1-antitrypsinand no clinical evidence of liver disease.CLINICAL COURSE OF THE PATIENTSThe clinical course of the patients was dependent onthe phenotype of α1-antitrypsin.Eight of 14 patients with α1-antitrypsin deficiency(PiZZ or PiSZ) had clinical evidence of liver disease atthe time of diagnosis, in 6 patients neonatal cholestasis,in two patients cholestasis in infancy (3 monthsand 6 months). Only 3 patients with liver disease dueto homozygous phenotypes developed progressive liverdisease leading to cirrhosis within 3 years. All ofthem had the phenotype PiZ. Two of these patientsdied of chronic liver failure, one patient underwent livertransplantation and died of severe fungal infection 2weeks after transplantation. Total serum α1-antitrypsinwas similar in all patients with the homozygousform of the disease (range: 16 - 45% of the normalvalue corrected for age). There was no difference inthe clinical outcome between patients with α1-antitrypsinlevels less than 20% and more than 20% ofnormal. The results of the homozygous group of patientsare summarized in Table 2, demonstrating theclinical status of the patients at the time of diagnosisand at the time of reevaluation, interestingly 11 patientswith phenotype PiZZ exhibited no signs of liverdisease.Eight of 24 patients with heterozygous α1-antitrypsindeficiency (PiMZ and PiMS) had neonatalcholestasis. Neonatal hepatitis was observed in 2 patientswith the phenotype PiMZ. As in patients withhomozygous α1-antitrypsin deficiency there was nodifference regarding the manifestation of liver diseasein patients with nearly normal serum levels of α1-antitrypsin(80-95% of normal age related value), comparedto those with serum levels of less than 70%. Nocirrhosis was observed in the heterozygous group. Liverfunction tests (AST, ALT, γGT, bilirubin, alkalinephosphatase) normalized in all patients but two withinthe first 6 months after birth. Only two patients of thisgroup had slightly elevated liver function tests (ALT of50 U/l in one patient, γGT of 65 U/l in another patient)2 and 3 years after diagnosis, respectively. Theresults of the heterozygous group of patients are summarizedin Table 3, demonstrating the clinical status ofthe patients at the time of diagnosis and at the time ofreevaluation.Table 2. Clinical status of patients with homozygous α1-AD at the time of diagnosis and at the time of reevaluation 1 – 3 yearsafter diagnosis.Healthy Cholestasis Hepatitis ZirrhosisAt diagnosis At reevaluation At diagnosis At reevaluation At diagnosis At reevaluation At diagnosis At reevaluationPiZZ 4 9 7 0 1 0 0 2PiSZ 1 1 0 1 1 1 0 0PiM palermo 2 2 0 0 0 0 0 0MQ0 0 1 2 1 2 1 0 1MP donauwörth 0 0 0 0 0 0 1 1Table 3. Clinical status of patients with heterozygous α1-AD at the time of diagnosis and at the time of reevaluation 1 – 3 yearsafter diagnosis.Healthy Cholestasis Hepatitis ZirrhosisAt diagnosis At reevaluation At diagnosis At reevaluation At diagnosis At reevaluation At diagnosis At reevaluationPiMZ 5 14 8 1 4 2 0 0PiMS 5 5 0 0 2 2 0 0PiMP 0 1 1 0 0 0 0 0PiVZ 1 1 0 0 0 0 0 0PiMM procida 1 1 0 0 0 0 0 0

512 EUROPEAN JOURNAL OF MEDICAL RESEARCHDecember 7, 2005The heterogeneous group of patients with rare variantsof the α1-antitrypsin protein shows a broad spectrumof clinical manifestations. The patients with thephenotypes PiMM procida , PiVZ, and PiM palermo showedno evidence of liver disease at the time of diagnosis, norat the time of reevaluation. Their serum levels of α1-antitrypsin however were below 30% in 5 patients, intwo of them less than 80%. In contrast all patients withthe phenotype PiM(Q0) and the patient with the phenotypePiMP donauwörth had severe liver disease at the timeof diagnosis (neonatal cholestasis in 2 patients, neonatalhepatitis in 3 patients) despite normal levels of α1-antitrypsin in serum. Two patients with the phenotypePiM(Q0) and the patient with the phenotype PiMPdonauwörthdeveloped rapidly progressive cirrhosis. Onlyone out of these 4 patients had normal liver functiontests at the time of reevaluation. The results of thevariant groups of patients are summarized in Table 2,demonstrating the clinical status of the patients at thetime of diagnosis and at the time of reevaluation.Unknown environmental factors in addition to themodified genes may play a certain role in the pathogenesisof liver disease.DISCUSSIONHomozygous α1-antitrypsin deficiency (phenotypePiZZ) is associated with neonatal cholestasis and maylead to cirrhosis in up to 13% of the affected patients[16, 21]. The heterozygous form of the disease as thephenotypes PiMZ and PiMS rarely coincide with liverdisease but single cases of cirrhosis are reported in theliterature [10, 14]. The homozygous as well as the heterozygousforms of the α1-antitrypsin deficiency areassociated with low serum levels of α1-antitrypsin. Inour series of 48 patients with α1-antitrypsin deficiencyincluding 10 patients with rare variants of the diseasewe found an all over prevalence of liver involvementin 67% of the patients at the time of diagnosis. Thehigh prevalence of liver disease in our series might beexplained by the design of the study. Patients were retrospectivelyevaluated and liver disease was the reasonof reduced α1-antitrypsin in 50 % of the patients.We detected liver involvement in 58% of our patientswith the PiZ homozygous form of the disease,leading to end stage liver disease in 13%. Sveger et al.prospectively screened 200.000 newborns in Swedenfor PiZZ α1-antitrypsin deficiency [19]. They foundclinical signs of liver disease in 18% of their patientswith homozygous α1-antitrypsin deficiency leading tocirrhosis in 7%. As the patients in our study were referredto our clinic for diagnostical workup of liverdisease or respiratory abnormalities this might explainthe higher prevalence of liver disease in our patients.However clinical outcome of our patients was similarto Sveger´s observations. Only 3 of 14 patients (PiZZor PiSZ) had clinical evidence of liver disease 1 - 2years after the diagnosis was established. In Sveger´sstudy 25% of the patients with PiZZ showed abnormalliver function tests 2 years after diagnosis [20]. Allof our patients with PiZZ had low serum levels of α1-antitrypsin according to the literature. Two of our patientswith homozygous α1-antitrypsin deficiencywere prematurely born infants with neonatal sepsis.Both children developed cirrhosis. In another child liverfunction tests deteriorated during a gastrointestinalinfection due to Salmonella enteritidis at the age of 2years. Additional factors as parenteral nutrition and severesystemic infections are known as risk factors forthe development of severe liver disease in these children.Environmental factors resulting in an enhancedsynthesis of abnormal α1-antitrypsin lead to a higheraccumulation rate of the protein in liver cells resultingin liver cell damage [12, 13]. These observations supportthe hypothesis that intracellular accumulation ofthe mutant protein in liver cells is responsible for liverdisease in homozygous α1-antitrypsin deficiency [11].In adults alcohol consumption and viral hepatitis alsoincrease the risk of cirrhosis in patients with homozygousα1-antitrypsin deficiency despite normal liverfunction tests during childhood [3].15 of 27 patients with heterozygous α1-antitrypsindeficiency showed clinical evidence of liver involvementin our series. In 12 of them liver function teststurned to normal within 6 months after the time of diagnosis.Only 2 of them had slightly elevated transaminasesduring the observation period of 1-3 years. Thiscoincides with other reports in the literature.Heterozygous α1-antitrypsin deficiency is known asbeing associated with an increased risk of neonatalcholestasis or neonatal hepatitis [1]. However clinicaloutcome of this group is different from patients withthe homozygous form of the disease. Only single reportsof cirrhosis due to heterozygous α1-antitrypsindeficiency exist [9]. As explained previously additionalfactors are necessary to increase the intracellular accumulationof mutant protein in liver cells. Factors as viralhepatitis and alcohol abuse are known to be associatedwith the development of cirrhosis even in patientswith heterozygous α1-antitrypsin deficiency [5,9]. In none of our patients cirrhosis was observed.However they still are at a higher risk for liver diseasecompared to healthy controls. Eight of 17 childrenwith heterozygous α1-antitrypsin deficiency and abnormalliver function tests were breast fed. Perlmutteret al. showed that breast feeding is associated with alower risk of mild neonatal cholestasis in children withheterozygous α1-antitrypsin deficiency [13]. The pathomechanismof this observation is still unclear. Liverfunction tests of the 4 patients returned to