Cephalad spread of epidural blockade in a 15‐degree head‐down position

A fit ASA grade 1 82-year-old woman presented for an ultra low anterior resection. She had no significant premorbid conditions and was fit for elective surgery. She was premedicated with temazepam 10 mg. On arrival in the operating theatre, she had a 16-gauge intravenous cannula sited in her right forearm and, in the sitting position, had an epidural catheter sited at the T9/10 level. The depth to the epidural space was 4 cm and 5 cm of catheter was left in the space. Anaesthesia was induced with a combination of fentanyl 100 μg, midazolam 2 mg and propofol 80 mg. She was then given vecuronium 8 mg and intubated with an 8.0-mm tracheal tube after 2 min. Right internal jugular and radial arterial lines were then sited. A test dose of 3 ml of bupivacaine 0.25% plus 1 : 200 000 epinephrine was given and, 10 min later, epidural block was established with 10 ml of bupivacaine 0.25%. In the operating theatre, she was placed in the Lloyd Davis position and tilted 15 degrees head down at the request of the surgeon. Anaesthesia was maintained with 0.5–1% isoflurane in 70% nitrous oxide/30% oxygen. She was ventilated to maintain an end-tidal carbon dioxide of 4–4.5 kPa. Body temperature was maintained with a forced air warming system and monitored with a nasopharyngeal temperature probe. The operation lasted 4.5 h. After 1.5 h, an infusion of bupivacaine 0.1% with fentanyl 2 μg.ml−1 was started at a rate of 10 ml.h−1. The patient remained stable throughout the operation and received no further opioids or neuromuscular blocking agents. At the end of the operation, she was ventilated with 100% oxygen and allowed to wake up. Despite being responsive and obviously awake, her respiratory effort was poor and her tidal volumes were 100–150 ml as measured with a Wright's respirometer. Neuromuscular blockade was assessed with double burst stimulation and there was no significant block detected. Her core temperature was 36.2 °C. Her respiratory effort appeared inadequate and a blood gas revealed a Paco2 of 10.1 kPa. She was given doxapram 50 mg with no effect and so was ventilated in recovery and reassessed. It became apparent that she was fully awake and aware but had poor respiratory effort. She was also noted to have grade 3+/5 power in her legs but only grade 1/5 power in her arms. She was well analgesed but it was difficult to assess the height of the sensory block as she was still intubated. The presumed diagnosis of a high epidural block was made. The epidural infusion was stopped and she was sedated with sevoflurane and ventilated for a further 2 h, after which the sevoflurane was stopped and she was reassessed. At this stage, she had grade 4/5 power in her arms and her legs and had tidal volumes of 300–400 ml. Blood gases at this stage showed Paco2 of 6.1 kPa and PaO2 of 16.2 kPa on 40% oxygen. She was extubated and the epidural infusion was restarted. After a further 4 h in recovery, she was sent back to the ward. The rest of her recovery was uneventful. One can only assume that the cause for her respiratory failure was cephalad spread of the epidural block caused by the infusion running in a 15 degree head-down position. I have used this technique in other patients without problems and would be interested to hear if anyone else has experienced a similar problem as I can find no cases reported in the literature. There have been many studies trying to establish the effect of gravity and posture on spread of epidural anaesthesia. Grundy reported that epidural anaesthesia induced in the lateral position appeared 2 min earlier, spread two segments higher and lasted 75 min longer [1]. Hodgkinson documented that keeping patients in the sitting position for 5 min after injection of 20 ml of bupivacaine 0.75% limited cephalad spread but only in obese patients [2]. Apostolou demonstrated slightly higher and slightly longer block on the dependent side when epidural anaesthesia was induced in the lateral position, although he felt these differences were of only minor significance [3]. In 1993, Ponhold documented a decrease in maximum cephalad spread of L1 vs. T10 when epidural anaesthesia was induced with patients who had 30 degrees trunk elevation [4]. However, other investigators [5678] have reported no significant difference in the eventual spread and distribution of block regardless of position. I can find no studies documenting the effects of infusions in the epidural space in various positions. It may well be that with bolus injections of local anaesthetic, the pressure generated by the injection is such that the effect of posture is relatively small. However, one could suppose that with an infusion running at a low rate, the pressure generated at the tip of the epidural catheter is lower and hence the relative effects of posture and gravity would be much more. I am not sure how this problem could be avoided. Siting the epidural lower would mean that cephalad spread would be less likely to affect respiratory function although cephalad spread is probably not predictable and postoperatively a higher infusion rate may be required to achieve adequate analgesia. Alternatively, giving boluses of local anaesthetic via the epidural catheter rather than an infusion may encourage more even spread around the tip of the catheter as discussed above. This would appear to be a rare but important complication of epidural use in the head-down position and should be considered if such a patient presents with respiratory difficulty postoperatively.