Sage Journals: Discover world-class research

Abstract

Study design

Retrospective cohort study.

Objectives

To elucidate incidence, risk factor, and clinical effect of bone regrowth after posterior cervical foraminotomy (PCF).

Methods

Ninety-eight patients who underwent PCF for the treatment of cervical radiculopathy and were followed up for >2 years were retrospectively reviewed. Foraminal dimension, sagittal gap at resected area, facet thickness, lamina length, and cervical range of motion (ROM) were measured. Neck pain visual analogue scale (VAS), arm pain VAS, and neck disability index (NDI) were recorded. Radiographic measures were compared between segments with foraminal narrowing of ≥20% at the 2-years follow-up (restenosis segments) and foraminal narrowing of <20% (patent segments).

Results

Sixty-nine patients with 109 segments were included. 73.4% (80/109) of foramens demonstrated foraminal narrowing and decrease of foraminal dimension of ≥20% occurred in 30.3% (30/109). Foraminal dimension at postoperative 2-days was significantly higher in the restenosis segments (P = .047). Furthermore, increase of foraminal dimension was significantly associated with foraminal restenosis of ≥20% (P = .018). Facet thickness was significantly higher in the restenosis segments compared to patent segments at postoperative 2-years follow-up (P = .038). Neck pain VAS was significantly aggravated only in the restenosis group at postoperative 2-years follow-up (P < .001).

Conclusions

Foraminal narrowing commonly occurs after PCF due to bone healing. Bone growth occurs in all directions while medial facet growth contributes more to foraminal restenosis. Greater widening of foramen during PCF is a risk factor for postoperative foramen restenosis. Therefore, amount of bone resection should be kept optimal and excessive resection should be avoided to prevent foramen restenosis.

Keywords

cervical radiculopathy posterior cervical foraminotomy restenosis segmental instability facet joint foraminal dimension

Introduction

Posterior cervical foraminotomy (PCF) is an effective surgical method for the treatment of cervical radiculopathy which revision rate and clinical efficacy are reported to be similar with that of anterior cervical discectomy and fusion (ACDF).^1-4 It also can avoid troublesome postoperative complications of ACDF including pseudarthrosis, adjacent segmental disease, loss of motion segments, and postoperative dysphagia.^5,6 Furthermore, recent reports have demonstrated that PCF results in shortened hospital stay and reduced blood loss or cost compared to ACDF which makes PCF a widely applied surgical option for cervical radiculopathy.^5,7

While many studies have been conducted regarding radiographic outcome of ACDF including fusion, subsidence, and pseudarthrosis, not much have been reported about what happens to the foramens that underwent PCF.^8-12 Fractured or resected bone has potential to heal, and it remains unclear whether decompressed foramen would remain patent permanently or restenosis would occur due to regrowth of facet and lamina.¹³ It has been reported that interlaminar bone fusion occurs after laminoplasty due to bone healing process although laminoplasty is originally known as a non-fusion, motion preservation technique.^14,15 Spontaneous lamina recovery has also been described for lumbar posterior decompression.^16,17 There is a possibility that similar, unintended bone healing phenomenon would occur after PCF.

While reviewing postoperative CT images of patients who underwent PCF, we frequently observed that previously resected lamina and facet had been spontaneously healed during the follow-up.¹¹ Since bone growth into the resected foramen is an unintended process after PCF, it might adversely affect clinical outcome or increase chance of radiculopathy recurrence. Whereas, it may have beneficial effect such as decreasing chance of segmental instability after PCF. Therefore, the present study was conducted to (1) describe incidence and pattern of bone healing after PCF; (2) demonstrate how such bone healing or restenosis affects clinical and radiographic outcome; and (3) elucidate factors affecting foraminal restenosis.

Materials and Methods

Study Design and Participants

This retrospective cohort study was approved by the institutional review board (Asan Medical Center Institutional Review Board, S2018-1580-0001). Informed consent was waived due to the retrospective nature of the study. The study was conducted in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement for cohort studies.

Ninety-eight patients who underwent PCF for the treatment of cervical radiculopathy between March 2014 and December 2018 were retrospectively reviewed. The inclusion criteria were (1) radiculopathy due to foraminal stenosis; (2) who underwent 1-,2-, or 3-level PCF; and (3) who were followed-up for >2 years. Patients who had (1) previous cervical spine operations; or (2) surgery due to infection, trauma, neoplasm, or ossification of the posterior longitudinal ligament were excluded (Figure 1).

Figure 1.

Patient selection process.

