Use of Spinal Anesthesia in Lower Thoracic Spine Surgery: A Case Series

Abstract

BACKGROUND:

Spinal anesthesia is a safe and effective alternative to general anesthesia for patients undergoing lumbar spine surgery, and numerous reports have demonstrated its advantages. To the best of our knowledge, no group has specifically reported on the use of spinal anesthesia in thoracic-level spine surgeries because there is a hypothetical risk of injuring the conus medullaris at these levels. With the advantages of spinal anesthesia and the desire for many elderly patients to avoid general anesthesia, our group has uniquely explored the use of this modality on select patients with thoracic pathology requiring surgical intervention.

OBJECTIVE:

To investigate the feasibility of performing thoracic-level spinal surgeries under spinal anesthesia and report our experience with 3 patients.

METHODS:

A retrospective chart review of medical records was undertaken, involving clinical notes, operative notes, and anesthesia records.

RESULTS:

Three spinal stenosis patients underwent thoracic laminectomy under spinal anesthesia. Two surgeries were performed at the T11-T12 level and 1 at the T12-L1 level. The average age was 82 years, average American Society of Anesthesiologists score was 3.3, and 1 identified as female. Two cases used hyperbaric 0.75% bupivacaine dissolved in dextrose, and 1 used isobaric 0.5% bupivacaine dissolved in water.

CONCLUSION:

Spinal anesthesia is feasible for thoracic-level spine procedures, even in elderly patients with comorbidities. We describe our cases and technique for safely achieving a thoracic level of analgesia, as well as discuss recommendations, adverse events, and considerations for the use of spinal anesthesia during lower thoracic-level spine operations.

ABBREVIATIONS:

BPH

benign prostatic hyperplasia

TURP

transurethral resection of the prostate.

Spinal anesthesia is emerging as a safe and effective alternative to general anesthesia for patients undergoing both simple and complex lumbar spine surgery. We previously reported a series of 345 consecutive lumbar spine procedures performed under spinal anesthesia, demonstrating that spinal anesthesia is a safe, viable, and effective option for patients across a wide range of age and health statuses.¹ Numerous other studies have confirmed these findings in lumbar spine surgery, demonstrating equivalent or potentially lower postoperative complications, lower cognitive impact in at-risk patients, lower postoperative pain, lower use of postoperative analgesia, and savings in operative times and costs.² Thoracic spinal procedures are less common than lumbar ones, but the use of spinal anesthesia is expected to confer similar advantages. However, no literature has specifically reported on the use of spinal anesthesia in lower thoracic spinal surgeries. Current use of thoracic segmental spinal anesthesia overall is limited, even in other surgeries such as laparoscopic cholecystectomies and breast lumpectomies.^3,4 Some clinicians warn against the use of spinal anesthesia for spinal decompression procedures at or above the level of the conus medullaris, typically the L1 level in the average adult.⁵ Their main concerns include increased risk of neurological deficits from injuring the spinal cord and increased difficulty in getting access for a spinal anesthetic.

We hypothesized that spinal anesthesia could be safely offered to patients with pathology at lower thoracic levels. Here, we report our experience with thoracic-level laminectomies performed under spinal anesthesia in 3 patients (2 at T11-T12 and 1 at T12-L1). To avoid injuring the conus medullaris, we describe techniques that allow for safe injection. We provide a description of the cases, describe our technique, and discuss considerations for the use of spinal anesthesia at lower thoracic-level spinal operations.

METHODS

A retrospective chart review of medical records was undertaken analyzing clinic notes, operative notes, and anesthesia records. Institutional Review Board approval is not necessary at this institution for reports analyzing 3 or fewer clinical cases. Signed patient consent is exempted by this institution as Health Insurance Portability and Accountability Act identifiers are removed, and patient identity is not disclosed or compromised. Data are not available because of patient confidentiality.

RESULTS

We summarize the characteristics of each patient case in this case series in Table 1. These surgeries were performed at a single academic research institution by a senior attending. This case series has been reported in line with the PROCESS guideline.⁶

TABLE 1.

Summary of Clinical Presentations for Thoracic-Level Operations Undergoing Spinal Anesthesia

Patient	Age	Sex	ASA	BMI	Comorbidities	Surgery levels	Surgery type	Spinal anesthetic
Case 1	76	Female	4	27.95	Prior spinal surgery (including T11-T12), limited neck mobility	T11-T12	Laminectomy	Hyperbaric bupivacaine at L1-L2
Case 2	76	Male	3	26.36	BPH s/p TURP	T12-L1	Laminectomy	Isobaric bupivacaine at L3-L4
Case 3	93	Male	3	26.55	Atrial fibrillation	T11-T12	Laminectomy	Hyperbaric bupivacaine at L1-L2

BMI, body mass index; BPH, benign prostatic hyperplasia; TURP, transurethral resection of the prostate.

