Multimodal Analgesic Management for Cervical Spine Surgery in the Ambulatory Setting

Patient Population

Following institutional review board approval (ORA No. 14051301), we performed a retrospective review of consecutive patients undergoing anterior cervical spine surgery, either ACDF or CDR, from a prospectively maintained surgical registry. All surgeries were performed between May 2013 and November 2018 by a single senior surgeon at our institution. Patients included in the study were treated for degenerative spinal pathologies. Empiric medical treatment for these conditions was attempted without symptom relief for all patients prior to operative management. Nonsurgical therapy included the use of anti-inflammatory medications, corticosteroid injections, local anesthetics, and physical therapy. Prior to undergoing cervical spine surgery, each patient was assessed and cleared for surgery by an anesthesiologist and primary care physician.

All patients underwent either a primary or revision single-level or multilevel ACDF or CDR using a standard anterior approach to the cervical spine. Prophylaxis for deep venous thrombosis was achieved through the use of ambulation, and no chemical prophylaxis was used. All patients received a unique MMA protocol specifically designed for the ambulatory setting (Table 1). The ASC did not allow for observation of patients for periods of time greater than 23 hours. All patients who underwent cervical surgery were required to receive a same-day discharge. Per state regulations, the ASC was located within 30 minutes of a hospital with inpatient and intensive care unit capabilities. In the event of an emergency that cannot be adequately treated at the ASC, patients would be transferred to this hospital via ambulance.

Data Collection

Patient demographic factors recorded included age, sex, body mass index (BMI), smoking status, Charlson Comorbidity Index (CCI), American Society of Anesthesiologists (ASA) score, and preoperative diagnosis. These diagnoses included cervical spinal stenosis, degenerative disc disease, degenerative spondylolisthesis, foraminal stenosis, herniated nucleus pulposus, myelomalacia, myelopathy, myeloradiculopathy, and radiculopathy. Preoperative medical conditions that were recorded at the time of the medical clearance appointment, including arthritis, asthma, cancer, diabetes, hyperlipidemia, hypertension, liver disease, and peripheral vascular disease, were also recorded.

Perioperative characteristics that were recorded included operative location, operative time, estimated blood loss, and ambulatory center length of stay. Patient-reported outcomes including the visual analog scale (VAS) pain score and neck disability index (NDI) score were recorded. Narcotic consumption prior to discharge was calculated following a standard conversion into units of oral morphine equivalents (OME). Postoperative complications were recorded, including acute renal failure, airway obstruction, altered mental status, atelectasis, atrial fibrillation, atrial flutter, aspiration, epidural hematoma, ileus, nausea and vomiting, postoperative anemia requiring transfusion, pulmonary embolism, pneumothorax, seizure of unknown origin, urinary retention, urinary tract infection, and venous thromboembolism.

Surgical Technique

Patients were intubated and placed on the operating table in the supine position. The surgical level was confirmed and localized with intraoperative radiographic fluoroscopy. Injection of local anesthetic, bupivacaine 0.5% with epinephrine 1:200 000, was made to each surgical site prior to incision. A 2- to 3-cm transverse incision, medial to the sternocleidomastoid muscle, was made through the skin and subcutaneous fat. The Smith-Robinson approach was used, and the platysma muscle was transversely dissected with an incision that aligned with that of the skin. Next, the sternocleidomastoid muscle and carotid sheath were laterally retracted, and the esophagus, trachea, and thyroid were medially retracted. Further blunt dissection was conducted down to the vertebral body and the pertinent disc space was identified.

Fluoroscopic radiographs were again used to confirm the surgical level. The annulus was then incised. Curved and straight curettes were used to remove the disc material and end-plate cartilage. Resection of the posterior longitudinal ligament was then accomplished with a Kerrison rongeur. Adequate preparation of the disc space was ensured. For patients undergoing ACDF, local autograft, allograft, or bone graft substitute was used to fill an appropriately sized interbody cage, and the cage was placed into the disc space. To help prevent interbody movement and subsidence, supplemental plate fixation was used. For patients undergoing CDR, following disc space preparation various size trials were used and checked under fluoroscopic guidance. Once the ideal size was agreed upon, the CDR implant was placed into the disc space. Once instrumentation was finished, the wound was thoroughly irrigated and evaluated for hemostasis.

