Measures to prevent acute airway obstruction after anterior cervical spine surgery: a retrospective cohort study from Japan and a review of the literature

ACSS procedures

While under general anesthesia, the patients were positioned supine with the head and neck tilted backward. Somatosensory evoked potentials were monitored intraoperatively in all patients. The procedure involved left-sided access. A Caspar cervical retractor was used to maintain the surgical field. Under surgical microscopy, corpectomy, discectomy, or hybrid decompression was performed, according to the pathologic condition and clinical symptoms [13].

To perform reconstruction, the surgeons used intervertebral cages packed with autologous iliac bone for the discectomy segments and autologous fibular strut grafts/iliac bone grafts for the corpectomy segments. Plate fixation was standard for decompressed segments, and for cases involving multilevel corpectomy of more than three vertebrae, a halo vest was applied. Patients were extubated immediately after surgery and were monitored overnight in private rooms in the general ward; drains were removed the next morning or the following day.

Retrospective analysis of patients

We reviewed patient characteristics (age, gender, and diagnosis) and surgical parameters (number of fixed disc segments, number of upper vertebra instrumented, simultaneous posterior fixation, and application of a halo vest) from medical records. Postoperative airway compromise that necessitated reintubation or hematoma evacuation was defined as AAO. Patients who underwent ACSS through January 2019 were divided into AAO and non-AAO groups, and the patient characteristics and surgical parameters of the two groups were compared to identify risk factors. Furthermore, detailed treatment courses for AAO were analyzed.

AAO preventive measures

In February 2019, we implemented the following 10 preventive measures in accordance with our experience with AAO cases and a review of the literature:

Comparison of patients before and after implementation of preventive measures

We compared outcomes of patients who underwent ACSS between April 2006 and January 2019 (before implementation of prevention measures) with those of patients who underwent ACSS between February 2019 and July 2024 (after implementation) to evaluate the effects of the preventive measures on AAO occurrence.

Statistical analyses

To perform statistical analyses, we used R ver. 3.4.3 (R Foundation for Statistical Computing, Vienna, Austria). Student t-test was used to compare data for continuous variables, and Fisher’s exact test, for categorical variables. We used the Cochran-Armitage trend test to assess associations between categorical variable. A p-value of <0.05 denoted statistical significance.

Retrospective analysis of data obtained before implementation of preventive measures

We analyzed the outcomes of 1,036 patients who underwent ACSS before the implementation of AAO preventive measures. In one patient, surgery was prolonged because of severe OPLL and restricted jaw opening; all the other patients were extubated immediately after surgery and monitored overnight in private rooms in the general ward.

AAO occurred in seven patients (0.68%; six men). The clinical characteristics of these seven patients are summarized in Table 1. The median age was 59 years (range, 46–70 years), and the number of fixed disc levels ranged from 2 to 5 (median, 4). One case (patient 3) involved simultaneous anterior and posterior fixation. Halo vests were applied in three cases (patients 2, 3, and 4). Surgical time ranged from 89 to 286 minutes (median, 157 minutes), and intraoperative blood loss ranged from minimal to 110 mL. The time interval from surgery to intervention for AAO ranged from 2 to 72 hours (median, 32 hours); in five patients (71%), drains were still in place when AAO occurred. The primary causes of AAO included hematoma (in patients 1, 6, and 7), airway edema (in patients 2 and 3), postoperative C4 segmental cord disorder (in patient 4), and mucus obstruction (patient 5).

Table 1

Cases with acute airway obstruction (patients who underwent surgery between April 2006 to January 2019)

Of these seven patients, 3 (50%; patients 4, 5, and 6) were reintubated successfully on the first attempt. Fiberoptic intubation by an anesthesiologist was performed successfully in two more (patients 2 and 3) after the initial oral intubation attempts failed. Hematoma evacuation before reintubation resulted in favorable outcomes in two of three cases (patients 1 and 6), but in one case (patient 7), intubation failed and the patient died. Another patient (patient 5) sustained hypoxic brain damage despite reintubation; thus, poor outcomes occurred in 0.19% of all patients.

According to the clinical symptoms (Table 2) [5,6,9], five patients (1, 2, 3, 4, and 6) were in early to middle-stage AAO, and two were in the late stage (patients 5 and 7) when intervention was initiated. Four patients (1, 2, 3, and 6) recovered fully, and a patent airway tracheostomy was achieved in one case (patient 4).

