Abstract

Recurrence or exacerbation of phantom limb pain induced by regional anesthesia in cluding spinal anesthesia, epidural anesthesia, and peripheral nerve block has been described in a few reports. This is a rare phenomenon, but it can occur in any am putee with or without a history of previous phantom limb pain. We describe a case whose phan tom pain of the amputated limb stump was twice induced by spinal anesthesia during two consecutive surgeries in the contralateral lower limb. It was revealed that midazolam was successful in treating this rare phantom limb pain after spinal anesthesia. Here, we discuss the management of phantom limb pain during spinal anesthesia and the anesthetic management for subsequent surgery in patients with previous spinal anesthesia-induced phantom limb pain.

Keywords

amputation; anesthesia, spinal; midazolam; pain; phantom limb;

1. Introduction

There have been several reports about phantom limb pain induced by neuraxial anesthesia in patients with previous lower-limb amputation.¹⁻⁹ In these reports, the effectiveness of different therapies was an important issue for discussion. We describe a case whose phantom pain of the amputated stump of a lower limb was twice induced by spinal anesthesia for two consecutive surgeries of the contralateral lower limb, and the pain was treated successfully with midazolam. The management of phantom pain during spi nal anesthesia and the choice of anesthesia for subsequent surgery are discussed in this report.

2. Case Report

A 68-year-old man, 63 kg in weight and 161 cm in height, suffered from peripheral artery occlusive disorder (PAOD) for which he was scheduled for belowknee amputation of the affected left leg under spinal anesthesia. The patient had a history of diabetes mellitus and hypertension for more than 20 years and had undergone a below-knee amputation of the right leg as a result of PAOD 10 years previously. Preoperatively, electrocardiography (ECG) disclosed a normal sinus rhythm and chest X-ray showed clear lungs without specific findings. Labo ratory data showed that his hemoglobin level was 8.3 g/dL, and all other parameters were within normal ranges.

Upon arriving in the operating room, standard monitors including ECG, pulse oximeter, and noninvasive blood pressure device were set up for use. In the left lateral position, using a 25-gauge needle, spinal anesthesia was induced at the L3−4 interspace with 10 mg bupivacaine. The patient was then placed in the supine position and preparation for the surgical procedure was begun. Ten minutes later, the patient started to complain of continuous severe tingling and burning sensations in his right amputated stump over an area about 2−3 cm below the cut end. He stated that he had never experienced phantom sensations or pain of his right foot before. The pain intensity was rated as 10 by visual analog scale (VAS) (0, no pain; 10, worst pain imaginable). Loss of sensation to cold, as tested, was up to the level of T10. He could not move either leg. Intravenous midazolam (3 mg) was immediately administered, but did not completely relieve the pain. Ten minutes after midazolam administration, he described the pain as a VAS score of 5. In an attempt to completely relieve his phantom limb pain, another 2 mg of midazolam was given and he became sedated. When he awakened about 15 minutes after the second dose of midazolam, the phantom limb pain had completely disappeared and did not recur. The operation lasted for 90 minutes, and two units of packed red blood cells were transfused because of postoperative anemia in spite of minimal blood loss of 150 mL.

He was sent to the postanesthesia care unit (PACU) where he stayed until complete restoration of sensory and motor functions of the legs. He was returned to the ordinary ward for further care. We visited him on the second postoperative day, and he no longer felt phantom pain.

Three weeks later, the patient was scheduled for debridement of the amputated stump of the left leg. After discussing the anesthetic options with the patient, we again performed spinal anesthesia with 10 mg bupivacaine, with which he complied. Unfortunately, 10 minutes after bupivacaine was administered, the phantom limb pain of his right leg occurred again. The character and location of the pain were the same as in the last operation 3 weeks previously. He described that the pain intensity was half as severe as on the previous occasion. Intravenous administration of 3 mg midazolam immediately relieved the pain with a VAS score ≤ 1 and no additional dose was necessary. The operation was completed in 1 hour. He was sent to the PACU and had an uneventful recovery. There was no recurrence of the phantom limb pain when we visited him on the second postoperative day.

3. Discussion

Several reports have highlighted the recurrence or exacerbation of phantom limb pain induced by spinal anesthesia in patients with previous lower-extremity amputations.¹⁻⁷ However, this phenomenon is rarely encountered in our clinical anesthesia practice.

Our report describes a patient who experienced two consecutive episodes of phantom limb pain in the stump of the amputated right leg after receiving spinal anesthesia for two consecutive operations of the left leg, one for below-knee amputation and one for stump debridement. We found that midazolam could successfully treat this phantom limb pain induced by spinal anesthesia.

The mechanism of phantom limb pain induced by spinal anesthesia has not been clarified completely. Melzack proposed a theory, the gist of which is that a portion of the brain stem reticular formation could exert a tonic inhibitory influence on transmission at all synaptic levels of the somatic sensory system. He explained that the occurrence of phantom pain after regional anesthesia could result from a decrease in the tonic inhibitory influence exerted by the brainstem reticular formation. Complete loss of sensory input after subarachnoid block may decrease the level of inhibition and may increase self-sustaining neural activity. Such pain may be generated from active spinal cord cells that are released from inhibitory control through the loss of afferent impulses.¹⁰

Besides spinal anesthesia, nerve block can also induce phantom limb pain.^11,12 Lee and Donovan described a patient who experienced reactivation of severe phantom limb pain in the PACU after combined interscalene brachial plexus block and general anesthesia.¹¹ This severe phantom limb pain was successfully treated with a small dose of intravenous lidocaine. The authors concluded that in patients with previous upper-extremity phantom pain, the anesthesiologist should be alert to the possibility of reactivation of the pain if brachial plexus block is performed. In another report, Martin et al presented a case in whom reactivation of severe phantom leg pain occurred after lumbar plexus block; the pain was treated successfully with a sciatic nerve block.¹² These two cases highlight the influence of peripheral nerves on the altered central processing that underlies the reactivation of phantom limb pain.

