1 Introduction
Anterior cervical discectomy and fusion (ACDF) is used to treat various cervical conditions such as myelopathy, disc herniation, and degenerative disc disease. It has been demonstrated to provide substantial reductions in pain and symptoms, with a patient-reported success rate between 85% and 95%.
[1] Nonetheless, adjacent segment degeneration and pseudarthrosis are known complications.
[1–
3] To minimize complications and associated need for reoperations, the clinical significance of the cervicothoracic junction (CTJ) has been explored.
[3–
7]As a transitional region of the spine, the CTJ is subject to significant static and dynamic stressors, potentially predisposing it to the development of pathology.
[3,
8] Some authors have proposed that extension of cervical constructs into the thoracic spine may minimize complication risk.
[5,
9] This idea has been explored in relation to posterior cervical fusion (PCF).
[5–
7,
9] Among a cohort of 438 patients undergoing PCF, extending constructs beyond the CTJ led to significantly lower rates of mechanical failure and revision surgery.
[5] Notably, extension to T2 led to superior outcomes than T1.
[5] Literature exploring this topic with regard to ACDF is far more limited, but Louie et al.
[3] found that among 83 patients undergoing ACDF, there were no differences in rates of adjacent segment degeneration or rates of reoperation between fusions terminating above or below the CTJ.
A notable gap persists in the existing literature regarding the significance of the CTJ in ACDF. To better inform surgical planning and patient counseling, we strove to better delineate the relationship of the CTJ in comparison to other spinal levels. Thus, among a cohort of patients undergoing ACDF, we aimed to determine the impact of C7-T1 fusion on patient-reported outcome measures (PROMs) and clinical outcomes.
2 Materials and methods
2.1 Study design
A retrospective cohort analysis was performed utilizing data obtained from the Quality Outcomes Database (QOD), a national multicenter spine surgery registry containing prospectively collected patient data.
[10] The QOD’s inclusion criteria consist of patients ≥ 18 years undergoing elective spinal surgery for degenerative spine diseases.
[10] Patients were excluded from the QOD if they underwent spinal surgery due to traumatic injury, tumors, spinal infections, or if they were incarcerated at the time of surgery.
[11] All data were obtained from and defined according to standardized protocols within the QOD.
2.2 Patient population
The cervical module of the QOD was queried for patients who underwent 1- to 2-level ACDF surgery. Inclusion criteria consisted of adult patients who were undergoing primary elective 1- or 2-level ACDF and had a surgical indication of degenerative pathology. Exclusion criteria included age under 18 years; diagnoses of infection, tumor, deformity, or trauma; and concomitant arthrodesis or instrumentation.
2.3 Exposure variables
The primary independent variable was fusion of C7-T1. Demographic variables and baseline characteristics consisted of age, sex, body mass index (BMI), comorbidities, smoking status, clinical characteristics, and prior spine surgery.
2.4 Outcome variable
The primary outcome of interest was 3-month and 12-month PROMs, including numeric rating scale (NRS) scores for neck and arm pain and quality-adjusted life year (QALY) scores.
[12] Secondary outcomes included 30-day complications and 3-month readmissions and reoperations.
2.5 Statistical analysis
Descriptive statistics were used to summarize demographic variables and baseline characteristics. Categorical variables were presented as frequencies and percentages, whereas continuous data were presented as means and standard deviations. Bivariate analysis was conducted to compare baseline variables and outcomes between patients with and without C7-T1 fusion, using pairwise deletion for missing data. Pearson’s chi-square was used to compare categorical variables. Wilcoxon rank sum test was used to compare continuous variables. Separate multivariable regression models were fitted for each of the endpoints on C7-T1 being fused, including all demographic variables, clinical characteristics (such as reoperation), and baseline PROMs as covariates. Logistic regression was used for binary outcomes and linear regression was used for continuous outcomes such as PROMs. Similarly, a subanalysis and separate multivariable regression model for myelopathy patients was further conducted with modified Japanese Orthopaedic Association (mJOA) scores.
[13] Listwise deletion was used for missing data in regression analysis. A
p value < 0.05 was considered statistically significant. All statistical analyses were performed using R version 4.1.
