PAIN caused by herpes zoster is classified as acute herpetic pain and postherpetic neuralgia (PHN). Acute herpetic pain is inflammatory and nociceptive, whereas PHN is a combination of neuropathic and maintained sympathetic pain, which contributes to ongoing pain.1Therefore, the treatment differs. Acute herpetic pain treatment includes antiviral drugs, nerve blocks, sympathetic blockade, and opioids, whereas PHN requires antidepressants, anticonvulsants, antiarrhythmic drugs, corticosteroids, gabapentin, topical capsaicin, topical clonidine, or local anesthetics and beside nerve blocks.1–6
Although there are various treatments for PHN, in some patients who are resistant to conventional therapies, intractable or poorly responsive symptoms can develop, which can lead to a devastating impact on a patient’s quality of life. For these cases, different approaches such as endoscopic transthoracic sympathicotomy,760 mg intrathecal methylprednisolone,8or even the removal of the painful skin9have been suggested.
We report six cases of patients with long-standing history of refractory PHN managed with spinal thoracic alcohol neurolysis after the spinal test block (tables 1–3).
A 82-yr-old woman weighing 72 kg was evaluated for reports of 2-yr excruciating and burning pain secondary to herpes zoster on the trunk with sequelae on the skin from right T7 to T10 but painful skin at T9–T12 on the same side. The scarred area was hypoesthetic; nevertheless, the patient reported severe allodynia to tactile stimuli and hyperalgesia from T9 to T12 extension area. She also reported paroxysmal lancinating pain occurring at least twice a week.
The symptoms were poorly responsive to the daily combination of 300 mg tramadol, 75 mg amitriptyline, 600 mg gabapentin, 600 mg carbamazepine, 2 g dipyrone, and 200 mg ketoprofen. Either topical application of clonidine cream or capsaicin alleviated 30% of the burning sensation for nearly 4 h. Topical application of 5 mg/day nitroglycerine did not have any effect. Sequential weekly intercostal blocks from T9 to T12 with 30 μg clonidine, 5 mg dexamethasone, and 70 mg lidocaine were also tried simultaneously, with poor response. An epidural catheter was then inserted at the T9–T10 interspace, and the patient was discharged home, taking 90 μg epidural clonidine daily combined with 180 mg lidocaine (divided into three applications per day). After 5 weeks, the catheter was removed because of patient dissatisfaction. Higher doses of clonidine resulted in somnolence.
After peripheral vein catheterization and standard monitoring, the patient underwent test thoracic spinal analgesia at the T9–T10 interspace, with 0.2-ml increment doses of 0.5% isobaric bupivacaine. The puncture was done at the lateral position, with the painful pain side facing up, and freedom from pain was obtained with 0.6 ml isobaric bupivacaine, 0.5%. The patient was kept in the lateral position for an additional 30 min and was sent home after eating and drinking.
A week later, the patient returned, enthusiastic and with no reports of side effects or pain. She underwent spinal neurolysis under the same conditions of the spinal test, with the final dose of 0.6 ml alcohol, 100%. The final alcohol volume was 0.6 ml because the previous bupivacaine test dose indicated ideal analgesia from T9 to T12, without side effects. Just after the alcohol injection, the patient reported a transient thoracic burning sensation; however, no medication was necessary because it was short lived—near 3 min. Just before withdrawal of the needle, 0.3 ml normal saline was injected into the needle to avoid any alcohol damage to the muscle or ligaments through the needle way. The patient was discharged home, taking 600 mg gabapentin and 75 mg amitriptyline daily. Thirteen months later, she had no allodynia or hyperalgesia, but she reported local thoracic hypoesthesia and pruritus (visual analog scale [VAS] score, 3 cm). There were no motor deficits, and the VAS score for residual steady burning pain was 2 cm.
The patient reported pain from T6 to T12 dermatomes and underwent test thoracic spinal analgesia at the T9–T10 interspace with 0.2-ml increment doses of 0.5% isobaric bupivacaine, while in the lateral position, with the painful pain side facing up. Freedom from pain was obtained with 0.8 ml isobaric bupivacaine, 0.5%. The patient was kept in the lateral position for an additional 30 min. Evaluation 1 h later revealed dermatome anesthesia varying from T4 to L1 dermatomes.