Segments which demonstrated narrowing of foraminal dimension of ≥20% at the 2 years follow-up compared to postoperative 2 days CT were classified as restenosis segments. Segments with no foraminal narrowing or narrowing of <20% at the 2 years follow-up compared to postoperative 2 days CT were defined as patent segments. A cutoff value of 20% was chosen based on the mean value of the foraminal narrowing rate demonstrated in the present study (20.0% ± 16.9%). Patients who had at least one restenosis segment were classified as restenosis group, while patients with no restenosis segment were classified as patent group.

Surgical Techniques

After confirming the operative level with lateral fluoroscopy, midline unilateral dissection is performed to expose indicated levels while care is taken not to injure facet joint capsule. Keyhole foraminotomy including partial laminectomy and partial resection of medial facet joint is performed using high-speed burr and Kerrison punch. The extent of laminectomy and facetectomy differs according to the locations of the disc material and foraminal stenosis. However, attention is paid not to resect greater than 50% of the facet to avoid iatrogenic instability.^18-20 Adequate decompression is confirmed by visualization of the nerve root and palpation of the foramen by angled curette.

Variables and Outcome Measures

Patient reported outcome measures including neck pain visual analogue scale (VAS), arm pain VAS, and neck disability index (NDI) were recorded preoperatively and at postoperative 3 months and 2 years.

CT was taken preoperatively for operative planning, 2 days after the operation for the detection of immediate complication and determine adequate decompression, and 2 years after the operation for the evaluation of delayed complications. Foraminal dimension was measured at each follow-up period using area measuring function of PACS software (PetaVision for Clinics, 3.1, Asan Medical Center, Seoul, Korea) in a sagittal image showing the inner margin of pedicles of the superior and inferior vertebrae (Figure 2A). Postoperative foraminal dimension was measured respecting the margin of remaining inferior articular process (Figure 2B). Sagittal gap, facet thickness, and lamina length of the index side were also measured in postoperative 2 days and 2 years CT to evaluate bone growth pattern. Facet thickness was measured by the length between medial and lateral margin of remained facet in the axial CT image targeting the middle of bone resection part (Figure 2C). Lamina length was measured as a remained lamina length after resection in the same axial CT image (Figure 2C). Sagittal gap between the remained superior articular process of inferior vertebra and inferior articular process of superior vertebra was also measured in the same sagittal image (Figure 2D). Dynamic cervical radiographs were taken preoperatively and at each postoperative follow-up period. Segmental range of motion (ROM) was measured as a change in the angle between the upper and the lower end plate of the respective disc on dynamic radiographs. Global cervical ROM was measured as the change in the angle between the lower margin of C2 and lower margin of C7 on dynamic radiographs.

Figure 2.

Radiographic measurement (A) Foraminal dimension measurement was performed by drawing lines along the margin of neural foramen (red lines). Area was automatically calculated by the image processing software. (B) Foraminal dimension measurement in postoperative images was performed respecting the margin of remaining inferior articular process (red lines). (C) Facet thickness was measured by the length between medial and lateral margin of remained facet (dashed line). Lamina length was measured as a remained lamina length after resection in the same axial CT image (straight line). (D) Sagittal gap was measured by the distance between the remaining superior articular process of inferior vertebra and inferior articular process of superior vertebra (red line).

Statistical Analysis

All statistical analyses were performed using SPSS software (version 22.0; SPSS, inc., Chicago, IL). The level of significance was set at P < .05. Comparison between restenosis segments and patent segments and also between restenosis group and patent group was performed using independent t-test and chi-square test. Parameters at different time periods were compared using a paired t-test. We conducted logistic regression analysis to identify risk factors for foraminal restenosis. Two orthopedic surgeons performed radiographic measurements. The intraclass correlation coefficient (ICC) was used to define interobserver reliability. The ICC for interobserver reliability was .881, .832, .795, and .827 for the measurement of foraminal dimension, sagittal gap, facet thickness, and lamina length. The ICC for segmental or global cervical ROM was .833.

Results

Total of 69 patients and 109 segments met the inclusion criteria and were included in the study (age, 61.1 ± 10.5 years; 48 male [69.6%]; follow-up rate, 74.2% [69/93]). Among the 109 segments analyzed, 80 (73.4%) demonstrated a narrower foraminal dimension at the 2-year follow-up compared to postoperative 2 days with a mean decrease rate of 20.0 ± 16.9%. Thirty-three segments (30.3%) demonstrated decrease of foraminal dimension of ≥20% and were classified as restenosis segments (foramen dimension decrease rate, 33.5 ± 17.1%) while remaining 76 segments (69.7%) were classified as patent segments (foramen dimension decrease rate, .4 ± 14.8%). 61.8% (47/76) segments that were classified as patent segments also demonstrated narrowing of foramen. Thirty-five patients (50.7%) had at least one restenosis segment and were classified as restenosis group (age, 61.6 ± 11.1 years; 24 male [68.6%]) and other 34 patients (49.3%) were classified as patent group (age, 60.7 ± 10.1 years; 24 male [70.6%]) (Table 1). All segments demonstrated foraminal stenosis due to uncinate process and facet joint hypertrophy on preoperative CT. Disc excision was performed in 3 patients (4.3%) and in 3 segments (2.7%). However, segments that required disc excision also had combined foraminal stenosis.