Case 1

This 76-year-old woman with a history of multiple prior spinal surgeries (including at the level of this operation), controlled asthma, and severely limited neck mobility secondary to prior fusion developed worsening leg weakness and difficulty walking over 6 months. Her myelopathy was deemed because of progressive thoracic stenosis and cord compression, with imaging supporting worsening of stenosis at T11-T12 with cord impingement and abnormal cord signal change (Figure 1). She elected to proceed with thoracic decompressive laminectomy and was motivated to avoid general anesthesia as possible. Her American Society of Anesthesiologists (ASA) score was 4, and body mass index (BMI) was 27.95. In the operating room, 2 cc of hyperbaric 0.75% bupivacaine dissolved in dextrose was injected into the thecal sac at L1-L2 with the patient sitting upright. The patient was then placed supine to allow the anesthetic to migrate rostrally and then transferred into prone position followed by normal prep and mild sedation with IV propofol. The surgery was completed in its usual manner without complications in 72 minutes, with the full operative time from the start of anesthesia to transfer to postanesthesia care unit taking 128 minutes. On postoperative day 1, she was noted to be confused with urinalysis revealing a urinary tract infection. Ciprofloxacin was started with symptomatic relief, and the patient was discharged on postoperative day 2 in stable condition after adequate ambulation and diet. She recovered for a week in a rehabilitation facility, followed with physical therapy at home. At her postoperative visit 1 month later, she reported improved ambulance and balance.

FIGURE 1.

Preoperative T2-weighted MRI of case 1 showing worsening of stenosis at T11-T12 with cord impingement and abnormal cord signal change.

Case 2

This 76-year-old man with a history of benign prostatic hyperplasia (s/p transurethral resection of the prostate and transurethral incision of bladder neck and urethral dilation) and L3-S1 transforaminal lumbar interbody fusion 4.5 years prior presents with several months of refractory and severe neurogenic claudication. Imaging showed severe T12-L1 stenosis with cord compression (Figure 2). Considering his severe and refractory symptoms, he elected for T12 to L1 decompressive laminectomy. His ASA score was 3 and BMI was 26.36. Preoperatively, there was difficulty with placing a Foley catheter because of scar formation from prior genitourinary surgeries, and urology was consulted for placement. In the operating room, 3 cc of isobaric 0.5% bupivacaine was injected at L3-L4 into the thecal sac with the patient in the sitting position. The patient was then placed in prone positioning, followed by normal prep and mild sedation with IV propofol, dexmedetomidine, and fentanyl. The surgery was completed in its usual manner without complications in 51 minutes, with the full operative time from the start of anesthesia to transfer to postanesthesia care unit taking 98 minutes. The patient was discharged on postoperative day 4 in stable condition with a Foley catheter (because of urology consultation). At his follow-up visit 1 month later, the patient reported symptomatic improvement.

FIGURE 2.

Preoperative T2-weighted MRI of case 2 showing showed severe T12-L1 stenosis with cord compression.

Case 3

This 93-year-old man with a history of prior lumbar laminectomy (L4-L5), atrial fibrillation, and coronary artery disease presented with a 3-month history of lower back pain and thoracic myelopathy with significant functional impairment. Imaging showed spinal cord compression at T11-T12 with abnormal cord signal changes (Figure 3). Considering this progressive myelopathy, he elected for T11-T12 decompressive laminectomy under spinal anesthesia. His ASA score was 3 and BMI was 26.55. In the operating room, 1.8 cc of hyperbaric 0.75% bupivacaine dissolved in dextrose was injected at L1-L2 while sitting upright. The patient was then placed supine to allow the anesthetic to migrate rostrally and then placed prone, followed by normal prep and mild sedation with IV propofol and dexmedetomidine. The surgery was completed without complications in its usual manner in 82 minutes, with the full operative time from the start of anesthesia to transfer to postanesthesia care unit taking 131 minutes. On postoperative day 1, he experienced a vasovagal event during which he felt dizzy, became hypotensive to a systolic pressure in the 60 seconds with a heart rate in the 30 seconds and unresponsive for 1 minute. He quickly returned to baseline vital signs and mental status. Cardiology workup supported the event to be likely be a vasovagal response in the setting of known first-degree atrioventricular block and right bundle branch block, unlikely to be related to the spinal anesthesia given the timing of onset. He was discharged on postoperative day 2 in stable condition after adequate ambulation and diet to a rehabilitation facility. At follow-up visit, he noted improvement in his pain and neurological function.

