To the Editors:—

A 20-g epidural catheter was uneventfully placed 4 cm into the epidural space at L3–4 in a patient in active labor (Perifix Continuous Anesthesia kit; B. Braun Medical, Bethlehem, PA). After the initial dose, local anesthetic was continuously infused using a syringe pump, through a 0.2-μm epidural filter.

During the course of the patient's 21 h of labor, she required several syringe refills and additional boluses of local anesthetic for breakthrough pain. Approximately 1 h after the fourth bolus had been given and the third 60-ml syringe had been connected, the anesthesiologist was informed that the epidural pump was signaling an “occlusion.” Although it was still possible to inject normal saline through the catheter, it was now requiring significantly more force than when the catheter had been initially placed. The force required for injection did not alter with a change in the patient's position or after the catheter had been pulled back 1 cm. However, after the epidural filter was removed from the infusion line, the force necessary to successfully inject was noted to have returned to its initial level. With the epidural filter removed, the epidural infusion was restarted. There were no further problems, and the patient completed her labor and delivery without incident.

Subsequently, the epidural filter was examined closely; no defect was grossly visible. No debris was seen inside except for what appeared to be a small amount of entrapped air. Although the small bubbles may have been insignificant individually, it is hypothesized the accumulated trapped bubbles reached a sufficient volume to obstruct the filter and the flow of infusate.

To test this hypothesis, air was deliberately introduced into an epidural filter after it was primed, and measurements were taken of the amount of air necessary to cause obstruction at an infusion rate of 15 cc per hour. As little as 0.3 ml of air in the Braun/Periflex epidural filter was sufficient to cause the infusion pump to signal obstruction. This air-lock phenomenon was also possible in another brand of epidural filter (Arrow International, Reading, PA, 0.7 ml air to obstruction). The amounts of air needed to obstruct the filters correspond to their priming volumes (Braun/Periflex 0.35 ml; Arrow 0.75 ml).

Complete or partial obstruction to epidural injections and infusions have been attributed to catheter-related problems such as kinking, knotting, blood clotting, or stretching, 1and to manufacturing defects in the catheter, the filter, 2and the connection screw-cap. 3 

Khalouf FK, Kunkel FA, Freeman J: Stretching with obstruction of an epidural catheter (letter). Anesth Analg 1987; 66: 1202–3
Clark VA, Dickson J: Blocked epidural filter (letter). A naesthesia 1993; 48: 537
Chandhok D, Vijayakumar E: An unusual case of epidural catheter obstruction (letter). A nesthesiology 1999; 91: 895–6