Table 1.

Patient Characteristics.

	Patent group	Restenosis group	P-value
N	34	35
Age	60.7 ± 10.1	61.6 ± 11.1	.780
Sex
Male	24 (70.6%)	24 (68.6%)	1.000
Female	10 (29.4%)	11 (32.4%)	1.000
BMI (kg/m²)	24.7 ± 2.7	24.7 ± 3.4	.971
DM	12 (35.3%)	9 (25.7%)	.387
HTN	11 (32.4%)	13 (37.1%)	.676
Smoking status	9 (25.0%)	7 (16.0%)	.524
Number of levels	1.8 ± .69	1.6 ± 0.8	.437
Follow-up period (m)	38.6 ± 25.8	37.0 ± 22.6	.828

n, numbers; BMI, body mass index; DM, diabetes mellitus; HTN, hypertension, m; months.

Age, BMI, number of levels, and follow-up period were analyzed using an independent t-test.

Sex, DM, HTN, and smoking status were analyzed using a chi-square test.

Demographic factors did not demonstrate significant difference between the patent group and the restenosis group (Table 1). Two patients (5.7%) in the restenosis group and 2 patients (5.8%) in the patent group underwent revision ACDF during the follow-up. Other complications including infection, neurologic deficit, or dura tear did not occur. No postoperative segmental instability or listhesis developed during the follow-up.

Radiographic Results

Preoperative foraminal dimension did not differ significantly between the patent segments and the restenosis segments (P = .928). Foraminal dimension increased significantly after the operation in both groups (P < .001, <.001, respectively). Increase rate was 56.3 ± 44.2% for the patent segments and 73.6 ± 48.1% for the restenosis segments with statistically significant difference (P = .044). Foraminal dimension significantly decreased at 2 years postoperative period compared to postoperative 2 days measurement only in the restenosis segments (P < .001) and was maintained in the patent segments (P = .136). Foraminal dimension at the 2 years follow-up was significantly greater in the patent segments compared to the restenosis segments (P < .001). Foraminal dimension at postoperative 2 years was significantly greater than preoperative values in both groups (P < .001, P = .034, respectively) (Table 2) (Figure 3A).

Table 2.

Radiographic Measurements.

		Patent segments	Restenosis segments	P-value^‡
		(n = 76)	(n = 33)	P-value^‡
Preoperative	Foraminal dimension	33.9 ± 10.5	34.1 ± 12.6	.928
Postoperative 2 days	Foraminal dimension	50.1 ± 12.2	55.9 ± 17.5	.047*
	P-value (pre- 2d)^†	<.001*	<.001*
	Sagittal gap	5.8 ± 1.7	5.6 ± 1.1	.605
	Facet thickness	7.1 ± 1.8	7.7 ± 1.7	.132
	Lamina length	15.9 ± 3.2	15.8 ± 3.5	.862
Postoperative 2 years	Foraminal dimension	48.9 ± 10.0	37.7 ± 14.6	<.001*
	P-value (2d-2y)^†	.136	<.001*
	Sagittal gap	2.6 ± 2.0	2.1 ± 2.0	.19
	P-value (2d-2y)^†	<.001*	<.001*
	Facet thickness	9.9 ± 2.1	10.9 ± 3.2	.038*
	P-value (2d-2y)^†	<.001*	<.001*
	Lamina length	20.4 ± 3.9	20.4 ±. 4.2	.957
	P-value (2d-2y)^†	<.001*	<.001*
Cervical ROM	Preoperative	41.0 ± 8.5	40.4 ± 11.6	.841
	Postoperative 3m	35.0 ± 10.7	34.6 ± 11.7	.902
	Postoperative 2y	35.3 ± 10.9	34.2 ± 8.9	.691
Segmental ROM	Preoperative	3.7 ± 3.0	3.2 ± 3.9	.408
	Postoperative 3m	2.8 ± 2.7	3.1 ± 3.7	.655
	Postoperative 2y	2.5 ± 2.3	1.9 ± 2.5	.259

n, numbers; d, days; m, months; y, years; ROM, range of motion.

†Comparison between 2 different periods was performed by paired t-test.

‡Comparison between 2 groups was performed by independent t-test.

