Tunn et al.: Sulphated glycosaminoglycan isomers in Dupuytren's disease

J. Clin. Chem. Clin. Biochem.Vol.26, 1988, pp. 7-14© 1988 Walter de Gruyter & Co.

Berlin · New York

The Distribution of Unsulphated and Sulphated Glycosaminoglycansin Palmar Fascia from Patients with Dupuytren's Disease and HealthySubjects1)

By Sabine Tunn2), E. Gurr, A. Delbrück

Institute for Clinical Chemistry II, Center for Laboratory Medicine, Hannover Medical School Hannover, FRG

T. Buhr

Institute for Pathology, Hannover Medical School, Hannover, FRG and

J. Flory

Department of Hand and Reconstruction Surgery, Hannover Medical School, Hannover, FRG

(Received August 24/November 11, 1987)

Summary: Eighty specimens from 20 patients with Dupuytren's disease and 7 biopsies of healthy palmar fasciawere analysed for their glycosaminoglycan isomer patterns with a combined enzymatic/HPLC method. Thediseased portions of palmar fascia tissue were characterized by elevated total glycosaminoglycans togetherwith a relative increase in the sulphated fractions. The macroscopic stages of nodules, bands and unaffectedtissue could be classified very well by multivariate statistical analysis on the basis of their glycosaminoglycanpatterns. The biochemical analysis provided evidence of the pathological process even in those specimens thatdid not yet show any clinical symptoms of the disease.

Because of the role played by sulphated glycosami-The mechanisms involved in the pathogenesis of Du- noglycans in the interaction with collagen (12), thepuytren's disease are still unknown (1). Recent studies question arises of whether there are changes in theon the morphological changes in diseased tissue have proportions of glycosaminoglycans and their unsul-demonstfated occlusioii of capillaries by cells that phated and sulphated isomers in affected palmar fas-migrate into the perivascular space. These cells are cia that may have an influence on the collagen patternthought to be the origin of dedifferentiated cells that and the macromolecular structure. The present studyproliferate and metabolize in such Way that they cause was conducted to determine the types and amountscontracture of the aponeurosis (2, 3). As to the pa- of the various sulphated and unsulphated glycosa-thobiochemistry, an increase in collagen type III pro- minoglycans and to reassess the data published so fartein has been reported (4-7), as has an increase in on the glycosaminoglycan content of both healthythe proportion of chondroitin sulphate (4, 8 — 11). human palmar fascia and tissue from Dupuytren's

^ contracture, by employing a method of determination') With support of the Deutsche Forschungsgemeinschaft (SFB more specific than any used previously. The new

54). method also enabled differentiation of the unsul-2) Present address: Institute for Clinical Chemistry and Labo- , , ~A . . - , n< e„i«i^t-/i ^mnr^ntcratory Medicine, university Hospital „BergmannsheiT Bo- Phated> C4-sulphated and C6-sulphated componentschum, FRG (11).

J, Clin. Chem. Clin. Biochem. / Vol. 26,1988 / No. 1

Tunn et al.: Sulphated glycosaminoglycan isomers in Dupuylrerfs disease

Materials and MethodsEighty biopsy specimens from 20 patients (aged 18 to 76 years;2 women, 18 men) with Dupuytren's disease of the hand weretaken during surgery and classified according to their macro-scopic appearance as(1) apparently normal,(2) tissue adjacent to bands or nodules,(3) bands, or(4) nodules.For reference, palmar fascia from healthy persons was takenduring reconstruction surgery following accidental injury(group Ο, η = 7, male) and during autopsies performed within15 hours post mortem (n = 8, female). The specimens werefreed from adjacent tissue and blood and stored at — 20 °Cuntil processing. A representative sample of each specimen wasset aside for microscopic examination.The total glycosaminoglycans were isolated and the variousfractions identified by specific enzymatic degradation, followedby a quantitative determination of the respective disaccharidemetabolites by high performance liquid chromatography asdescribed elsewhere (11). This method is specific and reliable,with an overall coefficient of variation of 6.8% at a substrateconcentration of 100 nmol per assay and a sensitivity of < 10nmol per assay. Uronic acid was determined with the methoddescribed by Bitter & Muir (13), hydroxyproline with that ofStegemann (14) and DKA with that of Burton (15).

