Epidural Blood Patch on the Resolution of Postdural Puncture Headache (PDPH)

Accidental dural puncture is the most common (rate 1 in 50-250 in most obstetric hospitals) and, arguably, important complication of epidural insertion. After breach of the dura with a 16-18 gauge epidural needle in adult patients aged 18-60 years, the incidence of post-dural puncture headache (PDPH) is 80-90%. This headache is frequently severe or incapacitating, markedly postural and of at least several days duration. Associated symptoms include nausea and vomiting, auditory disturbance and ocular symptoms. PDPH often interferes with maternal -infant interaction by restricting or preventing ambulation and in some cases it confines the new mother to bed in a supine position. It is a significant cause of increased nursing and anaesthetic staff workload and of prolonged hospitalisation. Accidental dural puncture and PDPH may also rarely also be associated with other morbidity, such as cranial nerve palsy (III, IV, V, VI, VII, VIII) pain and dysaesthesia; and subdural haematoma. Fatalities have been reported following brainstem "coning" after prolonged low intracranial pressure headache. Untreated, the headache may become chronic and persist for months or even years. The mechanism of PDPH is not clearly understood, although it is consistent with low intracranial pressure headache. Although imaging studies support that the precipitating event is loss of cerebrospinal fluid (CSF), the volume of loss is not directly related to the risk of headache. The change in intracranial CSF dynamics, with reduction of pressure, is considered more important. Reduction in CSF pressure may be influenced by a number of factors, including the volume of the lumbar subarachnoid CSF compartment, the rate of CSF loss and replenishment and the movement of CSF associated with patient positioning. The rate of change of intracranial CSF pressure may be more relevant than the specific pressure level, explaining the sudden onset of headache (mediated by vasodilation) when the patient assumes the upright position and hydrostatic forces redistribute intracranial CSF to the spinal CSF compartment. Of procedural factors that are associated with failure of EBP, it is not clear whether the volume of blood injected is relevant. Despite 40 years of clinical use, it remains uncertain as to how much blood to inject and practice varies. An apparent relationship between volume and efficacy has been mooted, based mainly on the historical features of EBP. The injection of autologous blood was first suggested by Gormley in 1960, inspired by his impression (since shown to be erroneous) that dural puncture accompanied by "blood tap" was less likely to result in headache. He claimed 100% cure with 2-3 ml of blood injected into 8 patients. In 1970 and 1972 DiGiovanni et al described two series of 108 patients in total, for whom a larger volume of 5-10 ml of blood lead to permanent cure of 90%. In 1974, an average volume of 10 ml was said to have cured 182 of 185 patients, with volumes less than 10 ml appearing to be less successful. Taivainen et al could not detect any advantage with 15 ml versus 10 ml, the volume being chosen according to height of the patient. Crawford described his experience with EBP in 1980. He started with 6-15 ml of blood in 16 patients and reported a recurrence of PDPH within 2 days in a third of patients. A repeat EBP with 20 ml cured all 4 of these, leading him to change to routine administration of 20 ml. The next 100 patients received 17-20 ml, with 'total success' claimed in 80 of 83 given 20 ml. However, it is now known that such a high success rate may have reflected an inadequate duration and quality of follow-up. Crawford reviewed further experience with EBP in 1985, claiming that complete and permanent relief was achieved in 91% when 20 ml of blood was injected and 83% if lesser volumes were used. Crawford's influence was such that, during the past 20 years, this recommendation to use up to 20 ml of autologous blood has been established practice. Further support for the use of larger volumes of blood came from a series by Brownridge. He found that 10 ml or less of blood produced permanent relief in only 75%. In addition a clinical and imaging study by Szeinfeld et al. reported that an average of about 15 ml (range 12-18 ml) was successful and resulted in mean segmental spread of blood over 9 (range 7-14) spinal segments. In vitro, EBP is capable of tamponading up to 40 mm Hg of pressure after dural puncture with small spinal needles, although is less effective when a hole is made with a Tuohy needle. The mass effect of EBP disappears after a few hours, although focal clot may remain adherent to the dura within the subarachnoid space.