Non-TIMI Major Bleeding Just as Important as TIMI Major Bleeding in Predicting Mortality in PCI Patients

September 03, 2009

September 3, 2009 (Barcelona, Spain) — In patients undergoing PCI with stable or unstable ischemic heart disease, the occurrence of a non-CABG major bleed within 30 days has an independent statistically significant impact on one-year mortality, comparable to that of having an MI, and non-TIMI major bleeding is just as important as TIMI major bleeding in predicting mortality, a new analysis shows.

Data from the pooled analysis of the three major bivalirudin trials--REPLACE-2, ACUITY, and HORIZONS--found that while bleeding meeting the TIMI major criteria has a large impact on subsequent mortality, major bleeding not meeting the TIMI criteria also increases mortality to a similar degree. But hematoma alone does not appear to increase subsequent mortality.

Presenting these data at the European Society of Cardiology 2009 Congress, Dr Roxana Mehran (Columbia University, New York, NY), explained that definitions of bleeding vary from trial to trial, which presents difficulty in attempts to compare bleeding rates in different studies. "We therefore wanted to investigate whether different types of bleeding after PCI have different impacts on subsequent mortality."

To heartwire she commented: "We have all these different definitions of bleeding used in the various trials, but we don't really know which one is best. With the three bivalirudin trials, TIMI major and minor bleeding definitions were adjudicated by an independent committee, and together the three trials had a large enough sample of patients to be able to look at how each different bleeding component related to mortality."

We have definitely shown that we should not be discounting the non-TIMI major bleeds.

She added: "We found that non-TIMI major bleeding was just as important as TIMI major bleeding in predicting mortality. This may surprise some people, as it is thought that TIMI major bleeds are the most severe, but we have definitely shown that we should not be discounting the non-TIMI major bleeds. If you discount them you are discounting a very important component of bleeding."

In contrast, the results showed that hematoma >5 cm, which is often included in major bleeding definitions, was not associated with increased mortality at one year. "We showed that these hematomas are perhaps just a nuisance. While it is certainly not pleasant for a patient to have a big lump in their groin, an isolated hematoma is not going to increase the risk of death," Mehran told heartwire .

Need to Reset the Clock

"We have to reset our clock and come up with a concise, uniform definition of bleeding to be used in future studies," she said. With this in mind, she and her colleagues are setting up a group known as the Bleeding Academic Research Consortium (BARC), which will include experts in the field, stakeholders, and regulators. They aim to develop a system under which each component of bleeding is weighted and an overall bleeding score can be worked out.

For the current study, Mehran et al conducted a patient-level pooled analysis of PCI patients in the three mega-trials, which included a total of 17 000 patients. They focused on four types of protocol-defined major bleeding in decreasing order of severity:

  • TIMI-defined major bleed.

  • Non-TIMI major bleed with blood transfusion.

  • Non-TIMI major bleed without blood transfusion.

  • Large (>5 cm) hematoma only.

They showed that any non-CABG related major bleed was associated with a similar hazard ratio for death at one year as the occurrence of an MI.

Independent Hazard Ratio of Non-CABG-Related Major Bleeding and MI Within 30 Days on Mortality Within One Year

Event Hazard ratio (95% CI) Deaths within 1 y, n p
Non-CABG major bleed 3.1 (2.4–3.9) 104 <0.001
MI 2.8 (2.2–3.6) 77 <0.001

In terms of the four defined components of major bleeding, TIMI major bleed was the most important individual type of bleed contributing to mortality, but non-TIMI major bleed with or without a transfusion together accounted for slightly more deaths. And large hematoma was not associated with many deaths at all.

Independent Hazard Ratio of Components of Non-CABG-Related Major Bleeding Within 30 Days on Mortality Within One Year

Event Hazard ratio (95% CI) Deaths within 1 y, n p
TIMI major bleed 4.85 (3.56–6.60) 53 <0.001
Non-TIMI major bleed with transfusion 2.98 (2.10–4.24) 40 <0.001
Non-TIMI major bleed without transfusion 1.79 (1.09–2.93) 17 0.021
Large (>5 cm) hematoma only 1.30 (0.58–2.92) 6 0.53

Designated discussant of the presentation, Dr William Wijns (Cardiovascular Center Aalst, Belgium), said this type of meta-analysis used the highest level of scientific evidence because it used specific statistical methods for pooling data.

He added that this new analysis confirmed that major bleeding is associated with increased mortality, which had already been shown in several previous trials, but it also showed very well that a standardized definition of bleeding is needed to be used across all trials.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.