Unintentional dural puncture. A survey of recognition and management.

A survey of the recognition and management of 21 unintentional dural punctures that occurred in a series of 3500 obstetric epidurals is presented. In seven cases, dural puncture was not recognised at the time of occurrence and two patients may have subsequently received mixed epidural and subarachnoid analgesia. In two patients with dural puncture in whom air had been used to locate the epidural space, cerebrospinal fluid did not drip from the hub of the needle. A test dose did not reveal incorrect catheter placement in seven patients. The provision of an epidural infusion of Hartmann's solution for 24 hours, together with bed rest, appeared to delay the onset of dural puncture headache rather than prevent it entirely. Blood patching was required only in five of the 21 patients but was entirely successful. The implications of these observations in relationship to obstetric epidural practice are discussed.