*P < .05.

Figure 3.

Radiographic results and patient reported outcome measures (A) Foraminal dimension, (B) cervical lordosis, (C) segmental lordosis, (D) neck pain visual analogue scale, (E) arm pain VAS, and (F) neck disability index. * Statistically significant difference between the 2 groups. ^† Statistically significant change compared to preoperative period. ^‡ Statistically significant change compared to the postoperative 2 days period.

Sagittal gap decreased significantly at the 2 years follow-up compared to postoperative 2 days measurement in both groups (P < .001, .001, respectively). Furthermore, facet thickness (P < .001, <.001, respectively) and lamina length (P < .001, <.001, respectively) also demonstrated significant increase at the 2 years follow-up measurement. Facet thickness was significantly greater in the restenosis segments compared to patent segments at the 2-years follow-up (P = .038), while sagittal gap (P = .190) and lamina length (P = .957) did not significantly differ between the 2 groups (Table 2).

Global cervical ROM significantly decreased after the operation in both groups (patent segments, P = .003; restenosis segments, P = .008). Furthermore, cervical ROM at the 2 years postoperative period was significantly lower than preoperative period in both groups (patent segments, P = .012; restenosis segments, P = .002). Cervical ROM at each follow-up point did not significantly differ between the 2 groups (Figure 3B). Segmental ROM significantly decreased after the operation, and segmental ROM at postoperative 2-years was significantly lower than the preoperative values in both groups (patent segments, P < .001; restenosis segments, P = .047). Segmental ROM at each follow-up period also did not demonstrate significant intergroup difference (Table 3) (Figure 3C).

Table 3.

Patient Reported Outcome Measures.

		Patent group	Restenosis group	P-value
Neck pain VAS	Preoperative	4.8 ± 2.9	3.8 ± 2.7	.211
	Postoperative 3m	1.7 ± 2.1	1.2 ± 1.6	.368
	Postoperative 2y	2.2 ± 1.9	3.2 ± 2.5	.141
Arm pain VAS	Preoperative	6.1 ± 2.0	5.9 ± 2.4	.687
	Postoperative 3m	2.0 ± 2.3	2.5 ± 2.3	.505
	Postoperative 2y	3.0 ± 2.4	3.6 ± 2.6	.407
NDI	Preoperative	12.1 ± 5.0	11.8 ± 6.2	.852
	Postoperative 3m	8.5 ± 7.1	8.2 ± 5.9	.880
	Postoperative 2y	6.1 ± 5.1	8.8 ± 6.9	.149

VAS, visual analogue scale; NDI, neck disability index; m, months; y, years.

Statistical analyses were performed by using an independent t-test.

Patient Reported Outcome Measures

Preoperative neck pain VAS, arm pain VAS, and NDI did not significantly differ between the 2 groups (P = .211, P = .687, P = .852, respectively). Neck pain VAS significantly improved after the operation in both groups (P < .001, <.001, respectively). However, neck pain aggravated at the 2 years postoperative period only in the restenosis group (P < .001), while it did not change in the patent group (P = .303). Neck pain at the postoperative 2 years follow-up remained significantly improved compared to preoperative scores in the patent group (P = .001), while it did not demonstrate significant difference in the restenosis group (P = .491). Neck pain VAS at each follow-up point was not significantly different between the 2 groups (Table 3) (Figure 3D).

Arm pain VAS also significantly improved after the operation in both groups (P < .001, <.001, respectively). Arm pain did not aggravate significantly at the 2 years follow-up in both groups (patent group, P = .060; restenosis group, P = .010) and remained improved compared to the preoperative scores in the 2 groups (patent group, P < .001; restenosis group, P = .004). Arm pain VAS at each follow-up period did not significantly differ between the 2 groups (Table 3) (Figure 3E).

Neck disability index significantly improved after the operation in both groups (patent group, P = .048; restenosis group, P = .008). Neck disability index at the postoperative 2 years was significantly less than the preoperative values in both groups (patent group, P < .001; restenosis group, P = .042). Furthermore, NDI at each follow-up point was not significantly different between the 2 groups (Table 3) (Figure 3F).

Factors Associated With Foramen Restenosis

Among perioperative demographic and radiographic parameters, increase of foraminal dimension was significantly associated with foraminal restenosis of ≥20% in the logistic regression analysis (odds ratio, 1.044; 95% confidence interval, 1.008–1.082; P = .018) (Table 4).

Table 4.

Results of Logistic Analysis Demonstrating Factors Related to Foraminal Restenosis.