normalwhen breast feeding was discontinued.Serum levels of α1-antitrypsin ranged betweennormal and 65% of the normal levels adapted to thepatients´age. We did not observe any correlation betweenthe serum levels of α1-antitrypsin and the appearanceof pathological liver function tests in patientswith heterozygous α1-antitrypsin deficiency. This coincideswith other reports in the literature [1, 19].The heterogeneous group of patients with rare mutantsof the α1-antitrypsin deficiency showed differentclinical outcome. The rare phenotype PiM(Q0) wasfound in 4 patients as diagnosed by isoelectric focusingin all patients. Two children were prematurely born(gestational age 28 weeks and 36 weeks, respectively).Both of them developed severe liver disease in theirfirst month of live and one child died of progressiveliver disease within 6 months. Their postnatal periodwas affected by several complications due to prematu-

December 7, 2005 EUROPEAN JOURNAL OF MEDICAL RESEARCH513rity, however, the reason for rapidly progressive liverdisease remains unclear. Liver biopsies were performedin 3 children, demonstrating cirrhosis in 1 caseand severe fatty degeneration of the liver in the othertwo cases. Only single reports exist about this rarevariant. It is characterized by a diminished band patternby isoelectric focusing of the protein. Mainly prematurepulmonary emphysema has been associatedwith different Pi(Q0) variants (PiQ0 Granite Falls,PiQ0 Bellingham, PiQ0 Mattawa, PiQ0 Riedenburg,PiQ0Ludwigshafen). All of them were homozygousfor the null-allele resulting in low serum concentrationsof α1-antitrypsin [2, 7]. Our patients withPiM(Q0) were heterozygous for the null allele resultingin a slightly reduced concentration for α1-antitrypsin.The pathomechanism in this group is unclearbut is seems possible that an abnormal structured proteinmight be responsible for liver cell damage.The rare variant PiMP donauwörth was found in one patientwith a normal serum concentration of α1-antitrypsinbut severe early liver disease. Histology revealedmicronodular cirrhosis in this child. This patient developeda rapidly progressive cirrhosis during the first3 years of live. This variant has also been found in thepatient´s having normal liver function tests. An abnormalstructured protein might have caused liver celldamage by accumulating in the liver cells. Howevernormal serum α1-antitrypsin stills remains unclear inthis patient [7].The lowest serum concentrations for α1-antitrypsinwere found in 2 siblings with the rare variantPiM palermo , first described in a Sicilian family [7, 18].None of them had abnormal liver function tests, norabnormal lung function tests. This coincides with thereport about this family. Possibly a normal proteinwith a decreased synthesis rate might explain the lowserum concentration and the lack of symptoms. Thisalso supports the theory that mainly intracellular accumulationof mutant proteins in the liver cells seems tobe responsible for liver cell damage. Low serum concentrationsof normally functioning α1-antitrypsin donot correlate with liver disease. Low activity of theprotease inhibitor could be responsible for pulmonaryemphysema in some patients but is not necessarily associatedwith lung disease.In one patient we found the rare variant phenotypePiVZ [8]. As in the patients with PiM palermo this patientalso had very low serum concentrations of α1-antitrypsinnot associated with liver disease. There is notvery much known about this variant type which is regardedas heterozygous for α1-antitrypsin deficiency.Patients reported in the literature having the allelePi*V do not have an increased risk for liver disease.However our female patient might be of a higher riskfor liver disease as she is heterozygous for the Z-allele.In summary we found an increased prevalence ofliver disease in patients with homozygous and heterozygousα1-antitrypsin deficiency. However, progressionof the disease was only noted in a subset ofpatients with the variant types PiZZ, PiM(Q0), andPiMPdonauwörth. The latter two phenotypes were notassociated with low serum levels of α1-antitrypsin.In the diagnostical workup of liver disease measuringserum α1-antitrypsin might not provide enoughinformation to rule out α1-antitrypsin deficiency. Rarevariants as shown in our study can have normal serumlevels of α1-antitrypsin but a structurally abnormalmutant