Carbohydrate-deficient transferrin in vitreous humour: a marker of possible withdrawal-related death in alcoholics

Adam Berkowicz, Sven Wallerstedt1,, Kerstin Wall2, and Hans Denison1,*,

Department of Forensic Medicine, Göteborg University, Göteborg,
1 Department of Medicine, Sahlgrenska University Hospital/Östra, Göteborg University, Göteborg and
2 Pharmacia & Upjohn Diagnostics AB, Uppsala, Sweden

Received 17 November 1999; in revised form 16 November 2000; accepted 12 December 2000


    ABSTRACT
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
The possibility of performing reliable post-mortem analysis of carbohydrate-deficient transferrin (CDT) concentration in vitreous humour (VH) by using a commercial assay designed for serum analysis (CDTectTM) as well as the usefulness of VH-CDT as a marker of alcohol misuse and possible withdrawal-related death were evaluated in a forensic sample. Detectable VH-CDT was found in 20 of 21 alcoholic subjects and in two of seven controls. By using the detection limit of the CDTectTM method (VH-CDT = 5 U/l) as cut-off level for a positive test, the alcoholic group was significantly separated from the control group (P = 0.0024, Fisher's exact test). The sensitivity and specificity of the test was 95% and 71%, giving a positive and a negative predictive value of 91% and 83%, respectively. Time-dependent changes of VH-CDT in the dead body could not unequivocally be excluded, which must be considered when selecting cases suitable for VH-CDT analysis. We conclude that adding VH-CDT analysis to ordinary alcohol tests may become useful in forensic medicine for establishing the so-called ‘alcoholic state’, which may provide a tool in research dealing with the relation between alcohol withdrawal and various causes of death in alcoholics.


    INTRODUCTION
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Complications to acute alcohol withdrawal, rather than to acute or chronic alcohol intoxication, may be responsible for some of the excess mortality in alcohol-dependent individuals (Denison et al., 1995Go, 1997bGo). Acute alcohol withdrawal should thus be considered a medical emergency during which alcoholics seem to be liable to sudden death from, e.g., withdrawal seizures and cardiac complications (Denison et al., 1994Go, 1995Go, 1997aGo). Due to the problem of establishing such causes of death at autopsy and the lack of biochemical methods to support the diagnosis of death during acute alcohol withdrawal, it has clearly been difficult to study the context of withdrawal-related death in alcoholics.

In the clinical setting, elevated serum concentration of carbohydrate-deficient transferrin (CDT) is used as a specific marker of potentially harmful alcohol consumption (Stibler, 1991Go). An increased serum CDT concentration corresponds to continuous ingestion of large amounts of alcohol (Salmela et al., 1994Go). Due to the long half-life of CDT in serum it is possible to detect high serum concentrations several days after an alcohol debauch (Stibler, 1991Go). The CDT method thus seems to be well-suited to indicate pre-mortal alcohol misuse in an alcoholic. Thus, an elevated serum CDT concentration in combination with negative blood and/or urine ethanol concentrations would support a withdrawal-related death. However, in post-mortem analysis, clinically used cut-off levels do not seem to be applicable to serum CDT (Sadler et al., 1996Go). High values of CDT may thus be false due to carbohydrate-chain dissociation from the transferrin molecule by enzymatic or bacterial interactions in blood after death (Sadler et al., 1996Go).

Our hypothesis was that some of these methodological problems could be overcome if a fenced body fluid such as vitreous humour was used for post-mortem CDT analysis. Consequently, we aimed to investigate whether in the present study reliable post-mortem analysis of vitreous humour CDT concentration (VH-CDT) could be achieved by use of a commercially available assay designed for serum analysis (CDTectTM). Moreover, we intended to evaluate the usefulness of VH-CDT as a marker of alcohol misuse and possible withdrawal-related death in forensic medicine.


    MATERIALS AND METHODS
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Subjects
A convenience sample of 21 alcoholic men aged 32–61 years (median 52 years) and seven non-alcoholic men aged 19–57 years (median 24 years) was recruited from a population of men who had died under circumstances necessitating forensic examination by decision of the police authority. In each case, the diagnosis of alcoholism was primarily based upon information about long-standing alcohol misuse in the police report. The diagnosis was further supported by non-specific but possibly alcohol-related autopsy findings in the liver (fatty liver, cirrhosis), in the pancreas (atrophy, fibrosis), in the brain (atrophy, contusions), and/or in the heart (fibrosis, hypertrophy). Detailed information concerning the past and recent alcohol history of each individual was not possible to obtain.