The microscopic examination of the specimens was performedon samples taken from the tissue portions that were analysed

biochemically. Samples were stained with hematoxylin and bythe van Gieson technique. The classification was made accordingto Millesi (16):Stage 1A = thickening and stretching of fibres,Stage IB = development of fibre bands,Stage 2 = cell proliferation, andStage 3 = few cells only, unoriented thick fibre bundles (fig. 1).

A discriminant analysis according to Schneider (17) was per-formed with the SPSS-X program "Discriminanf'-Release 2r.

Results

The results of the glycosaminoglycan determinationson the normal palmar fascia tissue and the fourgroups of diseased tissue are summarized in table 1and figures 2 and 3. In the most advanced stage ofthe disease the glycosaminoglycan content was doublethat found in the unaffected palmar fascia. In theearly phases of the disease the hyaluronate contentremained almost constant, with a decrease in the laterstages, whereas the chondroitin sulphate and the der-matan sulphate fractions increased with the severityof the disease, being 10 times greater in nodules thanin healthy fascia. It should be noted that the autopsy

χ.<^ **/>: vT^^t/J«-: /•'V-iVifc·' . > ·:\ ,· v · . · » < : . - ·· ' / « > · / . .'''Κ· ··,. 3*5&· ψ-f ί: · ,J ''&·'';-'·· · '* "·»· ^ · ^ λ

^Κ^^Μ^;Μ-Λ:ν .:ν·; · - ..-···; ·V ' ·:.·.-4 L·' .'^^>^^·^^'^ :. :VA >-·,. χ , · . . . · - < · - Λ · - ·^.·- . v r - ^^i^-b.^^-'V Λν -;·>7^.-.ν ·,. ^ · » , χ• \Λ ·*'.''ΐΜ. 'srt;~: τ „ ' - s - v · - - .W ..^1·...·^,. >·. . . · ν . ν - .

^ν:·-· v:.^;t^^,r->; ^p-^^c>':;· '<..,-··"-- :- ;*·, j; f^^v^^^A V-:«*^ -Wv,v^\ ...' . - *V

^· " " · ' ' - ;·/" ,. %^.·'··ν^ ^^V ' ν-^^'' ·':. <^% ~^^Ν · ·'££>, \ · ··, 'V· '"·· * Ρ^ ': · : - · ' : '-Μ^* ; t:;>£ ,;^;^^ν·;\ ν ·ν** ·>. ^V \- > ^^ ΛΒ

Fig. I. Dupuytrerfs disease according to Mitoz (16). The examples illustrate the proliferating stage 2 (A) and the«stage 3 (B, fewcells only, unonented thick fibre bundles). HE, lOx. *' Ι φ

J. Clin. Chem. Clin. Biochem. / Vol. 26,1988 / No. 1

Tunn et al.: Sulphated glycosaminoglycan isomers in Dupuylren's disease

s-s g £.Si! 'Bals. > ·>,= * . £ *

1 g U•a · 2 a „ _|.& ·§* rT

« 0 "S S 8

1|

Hilflilil*£? *g $j CL, Pg_ — 'S S §

o .2 S e>""* *5 Q

«2 O J2 <«_ cd5 * *S* ^^ W

• 2 r* * "* *""* Qj

«S i i» "* *3 ',«3

Hl 11.5 vf b -° 'Ss £ * § 's· | c| g^ fe "+3 *S s>> &· ce §> O

" "0 <ü -g

fcjjüjlil^jf'! O 2 IIcä t5 Q- -«isO wj «3 * §

O § *o c -ög .ß o cd

rt P OQ 3- . QJ cA

iflfi2 Q " iS § ;g §

·& > c "Bg .g g -2 g

1 «18^ 60 +g S v©

'Sn*' ^

J""1 ~ ^< o^ W*5 ^ «M C- l

ägu^^f

fÜK^» isn ö V 1*Zt G ® Ä <f1Q 'Bi^ä i-·'s 111 tO · *Ö O (£2"*3 «22 ·*·* ^*

§ 1 g | go § . 'S

i

11"o.3Mccdcd

«S

^D C3 ·"'^] ZL 'X

*O00

< 3.IX

160co^· w

41+'< =lix

•o^5- w

3 i+l2 §LIX

£

!'S

S"•3w

3CA

• S

S··*G

6

T3

» f -sa l+i< a. ix

£•

CO ^4 | + 1< ÜIX

1CO - W

4 i + i< a. ix

•a00^j. ee

S i. «

8 Jlo3 ^~-~ co

111 +lÄ sS. ix

O3 £" cd "°O Q 00

cd €H S» =L 'X

l »^g « «C3 " - «III -,:1i£

^* **

Q 'S ix

SöHHSö

soH-l

B

oöH-lSö

esesöH-lt*»

