Episode 21: Bleeding time


Author: Dr. Suneet Sood
Editor: Thong Yi Kun
Narrators: Thong Yi Kun, Ceceilia Ling


C: I can see you have an interesting story today

Yi Kun: Yes. I was recalling the case of one of my cirrhotic patients. This man was in the ICU, and had a platelet count of only 30,000 per microliter.

C: Ah yes, because of the splenomegaly, right?

Yi Kun: Partly, yes, the splenomegaly can cause a pancytopenia, including thrombocytopenia. However, it seems that in cirrhosis there are other mechanisms that lower the platelets. Anyway, we were wondering about the risk of spontaneous haemorrhage in this case.

C: Can it occur at a platelet count of 30,000?

Yi Kun: Not likely. Spontaneous haemorrhage tends to occur only if the platelet count falls below 20,000. [ref ] Anyway, one of my housemen asked me if we could request a bleeding time test and I agreed. So a BT was done, and it was reported as 3 minutes. The normal range is 2-8 minutes, so the team was happy. But the same houseman asked me how it was that the BT was normal despite such a low platelet count.

C: Valid question.

Yi Kun: Quite valid! So I asked my team if anybody knew the relationship between BT and platelet count, and my chief resident, a really bright chap, said he had read that there was an equation for this. This equation was valid at counts below 100,000 per microliter.

C: What equation?

Yi Kun: You divide platelet count by 3850. Subtract the result from 30.5, and you get the expected bleeding time in minutes.


Bleeding time (minutes) = 30.5 – (platelet count/3850)


C: So, for this patient…

Yi Kun: 30,000 divided by 3850 is 7.8. Subtract 7.8 from 30.5 and you get 22.7.

C: So this patient should have had a bleeding time of over 20 minutes?

Yi Kun: Yes, 22.7 minutes, and the lab reported 3 minutes, so there was a glaring discrepancy.

C: There sure was!

Yi Kun: …and nobody could explain it, so I asked the same girl how the test was done. She said the lab technician came and did the test by puncturing the finger-tip and seeing how long the bleeding continued.

C: I thought we were supposed to puncture the ear lobe?

Yi Kun: You are right, actually, but first let me tell you what happened. So I asked the houseman to do it herself using the Ivy method. She said she didn’t know what it was, and I asked her to Google it, so she did. In the Ivy method you apply a BP cuff above the elbow, and make a small puncture at the elbow. The exact dimensions of the puncture are written in the books. Anyway, she called me after 20 minutes and said that the bleeding is not stopping, what to do, and I said hold on. Finally, it turned out that the bleeding stopped after 24 minutes.

C: Wow!

Yi Kun: Then I asked her to look up the literature. The next day she reported that the first method was, as you rightly pointed out, by puncturing the ear lobe. This was the Duke method. Later the method was modified by Ivy, using a BP cuff, and this was much more sensitive.

C: Amazing! But why did the lab use the finger-tip to test the BT?

Yi Kun: My guess is that somewhere down the line people wondered why ear lobe, the finger should work as well, and modified the method without referring to the literature. You know how we all are, quick to take short cuts!

C: So the right method of doing bleeding time is the Ivy technique?

Yi Kun: Yes, it is. Of course, these in vivo tests are less often used now. Still, if one has to do the bleeding time, one should do it properly.

C: Can’t disagree with that!



Harker LA, Slichter SJ. The Bleeding Time as a Screening Test for Evaluation of Platelet Function. N Engl J Med 1972; 287:155-159

Mielke CH, Kaneshlro MM, Maher IA, Weiner JM, Rapaport SI. The standardized normal ivy bleeding time and its prolongation by aspirin. Blood 1969; 34 (2)204-215

Panzer S, Jilma P. Methods for testing platelet function for transfusion medicine. Vox Sanguinis (2011) 101, 1–9

Episode 20: Hereditary spherocytosis
Episode 22: Catch 22

Leave a comment

Your email address will not be published. Required fields are marked *