UMEM Educational Pearls

Category: Critical Care

Title: Thrombocytopenia and CVCs -- Are Platelet Transfusions Needed?

Keywords: thrombocytopenia, bleeding, hemorrhage, platelets, transfusions, central lines, CVCs (PubMed Search)

Posted: 5/30/2023 by Kami Windsor, MD
Click here to contact Kami Windsor, MD

Question

Background: In general practice, platelets are typically transfused for invasive procedures when the platelet count falls below 50 x 109/L. Regarding the placement of central venous catheters (CVCs), there is minimal data to support or refute decisions to transfuse platelets in these patients, although the 2015 Clinical Practice Guideline from the AABB (formerly, the American Association of Blood Banks) recommends deferring platelet transfusion until a platelet count of 20 x 109/L for CVC placement [weak recommendation, low quality evidence].1

In a study published this month in NEJM,2 van Baarle et al. performed a multicenter randomized controlled noninferiority trial comparing platelet transfusion to no transfusion in patients with platelets 10 to 50 x 109/L prior to US-guided CVC insertion. The primary outcome was the occurrence of catheter-related bleeding Grades 2-4 (Grade 1 = oozing; managed with <20 min of manual compression, not requiring RBC transfusion, & Grades 2-4 is everything else up to death) within 24 hours post-procedure. 

  • Noninferiority was not met, with primary outcome in 4.8% vs. 11.9% of transfused and nontransfused patients, respectively (RR 2.45, 90% CI: 1.27 to 4.70).
  • Major catheter-related bleeding (Grades 3-4) occured in 2.1% vs 4.9% (RR 2.43, 90% CI: 0.75 to 7.93).  
  • Other factors associated with higher bleeding risk included hematologic malignancy, platelets 10-20 x 109/L, and tunneled catheter placement.
  • Difference in bleeding rates between transfusion vs. no-transfusion groups was higher however, in patients with platelets 20-30 x 109/L (0 vs 15.7%), those receiving nontunneled lines (3.6% vs 10.8%), or CVCs placed in the subclavian vein (2.8% vs 18.6%). 

Bottom Line: The jury is still out on best platelet transfusion practices prior to CVC placement, but I would strongly consider prophylactic platelet transfusion in patients with platelets < 30 x 109/L, those with underlying hematologic malignancy, and patients receiving larger CVCs such as dialysis lines. How much to transfuse in those with more severe thrombocytopenia is uncertain.

Separately, I would also strongly recommend use of US-guidance for any CVC placement in this population as well, based on practical common sense and some supportive literature as well.5

Answer

Additional Background: Data in pediatric oncology patients indicates that CVC placement with platelets <50 x 109/L  is associated w/ increased occurence of minor but not major post-procedure bleeding,3 while adult data indicates that CVC placement can be performed until a threshold of 20 x 109/L before transfusions are needed to prevent severe bleeding.4

Additional Study Data:

  • Multicenter randomized controlled noninferiority trial
  • Adult patients in the Netherlands admitted to hematology ward or ICU, with platelet count 10-50 x109/L (total n for study = 373)
  • Transfusion of (1) unit of platelet concentrate vs no transfusion
  • Primary outcome: Any bleeding within 24h of US-guided CVC placement
  • Secondary outcomes:
    • Major bleeding (requiring procedural intervention or transfusion, or causing hemodynamic instability up to death)
    • # platelet and RBC tranfusions within 24 h of CVC placement
    • Hgb and platelet counts at 1h and 24h after CVC placement
    • Allergic transfusion reactions within 24h of CVC placement
    • Acute lung injury within 48h of CVC placement
    • ICU/hospital length of stay
    • Hospital mortality
    • Financial costs
  • Results:
    • Grade 2-4 bleeding in 4.8% transfusion group vs 11.9% no-transfusion
    • Noninferiority of no-transfusion strategy not met (AR difference 7.1%; RR 2.45 with 90%CI 1.27-4.70)
    • Higher risk of grade 2-4 bleeding with lower platelet counts 
    • Risk of grade 3 or 4 bleeding higher in no-transfusion group
    • Similar #RBC transfusions although more for CVC-related bleeding in no-transfusion group
    • Platelets higher at 1 and 24 hours in transfusion group
    • No transfusion group received more platelet transfusions after CVC placement
    • Similar rates of allergic tranfusion reactions/TRALI (low)
    • No transfusion group with slightly shorter ICU LOS
    • Similar mortality between groups
    • Prespecified subgroup analyses: higher bleeding % in hematology ward patients vs ICU, and in tunneled vs nontunneled CVC placement
    • Overall costs higher in transfusion group attributed to cost of prophylactic platelet transfusion -- study group notes transfusion costs within 24h of CVC placement higher in no-transfusion group due to bleeding related transfusions
  • Limitations include lack of full blinding, only transfusing 1 pack of platelets even if severe thrombocytopenia
  • Authors suggested personalized approach to patient rather than solely patient count

References

  1. Kaufman RM, Djulbegovic B, Gernsheimer T, et al. Platelet transfusion: A clinical practice guideline from the AABB. Ann Intern Med. 2015;162:205–313.
  2. van Baarle FLF, van de Weerdt EK, van der Velden WJFM, et al. Platelet Transfusion before CVC Placement in Patients with Thrombocytopenia. N Engl J Med. 2023;388(21):1956-1965. doi: 10.1056/NEJMoa2214322
  3. Stokes SCYamashiro KJBrown EG. Association of Thrombocytopenia With Bleeding Risk During Central Venous Catheter Placement in Pediatric Patients With Cancer. JAMA Surg. 2021;156(9):887–889.
  4. Zeidler  K?, Arn  K?, Senn  O?, et al?.  Optimal preprocedural platelet transfusion threshold for central venous catheter insertions in patients with thrombocytopenia. ? Transfusion. 2011;51(11):2269-2276.
  5. Cavanna L, Citterio C, Nunzio Camilla D, et al. Central venous catheterization in cancer patients with severe thrombocytopenia: Ultrasound-guide improves safety avoiding prophylactic platelet transfusion. Mol Clin Oncol. 2020;12(5):435-439. doi: 10.3892/mco.2020.2010.