	Odds ratio	95% confidence interval	P-value
Age	1.008	.955–1.064	.775
Sex	.875	.259–2.955	.830
Number of levels operated	.738	.348–1.564	.428
BMI	.966	.824–1.205	.970
Smoking status	.571	.139–2.348	.438
Preoperative foraminal dimension	1.002	.965–1.040	.927
Preoperative segmental ROM	.946	.829–1.079	.406
Postoperative foraminal dimension	1.030	1.000–1.061	.052
Increase of foraminal dimension	1.044	1.008–1.082	.018*
Postoperative segmental ROM	1.031	.901–1.180	.652
Postoperative sagittal gap	.928	.701–1.229	.602
Postoperative facet thickness	1.197	.946–1.513	.134
Postoperative lamina length	.989	.871–1.122	.860

BMI, body mass index; ROM, range of motion.

*P < .05.

Discussions

Posterior cervical foraminotomy has proven to be an effective alternative of ACDF for the treatment of cervical radiculopathy due to its minimal invasiveness, safety, and potential to preserve motion.^2,5,21 However, while patient symptoms have been proved to be favorable in long-term after PCF, not much is known regarding radiographic outcome.¹⁹ Mizouchi et al¹¹ reported that foramen reclosure occurs for patients who underwent concomitant PCF and laminoplasty. However, radiographic fate of decompressed foramens that underwent PCF alone is unknown. Therefore, the present study aimed to demonstrate foramen reclosure after PCF and its clinical implications.

The present study demonstrated that foraminal narrowing occurred in 73.4% of operated segments. Even in segments that were classified as patent segment, 61.8% demonstrated mild narrowing of foramen. Therefore, it seems that narrowing of foramen due to bone regrowth is a common phenomenon after PCF.¹¹ Furthermore, since sagittal gap, facet thickness, and lamina length at postoperative 2 years demonstrated significant difference compared to postoperative 2 days measurements, the bone regrowth seems to occur in both sagittal and axial directions by growth of inferior articular process, medial facet, and lamina. However, medial facet regrowth would be a factor that contributes more to foraminal narrowing since facet thickness at postoperative 2 years demonstrated significant difference between the 2 groups. Mizouchi et al¹¹ also demonstrated that medial facet regrowth is a major factor that causes foraminal narrowing after PCF.

Another interesting finding is that it was not the foramens that underwent less widening that demonstrated restenosis, but the foramens that had greater widening tend to have foraminal restenosis. Furthermore, logistic regression analysis also demonstrated that increase of foraminal dimension is significantly related to foraminal restenosis. Chen et al¹⁶ demonstrated that segmental instability is a factor that accelerates bone regrowth after lumbar decompression. Biomechanical studies also have reported that dilatational mechanical strain is a factor that promotes bone healing.^22-24 Increased micromotion due to more bone loss in the restenosis segments might have resulted in accelerated bone growth during the follow-up. However, since postoperative segmental ROM did not demonstrate significant intergroup difference, this suggestion would need further verification. Therefore, although sufficient decompression according to the location and size of the pathology should be performed, additional excessive resection of facet and lamina would not be helpful since it would increase the chance of foramen narrowing in long-term. Many previous reports have suggested that at least 50% of facet joint should be preserved to preserve cervical stability.^18,25,26 The result of the present study supports this suggestion in the point that excessive resection would lead to postoperative foraminal narrowing.

Segmental ROM significantly decreased both in the restenosis segments and the patent segments at 2-years follow-up. Cervical ROM also demonstrated significant decrease in both groups. Cho et al²⁷ also demonstrated that segmental ROM decreases after PCF although some degree of motion is preserved compared to ACDF. Since bone regrowth and foraminal narrowing is a common phenomenon as demonstrated in this study, bone healing at resected foramen could be a factor that causes ROM decrease. Segments that complete fusion have occurred due to bone healing were occasionally observed, which would have negative impact on cervical motion (Figure 4). Furthermore, while previous studies suggest that segmental instability can occur after PCF, the result of present study implies that segmental instability would not be a significant problem after PCF in long-term due to bone healing.^25,26

Figure 4.

Illustrative case 64-year-old male who was diagnosed with cervical radiculopathy underwent left side posterior cervical foraminotomy, C5-6-7-T1. (A) Postoperative 2 days CT demonstrates resected articular processes at C6-7. However, complete bone fusion occurred at C6-7 in postoperative 2 years CT (red circles). (B) 3D-reconstructed CT also demonstrates that complete bone fusion at C6-7 occurred at postoperative 2 years although adequate decompression was demonstrated at postoperative 2 days CT (red arrows).

