J Neurosurg Spine 21:727–731, 2014

727

©AANS, 2014

J Neurosurg: Spine / Volume 21 / November 2014

CerviCal spondylosis is a progressive disorder of the aging spine that results predominantly from degeneration of the intervertebral discs and verte-

brae. The formation of “bone spurs” or osteophytes may cause symptomatic compression of the spinal cord and/or nerve roots.4 Surgical intervention is indicated when patients experience neurological deficits or intractable pain,4,12 and dorsal approaches offer direct visualization

of the compressed spinal cord and/or nerve roots. A fo-raminotomy involves widening the foramen and indirect-ly decompressing the nerve roots; this type of procedure is often accompanied by a partial laminotomy and medial facetectomy.

Posterior cervical foraminotomy (PCF) was first described by Scoville in 1946.4 Although it is consid-ered to be a safe procedure with a low complication rate,1,8,11,12,15,17,20,21 radiculopathy symptoms may recur, in some cases necessitating reoperation. Many studies2,3,7,

9,10,13,14,16,18,21,23,24,26 have provided data on outcomes after PCF. In this article, we report the incidence and timing

Long-term patient outcomes after posterior cervical foraminotomy: an analysis of 151 cases

Clinical article

*MohaMad Bydon, M.d.,1,2 diMitrios Mathios, M.d.,1,2 MohaMed Macki, B.s.,1,2 rafael de la Garza-raMos, M.d.,1,2 daniel M. sciuBBa, M.d.,1 tiMothy f. WithaM, M.d.,1 Jean-Paul Wolinsky, M.d.,1 ziya l. Gokaslan, M.d.,1 and ali Bydon, M.d.1,2

1The Spinal Column Biomechanics and Surgical Outcomes Laboratory, and 2Department of Neurosurgery, Johns Hopkins Hospital, Baltimore, Maryland

Object. The authors conducted a study to investigate the rate and timing of reoperation due to symptom recur-rence after unilateral posterior cervical foraminotomy (PCF).

Methods. The authors retrospectively reviewed demographic, surgical, and clinical data from 151 patients who underwent unilateral PCF at their institution with an average follow-up of 4.15 years. The main outcome variables were reoperation rate, time to reoperation, and short- and long-term radiculopathy improvement rates. Kaplan-Meier analyses were conducted to assess risk of reoperation and recurrence of radiculopathy over time.

Results. After index PCF in 151 patients, the overall reoperation rate was 9.9% (15 patients). The average time until reoperation was 2.4 years, and the average last follow-up examination was 4.15 years after the first surgery. Patients who presented with preoperative neck pain in addition to radiculopathy had a higher risk for reoperation and a shorter time to reoperation. The majority of patients who underwent a reoperation had an anterior cervical discectomy and fusion (80%). A smaller number of patients had reoperation that included a repeat PCF (6.7%) or laminectomy with posterior cervical fusion (13.3%). The rate of same-level reoperation (6.6%, 10 patients) was significantly higher (p = 0.05) when compared with adjacent-segment (1.3%, 2 patients) or distant-segment (1.9%, 3 patients) reoperation. At last follow-up, the overall rate of improvement in radiculopathy was 85%, with the majority of patients (91.4%) experiencing resolution as early as 1 month after index surgery. Following the subgroup that ex-perienced initial symptom improvement, 16.1% of these patients experienced radiculopathy recurrence an average of 7.3 years after the initial operation. While the reoperation rate for the overall cohort in this series was 9.9%, patients with follow-up periods longer than 2 years had a reoperation rate of 18.3%. Moreover, patients with more than 10 years of follow-up had a reoperation rate of 24.3%.

Conclusions. PCF is a procedure performed to address nerve root compression in the cervical spine. The authors evaluated 151 patients who underwent unilateral PCF and found a reoperation rate of 9.9% at an average of 2.4 years after the initial surgery (6.6% at same level, 3.3% elsewhere). The reoperation rates reached 18.3% and 24.3% in patients with follow-up periods longer than 2 and 10 years, respectively. The authors’ analysis revealed that patients with no preoperative neck pain had the lowest rates of revision surgery after PCF.(http://thejns.org/doi/abs/10.3171/2014.7.SPINE131110)

key Words      •      cervical      •      degenerative      •      facetectomy      •      foraminotomy      •      laminectomy      •      posterior      •      reoperation      •      spine

Abbreviations used in this paper: ACDF = anterior cervical disc-ectomy and fusion; PCF = posterior cervical foraminotomy.