Demographic Characteristics

A total of 178 patients met inclusion criteria. Of these, 125 patients underwent a 1-level procedure, 52 patients underwent a 2-level procedure, and 1 patient underwent a 3-level procedure (71.3% underwent ACDF, 28.7% underwent CDR; Table 2). The overall cohort consisted of 63.5% men with a mean age of 46.7 ± 9.1 years. Mean BMI was 28.6 ± 4.4 kg/m², and 22 patients reported tobacco use at their preoperative appointment. The mean CCI was 0.46 ± 0.8, with the majority of patients (50.5%) having an ASA score of 2. Preoperative chronic medical conditions and comorbidities were as follows: hypertension (29), asthma (15), hyperlipidemia (3), cancer (2), uncomplicated diabetes mellitus (8), liver disease (1), and peripheral vascular disease (1). Of note, there were no patients in our cohort with a recorded medical history of myocardial infarction, chronic lung disease, renal failure, or gastrointestinal bleeding.

Perioperative and Postoperative Characteristics

A total of 127 patients underwent an ACDF, of which 92.7% were primary procedures (Tables 3 and 4). A total of 51 patients underwent a CDR, of which 90.2% were primary procedures. The most common preoperative diagnosis was myeloradiculopathy for both patients undergoing ACDF (59.8%) and those undergoing CDR (24.1%).

The most common operative level was C6-C7 (30.9%; Table 5). The longest surgical case was 95 minutes with 1 outlier for length of stay at 23 hours (first cervical procedure performed at the ASC). The mean length of stay was 6.1 ± 2.5 hours, and 2 patients required admission at a local hospital. All patients, aside from the first, were discharged within 10 hours of the procedure end. The mean postoperative VAS pain score prior to discharge was 5.1 ± 2.5. The mean narcotics consumed following surgery and prior to discharge was 31.4 ± 17.6 OME. The cohort reported considerable improvement in NDI during the postoperative period at 6 weeks (32.1 ± 19.3), 12 weeks (29.5 ± 18.9), 6 months (28.2 ± 19.4), and 1 year (26.4 ± 20.4).

A total of 6 complications were observed in patients during the immediate postoperative period in the ASC (Table 6). Postoperative nausea and vomiting constituted 4 of the complications. All 4 of these patients were discharged in less than 23 hours and did not require admission. Two patients required admission to the hospital following surgery. One of these patients used illicit drugs prior to surgery, a practice unbeknown to our team or the anesthesia providers, and experienced a postoperative seizure of unknown origin. The patient was admitted to a local academic hospital, and the seizures were subsequently determined to be secondary to serotonin syndrome. The second admitted patient developed an anterior cervical hematoma that was noted 5 hours postoperatively during the observational period. The hematoma caused shortness of breath concerning for airway obstruction and was immediately evacuated at the ASC. The patient was subsequently admitted to a local academic hospital for 23-hour observation. Both patients were uneventfully discharged the next day following surgery.

Patient Selection

Successful cervical spine surgery in the ambulatory setting begins with appropriate patient selection. Careful screening is essential due to the diminished number of care providers and emergency services compared with the inpatient hospital setting. Exclusion criteria such as obesity, history of chronic obstructive pulmonary disease, hypertension, and stroke have been previously suggested as a general guideline for outpatient surgery, but more specific criteria is needed for cervical surgery.¹⁴ Unfortunately, there has been a paucity of literature guiding successful patient selection specifically for cervical spine surgery in the ambulatory surgical setting.