r—oH-l

ö

»*00

HHcoö

s<DV

SÖ+1cocoo

sö+1s§öH-l

;

opes

+1

iesCO0H-l

P

ä si.|||tS ^J

MC

sö

Mc

•.osö

Mc

coC

oö

eoÖ

CAÖ

§ö

0§Ö

So

Ov

1<o o

** ..78p %ü fi

0öH-lo <Aö c

00ooöHH«tf· ex»es" c

soH-lsö c

VOONÖH-l\ir> »»

tn*^öH-l

ö c

es

H-l£ö a

*Q-Hes*-l «Q G

SG>-H^- CO

0 G

Ö+1

S SCO Ö

«.0

H-l

5 ;cvo^>

H-l00

S 2

o-4

HH ^00 000 OCO Ö

•9

g

Hl: i

gö+1

ö

00öH-l

es

göH-lSö

\ooö-H

H

voöH-lesö

.

H-l

ö

öH-l

ö

^p0HH

0

c-ooö+1esvoes

»ooov-t

H-l

W)

^^00

H-l

0

00öHH

5

1s!1s

Grou

p 2

Fasc

ia ad

to ba

nds <

VH8Ö

§0

5ö

sö

§ö

ooö

o0o

§0

iö

so

CO

CAc

« Vi i

OöH-lSö

£ö+1ooco

ö+1oö

es·»-*

H-lsco

r-

öH-lFiö

OSO

+1COesö

esö+1

ö

ooo

H-l?

coSH-lS^vo

H-lSsvo

ONCOr-i

-H

8

COooöHH55es

ro

gÖ

§O

§Ö

iö

iö

ooö

iö

(C

coC

8Ö

rQö

8Ö

IIc?

ö-Hn-coÖ

SöH-l

«o

esöHHVOesö

TJ-r^öH-l00

vd

covoöH-lS^

VO

HH

ö

coÖ+1

SvoO

+1»r»00es

3HHoovoi-«

co

"

etö*"*

cov^

H-l-;

OO

00'-'HHcoesvo

•\-il!J. Clin. Chem. Clin. Biocbem. / Vol. 26,1988 / No. l

10 Tunn et al.: Sulphated glycosaminoglycan isomers in Dupuytren's disease

10

Ea.

-5 0.1

0.01Healthy controls

-ι*-

Strang Nodules

10

0.1

0.01

Fig. 2. Glycosaminoglycan patterns of healthy palmar fascia and in specimens from Dupuytren's disease (bands, nodules).Combined specific-enzymatic-HPLC analysis.

glycosaminoglycan content

hyaluronate content

total condroitin sulphate fraction

unsulphated, C4- and C6-sulphated isomers

total dermatan sulphate fraction

unsulphated, C4- and C6-sulphated isomers

χ ± 2s

1Λ

Dis

acch

orid

es (

μιηο

ΐ/g

dry

wei

ght]

§ s

- s a«

i

·> 4

« E

ϊϊΘ :

*

|L

JLHealthy

m'614

controls

»1 1

n

jj 3 @ =<I i<6

M 1 ι!· iΙ ί= 0ι ι« \

tfyi//<:

Apparently normal fascia

m . . . . . ' _ _ .•

. . .. — |

1 ·· ·'

,',

|j*

-^1 \1 I1 . ι .- - ·-· ^ ·-; * $\$\4 .(W

'

Fascia adjacent to str ngand nodules ι

10

0.1

0;01

Fig. 3. Glycosaminoglycan patterns of healthy palmar fascia and in specimens from Dupuytren's disease (apparently normaltissue, tissue adjacent to bands and nodules). Combined specific-enzymatic-HPLC analysis.Explanation see figure 2. . f

J. Clin. Chem. Clin. Biochem. / Vol. 26,1988 / No. 1

\ et al.: Sulphated glycosaminoglycan isomers in Dupuytren's disease 11

-o ö

\·* 2

i,"0 S 'S

g e .J> .0.•slll?itliS

l00•s -H

< .K

l

5ex3-sS

< AI*

oo

< 3.IX

« «IS§

öO23- OT

|-H=LiX

-Wi'1

IX

.III -HH S =t ix

•R o § -HK §.§. .*

Q ß .*

il

sö+1S

CNÖ+1

pö+1Sö

o+1r*

o-flSö

sö-H

oöV

VO

ö+1

-H«0

<N

-H

•i«NcnÖ+1P

oö+1

+1

pö•fl

S Sö ö

§ sö -^

Sö+1

S Sö ö

sö•fl

p pö ö

Ö1 V

o^ -H

p ^O Ö

O

« +1g sÖ <»-»*

ö CM'