[14]3 Results
3.1 Demographics and clinical characteristics
Of the ACDF patients queried, 12,240 patients met the inclusion criteria. Patients had a mean age of 56.7 ± 11.8 years, a mean BMI of 30.43 ± 6.57, and 48.0% (3603) were male (Table 1). Among these patients, 38.7% (4736) underwent C7-T1 fusion. On bivariate analysis, patients who underwent C7-T1 fusion were younger, had lower BMI, and were more likely to be smokers, have comorbidities, and have a history of prior spine surgery compared with those who did not undergo C7-T1 fusion. Additionally, these patients exhibited less myelopathy and cervical instability.
3.2 PROMs and clinical outcomes
At 3 months, patients with C7-T1 fusion reported lower NRS-arm (1.94 ± 2.61 vs. 2.14 ± 2.75; p = 0.010) and neck pain (2.65 ± 2.53 vs. 2.79 ± 2.63; p = 0.039), neck disability index (NDI)% (20.0 ± 17.5 vs. 21.1 ± 17.6; p = 0.002), and QALY (0.793 ± 0.187 vs. 0.771 ± 0.194; p < 0.001) when compared with patients without C7-T1 fusion (Table 2). At 12 months, patients with C7-T1 fusion reported lower NDI% (17.8 ± 18.8 vs. 18.6 ± 18.8; p = 0.036) and higher QALY (0.800 ± 0.202 vs. 0.778 ± 0.210; p < 0.001) when compared with patients without C7-T1 fusion. Further, patients with C7-T1 fusion had fewer complications (8.7% vs. 9.9%; p = 0.036) and readmissions (3.5% vs. 5.3%; p < 0.001) than patients without C7–T1 fusion. There was no significant difference in patient satisfaction and reoperation rates between the cohorts (p > 0.05). To note, there was drop out from baseline to 3 and 12 months (Table 2).
On multivariable regression controlling for all demographics, clinical characteristics, and baseline PROMs (Tables 1 and 2), fusion of C7-T1 was significantly associated with a greater 3-month QALY score (odds ratio [OR] = 1.142; 95% confidence interval [CI]: 1.055–1.237; p < 0.001) and lower rate of readmission (OR = 0.760; 95% CI: 0.625–0.924; p = 0.006) (Table 3). There was no association between C7-T1 fusion and any other outcome (p > 0.05). Likewise, comparison was used for when there was missing data.
3.3 Subanalysis of myelopathy patients
Of the 4103 myelopathic patients in our cohort, 1151 (28.1%) had C7-T1 fusion (Table 1). On bivariate analysis, patients with C7-T1 fusion had greater 12-month patient satisfaction (69.0% vs. 63.9%; p = 0.039), better 3-month (14.46 ± 2.51 vs. 14.22 ± 2.54; p = 0.009) and 12-month (14.38 ± 2.61 vs. 14.11 ± 2.71; p = 0.045) mJOA scores, and fewer complications (9.3% vs. 11.6%; p = 0.038) and readmissions (4.1% vs. 6.9%; p = 0.001) than patients without C7-T1 fusion (Table 4). On multivariable regression, C7-T1 fusion was associated with lower odds of readmission (OR = 0.602; 95% CI: 0.420–0.864; p = 0.006), with no association to other outcomes (p > 0.05) (Table 5).
4 Discussion
The present study aimed to elucidate the significance of C7-T1 fusion in ACDF by assessing PROMs as well as rates of complications, readmissions, and reoperations. Although bivariate analyses suggested that C7-T1 fusion positively impacted postoperative PROMs, pain scores, and complication rates, only the 3-month QALY score and readmission were significantly affected by C7-T1 fusion following multivariate analysis. Similar results were noted when analyzing only myelopathic patients within the cohort; multivariate analysis revealed lower odds of 90-day readmission as the only significant difference following C7-T1 fusion. Consistent with the large sample size, both the primary and subgroup analyses were sufficiently powered (> 80%) to detect clinically meaningful differences in the reported outcomes. These findings suggest that C7-T1 fusion has minimal impact on most patient-reported and clinical outcomes in the context of single- and 2-level ACDF.