* Drs. M. Bydon and Mathios contributed equally to this work.

Unauthenticated | Downloaded 11/14/20 12:45 PM UTC

M. Bydon et al.

728 J Neurosurg: Spine / Volume 21 / November 2014

of reoperation after PCF and identify perioperative vari-ables associated with a higher reoperation rate.

MethodsPatient Collection

The study was approved by the Johns Hopkins In-stitutional Review Board. We retrospectively reviewed all PCFs for cervical degenerative disease performed by full-time neurosurgeons at our institution between January 1990 and January 2013. Our source population included 225 patients who underwent PCF, and we ex-cluded 6 cases with fewer than 3 months of follow-up. Of the 219 patients with follow-up of at least 3 months, we excluded 68 patients who underwent bilateral PCFs. Nondegenerative cases, anterior cervical operations, pos-terior fusions (instrumented and noninstrumented), and cases that involved foraminotomy and simultaneous lam-inectomy were also excluded. Thus, our study population was made up of 151 patients who underwent unilateral PCF with at least 3 months of follow-up. Patient data, in-cluding demographic information, presenting symptoms, operative results, and complications, were compiled in a retrospective database. Our main outcome variables were rate of reoperation, time to reoperation, and progression of radiculopathy.

Statistical AnalysisThe patient population was described with summary

statistics.We used the Kaplan-Meier method to assess the time

to reoperation for patients who underwent PCF, and the log-rank (Mantel-Cox) test was used to estimate the p val-ue. Cox proportional hazard ratios were used for our mul-tivariate analysis; we included only variables that were found to have a p value < 0.2 and used a forward-model building approach. A p value ≤ 0.05 was considered statis-tically significant. Principal component analysis was used to assess the variables that created the greatest variability in our data set. We calculated the Pearson correlation ma-trix among preoperative status, operative variables, and postoperative outcome to infer any relationship among the variables. All statistical analyses were performed us-ing Statistica software (StatSoft Inc.).

ResultsA total of 261 PCFs were performed in 151 patients.

Among the 151 patients, the mean age was 56 ± 13.6 years, and 70.8% (n = 107) were male (Table 1). The aver-age number of spinal levels decompressed was 1.7 ± 1, and the most frequent levels were C5–6 (31%) and C6–7 (35.6%) (Fig. 1, Table 2). The majority of the patients were treated for either spondylosis (51.6%) or disc herniation (37.1%), and the PCF was followed by removal of the os-teophyte (7.3%) or disc-osteophyte complex (4%). The preoperative diagnosis was not a predictor of reoperation based on a regression analysis (p = 0.62). The intra- and perioperative characteristics of the patients are described in Table 2.

Reoperation RatesIn an average follow-up period of 4.15 years, the

overall reoperation rate was 9.9% (Table 3, Fig. 2 upper). Kaplan-Meier analysis revealed the rates of reoperation to be 9%, 14.3%, and 24.3% at 1, 5, and 10 years, respec-tively.

The level of reoperation was also analyzed. Of 151 patients, 10 patients (6.6%) required reoperation due to same-level disease (Fig. 2 lower), and 2 patients (1.3%)

TABLE 1: Preoperative characteristics and operative indications for patients undergoing PCF for cervical spondylosis*

Variable Value (%)

no. of cases 151mean age in yrs (± SD) 56 ± 13.6no. of males 107 (70.8)comorbidities diabetes 10 (7) smoking history 35 (23.2) CHF 2 (1.3) CAD 12 (7.9) osteoporosis 2 (1.3) obesity 6 (4)presenting symptoms neck pain 64 (42.4) radiculopathy 151 (100)  motor deficit 75 (49.7)  sensory deficit 59 (39)PCF operative indication disc herniation 56 (37.1) osteophyte 11 (7.3) complex osteophyte–disc herniation 6 (4) spondylosis 78 (51.6)

* Values are number of patients unless otherwise noted. CAD = coro-nary artery disease; CHF = congestive heart failure.

Fig. 1. Distribution of PCFs based on the level of operation. The y-axis values represent the number of procedures.

Unauthenticated | Downloaded 11/14/20 12:45 PM UTC

Outcomes after posterior cervical foraminotomy

729J Neurosurg: Spine / Volume 21 / November 2014

required reoperation due to adjacent-segment disease. This rate was significantly lower when compared with the same-level rate as assessed with a log-rank test (p = 0.05). Lastly, 3 patients (1.9%) required reoperation at nonadja-cent segments.

