Donate For Public and Patients Store Search

American Academy of Dermatology Logo

Breaking up (with habits) is hard to do: The role of platelet transfusions when performing skin biopsies on profoundly thrombocytopenic patients

DII small banner By Warren R. Heymann, MD
May 12, 2017

Hypersensitivity reaction with hemorrhagic appearance in thrombocytopenic patient.
Credit: JAAD
We are taught at an early age to live by the rules. In an interesting essay entitled “Breaking the Rules for Better Care”, Berwick et al tell the story of the “bicycle book” at an English hospital, where those who arrived at the hospital by bicycle had to sign a daily register. Nobody understood why. With research , was found to be a holdover from World War II, where those who bicycled to work would receive an extra food ration, because they saved on gas. In short, the process became an institutional habit, despite the fact that it long outlived its value. The authors challenge us to break unhelpful rules that hinder optimal patient care and waste valuable resources (1).

We are often asked to assess patients on the oncology floor. A characteristic request would be to evaluate a leukemic patient with neutropenic fevers and a purpuric rash. The differential diagnosis includes leukemia cutis, drug-induced vasculitis, septic vasculitis, and eccrine hidradenitis, among other possibilities. A skin biopsy split for histology and cultures is warranted, but what if the platelet count is less that 20K/µL?
My practice “rule” is to avoid skin biopsies with that degree of thrombocytopenia. I request a platelet transfusion and time the biopsy accordingly. Honestly, I have never thought twice about that rule (it made sense) until I read the article by Xia et al who hypothesized that there may be a very low risk of hemorrhagic complications in this population. A total of 205 patients met their inclusion criteria (platelet counts between 20K and 150K). Only one bleeding complication was identified; it was in a patient with a 4-mm punch biopsy obtained from the calf by the primary (nondermatology) team that was left to heal by secondary intention. The patient had oozing at the biopsy site on the day of the biopsy procedure, which was stopped with a pressure dressing. The bleeding complication rate of 1 out of 205 (0.49%) was not statistically different from the bleeding complication rate of 0% in nonthrombocytopenic patients. None of the 75 patients with platelets < 20K had bleeding complications.
The authors concluded that, “this finding should diminish physician discomfort regarding skin biopsies in thrombocytopenic patients”. Old habits die hard — should we change our approach to performing biopsies on those with significant thrombocytopenia? I asked members of our division of hematology/oncology at MD Anderson/Cooper for their opinion of this article. They graciously offered the following comments:

Neil Lachant: Interesting. Unfortunately, no data given on whether this patient had epinephrine with the biopsy, nor is there any data on the size of the biopsies on the patients who didn’t bleed.

Andres Ferber: Unfortunately, the threshold for transfusion for procedures in thrombocytopenic patients is based more on patterns of practice and opinion rather than strong data. This paper seems to suggest there is no need to for aggressive transfusion for skin biopsy based on the 75 patients who had platelets below 20,000 since they did not bleed. For most hospitalized patient the threshold for transfusion is 10,000 if stable; therefore, from the practical point of view, many of those 75 would have been transfused prophylactically anyway the day of the biopsy, so why not do it before the skin biopsy?  In fact, in the paper, 21 % of patients got transfused the day of the biopsy. It is not clear if those with platelets lower than 20,000 were preferentially transfused more than those with platelets above 20,000.

Linda Devereaux: I would argue for a less rigid and more individualized approach to biopsies -especially for patients with counts between 10 and 20k. Some patients bleed more at a given platelet level. ITP patients are a classic example of those less apt to bleed. Obviously, if the patient is covered in bruises and having a nosebleed, this is a different story. A very low number of the patients in the study had other bleeding manifestations at the time of the biopsy. A reasonable approach might be: 1. Start looking at patients between 10-20k; 2. Select only those patients for non-transfused biopsy with minimal bruising and no bleeding history. If possible discuss with the hematology team. 3. Collect detailed data and look at outcomes after 10 or 20 patients; 4. Obviously, as Andres says, transfusion may already be planned for other reasons.  In which case it would make sense to wait.

I am truly fortunate to work with such thoughtful colleagues. As Dr. Ferber mentioned, it appears that 10K is the new threshold for platelet transfusions. Estcourt et al, in reviewing guidelines for platelet transfusions, refer to a meta-analysis (658 patients) that showed that a 10K threshold was not associated with increased bleeding in comparison with a higher threshold and also showed a significant reduction in the number of platelet transfusions given. However, this meta-analysis may not have been sufficiently powered to detect an increased bleeding risk (3).

There are certain rules of life that remain inviolate — do not drink and drive, look both ways before crossing the street, and say “please” and “thank you.” The rule of not doing a skin biopsy in a patient with a platelet count of < 20K may be on the verge of changing. Xia et al are to be congratulated on making us think about dogma, and how we can “break the rule” as Berwick et al suggest.

1. Berwick DM, et al. Breaking the rules for better care. JAMA. 2017 Apr 27 [Epub ahead of print].
2. Xia FD, et al. Low risk of hemorrhagic complications after obtaining diagnostic skin biopsy specimens in a cohort of thrombocytopenic inpatients. J Am Acad Dermatol 2017; 76: 1004-5.
3. Estcourt LJ, et al. Guidelines for the use of platelet transfusions. Br J Haematol 2017; 176: 365-94.

All content found on Dermatology World Insights and Inquiries, including: text, images, video, audio, or other formats, were created for informational purposes only. The content represents the opinions of the authors and should not be interpreted as the official AAD position on any topic addressed. It is not intended to be a substitute for professional medical advice, diagnosis, or treatment.

DW Insights and Inquiries archive

Explore hundreds of Dermatology World Insights and Inquiries articles by clinical area, specific condition, or medical journal source.

Access archive