A week later, the patient returned, with no side effects or pain; however, we decided to use the final dose of 0.4 ml alcohol, 100%, for the spinal neurolysis as the skin area of anesthesia after the bupivacaine spinal test block ranged from T4 to L1, and was unnecessary for this patient (who had neuropathic pain from T6 to T12). A volume of 0.3 ml normal saline was injected into the needle before withdrawal. The patient was discharged home, taking 2.5 mg methadone daily combined with 5 mg meloxicam, 150 mg ranitidine, and 50 mg amitriptyline. A week later, he was free from pain from T6 to T9; however, acute herpes zoster had developed at the T11 and T12 dermatomes. Daily oral acyclovir was initiated and continued for 15 days, with good response. Two months later, the patient reported pain at T11–T12, left side, with a VAS score of 10 cm, and he underwent intercostal neurolysis at the 11th and 12th thoracic levels with 0.5 ml phenol, 7.5%. After 11 months, he reported residual pain with a VAS score of 1 cm and had no sign of motor deficits.
Postherpetic neuralgia varied from the T5 to T8 dermatomes. The patient’s history included hypertension, angina, and acute myocardial infarction. He underwent test thoracic spinal analgesia at the T5–T6 interspace with 0.1-ml increment doses of 0.5% isobaric bupivacaine. Freedom from pain was obtained with 0.3 ml isobaric bupivacaine, 0.5%. The patient was kept in the lateral position for an additional 30 min. After evaluation at 6 h, he had no signs of chest pain or bradycardia. A week later, he returned, with no reports of side effects, motor deficits, or pain, and underwent spinal neurolysis with 0.3 ml alcohol, 100%. He was discharged home, taking 2.5 mg methadone daily combined with 5 mg meloxicam, 100 mg ranitidine, and 12.5 mg amitriptyline. After 10 months, the patient was completely free of pain.
Postherpetic neuralgia varied from the T2 to T5 dermatomes. The patient underwent test thoracic spinal analgesia at the T3–T4 interspace with 0.1-ml increment doses of 0.5% isobaric bupivacaine. Freedom from pain was obtained with 0.3 ml isobaric bupivacaine, 0.5%. The patient was kept in the lateral position for an additional 30 min. A week later, he reported that 24 h after the test block, the burning pain restarted (VAS score, 5 cm). He underwent spinal neurolysis with 0.3 ml alcohol, 100%. Currently, the patient has no complaints of shooting pain and is undergoing weekly stellate ganglion block combined with 400 mg carbamazepine daily plus 50 mg amitriptyline. He classifies the remaining burning pain as 4 cm on the VAS scale.
Postherpetic neuralgia varied from the T5 to T8 dermatomes. The patient had a history of hypertension and angina. He underwent test thoracic spinal analgesia at the T6–T7 interspace with 0.1-ml increment doses of 0.5% isobaric bupivacaine. Freedom from pain was obtained with 0.3 ml isobaric bupivacaine, 0.5%. The patient was kept in the lateral position for an additional 30 min. A week later, he returned, with no reports of side effects, motor deficits, or pain, and he underwent spinal neurolysis with 0.3 ml alcohol, 100%. He was discharged home, taking 50 mg amitriptyline. After 9 months, he is completely free of pain.
Postherpetic neuralgia varied from the T3 to T4 dermatomes. At the time the patient underwent test thoracic spinal analgesia at the T3–T4 interspace with 0.1-ml increment doses of 0.5% isobaric bupivacaine, he was confused about the intensity of the pain. The total dose of 0.2 ml isobaric bupivacaine, 0.5%, was injected at the time. The patient was kept in the lateral position for an additional 30 min. A week later, he returned, reporting pain with a VAS score of 8 cm, and did not undergo spinal neurolysis because of the poor response to the test block. After 4 months, he is still experiencing pain with a VAS score of 8 cm. He is currently under a sequential epidural blockade, taking oral amitriptyline, clonidine, and corticosteroids, and is poorly responsive.
