WE report a case in which spinal anesthesia at the cervico-thoracic level (C7–T1 intervertebral space) was performed successfully for a urologic procedure in a patient with muscular dystrophy and thoracolumbar kyphoscoliosis.
A 59-yr-old bed-ridden man with advanced Becker muscular dystrophy 1and severe thoracolumbar scoliosis was scheduled to undergo urgent laser lithotripsy of right renal pelvic and bladder stones. The patient had been admitted to the hospital 15 days previously for an upper urinary tract infection, which was treated with ampicillin and gentamycin. His left kidney had been nonfunctional for several years secondary to postrenal obstruction.
The patient had several general anesthetics in the past, often complicated by postoperative pneumonia requiring tracheal intubation and mechanical ventilation. Intubation of the trachea had required many attempts and had been performed either blindly via the nose or via the mouth with direct laryngoscopy, which revealed a very limited view of the larynx. Ventilation with face mask had been satisfactory.
Lumbar spinal anesthesia had been used for previous urologic procedures but did not provide adequate analgesia, probably because the spinal deformity interfered with the distribution of the local anesthetic. On another occasion, epidural anesthesia had been attempted but abandoned after multiple attempts.
The patient’s history included congenital cardiomyopathy and coronary artery disease with previous myocardial infarction. Home medication included 0.125 mg digoxin daily, 50 mg triamterene daily, 12.5 mg captopril twice daily, 40 mg furosemide daily, 20 mg lovastatin daily, sublingual nitroglycerin as required, and 1 mg lorazepam daily at bedtime.
The patient was alert and cooperative. His height and weight were 135 cm and 40 kg, respectively. Blood pressure was 100/60 mmHg, heart rate was 75 beats/min, respiratory rate was 14 breaths/min, and oral temperature was 38.0°C. The patient was unable to move any part of his body below the neck. He had a Mallampati class 4 airway 2with an 8-mm opening of the mouth and normal teeth. Extension of the neck was normal. Examination of the thorax revealed a severe right convex kyphoscoliosis and reduced breath sounds bilaterally.
Arterial blood gas testing with room air showed a pH of 7.43, an arterial carbon dioxide tension (Paco2) of 36 mmHg, and an arterial oxygen tension (Pao2) of 82 mmHg. The coagulogram and platelet count were normal. Forced expiratory volume in 1 s and forced vital capacity were 1.00 and 1.25 l, respectively (both 35% of predicted value). The chest radiogram (fig. 1) revealed a severe thoracolumbar scoliosis with normal lung parenchyma.
The patient did not want to have general anesthesia because of his experience with multiple postoperative pneumoniae. He fully appreciated the seriousness of his physical condition and coherently expressed his concerns about the distress resulting from prolonged mechanical ventilation.
We proposed spinal anesthesia at the cervico-thoracic level. We told the patient that there were two particularly serious risks. The first was trauma to the spinal cord from the needle with the possibility of permanent paralysis or of loss of feeling or both in one or more limbs. The possibility of severe chronic pain syndrome was also discussed. The second major risk was the development of total spinal anesthesia, which would require ventilatory support in exactly the same way as general anesthesia. This complication could also necessitate major drug support of heart function and blood circulation to prevent cardiac arrest. The patient understood and accepted the risks.
After placement of the electrocardiographic leads, pulse oximeter, and noninvasive blood pressure cuff, a right internal jugular catheter was inserted during local anesthesia. Oxygen was administered by face mask. The difficult airway cart and fiberoptic bronchoscope were brought into the room. While the patient was held sitting, an 18-gauge needle introducer was inserted in aseptic conditions at the C7–T1 spinal interspace (technically the most appealing space). A 27-gauge Quincke-Babcock spinal needle was placed in the introducer and advanced into ligamentous tissue. After removing the stylet obturator, the Quincke-Babcock needle was advanced slowly until cerebrospinal fluid appeared. The patient reported no paresthesia. Hyperbaric bupivicaine (0.75%; 15 mg) was injected in the subarachnoid space in 3 min. At the end of the injection, systolic blood pressure decreased from a baseline of 100 to 60 mmHg. Infusion of epinephrine was started, and systolic blood pressure returned to 90–100 mmHg. The patient was kept sitting for 10 min before being placed in the lithotomy position. The level of sensory block assessed with pin-prick test was C5 on the left side and T4 on the right. Heart rate remained between 60 and 80 beats/min.
