Successful treatment of post-dural-puncture headache with surgical dura repair two years after spinal anesthesia

Abstract

Background

Post-dural-puncture headache (PDPH) can be a highly disturbing complication of elective spinal anesthesia. The incidence of PDPH when small needles are used is estimated to be 0%–14.5%. PDPH usually resolves spontaneously within a few days, but there are rare cases that persist in spite of conventional and epidural blood patch therapy.

Case

A 59-year-old man suffered from persistent postural headache for more than two years after an otherwise uneventful spinal anesthesia. Conventional and minimally invasive treatments were unsuccessful.

Intervention/outcome

A neurosurgical procedure was performed, during which a dural leak was identified and repaired. The patient was immediately pain free after surgery and has remained so for the duration of follow-up (more than one year).

Introduction

Post-dural-puncture headache (PDPH) can be a highly disturbing complication of elective spinal anesthesia. PDPH was first described in 1898; in the early days of spinal anesthesia, as many as 66% of patients developed PDPH (1). Its incidence was greatly reduced by advances in needle design and size. Today, the incidence of PDPH when a 25-gauge pencil-point (Whitacre) or smaller needle is used is estimated to be 0%–14.5%. The reported risk factors for PDPH include not only the size and design of the spinal needle, but also the experience of the person performing the puncture and the age and sex of the patient. PDPH usually resolves spontaneously within a few days (2–4).

It is now generally accepted that PDPH is due to excessive leakage of cerebrospinal fluid (CSF) after dural puncture.

Case history

The patient, a 59-year-old man, had previously undergone surgery for herniated discs at L3/4 and L4/5. In 2008, he underwent elective orthopedic surgery on a foot in an outside hospital: Spinal anesthesia (bupivacaine) was administered by an experienced senior consultant in anesthesiology using a 25-gauge Whitacre needle inserted laterally from the left side at L3/4. The puncture was documented as uneventful and successful on the first attempt with good clinical effect. Three days later, the patient began to complain of severe occipital and frontal headache that worsened on sitting or standing. Infection at the puncture site was clinically excluded, and a first blood patch was performed 17 days after surgery. The headache persisted, as it did after a second blood patch, performed seven weeks after surgery. For two years thereafter, the patient suffered from ongoing postural and activity-dependent headache, with inadequate relief from treatment with nonsteroidal anti-inflammatory drugs as prescribed by his family doctor. The headache gradually became more severe and interfered with his everyday activities. The analgesics that were given with inadequate effect included paracetamol, ibuprofen, metamizol, tramadol, and oral and intravenous opioids.

Upon initial examination by our anesthesia pain service in 2010, the patient described his pain as throbbing, pressing, and present in the whole head. The pain was clearly worse on sitting, standing, or intense physical activity. The general physical examination and the neurological examination were normal, as were all routine blood tests. A head computed tomography (CT) scan was normal, and spinal magnetic resonance imaging (MRI) and myelographic CT revealed no pathological findings. For both diagnostic and therapeutic purposes, a third epidural blood patch was performed at the level of the suspected dural leak with 14 ml of blood under fluoroscopic guidance. Correct spreading was confirmed with the use of 2 ml of a contrast agent (Iopamiro®). This third blood-patch again brought no relief of pain.

The headaches worsened further: They began to be present even when the patient was supine, and nearly unbearable when he was upright. He was admitted to our hospital and an interdisciplinary pain board recommended performing radionuclide cisternography. This study was performed via lumbar puncture at L2/3, with a documented opening pressure of 70 mm H₂0. Evidence of a CSF leak was found at L3/4 on the left side (Figure 1), which was the documented site where the spinal anesthesia had been administered two years earlier. The patient gave his informed consent to a neurosurgical procedure. Under general anesthesia and in the prone position, the left L3/4 interlaminar space was fenestrated for exposure of the dura mater; a 2-mm hole in the dura was identified and sutured under the operating microscope (Figure 2). The headaches resolved immediately after surgery; the patient was rapidly mobilized from bed, without orders for bed rest or any further precautions, and he was discharged from the hospital one week later without any analgesic medication. He remained free of headaches and off medications at his first postoperative follow-up three months later and at all subsequent follow-up visits, the last one in 2012, 23 months after the dural repair procedure.

Figure 1.

Radionuclide cisternography, November 2, 2010: scintigraphic image one hour and 10 minutes after the injection of 34 mBq In-111-DTPA through a lumbar puncture at L2/3. Lower activity in the dural sac is noted at levels L3 through L5 (*→), corresponding to the known spinal canal stenosis at these levels (previously demonstrated by magnetic resonance imaging (MRI)). Radionuclide activity is seen to the left of the dural sac (picture at right), at a level estimated to lie between the laminae of L3 and L4 (**→). A leak was found at this level during surgery.

Figure 2.

View of the operative field under the microscope after fenestration at L3/4 on the left side (orientation of the picture: left = cranial, right = caudal). There is a 2 mm hole in the dura mater (*→), through which the arachnoid is visible. A metal suction tip is seen to the right of the dural hole (**→).

Discussion

Persistent PDPH after spinal anesthesia is a widely recognized problem (5). Multiple case reports have shown that it can last for years in rare cases. The case presented here documents the usefulness of addressing persistent PDPH with an interdisciplinary approach, with the participation of anesthesiologists, pain therapists, neurologists, and neurosurgeons. This is not the first case of successful neurosurgical treatment of PDPH; a relevant case report appeared in 1982 (6). In that case, as in ours, a neurosurgical procedure was considered the last available treatment option and led to the successful resolution of PDPH despite the long interval before surgery.

In our case, we suspected from the outset that a leaking hole in the dura mater would be found at L3/4 on the left side, as this was the site where the spinal anesthesia had been performed two years earlier. The poor results of multiple blood patches made it seem advisable, however, to confirm the site of leakage with a radionuclide study before operating, as we were reluctant to take the risk of a negative exploration or to make the surgical exposure any larger than necessary. Cisternography did, in fact, indicate leakage at the expected site, as was subsequently found at surgery. No more than an interlaminar fenestration at L3/4 was needed to expose and repair the leak. Preoperative leakage tracing would clearly have been even more important in this case, indeed indispensable, if the site of the earlier procedure had been unknown or inadequately documented.

Conclusion

Even today, there are patients who suffer from PDPH for months or years before receiving effective treatment. Physicians, and particularly those who perform dural punctures, should be aware that neurosurgical procedures can help as a last resort in the rare cases where less invasive treatments fail.

Clinical implications

Suffering from persistent post-dural headache for a long time after an otherwise uneventful spinal anesthesia is possible.

Conventional and minimally invasive treatments can be unsuccessful in these cases.

A neurosurgical identification of the dural leak and its repair was successful in our case.