Pathophysiology and Management of Post-dural Puncture Headache: A Current Review


It has been just over one hundred years since Dr. Bier experienced and wrote about the first reported postdural puncture headache. Dr. Bier's classic description of his severe postural headache would be familiar to anyone in practice today. A postdural puncture headache (PDPH) or "spinal headache" is usually described as a severe, dull, nonthrobbing pain, usually fronto-occipital, which is aggravated in the upright position and diminished in the supine position. It may or may not be accompanied by nausea, vomiting, visual disturbances and/or auditory disturbances. Patients who experience a postdural puncture headache should not be taken lightly. Data obtained from the ASA's closed claims analysis project show that this is the third most common reason for litigation in obstetric anesthesia. (1) Anyone being treated for a PDPH should receive reassurance as well as a full and frank discussion of treatment options.

A PDPH is usually a self-limiting process. If left untreated, 75% of them will resolve within the first week and 88% will have resolved by 6 weeks. (2) Most treatments are geared towards lessening the pain and symptoms until the hole in the dura can heal, or at least until it can close to the point where the symptoms are tolerable. So-called "conservative treatment" involves hydration, bedrest and analgesics.

The concept of hydration for a PDPH is often misunderstood. The purpose of the hydration is to ensure that the rate of CSF production is appropriate. Although the degree of CSF leak does not correlate with the severity of the symptoms in a PDPH, it is assumed that improvements in the ratio of CSF production to CSF leak will improve the clinical picture. Dehydration can result in a decrease in CSF production. However, if someone is appropriately hydrated, and the rate of CSF production is normal, there is no evidence that overhydration will increase the rate of CSF production any further. Therefore, there is no point in administering fluids to a patient who is already appropriately hydrated.

Bedrest is advised simply to lessen the severity of symptoms. At one time, it was thought that remaining supine after a dural puncture would help prevent the occurrence of a PDPH. This is not the case and the incidence of PDPH is not affected by bedrest. Bedrest is therefore no longer advised unless symptoms occur. When symptoms do occur, the patients are often at bedrest anyway, since it is the only way that they can obtain relief. For a mild PDPH, NSAIDs are often prescribed. For severe headaches, NSAIDs may not be sufficient, and narcotics may be necessary in the initial period. Epidural morphine has been shown to be effective, but is usually not a convenient treatment for an outpatient or a patient waiting to be discharged.

Most of us were taught the classic explanation for the pathophysiology of a PDPH. CSF leaking from a dural puncture would lead to a loss of CSF pressure in the spine and a loss of buoyancy supporting the brain. When the patient assumes an upright posture, the brain sags, and tension on the meninges and other intracranial structures creates the pain seen with PDPH. As Sechzer pointed out in the 1970s, this explanation is probably overly simplified. (3) It is likely that as the body assumes a vertical posture, the hydrostatic gradient across the brain increases, forcing more CSF to exit the dural puncture. The body then attempts to compensate for the loss of intracranial volume by vasodilation. Much of the pain in a PDPH would then be related to vascular distention. This process would reverse itself when the patient again became supine.

This difference in mechanisms is an important one. If some or all of the pain of a PDPH is the result of processes similar to those that occur during a vascular headache, then a PDPH should be susceptible to the treatments used for vascular type headaches. In fact, these treatments have met with varying degrees of success.

Intravenous caffeine sodium benzoate has been in use, off and on, for approximately half a century as a treatment for PDPH. One regimen was to put 500 mg of caffeine sodium benzoate in one liter of intravenous fluid and infuse this over one hour. It could be repeated every 8 hours. Presumably, the cerebral vasoconstrictive effects of caffeine help attenuate the vascular distention, providing analgesia. While some investigators reported very good results with intravenous caffeine sodium benzoate, others found it to be no better than placebo. These days it is difficult to obtain caffeine sodium benzoate in the United States.

Camann showed that 300 mg of oral caffeine would provide some temporary analgesia, but while the relief was better than placebo, it was often temporary and there was a high recurrence rate. (4) As a practical matter, oral caffeine is not always useful in this situation, since it involves taking multiple 300 mg doses of caffeine throughout the day. A time-release preparation of theophylline worked somewhat better, presumably because of its longer duration of action as well as its more potent vasoconstrictive effect on cerebral vessels. (5) Sumatriptan 6 mg subcutaneously has been very successful in some studies in relieving the pain of a PDPH. (6) It remains to be seen whether the oral form of this medication works as well as the subcutaneous route.

An epidural blood patch (EBP) remains the standard against which all other treatments for a PDPH are compared. By creating a blood clot over the hole in the dura, the CSF leak can be slowed or halted entirely. As with the more conservative treatments, the epidural blood patch buys time. While the clot is in place, the hole is closing. By the time the clot has resolved, the hole has usually reduced in size to the point where the symptoms of the PDPH have either disappeared or become tolerable. Of course, should the clot become dislodged during this period, the headache may return.

The pain relief from an EBP is often immediate. If the EBP worked simply by plugging the dural leak, one would expect that the CSF deficit would take a much longer time to be replenished. Therefore, a second mechanism of action may be at work. It is possible that the EBP also works by increasing pressure in the spinal cord.

