labor epidural and post-dural puncture headache (PDPH)


Post-dural puncture headache (PDPH)

Post-dural puncture headache occurs in between 1:1000 and 2:1000 in women receiving labor epidural. Most patients fully recover with conservative treatment, however small proportion require epidural blood patch.

Headache is a relatively common complication of neuraxial anesthesia. In the scenario involving epidural analgesia in labor post-spinal puncture headache usually happens after inadvertent puncture of the dura, also called dural tap. More rarely it may also happen after spinal anesthetic given for the caesarean section. Headache usually appears within the next day or two, though in some cases may be delayed for up to seven days. The patient complains of headache that typically involves the forehead and/or the back of the head and often irradiates to the neck. The headache gets worse within several hours and can become quite severe. It gets better when the woman lies down and worsens in the upright position. Bright light and loud sounds also worsen the headache. Other symptoms often accompanying post dural puncture headache are visual and hearing disturbances, as well as nausea and vomiting.

The reported incidence of post-dural headache varies. When spinal anesthesia introduced into clinical practice in 1898, about two thirds of all patients receiving it developed the headache. In modern times it is much less common. Among patients receiving spinal anesthetic for caesarean section the incidence of headache is between 0.3 and 3 percent, depending on the institution from which the data is coming. Teaching hospitals tend to have higher rates of complications due to many epidurals performed by trainees. In uncomplicated labor epidural the incidence is significantly lower. However, if inadvertent dural tap has been created during epidural placement the probability of a patient developing the headache is much higher, unless measures are taken to prevent it.

The mechanism of the headache following dural tap is not fully understood. In normal circumstances cerebrospinal fluid circulates within the space enclosed by the dura and cerebral ventricles at certain hydrostatic pressure within this system. If the dura is punctured a leak of cerebro-spinal fluid develops. In most cases, especially with small diameter of spinal needle, the puncture site closes and the circulation of the CSF resumes its normal course. However, if the puncture hole is large enough it does not spontaneously close, and the leakage of CSF continues. The decrease in CSF pressure is thought to cause the headache.

Previously it was believed that because of this decrease in pressure the force of gravity causes stretching of the cranial nerves – nerves originating directly from the brain – and this stretching is responsible for the symptoms. However, in a study where an MRI scan was done on patients with post dural spinal headache there was no visible displacement of the neural structures, and this theory has been abandoned.

Another factor contributing to the headache is the dilatation of the cerebral veins: while the hydrostatic pressure in the CSF falls, in the veins it remains unchanged, leading to the increase in their diameter.

It has also been theorized that during dural puncture small bubbles of air may enter the SCF and, in turn, contribute to the headache. Indeed, the results of one study indicated that headache has been more commonly observed in patients in whom epidural was performed using loss of resistance with air rather than normal saline, though the difference was small.

As the main cause of PDPH is the unsealed defect in the dura it is apparent that the most critical factors in prevention of this complication are the size and the type of the needle used for the spinal. The smaller the needle the more likely the defect in the dura inflicted by this needle is to spontaneously close and heal. Many studies confirm it, and it is currently accepted that spinal anesthesia must be performed by the thinnest needle possible. The sizes of spinal needles are expressed in gauges (G), the traditional unit of measurement of the diameter of the wire. The higher the gauge the smaller the needle. It is generally accepted to use 25G needles, and some anesthetists prefer even smaller 27G needles. Epidural needle sets for combined spinal-epidural block during labor also contain 27G.

The type of the spinal needle is also important. For any tissue damage to heal the body responds inflammatory reaction necessary for the process of healing. Generally tearing the tissue inflicts more inflammation response than cutting. Currently there are two type of spinal needles as determined by the shape of their tip: Whitaker and Quincke, named after their designers. Quincke needle is beveled and acts as a cutting tool when going through the dura. As the result the defect in the dura is a tidy cut that does not elicit much inflammation necessary for efficient healing. The tip of a Whitaker needle is shaped as the tip of a pencil and is relatively blunt. When pushed through the dura, it creates a small tears rather than cut, resulting in stronger inflammatory response, better healing process and reduces incidence of post-dural headache.