Postherpetic neuralgia is a difficult condition to treat, and effective management requires a multimodal approach, including pharmacologic and nonpharmacologic treatments.10Nevertheless, of the various treatments, none produce definitive effects, and unfortunately, patients can be resistant to conventional therapies, perhaps because of the new pattern of coetaneous innervations in the skin affected with PHN.11Immunofluorescence of the most painful area revealed a reduction in density of innervation of the epidermis, in the dermal–epidermal junction, and in the eccrine sweat glands of the compromised skin.11It has been suggested that a minimum of 650 neurites/mm2skin surface area is necessary to avoid PHN.12Destruction of the primary nociceptive neurons can reduce more than four times the skin density primary neurons,11and the intrathecal measurement of interleukin 8 could be indicative of the duration of neuralgia.8
We describe six patients with PHN, of whom five (cases 1–5) were successfully treated with correspondent thoracic spinal neurolysis of the affected skin, with individually incremented low doses of 100% alcohol. Subarachnoid neurolysis using ethanol to selectively destroy the posterior roots of the spinal cord has been described as a method for providing pain relief.3,4Apart from the selective destruction of dorsal roots, even at nontoxic concentrations, ethanol has a direct effect on insular free Ca2+smooth muscle, associated with vasoconstriction;13however, none of our patients showed any signs of motor injury.
In fact, the criteria for intrathecal neurolysis in our practice include (1) midthoracic neuralgia not responsive to conventional therapies and (2) previous intrathecal test block with isobaric bupivacaine (similar baricity compared with the alcohol solution) to direct to a more appropriate volume of alcohol to be used and possible side effects. In the presence of respiratory difficulty, cardiac arrhythmia, urinary or bowel incontinence, or absence of response, neurolysis is contraindicated, as in the patient in case 6, where the spinal test block had no benefit, suggesting that different ways to treat the pain should be attempted.
Complications after subarachnoid neurolysis with hyperbaric phenol reported in the literature include unexpected contralateral brachial weakness, which developed toward the conclusion of the procedure and fortunately resolved spontaneously.14Of importance, the site of administration for the neurolytic agent in our patients is always the midthoracic level (between the 5th and the 10th thoracic levels), responsible mainly for the motor activity of the intercostal muscles. Weakness or irreversible cord injury of the extremities is a complication of the procedure that has been attributed to spread of the neurolytic agent to the anterior roots.15Even in the presence of rostral spread and intercostal muscle paralysis, the diaphragm muscle that is innervated spinally at the 3rd to the 5th cervical level would maintain by itself appropriate respiratory function in most patients.
In addition to spinal neurolysis, acute herpes zoster developed in the second patient, in the two consecutive lower thoracic levels (T11 and T12). The varicella zoster virus remains dormant in dorsal root and can be reactivated by declining specific cellular immunity leading to herpes zoster, and patients aged older than 50 yr are at increased risk to develop such a complication.16In this specific case, intercostal neurolysis was successfully performed at the 11th and 12th thoracic levels with 0.5 ml phenol, 7.5%, after the antiviral treatment. It has been suggested that the antiviral agent acyclovir combined with epidural blocks is an effective treatment for pain.17A higher volume of phenol has been associated with irreversible complications. Immediate paraplegia has been described after intercostal neurolysis with 6 ml aqueous phenol solution, 7.5%, at the 10th thoracic level. A lumbar puncture performed 3.5 h after the intercostal block showed a phenol concentration of 87.5 mg/ml. The authors suggested that the most likely explanation was the diffusion of the phenol through the intervertebral foramina, reaching the spinal space and therefore damaging the motor and sensory roots.18A different alternative to the intrathecal route includes epidural assessment. After death secondary to the primary disease, necropsy of the spinal cord and roots demonstrated no abnormality, and only the laminar structure of the dura had been destroyed at the outer one third.19Patient 4 had sympathetically maintained pain, and was successfully treated with additional stellate ganglion block.20
In conclusion, we have described an easier alternative to treat irresponsive PHN, with no side effects. The importance of the technique lies in the previous test with a substance similar in baricity to alcohol (isobaric bupivacaine), guiding the proper indication, and in a more specific volume of 100% alcohol administered to avoid sequelae or side effects.