protein might cause liver cell damage by accumulatingin liver cells. Phenotyping and in special casesDNA analysis should be performed in each single patientwith unknown liver disease or neonatal cholestasis.In this study less correlation between α1-antitrypsinlevels in serum with the clinical phenotype was found.We conclude, therefore, that analysis of the biochemicalphenotype with appropriate methods is mandatoryin detecting disease causing variants.REFERENCES1. Aagenes O, Matlary A, Elgjo K, Munthe E, Fagerhol M:Neonatal cholestasis in α1-antitrypsin deficient children:clinical, genetic, histological and immunhistochemicalfindings. Acta Paediatr Scand 1972; 61: 632-642.2. Bamforth FJ, Kalsheker NA: α1-antitrypsin deficiencydue to Pi null: clinical presentation and evidence for molecularheterogeneity. J Med Genet 1988; 25: 83-87.3. Crystal RG: α1-antitrypsin deficiency, emphysema, andliver disease: genetic basis and strategies for therapy. JClin Invest 1990; 95: 1343-1352.4. Dycaico MJ, Grant SGN, Felts K, Nichols WS, GellerSA, Hager JH, Pollard AJ, Kohler SW, Short HP, JirikFR, Hanahan D, Sorge JA: Neonatal hepatitis induced byα1-antitrypsin: a transgenic mouse model. Science 1988;242: 1409-1415.5. Eriksson S: α1-antitrypsin deficiency and liver cirrhosis inadults: Acta Paediatr Scand 1987; 221: 461-467.6. Esquivel CO, Iwatsuki S, Gordon RD, Cox KL: Indicationsfor pediatric liver transplantation. J Pediatr 1987;111: 1039-1045.7. Faber JP, Poller W, Weidinger S, Kirchgesser M, SchwaabR, Bidlingmaier S, Olek K: Identification and DNA sequenceanalysis of 15 new α1-antitrypsin variants, includingtwo Pi*Q alleles and one deficient Pi*M allele. Am JHum Genet 1994; 55: 1113-1121.8. Faber JP, Weidinger S, Olek K: Sequence data of the raredeficient α1-antitrypsin variant PiZaugsburg. Am J HumGenet 1990, 46: 1158-1162.9. Hodges JE, Millward-Sadler GH, Barbatis C, Wright R:Heterozygous MZ α1-antitrypsin deficiency in adultswith chronic active hepatitis and cryptogenic cirrhosis. NEngl J Med 1981; 304: 357-360.10. Larsson C: Natural history and life expectancy in severeα1-antitrypsin deficiency, PiZ. Acta Med Scand 1978;204: 345-351.11. Lomas DA, Evans DL, Finch JT, Carrell RW. The mechanismof Z α1-antitrypsin accumulation in the liver. Nature1992; 357: 605-607.12. Nebbia G, Hadchouel M, Odievre M. Early assessment ofevolution of liver disease associated with α1-antitrypsindeficiency in childhood. J Pediatr 1983; 102:661-665.13. Perlmutter DH, Schlesinger MJ, Pierce JA: Synthesis ofstress proteins in increased in individuals with homozygousPiZZ α1-antitrypsin deficiency and liver disease. JClin Invest 1988; 84: 1555-1561.14. Perlmutter DH: The cellular basis for liver injury in α1-antitrypsin deficiency. Hepatology 1991; 12: 172-185.15. Pierce JA, Eradio BG: Improved identification of antitrypsinphenotypes through isoelectric focusing withdithioerythritol. J Lab Clin Med 1979; 94: 826-831.16. Psacharopoulos HAT, Mowat AP, Cook PJL: Outcomeof liver disease associated with α1-antitrypsin deficiencyPiZ. Arch Dis Child 1983; 58: 882-887.

514 EUROPEAN JOURNAL OF MEDICAL RESEARCHDecember 7, 200517. Rabin M, Watson M, Kidd V. Regional location of α1-antichymotrypsinand α1-antitrypsin genes on human chromosome14. Somat Cell Mol Genet 1986; 12: 209-214.18. Sefton L, Kelsey G, Kearny P. A physical map of humanPI and AACT genes. Genomics 1990; 7: 382-388.19. Sveger T: Liver disease in α1-antitrypsin deficiency detectedby screening of 200,000 infants. N Engl J Med1976; 294: 1216-1221.20. Sveger T: The natural history of liver disease in α1-antitrypsindeficient children: Acta Paediatr Scand 1988; 77:847-851.21. Wall M, Moe E, Eisenberg J: Long-term follow up of acohort of children with α1-antitrypsin deficiency. J Pediatr1990; 116: 248-251.22. Weidinger S: Reliable phenotyping of α1-antitrypsin byhubrid isoelectric focusing inan ultranarrow pH gradient.Electrophoresis 1992; 13: 234-239.23. Yunis EJ, Agostini RM, Glew RH. Fine structural observationsof the liver in α1-antitrypsin deficiency. Am JClin Pathol 1976; 82: 265-286.Received: August 15, 2005 / Accepted: October 17, 2005Address for correspondence:Thomas Lang, M.D.Pediatric GastroenterologyChildren´s University HospitalDr.v.Haunersches KinderspitalLindwurmstraße 4D-80337 Munich, GermanyPhone: +49-89 - 51602811FAX: +49-89 – 51607872e-mail: thomas.lang@med.uni-muenchen.de