To the Editor: The recent "heparin scandal" resulted from theuse of contaminated heparin that caused serious adverse eventsincluding death.1 The contaminant was identified as syntheticallyoversulfated chondroitin sulfate (OSCS).2 Despite the missingfinal proof of a cause-and-effect relationship, OSCS was shownto have pharmacologic effects that may contribute to the observedallergic-type reactions.3 Furthermore, OSCS is suspected tobe responsible for an observed increased incidence of heparin-inducedthrombocytopenia type 2.4 Revised monographs about heparin inU.S. and European pharmacopeias now include two mandatory teststo identify OSCS — nuclear magnetic resonance spectroscopyand capillary electrophoresis. However, these methods are technicallychallenging, not widely established, and not easily applicablein clinical practice.
We report on the use of prothrombin time to detect OSCS contaminationin both unfractionated heparin and low-molecular-weight heparins.As expected, only plasma concentrations of 5 µg per milliliter(about 1 IU per milliliter) or greater of unfractionated heparinand 10 µg per milliliter of unfractionated heparin contaminatedwith 17.4% OSCS (contaminated unfractionated heparin) slightlyprolonged the prothrombin time (by 9% and 12%, respectively)(Figure 1A). At concentrations of 5 µg per milliliteror less, contaminated unfractionated heparin, but not unfractionatedheparin, shortened the prothrombin time. However, it is essentialto use a prothrombin-time reagent that does not contain a heparin-neutralizingcompound such as polybrene, since this compound also antagonizesthe effects of OSCS.
Figure 1. Reduction of Coagulation Time in the Prothrombin-Time Assay by Heparins Containing Oversulfated Chondroitin Sulfate.
Pooled human plasma samples were treated with unfractionated heparin (Heparin Na, lot no. 73508019, Novartis), unfractionated heparin contaminated with 17.4% oversulfated chondroitin sulfate (OSCS) (Rotexmedica), or synthesized OSCS by mixing 10 µl of solution containing a concentration 10 times higher than the plasma concentration in 0.9% sodium chloride with 90 µl of plasma. The control was plasma supplemented with sodium chloride. Panel A shows the concentration-dependent effect of the samples on the prothrombin time. The reagent used was Thromborel S (Dade-Behring), and the instrument was a KC10 Coagulameter (Amelung). Similar results were obtained by turbidimetry (Behring Coagulation System, Dade-Behring). The standard deviations of the coagulation times were generally 2.5% or less. Maximum shortening of the prothrombin time was observed at a plasma concentration of 1 µg per milliliter. I bars denote standard deviations. Panel B shows the percentage reduction of the prothrombin time by unfractionated heparin, unfractionated heparin containing 4.4 to 52.2% OSCS, OSCS, enoxaparin, and enoxaparin containing 7.0% OSCS, each at a plasma concentration of 1 µg per milliliter. The percentage reduction was calculated with the formula 100–([mean sample coagulation time÷mean control coagulation time]x100).
To verify these observations, we synthesized OSCS by sulfatingchondroitin 4-sulfate.5 Whereas chondroitin 4-sulfate did notmodify the coagulation time, OSCS showed weak anticoagulantactivity at concentrations of more than 10 µg per milliliter,but it shortened the prothrombin time more than contaminatedunfractionated heparin. Unfractionated heparin supplementedwith 17.4% OSCS produced effects that were identical to thoseof contaminated unfractionated heparin. The reduction of thecoagulation time by unfractionated heparin containing OSCS increasedwith the increasing content of OSCS (Figure 1B). Contaminatedlow-molecular-weight heparins such as enoxaparin were also detectable(Figure 1B).
Highly sulfated polysaccharides such as OSCS are known to actas anticoagulants, but they also induce contact activation promotingin vitro coagulation.3 However, with activated partial-thromboplastintime assays (also used for the assay of heparin in the U.S.and European pharmacopeias) in which the coagulation is initiatedby contact activators, such effects cannot be recognized. Incontrast, the prothrombin time is suitable for determining suchprocoagulant effects for two reasons. First, the coagulationis induced by thromboplastin. Second, the prothrombin time isquite insensitive to heparins, so that any procoagulant effectof OSCS is not obscured.
In conclusion, the prothrombin time could be used in a validatedform as a sensitive screening test for the quality control ofheparins. In clinical practice it could serve as a simple andfast assay to check the applied heparin when heparin-inducedthrombocytopenia type 2 or allergic-type reactions develop ina patient who is receiving heparin.4
Susanne Alban, Ph.D. Susanne Lühn, Pharm.D. Christian-Albrechts University Kiel D-24118 Kiel, Germany salban{at}pharmazie.uni-kiel.de
This letter (10.1056/NEJMc0807070) was published at www.nejm.orgon December 3, 2008.
Dr. Alban reports receiving consulting fees from GlaxoSmithKlineand lecture fees from CSL Behring, GlaxoSmithKline, LEO Pharma,Novartis, and Pfizer. No other potential conflict of interestrelevant to this letter was reported.
References
2nd Workshop on the characterisation of heparin products. Strasbourg, France: European Directorate for the Quality of Medicines & HealthCare (EDQM), 2008. (Accessed November 24, 2008, at http://www.edqm.eu/site/Download-527.html.)
Guerrini M, Beccati D, Shriver Z, et al. Oversulfated chondroitin sulfate is a contaminant in heparin associated with adverse clinical events. Nat Biotechnol 2008;26:669-675. [CrossRef][Web of Science][Medline]
Kishimoto TK, Viswanathan K, Ganguly T, et al. Contaminated heparin associated with adverse clinical events and activation of the contact system. N Engl J Med 2008;358:2457-2467. [Free Full Text]
Greinacher A, Warkentin TE. Contaminated heparin. N Engl J Med 2008;359:1291-1292. [Free Full Text]
Maruyama T, Toida T, Imanari T, Yu G, Linhardt RJ. Conformational changes and anticoagulant activity of chondroitin sulfate following its O-sulfonation. Carbohydr Res 1998;306:35-43. [CrossRef][Web of Science][Medline]
Previous