° 3

gÖ+1

»n tCN io o

specimens of normal fascia contained smalleramounts of glycosaminoglycans than the biopsy spec-imens. The autopsy material proved to be unsuitablefor reference because of a post mortem drop in gly-cosaminoglycans, mainly hyaluronate (tab. 2). How-ever, both collagen and DNA contents remained un-changed in the autopsy specimens, compared with thebiopsy material. The chondroitin sulphate isomer pat-tern found in Dupuytren's contracture tissue was char-acterized by a very marked increase in unsulphatedchondroitin, which was 500 times higher than inhealthy fascia (fig. 2). The main portion of the chon-droitin sulphate fraction was the 6-sulphated com-ponent, which showed a concentration 10 times higherthan in healthy fascia. However, in the dermatansulphate fraction the elevation in diseased palmarfascia was due entirely to the 4-sulphated compound,which is the predominant isomer in the healthy palmarfascia (fig. 2). Regarding the unsulphated and 6-sul-phated compounds, it must be kept in mind that theconcentrations of these substances in palmar fasciaare at the lower limit of the range of measurement ofthe method applied. Therefore, no conclusive state-ment can be made as to whether changes in thedermatan sulphate isomer pattern occur during thecourse of the disease (fig. 2).

Although less pronounced, the changes in the glycos-aminoglycan patterns in tissue adjacent to nodules

-6 - 4 - 2 0 2 4Canonical discriminant function 1

Fig. 4. Classification of specimens from healthy palmar fasciaand Dupuytrerfs disease by multivariate statistical eval-uation (18) of the glycosaminoglycan patterns: 9 vari-ables (hyaluronate, total chondroitin sulphate, total der-matan sulphate, and the respective unsulphated, C4-and Co^sulphated isomers).• healthy control• apparent normal fasciaA fascia adjacent to sträng and nodules• sträng• nodulesThe open symbols indicate the respective group cen-troid.

J. Clin, Chem. Clin. Biochem. / Vol. 26,1988 / No. 1

12 Tunn et al.: Sulphated glycosaminoglycan isomers in Dupuytrerfs disease

Tab. 3. Clinical, historaorphological and biochemical classification of 46 specimens from Dupuytren's disease. Clinical classificationaccording to the macroscopic appearance of the specimen. Microscopic staging according to Millesi (16). Biochemicalclassification by multivariate statistical analysis (17) of the glycosaminoglycan patterns. Each number in the morphologicstage areas represents one classified specimen. Its size expresses the respective biochemical classification (stage 0: healthypalmar fascia, diseased fascia stages 1 through 4).

Microscopic staging1 A l B

Clinical classification

Group 1Apparently normal

Group 2Tissue adjacent to bands and nodules

Group 3Bands

Group 4Nodules

1 1

2 2 2 2 2 2 2 2

12

01 1 12 2 2 2

3 3 3 3 3 3 3 3 3 3 3 3 32 2

4 4 4 4 4 4 4

Tab. 4. Classification of results of specimens from healthy palmar fascia and Dupuytrerfs disease by multivariate statisticalanalysis (18) on the basis of 9 variables of biochemical parameters (see fig. 3). Definitions of groups predicted accordingto the macroscopic appearance of the specimens.

Classification of resultsActual group No. of Predicted group membership

cases

Group 0 1 4Healthy controls

Group 1 2 12Apparently normal fascia

Group 2 3 20Adjacent fascia

Group 3 4 19Band

Group 4 5 1 1Nodules

Percent of "Grouped" cases correctly classified:

1

4100.0%

00.0%

15.0%

00.0%

00.0%

81.82%

2

00.0%

866.7%

420.0%

00.0%

00.0%

3

00.0%

3f25.0%

1575.0%

00.0%

00.0%

4

00.0%

183%

00.0%

1894.7%

273.2%

5

00.0%

00.0%

00.0%

15.3%

981.8%

and bands as well as in apparently healthy portionsof diseased palmar fascia were still great enough toclassify these tissue as clearly diseased (fig. 3, tab. 1).These findings are compatible with the elevation inDNA and*in hydroxyproline concentrations (tab. 1)as well as with the increase in the number of cells (2,18) reported for these specimens of diseased palmarfascia.