We calculated the Pearson correlation matrix for our preoperative, demographic, and surgical variables in re-lationship to the time to reoperation. An inverse relation was noticed for patients who presented with neck pain (r = -0.141). This parameter was statistically significant with regard to the time to reoperation (p < 0.05). This indicates that patients who presented with preoperative neck pain required a reoperation sooner than patients without neck pain.

Radiculopathy RecurrenceThe improvement rate of radiculopathy at the 1-month

follow-up was 91.4% in our cohort. At an average last fol-low-up 4.15 years after treatment, 85% of patients expe-rienced radiculopathy improvement. We separately ana-lyzed data for 68 patients from our cohort who had more than 2 years of follow-up. The overall reoperation rate in

this group was 16.4%. Short-term outcome (at 1-month follow-up) revealed symptomatic improvement in 91.4% of patients. Following the subgroup of patients who expe-rienced an initial improvement of symptoms, 16.1% had recurrence of their radiculopathy an average of 7.3 years after the initial operation.

Multivariate Analysis of Reoperation RateWe sought to analyze variables that might confer a

higher risk for reoperation by performing multivariate analysis and only including patients who underwent reop-eration at the index level. All preoperative and operative variables with a p < 0.2 in univariate analysis were in-cluded in the model. We found that patients who present-ed with radiculopathy and preoperative neck pain had a 2.9-fold greater risk of requiring a reoperation after PCF compared with patients who presented with radiculopa-thy alone (p = 0.03).

DiscussionPosterior cervical foraminotomy has been well de-

TABLE 2: Intraoperative and perioperative characteristics of patients undergoing PCF

Variable Value (%)

mean no. of levels (± SD)* 1.7 ± 1.00blood loss (ml) 174intraop durotomy, no. of cases 1 (0.7)postop CSF leak, no. of cases 1 (0.7)mean length of stay in days (± SD) 2.7 ± 2.5discharged to rehab, no. of cases 9 (5.9)

* Mean number of spinal levels decompressed.

TABLE 3: Clinical outcome of patients after PCF and the operative indications for revision surgery

Variable Value (%)

postoperative outcome neck pain improvement 57 (92.5) radiculopathy improved 138 (91.4)  motor deficits improved  49 (65.3)  sensory deficit improved  49 (83) mean yrs of follow-up 4.15revision surgery overall reoperation rate 15 (9.9) ACDF 12 (80) PCF 1 (6.7) posterior cervical fusion 2 (13.3) average yrs to reoperation 2.4indication for revision surgery disc herniation 6 (40) spondylosis 8 (53.3) osteophyte 1 (6.7)

Fig. 2. Upper: Kaplan-Meier curve of patients who underwent PCF, plotting the time to reoperation after the index surgery (Day 0). Low-er: Kaplan-Meier curve showing the percentage of reoperated index-level, adjacent-level, or distant-level cases for patients who underwent initial PCF.

Unauthenticated | Downloaded 11/14/20 12:45 PM UTC

M. Bydon et al.

730 J Neurosurg: Spine / Volume 21 / November 2014

scribed in the literature. The immediate success rates for symptom resolution vary from 82% to 100%,19,25,26 but long-term outcomes are even more variable, with success rates ranging from 75% to 97%.2,3,9,10,13,14,16,18,23,24,26 In this study, we correlated the preoperative symptoms of pa-tients who underwent PCF with postoperative outcomes and documented the rate of and time to reoperation.

In this series, a majority of patients who underwent PCF experienced initial symptom improvement, at least in the short term. In patients with more than 2 years of follow-up, symptom relief reached 84% at the time of last follow-up. The overall rate of reoperation for the total cohort was 9.9%. Davis6 reported a 6% reoperation rate in a study of 170 patients; all of the cases involved re-operation to address pathology at the same level due to symptom recurrence. Wang et al.22 recently reported a 5% reoperation rate at the index level in a cohort of 178 pa-tients treated with PCF. Similarly, in the present study, 10 (6.6%) of 151 patients required reoperation due to same-segment disease, comparable to rates reported in other studies.5,6,15,22 Interestingly, the rates of reoperation due to adjacent-segment disease versus distant-segment disease were not significantly different (p = 0.99), which may be an argument in favor of the natural history of spondylosis.

