Epidural Blood Patches - Past and Present Considerations

by Donald Osborne, M.D.

It is well accepted that post dural puncture headaches (PDPHA) are a risk of spinal anesthesia or analgesia as well as other procedures in which the dura is punctured. The problem of these potential headaches in our current practice of anesthesiology continues to be a concern. The redesign of needles to minimize the risk of headache has certainly improved the likelihood of uncomplicated spinal anesthesia. The redesign has resulted in the availability of smaller gauge needles as well as needles with different shape tips including pencil point needles. The engineering goal has been to create a puncture which disrupts the structure of the dura as little as possible and allows this minimal puncture to heal spontaneously. When prevention fails we must treat the complication. There are various conservative therapies. The use of caffeine both orally and intravenously has become more popular also with substantial success. The invasive therapy of epidural blood patches generally follows failed conservative treatment.

Concern for PDPHAs exists on a daily basis for many of us. There are many operative procedures in which epidurals may be favored simply because of the reduced chance of PDPHAs. That is, the possibility of subarachnoid puncture is certainly less when an epidural needle placement is the goal. Consequently, if regional anesthesia is to be used, epidurals may be chosen in spite of a somewhat increased time for establishment of the block which may translate to an increase in cost. Spinal anesthesia would institute the block with faster onset and perhaps greater density of block, and hence, with more reliability. The recent interest in combined spinal - epidural techniques allows us the advantages of both epidurals and spinals but brings us back to a risk of PDPHA which might be greater than either technique alone. It may well be that needle design has been optimized and that we must accept the improved but prevailing risk of PDPHAs as inherent in our current practice.

It is interesting how the etiology of spinal headaches was understood and investigated many years ago. The evolution of our understanding and our application of blood patches is equally fascinating. The hypothesis in the early part of this century was that a leak did indeed exist in the dura. The loss of CSF pressure with PDPHA was demonstrated by manometer studies as early as 1923 (Jacobaeus and Frumerie). They also restored the subarachnoid pressure to normal by the subarachnoid injection of fluid and the headache disappeared instantly but the relief generally lasted only a few hours. This CSF pressure was also found to be responsive to the placement of epidural fluid. Rice and Dabbs studied this in 1948 and hypothesized that the epidural fluid stops the leak by eliminating the pressure gradient and also compresses the spinal dura thereby increasing the CSF pressure. These investigators placed a subarachnoid needle connected to a manometer as well as an epidural needle one space lower. They injected 10 ml epidurally and observed a prompt 100 mm CSF pressure rise. This pressure fell over one minute to a level 30 mm over baseline. Subsequent further injections had similar results with 30 mm CSF pressure added each time to the baseline and relief or diminution of the headache. Their average total injection was approximately 80 ml, incrementally. Further evidence for the leak theory came from a 1923 study by Pool in which he was doing myeloscopic studies of the epidural space and found that many patients who had diagnostic taps in the preceding few days had large collections of fluid in the epidural space. There had also been several autopsy findings of a patent dural tract in patients who had headaches and died in the subsequent two weeks. An initial unsuccessful approach to sealing this tract included the placement of catgut in the dural puncture which would swell on exposure to leaking CSF. The catgut would be introduced through the lumbar puncture needle. It did reduce the incidence of headache but caused cauda equina syndrome in half of the patients.

The background of blood patches is more established in terms of technique than nature. Blood in the epidural space may adhere to a dural defect but it also is a fluid in the epidural space which affects the pressure within the subarachnoid space as well as hydraulically stopping a fluid leak through a defect. The suspicion that blood in the epidural space might be useful was reported in 1960 (Gormley). Specifically, Dr. James B. Gormley believed that the 20% prevailing incidence of PDPHAs was not realized if the tap "happened" to be a bloody tap. He theorized "...that a higher incidence of repair with subsequent prevention of cerebrospinal fluid loss occurs with blood present". His goal was to restore the lost CSF volume and then supplement with blood "...adjacent to the puncture...". His first patient had a headache for the four days following a tap. Dr. Gormley placed a subarachnoid needle at the same location and was not able to measure a pressure by manometry. He instilled 15 ml of normal saline into the subarachnoid space and provided immediate relief. He withdrew the needle approximately 2 mm until CSF was no longer apparent. He then injected two ml of the patient's blood. Also, one liter of IV fluid was administered and the patient remained recumbent for a short while. Several other patients were similarly treated, all gaining permanent relief. One of the subsequent patients so treated developed their headache following a tap for a myelogram. That was Dr. Gormley himself. Another approach included the injection of partially clotted blood through the lumbar puncture needle as the needle was withdrawn. The hope was that the blood would physically plug the dural hole.
The hydraulic effects of epidural saline alone continues to be an option. It has been used prophylatically (Craft et al.) as well as following a failed series of epidural blood patches (Baysinger et al.). These latter case reports described patients in whom an epidural catheter was placed for approximately 24 hours. Saline was infused by way of a PCA pump. The initial bolus was 30-50 ml and then a patient controlled infusion of 15-30 ml/hour. Both the load and the infusion were limited by pain, especially ocular. Both patients reported obtained permanent relief.