The patient occasionally reported pain. His statements were sometimes vague (“My back hurts. Could you move me a little?”), sometimes specific to the site of surgery. If position adjustment failed to relieve the pain or if the discomfort was at the surgical site, we gave small (25–50 μg) boluses of fentanyl (total dose of 350 μg over 2 h 45 min). Systolic blood pressure was maintained between 90–110 mmHg by titration of the epinephrine drip. Oxygen saturation remained greater than 98%. The epinephrine drip was tapered down and discontinued at the end of surgery. The patient was transferred to the surgical ward the next day after an overnight stay in the surgical intensive care unit.
Five days later, the patient came back for right percutaneous nephrolithotomy and removal of the double-J tube. Spinal anesthesia at the C7–T1 level was performed as described. However, the dose of hyperbaric bupivicaine was increased to 22.5 mg to establish a more complete block. Epinephrine infusion was started just before spinal injection to prevent arterial hypotension. The level of sensory block was T4 on both sides. Systolic blood pressure was maintained at 90–100 mmHg, and oxygen saturation was maintained above 98%. Fentanyl (two 50-μg boluses) was administered to relieve occasional pain. Propofol (80 mg in 10-mg boluses) was administered for sedation. The procedure lasted 2.5 h. After spending a night in the surgical intensive care unit, the patient was transferred to the surgical ward. The patient went home 4 days later. He subsequently expressed his gratitude for the quality of intraoperative analgesia.
Although spinal anesthesia at the cervico-thoracic level is, and must remain, an exceptional procedure, we felt that it was a suitable option for this patient. The surgical procedure required general or neuraxial anesthesia. There were two reasons against choosing general anesthesia: patient refusal and likelihood of postoperative pneumonia. Lumbar spinal anesthesia would likely not have provided sufficient analgesia, based on past experience. The spinal deformity made placement of an epidural needle in the lumbar or mid-thoracic area very difficult or impossible. Even with successful placement of the epidural catheter, the spinal deformity would have likely interfered with the spread of the anesthetic solution. A high thoracic (T1 or T2) epidural would have been easy to perform but would probably not have provided adequate anesthesia because of the distance from the dermatomal level of surgical stimulation. Abnormal spreading of the anesthetic solution also remained a concern. Furthermore, the large volume of local anesthetic required with a high thoracic epidural could have produced an unacceptably high block.
Although this patient had favorable course, this procedure involves serious risks and must not be viewed with complacency. Damage to the spinal cord from the needle could have led to severe, permanent sequelae. The neuroradiologic literature 3,4nevertheless provides some reassurance. The experience with intentional puncture of the cervical or thoracic cord for diagnosis or treatment of cystic lesions shows that the appearance of cerebrospinal fluid reliably precedes contact with the cord. Furthermore, paresthesia usually occurs when the needle passes between the posterior columns. We therefore expected that free flow of cerebrospinal fluid would indicate correct needle placement and that paresthesia would reveal accidental contact with the spinal cord. To facilitate immediate detection of cerebrospinal fluid, we advanced the Quincke-Babcock needle through the interspinous ligament without the obturator stylet. We believed that this would give an earlier and safer indication of reaching the intrathecal space than would stepwise advances with intermittent obturator removal.
The risk of total spinal anesthesia was also a serious concern, mainly because of the requirement for mechanical ventilation. We judged that this risk was acceptable despite anticipated difficulties with endotracheal intubation because we were sure that adequate ventilation could be achieved with a face mask or with a laryngeal mask airway.
Total spinal anesthesia did not develop, but we nevertheless observed a rapid onset of severe arterial hypotension, which was easily corrected with infusion of epinephrine. Hypotension likely resulted from the combination of the sitting position and almost immediate thoracic and lumbar sympathectomy. There was no bradycardia although the cardioaccelerator fibers (T1–T4) were probably affected.
Spinal anesthesia at the thoracic or cervical level may offer a solution under exceptional circumstances when general anesthesia or conventional regional techniques are contraindicated or impossible to perform. The patient should be fully informed of the risks and of the exceptional nature of this procedure. There must be no doubt about the ability to achieve adequate ventilation of the lungs in the event of total spinal anesthesia. A hyperbaric solution of local anesthetic should be used The patient should be left sitting for at least 10 min to avoid total spinal anesthesia and to establish an adequate block. An infusion of epinephrine should be immediately available to treat arterial hypotension, bradycardia, or both.