Table 2

Stages of acute airway obstruction [6,9]

The comparison between the AAO group (n=7) and the non-AAO group (n=1,029) revealed that the incidence of AAO was significantly associated with the number of fixed disc segments (p=0.031) and the number of the more cephalad vertebra that were instrumented (p=0.007) (Table 3). Patients in the AAO group had significantly more fixed disc segments (3.6±1.3) than did those in the non-AAO group (2.8±0.9; p=0.031), and the proportion of patients who received a halo vest was significantly higher in the AAO group (42.9%) than in the non-AAO group (2.0%; p<0.001). These findings are illustrated in Fig. 1.

Table 3

Non-AAO vs. AAO (patients who underwent surgery between April 2006 and January 2019)

Fig. 1

Case presentation (case 7). A 46-year-old male presented with neck pain and left arm paresthesia. (A) Preoperative magnetic resonance imaging showed a left-sided herniated disc at C4–C5 and C5–C6. (B) He underwent two-level anterior cervical spine surgery and was extubated immediately. After drain removal the next morning, his condition initially remained stable but gradually worsened, with increasing throat obstruction. By 31 hours post-surgery, he developed severe respiratory distress with stridor, neck swelling, facial flushing, and worsening voice obstruction. Immediate bedside hematoma evacuation was performed, but respiratory distress persisted, requiring emergency oral intubation. Severe retropharyngeal edema obstructed glottic visualization, preventing successful intubation. The patient subsequently went into cardiac arrest. (C) Percutaneous cricothyrotomy restored ventilation 5 minutes after arrest, but he suffered hypoxic brain injury and later died. Bleeding from the drain tract after removal led to hematoma formation, and the resulting sustained circulatory disturbances likely caused secondary airway edema, thus, oral intubation was impossible.

Analysis of data obtained after implementation of preventive measures

After the implementation of preventive measures, 156 patients (the countermeasure group) underwent ACSS. The number of fixed disc segments was significantly lower in the countermeasure group (2.2±0.9) than among the 1,036 earlier patients (2.8±0.9; p<0.001). The frequency of halo vest application was also lower in the countermeasure group (0.6%) than in the earlier cohort (2.3%), but the difference did not reach statistical significance (p=0.237). Of note is that in the countermeasure group, no cases of AAO were reported; however, because of the limited number of cases, statistical significance regarding the effectiveness of the preventive measures could not be determined (Table 4).

Table 4

Comparison of patients who underwent ACSS before the introduction of AAO preventive countermeasures (between April 2006 and January 2019) and those who underwent ACSS after the introduction (between February 2019 and July 2024)

Surgery-related risk factors

In accordance with previous studies [2,5,14,24], our results showed that a greater number of fixed disc segments (p=0.031), a more cephalad surgical range (p=0.007), and the application of a halo vest (p<0.001) were significantly associated with AAO. Halo vests restrict head and neck movement, thereby hindering deep breaths and secretion clearance, which increases the risk of airway obstruction. In addition, frequent suctioning may irritate the laryngeal mucosa and lead to bleeding and edema.

These findings highlight the importance of minimizing multilevel exploration and avoiding cephalad surgical extensions when feasible. During the surgical planning stage, as part of our preventive measures, we selected the posterior approach for cases involving C2 or multilevel fixation of more than three segments. Furthermore, by selecting the more stable multilevel discectomy over the multilevel corpectomy, we reduced the application of halo vests. When the aforementioned surgical approaches are not feasible, we consider delaying extubation.

Postoperative hematoma

AAO caused by hematoma typically develops 6–12 hours postoperatively, whereas airway edema tends to appear 12–72 hours after surgery [5,27]. Hematoma is a significant cause of AAO and can result from insufficient intraoperative hemostasis, excessive coughing during extubation, or postoperative hypertension [5,7,9]. In our series, patients 1 and 6 exhibited persistent postoperative agitation and hypertension, which probably contributed to the postoperative formation of hematomas.

Patients with hematomas often exhibit anterior neck swelling and possible tracheal deviation [9]. Severe swelling can hamper oral intubation and emergency cricothyrotomy [7]; therefore, the hematoma should be evacuated immediately through the surgical incision. According to one review, seven of 13 patients underwent hematoma evacuation before reintubation [6].

In comparison with emergency oral intubation, hematoma evacuation can be performed more immediately, regardless of location or equipment, and its decompression effect may relieve respiratory distress. In our series, patients 1 and 6 showed improvement after hematoma evacuation. The procedure typically involves the removal of sutures from the surgical wound, often without requiring local anesthesia. If the patient’s respiratory status remains compromised, intubation should be attempted promptly.