When blood patches were first performed, small volumes (3 - 8 ml) of autologous blood were used. These days, the common practice is to use 15 to 20 ml of blood in the patch. This enables a greater spread of the blood through the epidural space to ensure that it covers the dural puncture. With these volumes, the EBP is said to be effective more than 95% of the time. (7) A headache may recur following an EBP, but the EBP is so effective in the treatment of a PDPH, that if the patient fails to obtain any relief, the original diagnosis should be questioned. Headaches that result from other etiologies are sometimes mistaken for a PDPH. (8)

Although the EBP is usually thought of as a benign procedure, it is not without its complications. Fortunately, most of these are relatively minor. Approximately 35% of patients who receive an EBP report back pain. (9) Neck pain, leg pain, paresthesias, radiculitis, fever, and temporary cranial nerve palsies have all been reported following the administration of an EBP. It is not uncommon to obtain a second wet tap when attempting to place an EBP.

Many practitioners have the patient remain in the decubitus position for 30 minutes after placing the EBP before they are allowed to ambulate. There is some evidence that if the patient remains in the decubitus position for longer periods (1 to 2 hours), the EBP might be more successful. (10)

There was a time when EBPs were avoided during the first 24 hours of a PDPH. Their effectiveness during this early stage was thought to be low and the PDPH recurrence rate was believed to be high. Reevaluation of early EBPs in the 1980s and 1990s showed that with the administration of 15 to 20 ml of blood in the patch, the EBP could be very successful during this period. In fact, when an EBP was placed prophylactically for a wet tap before the development of any symptoms, only 10% of the patients developed a headache. (11)

This leads to a treatment dilemma with differing points of view. If a wet tap occurs during the attempted insertion of an epidural catheter, many anesthesiologists believe that a prophylactic blood patch should be inserted to avoid the development of a PDPH with its distressing symptomatology. Others feel that a patient should not be subjected to the possible complications of an EBP unless a PDPH develops and the symptomatology justifies the risk/benefit ratio.

Other substances besides autologous blood have been used in the epidural space in an attempt to either "patch" the hole in the dura or diminish the CSF leak. Historically, the oldest of these is normal saline, the "saline patch." There have been many regimens proposed for this therapy. One is to inject 40 ml of saline into the epidural space, and follow this with an infusion of 40 ml/hr over the next 12 to 24 hours. Others simply use the initial injection and forego the subsequent infusion. Although effective in the short term, a high recurrence rate of PDPH symptoms is often reported with this method.

Low molecular weight dextran has been used in some studies as a substitute for the EBP. Injecting 20 - 30 ml of dextran has been highly successful in treating the headaches. (12) In one study, a 20 ml of dextran given prophylactically as a "patch" was effective in preventing PDPHs. (13)

Gelatin powder (Gelfoam) and fibrin glue have both been used as epidural patches for postdural puncture headaches. They may be effective, but are significantly more difficult to administer than dextran.

Steve Schwalbe, MD
Education Committee


1. Chadwick HS. An analysis of obstetric anesthesia cases from the American Society of Anesthesiologists' closed claims project database. Int J Obstet Anesth 1996,5:258-63.

2. Vandam LD, Dripps RD. Long term follow up of patients who received 10,098 spinal anesthetics. JAMA 1956,161:586-90.

3. Sechzer PH. Post-spinal anesthesia headache treated with caffeine. Part II: Intracranial vascular distention, a key factor. Curr Ther Res 1979,26:440-448.

4. Camann WR, Murray RS, Mushlin PS, Lambert DH. Effects of oral caffeine on postdural puncture headache. A double-blind, placebo-controlled trial. Anesth Analg 1990,70:181-4.

5. Schwalbe SS, Schiffmiller MW, Marx GF. Theophylline for post-dural puncture headache (abstract). Anesthesiology 1991,75:A1082

6. Carp H, Singh PJ, Vadhera R, Jayaram A. Effects of the serotonin receptor agonist sumatriptan on postdural puncture headache: report of six cases. Anesth Analg 1994,79:180-2.

7. Crawford JS. Experiences with epidural blood patch. Anaesthesia 1980,35:513.

8. Gewirtz EC, Costin M, Marx GF. Cortical vein thrombosis may mimic postdural puncture headache. Reg Anesth 1987, 12:188-90.

9. Abouleish E, Vega S, Blendinger I, Tio TO. Long term follow up of epidural blood patch. Anesth Analg 1975, 54:459-63.

10. Martin R, Jourdain S, Clairoux M, Tétrault JP. Duration of decubitus position after epidural blood patch. Can J Anaesth 1994, 41:23-5.

11. Cheek TG, Banner R, Sauter J, Gutsche BB. Prophylactic extradural blood patch is effective: A preliminary communication. Br J Anaesth 1988, 61:340-2.

12. Barrios-Alarcon J, Aldrete JA, Paragas-Tapia D. Relief of post-lumbar puncture headache with epidural Dextran 40: A preliminary report. Reg Anesth 1989, 14:78-80.

13. Salvador L, Carrero E, Castillo J, Villalonga A, Nalda MA. Prevention of post dural puncture headache with epidural administered Dextran 40 (Letter) Reg Anesth 1992, 17:357-8.

illalonga A, Nalda MA. Prevention of post dural puncture headache with epidural administered Dextran 40 (Letter) Reg Anesth 1992, 17:357-8.