Medical treatment is the first choice when phantom limb pain is induced by spinal anesthesia. Koyama et al reported three cases of phantom limb pain following spinal anesthesia and thought that intravenous thiopental should be tried immediately.³

Wajima et al described a 48-year-old male patient with spinal cord injury-induced neuropathic pain who showed newly developed severe lightning pain after a therapeutic subarachnoid block.¹³ The authors performed subarachnoid block 16 times in this patient and tried to determine which drug was best for treating this neuropathic pain. The drugs they used included thiopental, benzodiazepines, opioids, ketamine, lidocaine, droperidol, and sevoflurane inhalation. Intravenous thiopental stopped the pain quickly at a subanesthetic dose (1 mg/kg), and there was no adverse effect.¹³ Mackenzie stated that intravenous administration of 10 mg diazepam, 100 μg or 150 μg fentanyl, and nitrous oxide/oxygen could relieve recurring phantom pain effectively in two patients following spinal anesthesia.⁵ In our case, midazolam was effective for relief of the pain. Both barbiturates and benzodiazepines facilitate the inhibition of the gamma aminobutyric acid receptor in the central nervous system. These drugs can suppress the central nervous system and have been thought to provide relief of pain induced by spinal anesthesia. Apart from medical treatments, application of transcutaneous electrical nerve stim ulation (TENS) can also provide excellent effects during the postoperative period.^4,9 Bulder and Smelt described one patient who experienced phantom limb pain during the onset and regression of epidural anesthesia.⁹ TENS was applied paravertebrally (L3−5) during postoperative regression of epidural anesthesia, and a definite regression of the phantom pain was noted. Therefore, TENS is a non-medicinal option that may be considered during the postoperative period.

Interestingly, our patient stated that he had never experienced phantom sensations and limb pain after his previous below-knee amputation of the right leg. Most of the reported cases did suffer phantom limb pain of differing intensity and durations after lower-limb amputation.³⁻⁶ In these reports, one patient suffered from severe phantom pain once a month³ and others were pain-free.³⁻⁶ The durations of the pain-free state in these patients varied, ranging from 2 weeks to 40 years.

This implies that neither the occurrence of previous phantom pain nor the duration of the pain-free state afterwards has any effect on the occurrence or reactivation of spinal anesthesia-induced phantom limb pain in the amputees. Tessler and Kleiman designed a prospective study to determine the incidence of phantom limb pain during spinal anesthesia and to define its predisposing factors.¹⁴ They studied 23 administrations of spinal anesthesia in 17 patients with previous lower-limb amputations. How ever, only one patient developed significant phantom limb pain during spinal anesthesia. Because of this low incidence of spinal anesthesia-induced phantom limb pain and the limited number of cases, they were unable to define any predisposing factors.

Some reports have suggested that a previous history of phantom pain after lower-extremity amputation would be a relative contraindication for spinal anesthesia.⁴⁻⁷ However, consistent with Wajima et al’s comment,¹³ we suggest that spinal anesthesia should not be contraindicated in these amputees because spinal anesthesia-induced phantom limb pain is a comparatively rare phenomenon in amputees, and there are drugs that can treat the pain successfully should it occur. In addition, the phantom pain will disappear without any sequelae after recovery from spinal anesthesia. Furthermore, spinal anesthesia might have greater benefits than general anesthesia in some of these patients. Therefore, we suggest that spinal anesthesia is not contraindicated on the basis of a previous history of phantom pain after lower-extremity amputation.

Our patient suffered from two consecutive episodes of phantom limb pain following spinal anesthesia in 1 month. Uncles et al also reported a patient who underwent two cesarean sections 5 years apart under epidural anesthesia.⁸ Phantom limb pain developed upon regression of the epidural block in these two operations. This suggests that an amputee who has suffered neuraxial anesthesia-induced phantom limb pain in the past is at higher risk of recurrent phan tom lower limb pain if neuraxial anesthesia is performed again. Therefore, communication is necessary between the anesthesiologists and such patients regarding previous experiences of spinal anesthesia-induced phantom limb pain and the different methods of anesthesia available. In this case, we discussed the potential recurrence of phantom limb pain and different methods of anesthesia with the patient when the second operation was scheduled. He decided to receive spinal anesthesia again, and requested that we manage the pain as early as we could.

In summary, with the increase in the number of amputees, it is not uncommon for this kind of patients to one day require some operations under spinal anesthesia. The anesthesiologist should be aware of the possibility of spinal anesthesia-induced phantom pain in such patients and be ready to manage it. Our case shows that intravenous administration of midazolam is effective in relieving severe phantom limb pain during spinal anesthesia.