For each subject, the time-points were noted regarding when the person had been seen alive for the last time and when the body was found (see Table 1Go). Cause of death was established based on clinical and biochemical investigations according to the routines at the forensic department (Table 1Go). The study was approved by the Research Ethics Committee of Göteborg University.


View this table:
[in this window]
[in a new window]
 
Table 1. Demographic data and results from analyses of vitreous humour carbohydrate-deficient transferrin and urine/blood alcohol
 
Sampling procedures
After the dead subjects were found, they were transported to the forensic department and kept in a supine position at +4°C. At the start of the autopsy, vitreous humour was aspirated from the eye by means of a sterile syringe and needle. The vitreous humour was then collected in sterile tubes and kept at –70°C at the forensic department. When all specimens had been collected the frozen tubes were sent to Pharmacia & Upjohn Diagnostics Sverige AB for analysis. The specimens had been assigned codes, thus making the laboratory staff blinded to whether the samples were taken from alcoholics or controls. Blood samples from the femoral vein and urine samples from the bladder were collected at autopsy for measurement of ethanol concentrations and for drug screening tests.

Biochemical assay
Carbohydrate-deficient transferrin concentration in vitreous humour (VH-CDT). One tube of vitreous humour from the eye from each subject was thawed, well mixed and tested in double duplicate in the same assay run using CDTectTM radioimmunoassay (Pharmacia & Upjohn Diagnostics AB, Uppsala, Sweden, nowadays Axis-Shield PLC, Oslo, Norway) according to the manufacturer's directions as detailed by Stibler et al. (1991). Briefly, after iron saturation of the samples, separation of the transferrin isoforms with pI >5.7 was accomplished by anion-exchange chromatography on microcolumns. Thereafter, these carbohydrate-deficient fractions of transferrin were quantified by a double-antibody radioimmunoassay with a standard curve comprising values from 5 to 300 U/l. The mean value was reported and used for all calculations. The intra-assay and inter-assay coefficients of variation for CDT analyses were <5 and 16.8%, respectively.

Cut-off level for VH-CDT. Since the transferrin content in vitreous humour is <10% of that in serum (Devgun and Dunbar, 1986Go), it was unknown at the planning of the study whether CDT could be detected in vitreous humour. It thus seemed appropriate to use the detection limit of the commercially available test, which is 5 U/l, as the highest acceptable cut-off level of VH-CDT for recent alcohol abuse.

Ethanol concentrations . Ethanol concentrations were measured by headspace gas chromatography (Jones and Schuberth, 1989Go) at the Department of Toxicology, National Institute of Forensic Medicine (Linköping, Sweden).

Statistics
Due to the small sample size, Fisher's exact test was used to compare categorical data between alcoholics and controls. P < 0.05 was considered as statistically significant.


    RESULTS
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Twenty of the 21 alcoholic subjects and two of the seven controls had detectable VH-CDT (Fig. 1Go). By using a VH-CDT of 5 U/l as a cut-off level for a positive test, the alcoholic group was significantly separated from the control group (P = 0.0013, Fisher's exact test). The sensitivity of the test was 95% and the specificity 71%, giving a positive and a negative predictive value of 91 and 83%, respectively.



View larger version (6K):
[in this window]
[in a new window]
 
Fig. 1. Distribution of carbohydrate-deficient transferrin concentrations in vitreous humour (VH-CDT) in a sample consisting of alcoholics and controls undergoing forensic examination. Of the subjects with positive result (VH-CDT >=5 U/l), seven alcoholics but no controls were tested within 72 h of death. The corresponding numbers in those with levels of <5 U/l were one alcoholic and five controls.