Although arm pain VAS and NDI demonstrated significant improvement at 2-years follow-up compared to preoperative scores both in the patent group and the restenosis group, neck pain VAS demonstrated different trend in the 2 groups. Restenosis group demonstrated significant aggravation of neck pain during the follow-up, and neck pain VAS at the 2-years follow-up was not significantly different from that of preoperative values. Therefore, restenosis of foramens would be a factor that adversely affects clinical outcomes. While arm pain is the main symptom of radiculopathy caused by foraminal stenosis, arm pain did not aggravate even in the restenosis group.¹ Dynamic neural foramen narrowing by segmental motion is closely related to radiculopathy, and recurrence of arm pain might not have occurred due to postoperative decrease in segmental motion despite foraminal restenosis.²⁸ Furthermore, foraminal dimension at the final follow-up was still greater than the preoperative value in both groups. Therefore, arm pain recurrence due to foramen restenosis would not be of great concern. However, the present study failed to demonstrate significant inter-group differences of patient reported outcome measures between the patent group and the restenosis group and no significant results were demonstrated with arm pain VAS or NDI. This might be due to small sample size at each group which warrants further clarification.

The present study involved a patient group who underwent PCF using a mini-open approach. Previous studies comparing open and minimally invasive approaches using tubular retractors demonstrated no significant difference in long-term outcomes between the 2 techniques, including neck and arm pain.^19,29,30 While an additional muscle dissection is performed with open PCF, the amount of facet capsule injury and bone resection are similar between the 2 techniques.^19,29,30 Therefore, foraminal restenosis is expected to occur regardless of approach, although further verification is needed. Although Dunn et al² suggested that minimally invasive PCF could prevent undesired bone growth and fusion, this study did not directly compare open and minimally invasive techniques.

In contrast to PCF, no further osteophyte growth or foraminal narrowing occurs after ACDF.^31-33 ACDF eliminates segmental motion, which resists further degeneration and osteophyte enlargement, demonstrating a distinct radiographic course with PCF.^31-33 Since PCF is a motion preservation surgery, it carries the potential for further degeneration and foraminal restenosis.¹¹ Furthermore, the present study’s findings suggest the possibility that foraminal restenosis after PCF might be related to the aggravation of neck pain. Nevertheless, previous studies have proven that PCF and ACDF are equally valuable treatment options for cervical radiculopathy and that long-term neck and arm pain do not differ significantly between them.^3,5 Therefore, there is no need to avoid PCF and perform ACDF due to fear of foraminal restenosis after PCF, although further verification is needed to clarify whether distinct radiographic prognosis between the 2 surgical techniques would lead to different clinical outcomes.

The present study suggests a few clinical implications when posterior revision surgery is needed for patients who have undergone PCF. Firstly, covering the posterior surface of facet resulting from the restoration of resected inferior articular process might provide enough thickness for the insertion of lateral mass screws. Secondly, recovery of lamina thickness and length would decrease the possibility of hinge fracture when secondary laminoplasty is needed.¹¹ Therefore, posterior cervical fusion or laminoplasty need not be precluded due to previous history of PCF, and preoperative CT evaluation would be needed to confirm whether revision surgery could be safely performed (Figure 5).

Figure 5.

Feasibility of posterior revision surgery after posterior cervical foraminotomy 3-dimensional reconstructed CT images of a 56-years-old male who underwent right side posterior cervical foraminotomy, C5-6-7. (A) Postoperative 2-day CT. The generation of lateral slits for laminoplasty (red area marked a) would increase the risk of hinge fracture due to the resected lamina by foraminotomy. The space for inserting lateral mass screws would also be limited due to the removed facet (blue lines). (B) Postoperative 2-year CT. Due to bone healing of the lamina, the lateral slits for laminoplasty (red area marked b) could be performed without the risk of hinge fracture. Furthermore, lateral mass screws could be inserted with more ease due to the recovery of the lateral mass posterior surface (blue lines).

There are a few limitations in this study. Firstly, the measurement in the present study is based on 2-dimensional CT images only. Further evaluation using 3-dimensional volumetric approaches for measurement of bone growth could provide further information. Secondly, as previously mentioned, the study is with relatively small sample size in the patent group and the restenosis group which limits its efficacy to demonstrate statistical significance of patient reported outcome measures. Cervical radiculopathy caused by foraminal stenosis is more directly related to arm pain rather than neck pain, although the result of present study demonstrated significant result only for neck pain. While the present study suggests the possibility that foraminal restenosis after PCF is a factor that could adversely affect patient symptoms, further verification with larger sample size would be needed. Finally, the cutoff value of 20% was selected based on the mean value of foraminal restenosis rate demonstrated in the current study. Since not many studies have described foraminal restenosis phenomenon after PCF, we could not find previous reports that could provide reference point for analysis.