A multivariate statistical analysis (17) of the datayielded a classification that clearly distinguished be-tween the specimens taken from nodules or bands andthose from apparently normal or adjacent tissue, orfrom the healthy reference specimens (fig. 4, tab. 3).This biochemical classification was thus im goodagreement with the macroscopic classification. A lesssatisfactory agreement, however, was found with theresults of the microscopic examinatior>sf(tab. 4).

J. Clin. Chem. Clin. Biochem. / Vol. 26,1988 / No. 1

Tunn et al.: Sulpha ted glycosaminoglycan isomers in Dupuy trends disease 13

Discussion

Data on the total glycosaminoglycan content, as wellas on the distribution pattern of their fractions ofpalmar fascia from healthy subjects and from individ-uals with Dupuytren's contracture are sparse and formethodological reasons of limited value. Using elec-trophoresis for differentiation of the glycosaniino-glycans, Flint and coworkers (8) showed that theelevation of the total glycosaminoglycans seen in Du-puytren's tissue is based mainly on the increase in thechondroitin sulphate fraction. The findings in thepresent study are in general agreement with the ob-servations of Flint et al., although no detailed com-parisons can be made because the Flint findings arebased on relative concentrations of glycosaminogly-cans and the use of autopsy specimens for reference.In contrast, our data were obtained by specific en-zymatic degradation of the glycosaminoglycans andquantitative determination of the respective disac-charide metabolites by high performance liquid chro-matography. The specificity, sensitivity, and reliabilityof this procedure allow quantitative analyses on verysmall specimens from various parts of the afflictedpalmar fascia and differentiation between the sul-phated and unsulphated isomers of the individualglycosaminoglycans. The glycosaminoglycan isomerpattern found in the Dupuytren's contracture tissueshowed an increase in the 6-sulphated compounds,and thus a change in the ratio of the 4- to the 6-sulphates (fig. 2). With regard to the high synthesisrate of chondroitin-6-sulphate, it seems likely that theconcentration of the unsulphated precursors shouldalso increase. The overall ratio of unsulphated tosulphated glycosaminoglycans shifts toward the sul-phated compounds because of the decrease in thehyaluronate concentration and the increase in bothchondroitin and dermatan sulphate as the diseaseprogresses. Mori & Honda (12) presented experimen-tal evidence for the functional importance of the pro^teoglycan sulphate groups iri fibrillogenesis in heartvalve fascia. These findings support the hypothesisthat changes in the relation of sulphated to unsul-phated glycosaminoglycans in Dupuytren's contrac-ture give rise to aberrations in the structure of fibreproteins and the subsequent contracture. The bio-chemical changes found in the glycosaminoglycansallowed staging of the disease in which stages 3 (band)and 4 (nodules) could be clearly identified, but stages1 (apparently normal) and 2 (adjacent tissue) couldiiot be separated from each other or from healthypalmar fascia. However, the biochemical changes instages 1 and 2 did provide clear evidence of the onsetof the pathological process in these portions of tissuethat appeared to be unaffected when examined ma-

croscopically. The same was true for the microscopicexamination of both stage 1 and stage 2 specimens,although the staging of a given specimen was some-times different from the macroscopic and/or biochem-ical classification (tab. 4). This difference resultedfrom the morphological inhomogeneity of the dis-eased tissue and the consequent effect of sample se-lection on the classification. In contrast, the biochem-ical analysis provided a finding that was valid for thewhole specimen. In clinically and macroscopically un-suspicious palmar fascia (stage 1 in the present study)Kischer & Speer (2) and Mohr & Vossbeck (18) dem-onstrated, by electron microscopy and autoradiogra-phy respectively, a proliferation of endothelial cells incapillaries, migration of these cells and perivascularaccumulation. Kischer & Speer (2) term this earlystage of the disease "pre-Dupuytren", which may cor-relate with our biochemically defined stage 1 (appar-ently normal). Cell culture experiments with cell linesderived from cells isolated from Dupuytren's contrac-ture specimens showed an abnormal capacity to syn-thesise sulphated glycosaminoglycans and collagen,which the authors attribute to a permanent modula-tion of metabolic cell characteristics, which can bepropagated in cell culture (19). In addition to thededifferentiation of palmar fascia cells or cells of otherorigin,· continuous stimulation of the fascia cells byabnormal matrix components or exogenous sub-stances such as growth factors may also induce andmaintain metabolic malfunctions.