As a part of preventive measures, blood pressure is maintained within the normal range to ensure intraoperative hemostasis, which reduces the risk of bleeding after surgery [21]. To minimize risk of bleeding after removal, surgical drains are positioned along the approach without penetrating muscle layers. Postoperatively, wound dressings are minimal so as to facilitate the early detection of bleeding or swelling. Remifentanil infusion is maintained until extubation to prevent emergence coughing without causing respiratory depression [22,28,29]. After extubation, low-dose dexmedetomidine is administered overnight to agitated patients for moderate sedation with minimal respiratory effects [23,30].

Airway edema

Laryngopharyngeal edema resulting from surgical manipulation and retraction is a significant contributor to postoperative respiratory distress. Airway edema often does not manifest with visible neck swelling, but it can make reintubation difficult; thus, alternative airway management methods should be considered during surgical planning [7]. In our study, airway edema occurred in patients 2 and 3, manifesting in both more than 60 hours postoperatively. Both patients 2 and 3 received halo vests; patient 3 had undergone simultaneous anterior and posterior fixation. In both patients, oral intubation failed, but fiberoptic intubation by an anesthesiologist was successful. Simultaneous anterior and posterior fixation increases the risk of airway edema because of the manipulation of ventral soft tissue and prolonged prone positioning [4].

To reduce the risk of airway edema, we implemented several strategies. In the surgical planning stage, simultaneous anterior and posterior fixation was avoided when feasible. The head was elevated slightly forward with stacked towels to reduce tracheal tension, thereby alleviating the retractor-induced pressure during surgery [15]. Dexamethasone was applied locally before wound closure to reduce prevertebral soft tissue swelling because studies have shown that corticosteroids safely inhibit inflammatory prostaglandins and cytokines [16–20]. Postoperatively, the head of the bed was elevated by 30° to improve venous drainage and ventilation [7,15].

Timing and management of extubation

Some authors have advocated prolonged intubation and ICU management to mitigate the risk of AAO [4,5,11,12]; however, the timing of extubation after ACSS remains controversial. In our series, interventions for AAO occurred at a median of 32 hours postoperatively; 71% of cases were identified after 24 hours. These results suggest that overnight intubation alone does not effectively prevent AAO. At the same time, prolonged intubation has risks, including ventilator-associated pneumonia, tracheal injury, and hemodynamic instability [31,32]. We generally perform immediate extubation but consider prolonged intubation for patients with previously reported high-risk comorbidities or surgery-related risk factors [3–5].

Airway management after extubation

AAO can progress rapidly from asymptomatic to complete obstruction within minutes [8,9]; thus, early recognition and timely intervention are critical. For the systematic evaluation, Song et al. [9] classified AAO according to respiratory status, surgical site condition, and patient behavior. Similarly, Yamada et al. [6] categorized AAO symptoms into three stages, in reference to prior reports. Early- or middle-stage interventions are often effective, whereas late-stage progression is associated with poor outcomes, including death or hypoxic brain injury. In accordance with these classifications, the stages of AAO symptoms are summarized in Table 2.

In our series, five of seven cases of AAO were treated successfully in the early or middle stages; the two that progressed to the late stage resulted in poor outcomes. To improve the monitoring of airway obstruction after extubation, medical staff should regularly measure the patient’s neck circumference for early detection of swelling and assess the respiratory rate in distressed patients. According to Yamada et al. [6], a neck circumference increase of >2 cm and a respiratory rate of >25 breaths/min are the criteria for emergency intervention. Fig. 2 depicts our perioperative airway management algorithm for AAO prevention in patients who undergo ACSS.

Fig. 2

Airway management algorithm for patients undergoing anterior cervical spine surgery.

Limitations

This study had several limitations. First, it was conducted at a single institution, and the number of cases of AAO was relatively small; thus, the generalizability of our findings is limited. Moreover, the low incidence of AAO reduced the statistical power of our analyses, which precluded definitive conclusions about the effectiveness of our preventive measures. Second, previously reported risk factors are important for comprehensive risk stratification but were not systematically assessed because of data limitations. In future research, investigators should include these factors to better tailor perioperative protocols. Third, the application of local corticosteroids to the retropharyngeal space after ACSS remains controversial. Further studies are necessary to validate the dose and efficacy of this approach.