FIGURE 3.

Preoperative T2-weighted MRI of case 3 showing spinal cord compression at T11-T12 with abnormal cord signal changes.

DISCUSSION

Key Results

Thoracic spinal pathology is not a contraindication to spinal anesthesia. Here, we describe our experience successfully using spinal anesthesia in 3 patients requiring lower thoracic laminectomy. To the best of our knowledge, no other study has specifically reported on the use of spinal anesthesia in thoracic-level spine surgeries. These patients were motivated to undergo spinal anesthesia because they were significantly older than our previously published cohort (82 vs 63 years)¹ and concerned about the cognitive risks of general anesthesia among other factors. These patients had ASA scores of two 4 and one 3, in contrast to studies where spinal anesthesia is typically selected for healthier patients with ASA scores of 2 to 3.^1,5 In addition, 1 patient had extreme neck immobility with potential for difficult intubation. Although these factors were originally believed to exclude patients from consideration for spinal anesthesia, we feel that a wide range of patients are suitable and may in fact benefit from the use of spinal anesthesia.

To safely induce spinal anesthesia at the thoracic level, we recommend the following steps (Table 2). First, desire for spinal anesthesia should be evaluated, along with discussion of the risks, benefits, and alternatives. Specifics of what to expect and feel during the awake anesthesia should be reviewed to reduce anxiety during the procedure. MRI scans of the thoracic and lumbar spine should be reviewed to ascertain the level of the conus medullaris and to assess for scoliosis or abnormal spinal anatomy. During injection of the spinal anesthesia, the patient is first seated upright, and the needle is carefully advanced. Once the needle is engaged in ligament, it should be moved slowly and gently until just penetrating the dura and into the intrathecal space. As a check, cerebrospinal fluid should flow out of the needle, without evidence of blood or abnormal paresthesias. After injection of the anesthetic, the patient is placed in supine or prone positioning depending on the anesthetic used. If isobaric bupivacaine is used, the patient can be placed immediately into prone Trendelenburg. If a hyperbaric bupivacaine is used, the patient needs to be first placed supine, assessed for the level of sensory block with pinprick testing and then place in prone positioning. Rationale for this positioning is explained in the next paragraph. Because it takes 10 to 15 minutes for the spinal anesthetic to take effect, this interim time can be efficiently used for surgical prep (C-arm positioning, surgical site marking, prepping, and draping). Anesthesia at the surgical site is then confirmed, and adjunct mild sedation may be added with single or combined use of intravenous propofol, ketamine, dexmedetomidine, fentanyl, or midazolam.

TABLE 2.

Steps for Using Spinal Anesthesia During Spine Surgeries at the Thoracic Level

Step 1	Assess desire for spinal anesthesia and discuss risks, benefits, and alternatives.
Step 2	Review imaging and relevant anatomy
Step 3	Carefully insert the spinal needle and check for CSF flow without blood or paresthesias
Step 4	Proceed with injecting isobaric/hyperbaric bupivacaine while patient is sitting upright
	If using hyperbaric bupivacaine, place patient in supine position for 10 min to allow for anesthetic to “travel” to desired level. Pinprick testing for level of sensory block.
Step 5	Place patient in prone position and wait 10-15 min while the surgical and anesthesia teams are getting ready for surgery (prep, drape, and hold sedation)
Step 6	Assess for pain at operative level, if none, proceed with adjunctive IV anesthetic (eg, propofol)
	If there is pain, determine the current level of anesthesia and whether it is ascending. If it is patchy, wait for 3-5 min more. Reassess need for second dose or conversion to general anesthesia.
Step 7	Proceed with the operation

CSF, cerebrospinal fluid.

To induce spinal anesthesia at the thoracic level without injuring the conus medullaris, our technique relies on the baricity/density of the anesthetic. Intraoperatively, MRI spine scans should be reviewed for the level of the conus. Then, 1 of the 2 options may be used. First, an isobaric 0.5% bupivacaine dissolved in water can be injected into the subdural space for spinal anesthesia. This isobaric solution will simply stay at the level it is injected without being affected by gravity, even when body position changes (Figure 4). Second, a hyperbaric solution of 0.75% bupivacaine dissolved in dextrose can also be used. After injecting the hyperbaric anesthetic at the lumbar level, the patient is placed immediately in supine positioning. Because of the kyphotic curvature of the thoracic spine, the hyperbaric anesthetic will be carried by gravity toward the higher thoracic levels (Figure 4) because hyperbaric solutions gravitate to lowest dependent point of the body. A wait time of 5 to 10 minutes is advised to allow the solution to travel. Although the patient's anatomy can help guide the choice of approach between the use of hyperbaric or isobaric anesthetic, our case series does not currently support a rigid algorithm for this choice, and both options are feasible. Future studies may further investigate the optimal use of each type of anesthetic in specific situations.