Patients with symptom recurrence after PCF un-derwent repeat surgery an average of 2.4 years after the index surgery. A Kaplan-Meier analysis showed that 24 patients who did not experience radiculopathy recurrence 10 years after the initial PCF would not have any recur-rence for the rest of their follow-up time (10–22 years from the initial operation).

When analyzing preoperative symptoms and corre-lating them with postoperative outcomes, we found that patients with preoperative neck pain were more likely to require earlier reoperation. In addition, the multivariate analysis of the pre- and intraoperative variables showed that preoperative neck pain was correlated with an in-creased risk of revision surgery. In this study, patients who presented with radiculopathy and neck pain had a higher incidence of reoperation and required reoperation earlier than patients who presented with radiculopathy alone. In our institutional experience, these results have a particularly strong impact on young adults or athletes who opt out of a cervical decompression and fusion in favor of a less invasive foraminotomy. The operation is intended to preserve the cervical range of motion in these active patients. Within the general population, patients may be advised that PCF is a safe procedure associated with favorable outcomes. Patients who have neck pain in addition to radiculopathy may be consulted regarding their elevated risk of reoperation.

Both PCF and anterior cervical discectomy and fu-sion (ACDF) represent surgical options to treat cervical radiculopathy.22 Certain cervical pathologies may ben-efit more from ACDF (e.g., central-lateral disc hernia-tion, bilateral foraminal stenosis, osteophyte anterior to the nerve root, or facet hypertrophy). On the other hand, patients who wish to maintain cervical range of motion may be good candidates for PCF. Moreover, patients who are poor surgical candidates or who are at increased risk for nonunion (e.g., those receiving chronic corticosteroid

treatment or those who smoke) may have improved re-sults with PCF.1,4,9

LimitationsAlthough we presented statistically and clinically sig-

nificant data on patients undergoing PCF, our retrospec-tive study has inherent limitations. Retrospective cohort studies are limited to a single outcome while measuring multiple exposures. Future prospective studies are best suited to thoroughly analyze outcomes following PCF.

ConclusionsPosterior cervical foraminotomy is a procedure that

addresses foraminal nerve root compression. In this series, patients who underwent PCF had an overall 9.9% reopera-tion rate with an average time to reoperation of 2.4 years after the index operation. The rate of reoperation at the in-dex segment was 6.6.%, which was statistically significant-ly higher than reoperation at adjacent or distant segments (p = 0.05). Patients with preoperative neck pain had a high-er risk for reoperation and a shorter time to reoperation. At the average last follow-up of 4.15 years, the overall rate of improvement in radiculopathy was 85%, with the major-ity of patients (91.4%) experiencing resolution as early as 1 month after the index surgery. Following the subgroup of patients who experienced initial symptom improvement, 16.1% had recurrence of their radiculopathy an average of 7.3 years after the initial operation. While the reoperation rate was 9.9% for the overall cohort in this series, patients who had more than 2 years of follow-up had a reoperation rate of 16.4%. Moreover, patients who had more than 10 years of follow-up had a reoperation rate of 24.3%. Accord-ing to the results of this series, the majority of patients ex-perience symptomatic improvement of radiculopathy after PCF; therefore, patients may be advised that PCF is a safe operation with favorable outcomes. Patients who have neck pain in addition to radiculopathy may be consulted regard-ing their elevated risk of reoperation.

Disclosure

The authors report no conflict of interest concerning the mate-rials or methods used in this study or the findings specified in this paper. General disclosures (unrelated to this research) are as follows: Ali Bydon is the recipient of a research grant from DePuy Spine. He serves on the clinical advisory board of MedImmune, LLC. Daniel Sciubba is the recipient of a research grant from DePuy Spine. He has consulting relationships with Medtronic, NuVasive, Globus, and DePuy. Timothy Witham is the recipient of a research grant from Eli Lilly and Company. Ziya Gokaslan is the recipient of research grants from DePuy Spine, AOSpine North America, Medtronic, Neuro-surgery Research and Education Foundation (NREF), Integra Life Sciences, and K2M. He receives fellowship support from AOSpine North America and holds stock in Spinal Kinetics and US Spine. The remaining authors have nothing to disclose.

Author contributions to the study and manuscript preparation include the following. Conception and design: M Bydon. Acquisi-tion of data: Mathios. Analysis and interpretation of data: Mathios, Macki. Drafting the article: M Bydon, Mathios. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: A Bydon. Statistical analysis: Mathios, Macki. Admin-

Unauthenticated | Downloaded 11/14/20 12:45 PM UTC

Outcomes after posterior cervical foraminotomy

731J Neurosurg: Spine / Volume 21 / November 2014

istrative/technical/material support: A Bydon, Sciubba, Witham, Wolinsky, Gokaslan. Study supervision: A Bydon, Sciubba, With-am, Wolinsky, Gokaslan.