The least frequently required but most successful therapy remains that of epidural blood patches. The immediate relief obtained from blood patches may well be from the fluid presence and resultant increase in CSF pressure and prevention of leakage just by pressure. This alone would not last long as has been discovered by the manometer studies performed long ago. However, it appears that the interaction of CSF and blood may result in a rapid clot formation. Controlled models simulating the mixing of these two fluids and utilizing a thromboelastograph have demonstrated that a clot will form in only 22 seconds (Cook et al.). The puncture site itself will be the immediate CSF / blood interface and clot formation may well originate at that site, thereby forming a plug at the optimal location. This could explain the continued benefit of a blood patch soon after the fluid effect wears off. A more recent study (Djurhuus et al.) utilized CT epidurography to examine blood patches. They examined four patients who received blood patches in which 2 ml of contrast agent was added to the blood. CT studies were performed immediately after patching and at 24 hours. They found that some distributed clot would form within the epidural space but not exert a mass effect on the dural sac. They also found that there was a clear affinity of the blood to adhere to the dural surface and form a thin clot. Additionally, the spread was predominantly cephalad indicating that the epidural injection might have a more successful distribution if it were below the original dural puncture. The investigators also saw various septae within the epidural space and a clear plica dorsalis in one of their patients. The presence of septae may often play a role in our incomplete blocks and also play a role in unsuccessful blood patches. That is, just as the septae may "protect" nerve roots from local anesthetics, they may also protect the dural puncture site from blood administered epidurally.

The evolution of treatment for PDPHAs is as interesting as the origin of spinal anesthesia itself. We have so much more knowledge now through the use of thrombo- elastograms as well as CTs. Yet, the clinical thinking earlier this century was so well directed. The complication of post dural puncture headaches, like many other anesthetic complications, has become much less common as our science has evolved. However, we have not overcome it entirely.

Bibliography

Rice, Glen G., M.D. and C. Harwell Dabbs, M.D. "The Use of Peridural and Subarachnoid Injections of Saline Solution in the Treatment of Severe Postspinal Headache", Anesthesiology, 11: 17-23, (January) 1950
Gormley, J. B. "Treatment of Postspinal Headache", Anesthesiology, 21: 565-566, 1960
Craft, John B., M.D., Burton S. Epstein, M.D. and Charles S. Coakley, M.D. "Prophylaxis of Dural-Puncture Headache With Epidural Saline", Anesthesia and Analgesia, 52: 228-231, (March-April) 1973
Baysinger, Curtis L., M.D., Emil J. Menk, M.D., Edward Harte, M.D. and Rjobert Middaugh, M.D. "The Successful Treatment of Dural Puncture Headache after Failed Epidural Blood Patch", Anesthesia and Analgesia, 65: 1242-1244 1986
Cook, Mark A. and J. M. Watkins-Pitchford, M.D. "Epidural Blood Patch: A Rapid Coagulation Response" Anesthesia and Analgesia, 70: 567, 1990
Djurhuus, H., M. Rasmussen and E. H. Jensen "Epidural blood patch illustrated by CT - epiduragraphy", Acta Anaesthesiologica Scandinavica, 39: 613-617, 1995

Questions & Answers

CME Links

 

	Continuing Medical Education - Main Page
	Fibromyalgia and Myofascial Pain
	Epidural Blood Patch
	Stress Management