Conclusions

In conclusion, foraminal restenosis commonly occurs after PCF due to bone healing. Bone growth occurs in all directions while medial facet growth contributes more to foraminal restenosis. Greater widening of foramen during PCF is a risk factor for postoperative foramen restenosis. Therefore, amount of bone resection should be kept optimal and excessive resection should be avoided to prevent foraminal restenosis.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Dong-Ho Lee

Jae Hwan Cho

Sehan Park

References

Kang

Lee

. Cervical radiculopathy focus on characteristics and differential diagnosis. Asian Spine J . 2020;14(6):921-930. doi:10.31616/asj.2020.0647.

Dunn

Moore

Sahai

, et al. Minimally invasive posterior cervical foraminotomy with tubes to prevent undesired fusion: a long-term follow-up study. J Neurosurg Spine. 2018;29(4):358-364. doi:10.3171/2018.2.SPINE171003.

Kim

Seo

Lee

Bae

. Comparison of anterior cervical foraminotomy and posterior cervical foraminotomy for treating single level unilateral cervical radiculopathy. Spine. 2019;44(19):1339-1347. doi:10.1097/BRS.0000000000003081.

Maduri

Cossu

Aureli

Wüthrich

Bobinski

Duff

. Transtubular anterior cervical foraminotomy for the treatment of compressive cervical radiculopathy: surgical results and complications in a consecutive series of cases. Asian Spine J . 2020;16:673-681. doi:10.31616/asj.2020.0323.

Emami

Coban

Changoor

, et al. Comparing mid-term outcomes between ACDF and minimally invasive posterior cervical foraminotomy in the treatment of cervical radiculopathy. Spine. 2021. Publish Ahead of Print. doi:10.1097/BRS.0000000000004140.

Skovrlj

Gologorsky

Haque

Fessler

Qureshi

. Complications, outcomes, and need for fusion after minimally invasive posterior cervical foraminotomy and microdiscectomy. Spine J. 2014;14(10):2405-2411. doi:10.1016/j.spinee.2014.01.048.

Alvin

Derakhshan

Lubelski

, et al. Cost-utility analysis of 1- and 2-level dorsal lumbar fusions with and without recombinant human bone morphogenic protein-2 at 1-year follow-up. Clin Spine Surg . 2016;29(1):E28-E33. doi:10.1097/BSD.0000000000000079.

Lee

Cho

Hwang

, et al.

What is the fate of pseudarthrosis detected 1 year after anterior cervical discectomy and fusion?

Spine. 2018;43(1):E23-E28. doi:10.1097/BRS.0000000000002077.

Yang

Park

Kim

H-J

Yoon

. Segmental height decrease adversely affects foraminal height and cervical lordosis, but not clinical outcome after anterior cervical discectomy and fusion using allografts. World Neurosurg. 2021;154:e555-e565. doi:10.1016/j.wneu.2021.07.088.

10.

Igarashi

Hoshino

Omori

, et al. Factors influencing interbody cage subsidence following anterior cervical discectomy and fusion. Clin Spine Surg . 2019;32(7):297-302. doi:10.1097/BSD.0000000000000843.

11.

Mizouchi

Watanabe

Izumi

, et al. Quantitative radiographic analysis of foraminal re-stenosis after posterior cervical foraminotomy with laminoplasty. J Clin Neurosci . 2019;67:99-104. doi:10.1016/j.jocn.2019.06.012.

12.

Alhashash

Allouch

Boehm

Shousha

. Results of four-level anterior cervical discectomy and fusion using stand-alone interbody titanium cages. Asian Spine J . 2021. Online ahead of print. doi:10.31616/asj.2020.0463.

13.

Cottrell

Turner

Arinzeh

O'Connor

. The biology of bone and ligament healing. Foot Ankle Clin. 2016;21(4):739-761. doi:10.1016/j.fcl.2016.07.017.

14.

Lee

Cho

Hwang

Lee

Cho

. Can C3 laminectomy reduce interlaminar bony fusion and preserve the range of motion after cervical laminoplasty? Spine. 2016;41(24):1884-1890. doi:10.1097/BRS.0000000000001852.

15.

Lee

Cho

Shin

Ahn

. Which technique is better option for c3 segment in multilevel open-door laminoplasty of the cervical spine?: laminectomy versus laminoplasty. Spine. 2017;42(14):E833-E840. doi:10.1097/BRS.0000000000001974.

16.

Chen

Baba

Kamitani

Furusawa

Imura

. Postoperative bone re-growth in lumbar spinal stenosis. A multivariate analysis of 48 patients. Spine. 1994;19(19):2144-2149. doi:10.1097/00007632-199410000-00003.

17.

