Carbohydrate-deficient transferrin did not show significant group changes with delayed sampling or collection site in a 70-autopsy study by Rainio and colleagues. That finding supports relative stability under the conditions studied. It does not prove that every postmortem specimen is usable or that a CDT result establishes acute intoxication.
A marker of drinking history, not a blood ethanol measurement
CDT concerns less glycosylated forms of the iron-transport protein transferrin. The study quantified disialotransferrin relative to selected transferrin isoforms, reporting a percentage index. This is analytically different from measuring ethanol in g/L or mg/100 mL.
The research question concerned whether a marker used in assessing chronic excessive alcohol consumption remained stable after death. The investigators did not validate CDT as a measure of ethanol concentration at death, the time of a last drink, behavioral impairment or cause of death.
How sequential specimens were collected
Blood was obtained from 70 consecutive corpses at the Karolinska forensic medicine department. An initial femoral sample was collected transcutaneously as soon as practicable after arrival at the mortuary. At autopsy one to six days later, separate samples were obtained by cutting the left and right femoral veins and inferior vena cava.
The resulting 280 specimens were centrifuged. Serum was stored without preservative at −20°C until analysis approximately two months later. The delay while the body remained in cold storage and the later storage of separated serum are different parts of the specimen history.
What the paired comparison showed
Initial and autopsy values could be compared in 29 cases. The interval between collections ranged from 24.8 to 145.3 hours, with a median of 73 hours. Estimated time in warm conditions before mortuary transport ranged from 1.7 to 38.0 hours; the initial sample was collected 4.5–66.1 hours after death.
For capillary zone electrophoresis, reported changes ranged from −0.8 to +0.9 percentage points, with a median of −0.1. For HPLC, changes ranged from −0.6 to +0.4 percentage points, with a median of −0.2. A paired t-test found no significant difference at the study’s 0.05 threshold.
These are changes in the percentage index, not relative percentage losses of the original value. A nonsignificant group result also does not establish that every individual value remained identical.
Site comparisons supported the same limited conclusion
ANOVA found no significant differences among left femoral, right femoral and inferior-vena-cava results. The authors interpreted the findings as showing no major redistribution of CDT under the tested conditions. This cannot be generalized to unrelated drugs, whose distribution and postmortem behavior differ.
The methods were capillary zone electrophoresis with ultraviolet detection and anion-exchange HPLC with visible detection. Quantitation used the disialotransferrin peak area divided by the summed areas of specified transferrin isoforms. The article did not provide a full numerical detection-limit, quantification-limit and precision dataset for this postmortem application.
Missing results limit the inference
Only 115 of 280 electrophoresis analyses were interpretable. At least one usable result was obtained in 57 of 70 autopsies. Uninterpretable specimens cannot be treated as normal CDT values, and the stability findings necessarily depend on the subset that could be measured.
The study did not compare CDT with independently established drinking histories to calculate diagnostic sensitivity or specificity. Its evidence concerns analytical feasibility, stability and site comparisons. For a contrasting analyte-specific distribution problem, see different postmortem distributions of co-detected drugs.
Primary source
Rainio J, De Paoli G, Druid H, et al. Post-mortem stability and redistribution of carbohydrate-deficient transferrin (CDT). Forensic Science International. 2008;174:161–165. doi:10.1016/j.forsciint.2007.03.020.
Leave a Reply