 
The highest VH-CDT value was recorded in an alcoholic who was examined almost 10 days after the last time he had been seen alive (Table 1Go). In the two controls with detectable VH-CDT the examination was carried out after a considerably longer time span since they were last seen alive, compared to other controls (Table 1Go). Since it could not unequivocally be excluded that the raised VH-CDT was related to time-dependent post-mortal biological processes, Fisher's exact test was repeated in eight alcoholics and five controls who undoubtedly had been examined within 72 h after death (Table 2Go). The result of this analysis shows a consistent significant difference between the groups (P = 0.0093). The specificity and positive predictive value of the test with this selection modification were both 100%, and the sensitivity and negative predictive values 88 and 83%, respectively.


View this table:
[in this window]
[in a new window]
 
Table 2. Distribution of positive and negative tests among alcoholics and controls with use of vitreous humour carbohydrate-deficient transferrin (VH-CDT) concentration of 5 U/l as cut-off value
 
Table 3Go defines alcoholic states of post-mortem subjects on the basis of a comparison of VH-CDT and urine/blood alcohol data.


View this table:
[in this window]
[in a new window]
 
Table 3. Clinical application of combining the results of analyses of vitreous humour carbohydrate-deficient transferrin (VH-CDT) and urine/blood alcohol concentrations
 

    DISCUSSION
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
The rationale for the use of vitreous humour for post-mortem analysis of CDT is that this body fluid seems to be more protected from bacterial action during putrefaction than blood (Harper, 1989Go). It thus seems possible that determination of CDT in vitreous humour may reflect more adequately the biochemical situation at the time of death, than analysis of CDT in serum. In a post-mortem analysis of CDT in serum, Sadler et al. (1996) found that not only alcoholics but also controls had CDT levels well above the clinically used in vivo reference level, probably due to post-mortem changes. Such processes may be time-dependent, which causes problems in setting reference levels and thus limits the usefulness of testing CDT in serum post mortem.

The choice of CDTectTM for analysis of VH-CDT in our study turned out well because, by chance, it seemed to be possible to use the detection limit of this test in a semiquantitative way to discriminate between alcoholics and non-alcoholics. The highly significant result in the Fisher's exact test was somewhat surprising, because it could hardly be expected that all of the alcoholics would have been in a drinking phase. However, the outcome supports that most of the alcoholics had been drinking heavily prior to death, because a raised VH-CDT level is related to alcohol consumption and not to alcoholism per se.

Two of the controls had detectable levels of VH-CDT. This may be due to selection failure; that is, they may in fact have been high-consumers of alcohol. However, since these two controls had a much longer post-mortem interval before the vitreous humour specimens were collected, time-dependent changes in CDT concentrations may have occurred also in vitreous humour, giving a false positive test. Theoretically, this hypothesis could easily be investigated by taking repeated samples of vitreous humour from some subjects at various time points for CDT analyses. However, this procedure was not considered appropriate, because it must be assumed that the barrier surrounding the vitreous humour in this way is destroyed and thus the VH-CDT will be influenced in the same manner as is noted in post-mortal serum specimens. Another cause of a false positive result could be certain eye disorders in which the protein content may increase and the protein composition may change as a consequence of the breakdown of the blood–retinal barrier (Bresgen et al., 1991Go).

Until more data are available on this subject, it seems reasonable to restrict the post-mortem interval for the use of VH-CDT analysis. Nevertheless, when an arbitrary, but in forensic medicine practical, post-mortem interval of 72 h for VH-CDT analysis was applied, the outcome was not considerably changed.

The conclusions from the present preliminary study, with a rather small sample, must be drawn with some caution. There is a need for more studies to compare the various CDT methods (Keating et al., 1998Go; Viitala et al., 1998Go). If the findings of this study can be reproduced and established, analysis of CDT in vitreous humour can be useful in forensic medicine with at least two applications. First, it may make it possible to detect heavy alcohol consumption before death in forensic cases in general. Such information may explain certain confusing autopsy findings, and may also be helpful in the process of establishing cause of death in obscure cases. Secondly, in persons with known alcohol-dependence, analysis of CDT in vitreous humour, in combination with blood and urine ethanol tests, may offer an opportunity to define an ‘alcoholic state’ during which an individual has died (Table 3Go). This approach would make it possible to link various causes of death to certain phases of alcohol misuse and thus provide a tool to identify withdrawal-related death in alcoholics.