In conclusion, glycosaminoglycan analysis with thecombined enzymatic/HPLC method used in the pres-ent study allows characterization and classification ofbiopsy specimens from Dupnytren's contracture. Thetypical changes in the isomer fractions suggest meta-bolic aberrations in cells of so far unknown originand dedifferentiation. Malcomposition of the extra-cellular matrix carbohydrates suggests that these car-bohydrates may interact with the cell surface, therebyinducing and maintaining aberrations of metabolismand proliferation of the cells. The interactions betweenfibre proteins and proteoglycans are probably dis-turbed, with subsequent development of the contrac-ture. Regarding the latter process, a study on the roleof proteoglycans in the pathogenesis of the contrac-ture is in progress.

AcknowledgementThe authors are greatly endebted to Miss M. Arlart and MissU. Poehls for their skillful technical assistance and to Dr. M.Török of the Institute for Biometrics, Hannover MedicalSchool, for her support in the statistical evaluation of the data.

J. Clin. Chem. Clin. Biochem. / Vol. 26,1988 / No. 1

14 Tunn et al.: Sulphated glycosaminoglycan isomers in Dupuytren's disease

References1. McFarlane, R. L. (1983) J. Hand Surg. 8, 703-709.2. Kischer, C. W. & Speer, D. P. (1984) J. Hand Surg. 9A,

58-62.3. Shura, D. T. (1985) personal communication.4. Bazin, S., Le Lous, ML, Duance, V. C., Sims, T. J., Bailey,

A. J., Gabbiani, G., D'Andiran, G., Pizzolato, G., Browski,A., Nicoletis, C. & Delaunay, A. (1980) Europ. J. Clin.Invest. 70,9-16.

5. Menzel, E. J., Piza, H., Zielinski, C., Endler, A. T., Steifen,C. & Millesi, H. (1979) Hand 11, 243-248.

6. Gelberman, R. H., Amiel, D., Rudolph, R. M. & Vance,R. M. (1980) J. Bone Jt. Surg. 62, 425-432.

7. Brickley-Parsons, D., Glimcher, M. J., Smith, R. J., Albin,R. & Adams, J. P. (1981) J. Bone. Jt. Surg. 63A, 787-797.

8. Flint, M. H., Gillard, G. C. & Reilly, H. C. (1982) Conn.Tiss. Res. 9, 173-179.

9. Can-, T. L. (1970) Hand 7, 50-55.10. Viljanto, J., Seppälä, P. O. & Lehtonen, A. (1971) Ann.

Rheum. Dis. 30, 423-427.11. Gurr, E., Pallasch, G., Tunn, S., Tamm, C. & Delbrück, A.

(1985) J. Clin. Chem. Clin. Biochem. 23, 77-87.

12. Mori, . & Honda, A. (1982) In: Glycosaminoglycans andProteoglycans in Physiological and Pathological Processesof Body Systems (Varma, R. S. & Varnaay R., eds.) KargerVerlag, Basel, pp. 187-198;

13. Bitter, A. & Muir, H. (1962) Anal. Biochem. 4, 330-334.14. Stegemann, H. (1958) Hoppe-Seyler's Zj Physiol. Chem.

377,41-45.15. Burton, K. (1956) Biochem. J. 62, 315-323.16. Millesi, H. (1981) In: Handchirurgie, Vol. 1 (Nigst, H.,

Buck-Gramko, D. & Millesi, H., eds.) Georg Thieme VerlagStuttgart—New York, pp. 15.0-15.57.

17. Schneider, B. (1970) Mathematische Grundlagen der medi-zinischen Diagnostik, In: Computer: Werkzeug der Medizin(Ehlers, C. Th., Hollberg, N. & Proppe, A., eds.) SpringerVerlag, Berlin, Heidelberg, New York, pp. 160-182.

18. Mohr, W. & Vossbeck, G. (1985) Z. Rheumatol. 44, 226-230.

19. Delbrück, A. & Schröder, H. (1983) J. Clin. Chem. Clin.Biochem. 27, 11-17.

Prof. Dr. med. A. DelbrückInstitut für Klinische Chemie IIder Medizinischen Hochschule HannoverKrankenhaus OststadtPodbielskistraße 380D-3000 Hannover

J. Clin. Chem. Clin. Biochem. / Vol. 26,1988 / No. l