Failed spinal anaesthesia after intrathecal chemotherapy

EDITOR: We report the case of a 60-yr-old male suffering from meningeosis lymphomatosa with subsequent facial nerve palsy. Within 2 weeks, he received five intrathecal injections for chemotherapy, each injection consisting of methotrexate 15 mg, alexan 40 mg and fortecortin 4 mg. One week later, the patient developed a scrotal swelling of unknown origin, and therefore had to undergo an elective orchiectomy. Since the preanaesthetic examination revealed a patient without any obvious neurological symptoms, he was scheduled for spinal anaesthesia. Using a 27-G Whitacre® needle (Becton Dickinson, Barcelona, Spain), the subarachnoid space was easily punctured in the sitting position (L3-4 interspace). The stylet was removed and correct needle placement was confirmed by free flow of cerebrospinal fluid (CSF) into the hub of the needle. The syringe containing the predetermined dose of local anaesthetic (hyperbaric bupivacaine 0.5% 2.8 mL) was connected to the needle. Aspiration of CSF into the syringe was seen to have a birefringent effect, reconfirming correct needle placement. A reaspiration test was made after injection of half the dose of bupivacaine and again at the end of injection to confirm that the point of the needle was still within the subarachnoid space; the needle was then removed. Despite this uneventful approach, the patient showed no signs of an effective spinal block. He only described a patchy paraesthesia around his bottom after an interval of 30 min. We then decided to repeat the spinal anaesthetic. As on the previous occasion, we easily punctured the subarachnoid space using a 27-G Whitacre® needle (L2-3 interspace). Again, a back-flow of CSF was immediately observed, and we administered hyperbaric bupivacaine (0.5% 3.0 mL). Again, we confirmed that we could aspirate fluid, after half of the predetermined dose of local anaesthetic had been injected and again after we had completed the injection to confirm that the needle tip was in the correct position in the subarachnoid space. Despite a total dose of hyperbaric bupivacaine 5.8 mL given intrathecally, the patient did not develop a spinal block. We therefore had to induce general anaesthesia. Afterwards, in the recovery room, the patient, who had been extubated, still demonstrated no signs of a spinal block. He was then transferred to an Intermediate Care Unit overnight to guarantee adequate therapy just in case he developed late-onset high spinal anaesthesia. Spinal anaesthesia is one of the basic techniques in the arsenal of modern regional anaesthesia. Although it is generally easy to perform, a failure rate of about 5% has been reported [1]. However, this rate includes all cases: either where the anaesthetist was unable to locate the subarachnoid space, or where the spinal needle was dislodged from the space during injection of the local anaesthetic, as indicated by the inability to obtain free flow of CSF during or following the procedure [1,2]. If, as in our patient, free flow of CSF can be obtained before, during and after injection of the local anaesthetic, failure of spinal anaesthesia is extremely rare. A few reasons for such failure have been suggested: Render reported that the aperture of the Whitacre needle (0.8 mm) was long enough for the distal part to penetrate the dura mater, while the proximal part still remained in the epidural space [3]. Such a position of the needle point would permit the aspiration of CSF, although the local anaesthetic may, in part, be injected into the epidural space. In our case, this is very unlikely to be reason for failure because we punctured the subarachnoid space twice. Ben-David and colleagues published a case of failed spinal anaesthesia where the local anaesthetic left the spinal needle through a proximal hole [4]. Again, since we used two different needles, such a rare case of malmanufacturing is an extremely unlikely explanation. Since all previously published reasons for failed spinal anaesthetics can be excluded in our patient, and since we twice obtained a free flow of CSF before, during and after the injection of bupivacaine, we suggest that the previously administered chemotherapy - into the subarachnoid space - may be the reason for the ineffective block. Aggressiveness is an imminent ingredient of any chemotherapy regimen, and it may well have been that a non-infectious, chemotherapy-induced intrathecal inflammation - causing either the generation of septal walls, or an increased tissue barrier (e.g. increased thickness of the perineurium or generation of surrounding scarring tissue) - prevented the local anaesthetic from diffusing into the nerve roots. Such histological changes have repeatedly been reported in patients receiving methotrexate intrathecally [5,6]. Since these morphological changes were not related to any clinical symptoms, even the prophylactic treatment of non-Hodgkin lymphomas with intrathecal methotrexate is a clinical routine. The anaesthetist should therefore be aware of this potential problem when performing spinal anaesthesia in patients previously treated with intrathecal chemotherapy. M. Westphal T. Götz M. Booke Department of Anaesthesiology and Intensive Care, University of Münster, Münster, Germany