The size of epidural needle has to be of sufficient size so that epidural catheter can be passed through it, and needles currently used are 18 or 16 gauge. They must also be sharp to make it possible to pass them through the ligaments, as pushing a blunt needle of such large diameter would inflict too much tissue damage. Because of this, if the dura is accidentally damaged by the epidural needle the dural wound is relatively large and does not close as easily as that inflicted by a spinal needle. The CSF leak develops and the headache follows.

The technique is also important. The person doing the spinal or epidural must be well trained and follow basic safety rules. The patient must be adequately positioned. As labor progresses the pain gets worse, and it is easier – and potentially safer – to perform labor epidural at early stages of cervical dilatation, when the patient can sit still longer and is not overly distressed. At later stages of labor when the awareness of the anesthetist of excruciating pain and distress the woman is experiencing adds to the feeling of urgency and may lead to mistakes and possibly complications, including dural tap and resulting post-dural puncture headache.

On a technical note, some data suggests that when using Quincke spinal needles it is important to keep its bevel vertical, even with very small needles. The logic is that the needle advanced in this manner may cause less trauma to the ligaments (which fibers run in vertical direction), and better integrity of these ligaments help to close the damage inflicted by the needle to the dura.

Probably because the post-dural puncture headache gets worse in the upright position, for a long time it was believed that it can be prevented if the patient stays in bed for the first 24 hours after spinal block. Recent evidence does not support this. Bed rest does not prevent the development of a headache after uncomplicated spinal block. Moreover, reduced mobility after the caesarean may increase the chances of other complications, such as pneumonia and deep venous thombosis, or formation of blood clots in the veins of legs. The latter can lead to potentially fatal pulmonary embolism, and early mobilization – sitting and walking – is beneficial after surgery.

As already mentioned, incidental dural puncture by epidural needle during epidural placement for labor is almost a guarantee of post-dural puncture headache, simply because the defect inflicted by large bore needle in the dura is relatively large. Older guidelines recommended that in case of incidental dural tap epidural needle should be taken out and the procedure repeated at higher or lower level. However recently new strategy has been recommended that reduces the chances of PDPH in such scenarios. Current evidence suggests that the best action in such situation is to pass the epidural catheter through the needle into the subarachnoid space and establish analgesia using low amounts of local anesthetic mixture. Foreign body, in this case epidural catheter, placed in human tissue causes inflammatory reaction, which facilitates the healing of dura and significantly reduces chances of PDPH.

This has been demonstrated in recent study. If incidental dural puncture occurred the patients were treated in one of the following ways. In control group the epidural needle was removed and the catheter re-cited at different level, as per old recommendations. In the second group epidural catheter was inserted through the dura, used for pain relief and removed immediately after delivery. Finally, in the third group of patients the catheter was also inserted through the dura and used for pain relief, but was only removed twenty four hours after delivery. The incidence of PDPH was 91.1% in the first group, 51.4% in the second and only 6.2% in the third. In other words, placing the catheter in subdural space and keeping it there for 24 hours reduced the incidence of PDPH almost 15-fold. Obviously, reduced doses of local anesthetics and opioids are used when using subarachnoid catheter for pain relief in this clinical scenario.

The treatment of PDPH is depends on the severity. During the first twenty four hours it is conservative. Bed rest is recommended. It helps lower the pressure of the CSF and reduce the severity of the headache, as well as facilitate healing of the dural defect. Increased intake of fluids is recommended, either oral or intravenous, in order to facilitate the production of the CSF and compensate for CSF losses throught the defect. The main symptom, headache, is controlled by medication: paracetamol (acetaminophen), anti-inflammatory drugs and, in more severe cases, narcotics such as morphine. Cafeine has been shown to reduce the severity of the headache, and the patient is encourage to increase its intake via caffeinated drinks, coffee, tea or coca-cola.