In light of this, we have proposed multifaceted recommendations for patient selection for cervical surgery in the ASC setting (Table 7). These are based not only on our clinical experience with 178 patients but also the available literature from the inpatient and nonambulatory setting. For example, Bovonratwet et al⁸ observed that patients with the following risk factors were more likely to experience postoperative hematoma requiring reoperation following ACDF: preoperative international normalized ratio (INR) > 1.2 relative risk [RR] = 2.85, lower BMI (≤24) [RR] = 2.11, ASA ≥ 3 [RR] = 1.67, and male sex [RR] = 1.67. Similarly, patients with BMIs greater than 40 kg/m² undergoing cervical spine procedures have an increased risk of experiencing postoperative pulmonary embolism or deep vein thrombosis (odds ratio = 3.34).²⁰ Adamson et al¹⁰ performed a retrospective review of patients undergoing ACDF both in the outpatient ambulatory center and inpatient setting. They reported¹⁰ a mean age of 49.5 years, a female majority (51.6%), and the majority having ASA ≤ 2. They found that patients undergoing 3 or more levels had a higher risk of experiencing cervical hematomas and dysphagia,¹⁰ a finding that has been supported in other research.¹⁶ Efforts made to minimize operative time (<5 hours) and blood loss (<300 mL) may successfully reduce the risk of cervical hematoma and dysphagia, which can result in prolonged intubation and reintubation when present.²²

The patient cohort in the present study was consistent with those of similar research in that patients were generally non-obese (mean BMI of 28.6 kg/m²), healthy (majority of patients with ASA score ≤ 2), mostly women, aged less than 50 years, and undergoing 1- or 2-level procedures. Considering the findings of the present study and the recent pertinent literature, we have proposed the following exclusion criteria for cervical spine surgery in the ambulatory setting: age greater than 50 years; BMI less than 24 kg/m² or greater than 40 kg/m²; INR greater than 1.2, ASA score greater than or equal to 2, and preoperative comorbid conditions such as asthma, New York Heart Association grade ≥3 congestive heart failure, myocardial infarction within 6 months, angina pectoris, or nonadherent obstructive sleep apnea (Table 7).⁹

ASC Considerations

Performing cervical surgery within the ASC raises a number of unique considerations and challenges. Surgeons must consider their patient population, the extent of surgery performed in regards to the degree of preoperative pathology and number of levels operated on, whether the facility has contracts with specific medical device companies, and personal experience of the surgeon. There is also concern regarding internal and financial bias, because several studies have suggested that physician ownership of ASCs may influence practice patterns and surgical efficiencies.^23,24 Despite potential shortcomings and conflicting interests, inherent attributes of the ASC appear to contribute to improved efficiency. Dedicated staff and operating rooms have contributed to decreased operative duration, estimated blood loss, and increased efficiency.^25–27 Furthermore, single-specialty ASCs have been associated with even lower rates of surgical site infection than multispecialty facilities.²⁸

Number of Levels Involved in Operation

When assessing cervical spine surgery, both ACDF and CDR procedures are generally well-tolerated with positive outcomes. For either procedure, a critical presurgical consideration is the number of levels involved in the operation. Single-level ACDF, for example, is one of the most common spine surgeries performed, with generally good outcomes. Given its generally short operative duration, manageable postoperative pain, and a relatively lower requirement for postoperative care, it has been successfully adopted in the outpatient setting.²⁹ However, considerable concern remains about performing multilevel surgery.

The potential for a postoperative retropharyngeal hematoma is one of the most feared complications associated with ACDF procedures, particularly those involving multiple operative levels. Few studies have investigated the association of multilevel ACDF and retropharyngeal hematoma in the outpatient setting. Although hematomas can occur days after surgery, studies have suggested that these clinically severe hematomas tend to be detected within 4 to 6 hours postoperatively.^30,31 Whereas research suggests that a 6-hour observational period is the maximum time required after ACDF procedures, the same studies either do not specify the number of levels operated upon or are based on single-level procedures. Unfortunately, studies examining the relationship between multilevel ACDF and postoperative complications such as retropharyngeal hematoma tend to be based on large institutional datasets,^32–34 and 3-level case investigations are especially limited.²⁹ On the basis of our findings, when patients are selected and counseled appropriately, a multilevel ACDF can be safely performed in the ASC setting.

Postoperative Consideration