Use of a 'blood patch' to treat severe headache due to cerebrospinal fluid leakage after epidural anesthesia

A colleague reports that her hospital's transfusion service received a request to dispense a 30 mL aliquot of ABO group-specific RBCs for use as a "blood patch" for an obstetrics patient who was experiencing a post epidural anesthesia headache. The term "blood patch" was totally new to the inquiring colleague's transfusion service, including those with years of experience. Her investigation revealed that a 'blood patch' is a treatment for epidural anesthesia that has been complicated by severe headache due to CSF leakage, and entails drawing 15-30 mL of autologous blood from a patient and immediately injecting it into or near the anesthesia injection site in the vicinity of the dura.

Here are some references to the use of an epidural blood patch:

• Woodward WM et al, 1994
• Costigan SN, Sprigge JS, 1996
• Yentis SM & Haire K, 1997
• Davies JM 2001

The injected blood clots, which seals the leak, after which the CSF volume is spontaneously restored and the headache subsides. Because autologous blood is used, the hospital transfusion service may have no knowledge that this practice is occurring. In this particular case, the physician who ordered the RBC aliquot explained that his patient had hepatitis B and he did not want to handle her autologous blood. The blood bank staff explained that the request for banked RBCs did not seem appropriate because AS-1, AS-3 or AS-5 RBCs contain so little plasma that to inject an aliquot of 'banked red cells' might not result in the desired clotting effect.

The inquiring colleague wonders if others have had similar requests for "blood patches", and/or if other hospitals are using sterile, virus-free, human-derived fibrin sealants to manage headaches in this situation.

The following comments have been received.

ADDENDA Feb. 24, 2004

1. An anesthesiologist in Ohio reports that he is totally opposed to any request for an aliquot of allogeneic RBCs to be used as a blood patch. In his opinion, there is absolutely no reason that a patient's own whole blood could not be used, even if the patient had a viral infection, such as hepatitis B or C. If the ordering physician is afraid of handling the blood from a Hepatitis B infected patient, he/she should get the Hepatitis B vaccine and follow universal (standard) precautions. He reminded us that as physicians, especially surgeons and anesthesiologists, frequent contact with the blood of infected individuals is part of the job. Finally, the Ohio colleague believes that an allogeneic RBC blood patch would not work due to the fact that allogeneic RBCs lack sufficient plasma and functional platelets. He adds that it would be better to ask the blood bank to dispense a unit of platelets which could be activated with thrombin and calcium in order to form a platelet gel. The Ohio anesthesiologist concludes saying 'I am going to have to share this one with my residents!'
   
   
2. An anesthesiologist in New Jersey reports that in his opinion, the use of an allogeneic aliquot of blood for "post dural puncture heachache" makes no sense, and may actually contribute to iatrogenic disease. In his community practice a blood patch is routinely (always) comprised of autologous blood, for obvious reasons.

ADDENDA Feb. 25, 2004

3. An anesthesiologist at an academic center in San Francisco reports that the blood patch technique has been used for more than 30 years. It is well-established practice, and fresh autologous whole blood is used. The San Francisco physician has never heard of stored allogeneic blood being used, since stored blood would not be expected to clot, thus, it does not make sense to use stored blood for this purpose. He adds that some have, at times, injected saline, with suboptimal good results. He would not approve a request for allogeneic blood for this purpose. Although he is unaware of the use of fibrin sealants for this purpose in humans, he would not be surprised if someone had tried it, in as much as it has been shown to be effective in a swine model.

ADDENDA Feb. 26, 2004

4. An anesthesiologist in Texas concurs that allogeneic RBCs probably would not coagulate to effectively seal the type of leak that is under discussion. He reports that experience is that there may be some temporary relief of headache purely from the volume effect, because merely injecting saline into the epidural space also sometimes relieves the headache temporarily. He adds that the comment about using platelets to make platelet gel is interesting, and that using a commercially available device like the Magellan Platelet system, one can make platelet gel from the patient's blood, and that surgeons actually use this system to seal CSF leaks created during back surgery. Finally, he comments that some anesthesiologists are concerned about performing neuraxial procedures in patients with active viral disease processes such as hepatitis and HIV. For example, he asks, "if one took an aliquot of infected blood from that patient, and attempted to perform a blood patch, and inadvertently some of the infected blood product leaked into the CSF, would the patient be susceptible to encephalitis?"
   
   
5. Editor's NOTE: A colleague has submitted the following link for consideration by the e-Network Forum as we discuss the merits of the blood patch system. This link addresses the issue of possible viral spread to the nervous system by using infected autologous blood for a blood patch. (Webmaster's Note- the link no longer exists)

Submit comments to the e-Network Forum at enetworkforum@cbbsweb.org

Ira A. Shulman, MD
CBBS e-Network Forum Senior Editor & Moderator

W. Tait Stevens, MD
CBBS e-Network Forum Editor & Moderator

Elizabeth M. St. Lezin, MD
CBBS e-Network Forum Associate Editor & Moderator