    ACKNOWLEDGEMENTS
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Thanks to Stina Söderlund at Pharmacia & Upjohn for skilful laboratory assistance, to Pharmacia & Upjohn Diagnostics, Sverige AB for supplying CDTectTM kits, and to Göteborg University for financial support (research grants LUA 631407).


    FOOTNOTES
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
* Author to whom correspondence should be addressed at: Department of Medicine, Sahlgrenska University Hospital/Östra, Plan 2, S-416 85 Göteborg, Sweden. Back


    REFERENCES
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Bresgen, M., Martiny, B., Weller, M., Heimann, K. and Wiedemann, P. (1991) Analyse des Proteinmusters in physiologischen und pathologischen Glaskörpern durch Elektrophorese und immunologische Identifizierung. Forschritte der Ophtalmologie 88, 665–670.

Denison, H., Jern, S., Jagenburg, R., Wendestam, C. and Wallerstedt, S. (1994) Influence of increased adrenergic activity and magnesium depletion on cardiac rhythm in alcohol withdrawal. British Heart Journal 72, 554–560.[Abstract]

Denison, H., Berkowicz, A., Wendestam, C. and Wallerstedt, S. (1995) Ischemic heart disease and epilepsy: two major causes of out-hospital natural death in male alcoholics. Forensic Science International 73, 19–31.[ISI][Medline]

Denison, H., Jern, S., Jagenburg, R., Wendestam, C. and Wallerstedt, S. (1997a) ST-segment changes and catecholamine-related myocardial enzyme release during alcohol withdrawal. Alcohol and Alcoholism 32, 185–194.[Abstract]

Denison, H., Berkowicz, A., Odén, A. and Wendestam, C. (1997b) The significance of coronary death for the excess mortality in alcohol-dependent men. Alcohol and Alcoholism 32, 517–526.[Abstract]

Devgun, M. S. and Dunbar, J. A. (1986) Biochemical investigation of vitreus: Applications in forensic medicine, especially in relation to alcohol. Forensic Science International 31, 27–34.[ISI][Medline]

Harper, D. R. (1989) A comparative study of the microbiological contamination of postmortem blood and vitreous humour samples taken for ethanol determination. Forensic Science International 43, 37–44.[ISI][Medline]

Jones, A. W. and Schuberth, J. (1989) Computer-aided headspace gas chromatography applied to blood-alcohol analysis: importance of online process control. Journal of Forensic Science 34, 1116–1127.[ISI]

Keating, J., Cheung, C., Peters, T. J. and Sherwood, R. A. (1998) Carbohydrate deficient transferrin in the assessment of alcohol misuse: absolute or relative measurements? A comparison of two methods with regard to total transferrin concentration. Clinica Chimica Acta 272, 159–169.[ISI][Medline]

Sadler, D. W., Girela, E. and Pounder, D. J. (1996) Postmortem markers of chronic alcoholism. Forensic Science International 82, 153–163.[ISI][Medline]

Salmela, K. S., Laitinen, K., Nyström, M. and Salaspuro, M. (1994) Carbohydrate-deficient transferrin during 3 weeks heavy alcohol consumption. Alcoholism: Clinical and Experimental Research 18, 228–230.[ISI][Medline]

Stibler, H. (1991) Carbohydrate-deficient transferrin in serum: a new marker of potentially harmful alcohol consumption reviewed. Clinical Chemistry 37, 2029–2037.[Abstract/Free Full Text]

Stibler, H., Borg, S. and Joustra, M. (1991) A modified method for the assay of carbohydrate-deficient transferrin (CDT) in serum. Alcohol and Alcoholism 26 (Suppl. 1), 451–454.[ISI][Medline]

Viitala, K., Lahdesmaki, K. and Niemala, O. (1998) Comparison of the Axis %CDT TIA and the CDTect method as laboratory tests of alcohol abuse. Clinical Chemistry 44, 1209–1215.[Abstract/Free Full Text]

World Health Organization (1978) Manual of the International Classification of Diseases, Injuries and Causes of Death (ICD-9). World Health Organization, Geneva.





This Article
Abstract
FREE Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (3)
Request Permissions
Google Scholar
Articles by Berkowicz, A.
Articles by Denison, H.
PubMed
PubMed Citation
Articles by Berkowicz, A.
Articles by Denison, H.