FIGURE 4.

The effect the baricity of spinal anesthesia has over time in supine positioning. Top: An isobaric solution of spinal anesthetic will remain at the same level after injection. Bottom: A hyperbaric solution of spinal anesthetic will travel to the lowest dependent point of the body because of gravity.

With proper training, our clinical practice is comfortable using spinal anesthesia up to the T10 level, although hyperbaric spinal anesthesia could theoretically reach up to the level of T4. We have not used spinal anesthesia above the T10 level. Above this level, this modality likely becomes less feasible, consistent, and reliable, although this may be a topic for future exploration. Furthermore, there may be potential for injection of anesthetic rostral to the conus medullaris, with anatomic studies revealing that there is a greater depth of posterior subarachnoid space at the mid to lower thoracic segments of the spine where the cord lies anteriorly.³ Intrathecal injections directly at these levels may have a minimum safe distance before contact with the cord. Nevertheless, even expanding the use of spinal anesthesia to low thoracic levels as shown in these cases already allows for inclusion of a greater number of patients where stenosis and disk pathologies are more common.

Limitations

There are several considerations for spinal anesthesia during spinal surgery at any level. A foremost concern is intraoperative airway loss, especially problematic when the patient is in prone positioning without easy airway access. The team must be prepared to insert a supraglottic airway intraoperatively if necessary. Although extra caution must be taken when patients have obstructive pathologies such as obstructive sleep apnea or limited neck mobility, they are not absolute contraindications, as demonstrated in case 1. Another consideration is the potential for spinal anesthesia to incite a sympathectomy and resultant hemodynamic dysregulation. Perioperatively, some patients may experience hypotension, bradycardia, or rarely transient cardiac arrest.⁷ Patients should be assessed for history of sympathetic disorders such as vasovagal episodes and should be carefully monitored perioperatively. Although the patient in case 3 experienced a vasovagal response, the late timing of this event was well after the effects of spinal anesthesia had worn off and thus the response was unlikely because of the anesthetic choice. A further consideration is the length of surgery suitable before the anesthetic effect wears off. The most commonly used anesthetic is bupivacaine, which lasts between 90 and 150 minutes depending on the dose. Therefore, spinal anesthesia should be used for thoracic spine surgeries that can be finished within this timeframe, such as laminectomies and discectomies. Spinal anesthesia is not currently indicated for intradural procedures because there is a theoretical loss of anesthetic effect, although not yet proven. Other contraindications to consider for spinal anesthesia include patient refusal, allergy to the anesthetic, local site infection, and increased intracranial pressure.³ Finally, this preliminary case series of 3 patients represents an initial proof-of-principle that should be followed up with further studies on the use of spinal anesthesia at lower thoracic-level spine surgeries.

Generalizability

Spinal anesthesia has many benefits that may be extended to surgeries at the lower thoracic level. It is very simple to perform, with a fast onset and profound sensory block. With a small dosage of anesthetic used, there is a lower risk of systemic toxicity. Besides use in healthy patients, spinal anesthesia can also be beneficial for higher-risk patients. This includes older patients who are at higher risk of perioperative morbidity under general anesthesia; those who are frailer, with more comorbidities and cognitive impairment; those who prefer not to undergo general anesthesia; and those who would like to avoid polypharmacy. A systematic review of available studies has found that spinal anesthesia in lumbar spine surgery is safe and effective for a wide range of patients, with benefits including reduced postoperative complications, duration of anesthesia and surgery, total cost, hospital length of stay, intraoperative blood loss, postoperative pain scores, and use of postoperative analgesics.² These benefits likely extend to the use of spinal anesthesia in thoracic-level spine surgeries.

CONCLUSION

Spinal anesthesia is a feasible alternative to general anesthesia for patients undergoing lower thoracic-level spine surgeries. We present our experience with 3 cases, describe our technique for inducing spinal anesthesia, and discuss safety considerations as well as candidacy for its use across a variety of patients. Future studies may further explore the use of this anesthetic modality during thoracic-level spine procedures.

Funding

Andy Y. Wang was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, Award Number TL1TR002546. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Disclosures

The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.