Secondary Logo

Journal Logo


Pitfall in interpretation of TEG results

Jeeji, R.*; Dutta, S.*

Author Information
European Journal of Anaesthesiology: August 2008 - Volume 25 - Issue 8 - p 693-694
doi: 10.1017/S0265021508004146


Thromboelastograph (TEG) displays the viscokinetic properties [1] of the clot formed, which in turn reflects the functions of different factors involved in the coagulation process [2] and also the interaction between these factors. TEG is considered to be better than the conventional coagulation tests [3] and is becoming more popular among perioperative clinicians in managing coagulopathies. TEG machines are often located away from the theatre where surgery takes place (often placed near cardiac/vascular theatre or cardiac ICU). When TEG is performed away from where patients are treated, results are often conveyed to treating clinicians by phone, especially in emergency situations, as it is not possible for clinicians to review the TEG graph themselves (which may be still ongoing). I would like to bring to attention a potential error that can happen with this practice.

Case report

A 65-yr-old male who had undergone aortic valve replacement had postoperative bleeding in cardiac surgical intensive care. TEG was performed to find out whether the bleeding was secondary to coagulation dysfunction or due to surgical failure to attain adequate haemostasis. As this patient had undergone heparinization for cardiopulmonary bypass, TEG was performed using a plain cup as well as a heparinase cup. The TEG results were normal with the heparinase cup (Fig. 1a) and abnormal with the plain cup (Fig. 1b), even though it was done with the same sample of blood.

Figure 1.
Figure 1.:
TEG graphs: (a) normal – heparinase cup (b) abnormal – plain cup.


The abnormal graph generated by TEG using the plain cup did not fit with the typical graph generated with coagulopathies. This could have been attributed to abnormal coagulation (effect of residual heparin in this case, even though it did not fit into the typical effect of residual heparin) by clinicians who have limited experience with TEG. Values generated from the plain cup in this case could also be interpreted as abnormal clot lysis, if plain cup result was interpreted alone.

The abnormal TEG in the plain cup was due to a mechanical disturbance to the TEG machine, resulting in the breakage of the formed clot generating an abnormal graph and values. A mechanical disturbance typically occurs if someone tries to load another sample in the second channel of the same TEG machine. In these circumstances if the clinician is not able to review the TEG graph and a wrong diagnosis of coagulopathy is made, inappropriate treatment might be initiated with potentially disastrous consequences. This case highlights the importance of reviewing the TEG graph along with the values before embarking on treatment based on TEG results.


1. Wenker OC, Wojciechowski Z, Sheinbaum R, Zisman E. Thrombelastography. Internet J Anesthesiol 2000; 1(3).
2. Samama CM. Should a normal thromboelastogram allow us to perform a neuraxial block. Can J Anesth 2003; 50: 761–763.
3. Mallett SV, Cox DJ. Thrombelastography. Br J Anaesth 1992; 69: 307–313.
© 2008 European Society of Anaesthesiology