Design and implementation of a bleeding assessment score for idiopathic thrombocytopenic purpura (itp)

George R. Buchanan, Leah M. Adix

Research output: Contribution to journalArticlepeer-review


Most studies of childhood ITP have examined platelet count as the only outcome measure. However, other factors are important with regard to bleeding risk and in planning of appropriate treatment of ITP patients. Bleeding tendency, the major clinically relevant event in ITP, has been assessed quantitatively in only one prior study (Am J Pediatr Hematol/Oncol 1984; 6:355-61). We have modified and expanded on this grading system to include assessment by a physician of bleeding in children with ITP on a 0 to 4 scale (0 = no bleeding, 1 = minor, 2 = mild, 3 = moderate, and 4 = severe hemorrhage). Four bleeding scores are recorded at each patient encounter, reflecting overall clinical bleeding, epistaxis, oral bleeding, and cutaneous hemorrhage. The scores measure hemorrhage occurring during the prior 24 hours based upon history and physical examination. Between February and August 2000, bleeding score assessments were performed on 94 occasions in 40 children with ITP aged 1 to 16 yr (median 6 yr), irrespective of therapy or time from diagnosis. Platelet counts ranged from 1,000 to 98,000/mm' (median 13,000/mmJ). Interobserver variability in bleeding score measurements was determined in 15 patients assessed during the same encounter by 2 to 4 (median 2) observers. The overall and cutaneous bleeding scores correlated inversely with platelet count, i.e., the lower the platelet count the higher the bleeding score. With the exception of cutaneous bleeding, scores above 1 were observed in only 21 of 135 (15.5%) measurements in patients with platelet count > 10,000/mm1. In 48 patients with platelet count < 10,000/mm', 124 of the 144 (86%) overall, epistaxis, and oral bleeding scores were rated 2 or less (mild, minor, or no hemorrhage). A score of 4 was recorded just four times (2.8%), as a result of menorrhagia (3 measurements, 1 patient) and hematuria (1 measurement). No patient had intracranial hemorrhage. Interobserver variability in bleeding score measurements was minimal. We conclude that use of such a semi-quantitative bleeding assessment tool should accompany platelet count and other outcome measures in future studies of children with ITP receiving diverse therapies. Such scoring might also prove useful in other thrombocytopenic conditions such as acute leukemia and following cancer chemotherapy to better define candidates for platelet transfusions and other intervention. Further studies of the bleeding score are required to better determine its usefulness as a predictor of serious hemorrhage.

Original languageEnglish (US)
Pages (from-to)63b
Issue number11 PART II
StatePublished - 2000

ASJC Scopus subject areas

  • Biochemistry
  • Immunology
  • Hematology
  • Cell Biology


Dive into the research topics of 'Design and implementation of a bleeding assessment score for idiopathic thrombocytopenic purpura (itp)'. Together they form a unique fingerprint.

Cite this