Our study found that ACDF constructs through the CTJ did not impact complication rates. While literature investigating fusion of the CTJ via an anterior approach is sparse, the majority of existing literature has focused on adjacent segment disease. For example, Hilibrand et al.
[4] found that adjacent segment disease occurs most commonly at the C4–C5, C5–C6, and C6–C7 levels, with risk decreasing notably at the C3–C4 and C7–T1 levels. The authors also noted that with increased length of the overall ACDF construct, the risk of adjacent segment disease decreased.
[4] Concurrent findings were published by van Eck et al.
[15] among their cohort of 672 patients. In an
in vitro study, Cheng et al.
[9] noted that following anterior cervical fusion procedures spanning the CTJ, bending motion resulted in the least intradiscal pressure at the T2 to T3 level. Based on this finding, the authors suggested that termination of anterior constructs at the T2 level may be most beneficial in decreasing adjacent segment pathology.
[9] The biomechanical rationale has been proposed by several authors, suggesting that without including the CTJ in the fusion construct, 2 rigid levers are created between the high-motion cervical spine and lower-motion thoracic spine, potentially predisposing to the development of pathology.
[9,
16,
17]There were no observed differences in PROMs, pain scores, or patient satisfaction between those who underwent ACDF with C7-T1 fusion and those without. Our findings are corroborated in the setting of PCF,
[5,
7,
18] with scant literature addressing ACDF in this context. Lee et al.
[7] noted that among 46 patients undergoing PCF, there were no differences in arm and neck pain between patients receiving constructs terminating at C7 compared with those extending to T1. Similarly, both Okamoto et al.
[18] and Labrum et al.
[5] found no differences in mJOA, NRS-neck pain, NDI, and other PROMs between fusions that did or did not cross the CTJ at 1-year follow-up post-PCF. Nonetheless, the potential increased invasiveness of extending fusion constructs into the thoracic region remains another factor to consider. For example, even in the setting of similar PROMs, the Okamoto cohort of patients receiving fusion crossing the CTJ experienced significantly longer surgical times, increased blood loss, and higher rates of peri- and postoperative complications than those with constructs terminating at C7.
[18] Our findings are some of the first to report on PROMs, pain scores, and patient satisfaction following ACDF, investigating the CTJ. While existing literature focused on posterior approaches provides valuable insight, future work focused on anterior approaches is needed.
The present research must be considered with some limitations. Primarily, the retrospective nature of this work using QOD data makes it a nonrandomized study with patient and procedure selection being surgeon-dependent, introducing bias. Further, our included follow-up was only 12 months, potentially missing clinical and patient outcomes over a longer timeframe. We were also limited by the granularity of the data in each group, with exact indications and specifics of each operation unknown, such as the number of fixed segments. Lastly, mixed-effects models for the PROM regression analysis could have offered additional insight; however, due to the structure of the QOD and scope of analysis, separate regression models at each time point were selected instead. Similarly, no formal correction for multiple comparisons was conducted. While multiple comparisons may increase the risk of type Ⅰ error, the primary outcomes were predefined based on clinical relevance, which helps mitigate this concern. Considering these limitations and the current dearth of literature examining fusion of the CTJ in ACDF, it is important to continue examining this subject to determine any potential advantages and drawbacks.
5 Conclusion
The C7-T1 segment presents difficulty in spinal surgery due to its biomechanical transition from the mobile cervical spine to the rigid thoracic spine. The impact of fusing C7-T1 on outcomes remains unclear in the literature relative to other levels during ACDF. Our analysis demonstrated that C7-T1 fusion was associated with some improved short-term quality and clinical outcome measures. C7-T1 fusion versus without fusion were not significantly different from any other outcome. Our results suggest comparable outcomes when fusing C7-T1, which may be useful for surgeons in determining appropriate treatment and counseling patients facing surgery for pathology at this level.
© The Author(s) 2026. Published by Wolters Kluwer Health, Inc. on behalf of Higher Education Press.
PDF
(157KB)