Postacchini

Cinotti

. Bone regrowth after surgical decompression for lumbar spinal stenosis. J Bone Joint Surg Br . 1992;74(6):862-869. doi:10.1302/0301-620X.74B6.1447247.

18.

Lee

Cho

Hwang

Lee

Kim

. Multilevel posterior foraminotomy with laminoplasty versus laminoplasty alone for cervical spondylotic myelopathy with radiculopathy: a comparative study. Spine J . 2018;18(3):414-421. doi:10.1016/j.spinee.2017.08.222.

19.

Platt

Gerard

O'Toole

. Comparison of outcomes following minimally invasive and open posterior cervical foraminotomy: description of minimally invasive technique and review of literature. J Spine Surg . 2020;6(1):243-251. doi:10.21037/jss.2020.01.08.

20.

Dodwad

Prasarn

Savage

Patel

Hsu

. Posterior cervical foraminotomy: indications, technique, and outcomes. Clin Spine Surg . 2016;29(5):177-185. doi:10.1097/BSD.0000000000000384.

21.

Sahai

Changoor

Dunn

, et al. Minimally invasive posterior cervical foraminotomy as an alternative to anterior cervical discectomy and fusion for unilateral cervical radiculopathy: a systematic review and meta-analysis. Spine. 2019;44(24):1731-1739. doi:10.1097/BRS.0000000000003156.

22.

Iobst

Milne

Khoury

Ingwer

Latta

. A novel way to dynamize a spatial frame and optimize fracture healing. Injury . 2021;52(1):106-108. doi:10.1016/j.injury.2020.10.007.

23.

Morgan

Longaker

Carter

. Relationships between tissue dilatation and differentiation in distraction osteogenesis. Matrix Biol . 2006;25(2):94-103. doi:10.1016/j.matbio.2005.10.006.

24.

Steiner

Claes

Ignatius

Niemeyer

Simon

Wehner

. Prediction of fracture healing under axial loading, shear loading and bending is possible using distortional and dilatational strains as determining mechanical stimuli. J R Soc Interface. 2013;10(86):20130389. doi:10.1098/rsif.2013.0389.

25.

Cossu

Messerer

Barges-Coll

. Cervical posterior foraminotomy: how i do it. Acta Neurochir. 2020;162(3):675-678. doi:10.1007/s00701-020-04221-z.

26.

Zdeblick

Abitbol

Kunz

McCabe

Garfin

. Cervical stability after sequential capsule resection. Spine. 1993;18(14):2005-2008. doi:10.1097/00007632-199310001-00013.

27.

Cho

Kim

Park

. Long term effect on adjacent segment motion after posterior cervical foraminotomy. Korean J Spine . 2014;11(1):1-6. doi:10.14245/kjs.2014.11.1.1.

28.

Yeni

Baumer

Oravec

, et al. Correlation of neural foraminal motion after surgical treatment of cervical radiculopathy with long-term patient reported outcomes. J Spine Surg . 2020;6(1):18-25. doi:10.21037/jss.2020.03.02.

29.

Eicker

Rhee

Steiger

Herdmann

Floeth

. Transtubular microsurgical approach to treating extraforaminal lumbar disc herniations. Neurosurg Focus . 2013;35(2):E1. doi:10.3171/2013.4.FOCUS13126.

30.

Winder

Thomas

. Minimally invasive versus open approach for cervical laminoforaminotomy. Can J Neurol Sci . 2011;38(2):262-267. doi:10.1017/s0317167100011446.

31.

Seo

. Fate of posterior osteophytes in fused segments after anterior cervical discectomy and fusion. Spine. 2012;37(9):741-747. doi:10.1097/BRS.0b013e318231d057.

32.

Vodičar

Košak

Vengust

. Long-term results of surgical treatment for symptomatic anterior cervical osteophytes: a case series with review of the literature. Clin Spine Surg . 2016;29(9):E482-E487. doi:10.1097/BSD.0b013e31829046af.

33.

Liu

Luo

Zhou

, et al. Prognosis of posterior osteophyte after anterior cervical decompression and fusion in patients with cervical spondylotic myelopathy using three-dimensional computed tomography study. Eur Spine J . 2016;25(6):1861-1868. doi:10.1007/s00586-016-4390-y.

Foraminal Restenosis After Posterior Cervical Foraminotomy for the Treatment of Cervical Radiculopathy

Abstract

Study design

Objectives

Methods

Results

Conclusions

Keywords

Introduction

Materials and Methods

Study Design and Participants

Surgical Techniques

Variables and Outcome Measures

Statistical Analysis

Results

Radiographic Results

Patient Reported Outcome Measures

Factors Associated With Foramen Restenosis

Discussions

Conclusions

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iDs

References