One of the factors thought to contribute to the severity of the headache is the dilation of the cerebral veins. Similar mechanism is at least partially responsible for headaches associated with migraines, and drugs used to treat migraines have been successfully used to treat PDPH.

In most patients conservative treatment is sufficient for getting rid of PDPH during the first 24 hours of its onset. If the symptoms persist longer that 24 hours epidural blood patch is recommended.

Obstetric patients are not those who most commonly develop post-dural puncture headache. PDPH is much commonly seen in patient who undergo lumbar puncture as the test for making a neurological diagnosis, such as meningitis. Spinal needles used for neuraxial blocks in obstetric anesthesia are of small diameter, 25 – 27 gauge. In contrast, spinal needles used for diagnostic lumbar puncture are generally larger, 20 – 22 G. Moreover, many patients in whom neurological disease is suspected already have a headache, often severe, and PDPH among them may be overlooked. In 1960 one Dr Gormley noticed that if blood was present in the CSF obtained during diagnostic lumbar puncture, the chance of developing a headache was lower. He reasoned that blood may act as the sealing material of the puncture site. The theory was tested: in seven patients who developed the headache after lumbar puncture (one of the patients was the doctor himself) injecting 2 to 3 milliliters of their own blood at the site of lumbar puncture relieved the headache. The technique of the epidural blood patch (EBP) was born. It was refined over the next decades, and in 1980 the methodology that is currently in use was finalized.

Blood patch is used in cases where conservative treatment has not been successful. The procedure is usually performed in the operating theatre in sterile conditions and requires an anesthetist and an assistant, both scrubbed and dressed in sterile surgical gowns. The patient is placed in the same position as for performing epidural block, sitting or on the side, and epidural needle is inserted either at the site of previous injection or at one level below or above. As soon as the epidural space is identified, the assistant takes 20 milliliters of blood from patient’s vein. This blood is then injected into the epidural space. Full amount is injected through the epidural needle, unless the patient complains of back pain, in which case the injection is stopped.

Epidural blood patch is technically simple and very efficient, and its success rate in treating the headache is close to 100 percent. Many patients experience complete relief from headache during the actual injection of blood. Complications are rare. As blood is fertile medium for growing microorganisms, following strictly sterile technique is important in order to prevent septic complications, such as epidural abscess or meningitis. Some anesthetists advocate prophylactic use of antibiotics. The inadvertent spinal injection of blood during blood patch has been reported in the literature, with the symptoms similar to meningitis. Fortunately, this is rare, and all patients recovered without special intervention or long-term consequences.

The time frame for performing epidural patch after the onset of PDPH is still debated. The proponents of performing blood patch early point out that because it is so efficient and virtually guarantees the relief it should be done straight away after the diagnosis of PDPH has been made. On the other hand, clinical evidence does not support early intervention, and it has been demonstrated that early EBP has considerably lower success rate. On the other hand, delaying EBP does not provide benefits either, and it seems that 24 hours is optimal time for conservative treatment.

There have also been debates regarding prophylactic EBP in cases of inadvertent dural puncture. Once again, its efficiency is not supported by clinical data, and prophylactic EBP does not seem to reduce the incidence of PDPH.

Post-dural puncture headache can be fairly distressing complication of the epidural in labor. However, the incidence of PDPH in experienced hands is low. Conservative treatment of PDPH is effective in most patients. In the remaining proportion of patients epidural blood patch is required, and it is effective and associated with low rate of complications.


1. Thew M, Paech MJ. Management of postdural puncture headache in the obstetric patient. Curr Opin Anaesthesiol. 2008, 21: 288–292.



Dr. Eugene Smetannikov is a practicing anesthesiologist with the interest in obstetric anesthesia. He is the author of the most comprehensive book on the subject, The Truth About Labor Epidural