References

1. Albert TJ, Murrell SE: Surgical management of cervical ra-diculopathy. J Am Acad Orthop Surg 7:368–376, 1999

2. Aldrich F: Posterolateral microdisectomy for cervical mono-radiculopathy caused by posterolateral soft cervical disc se-questration. J Neurosurg 72:370–377, 1990

3. Cağlar YS, Bozkurt M, Kahilogullari G, Tuna H, Bakir A, Torun F, et al: Keyhole approach for posterior cervical discec-tomy: experience on 84 patients. Minim Invasive Neurosurg 50:7–11, 2007

4. Caridi JM, Pumberger M, Hughes AP: Cervical radiculopa-thy: a review. HSS J 7:265–272, 2011

5. Clarke MJ, Ecker RD, Krauss WE, McClelland RL, Deku-toski MB: Same-segment and adjacent-segment disease fol-lowing posterior cervical foraminotomy. J Neurosurg Spine 6:5–9, 2007

6. Davis RA: A long-term outcome study of 170 surgically treat-ed patients with compressive cervical radiculopathy. Surg Neurol 46:523–533, 1996

7. Dubuisson A, Lenelle J, Stevenaert A: Soft cervical disc her-niation: a retrospective study of 100 cases. Acta Neurochir (Wien) 125:115–119, 1993

8. Epstein NE: A review of laminoforaminotomy for the man-agement of lateral and foraminal cervical disc herniations or spurs. Surg Neurol 57:226–234, 2002

9. Hamburger C, Festenberg FV, Uhl E: Ventral discectomy with PMMA interbody fusion for cervical disc disease: long-term re-sults in 249 patients. Spine (Phila Pa 1976) 26:249–255, 2001

10. Harrop JS, Silva MT, Sharan AD, Dante SJ, Simeone FA: Cer-vicothoracic radiculopathy treated using posterior cervical fo-raminotomy/discectomy. J Neurosurg 98 (2 Suppl):131–136, 2003

11. Heary RF, Ryken TC, Matz PG, Anderson PA, Groff MW, Holly LT, et al: Cervical laminoforaminotomy for the treatment of cervical degenerative radiculopathy. J Neurosurg Spine 11:198–202, 2009

12. Herkowitz HN, Kurz LT, Overholt DP: Surgical management of cervical soft disc herniation. A comparison between the anterior and posterior approach. Spine (Phila Pa 1976) 15: 1026–1030, 1990

13. Jagannathan J, Sherman JH, Szabo T, Shaffrey CI, Jane JA: The posterior cervical foraminotomy in the treatment of cer-vical disc/osteophyte disease: a single-surgeon experience with a minimum of 5 years’ clinical and radiographic follow-up. Clinical article. J Neurosurg Spine 10:347–356, 2009

14. Jödicke A, Daentzer D, Kästner S, Asamoto S, Böker DK: Risk factors for outcome and complications of dorsal forami-notomy in cervical disc herniation. Surg Neurol 60:124–130, 2003

15. Kumar GR, Maurice-Williams RS, Bradford R: Cervical fo-raminotomy: an effective treatment for cervical spondylotic radiculopathy. Br J Neurosurg 12:563–568, 1998

16. Lam S, Khoo LT, Cannestra A, Holly L, Shamie AN, Wang J, et al: A long-term clinical outcome analysis of minimally invasive cervical foraminotomy for the treatment of cervical radiculopathy. Eur Spine J 14 (Suppl 1):S2, 2005 (Abstract)

17. Nikolaidis I, Fouyas IP, Sandercock PA, Statham PF: Surgery for cervical radiculopathy or myelopathy. Cochrane  Data-base Syst Rev (1):CD001466, 2010

18. Onimus M, Destrumelle N, Gangloff S: [Surgical treatment of cervical disk displacement. Anterior or posterior approach?] Rev Chir Orthop Reparatrice Appar Mot 81:296–301, 1994 (Fr)

19. Parker WD: Cervical laminoforaminotomy. J Neurosurg 96 (2 Suppl):254–255, 2002 (Letter)

20. Schebesch KM, Albert R, Schödel P, Proescholdt M, Lange M, Brawanski A: A single neurosurgical center’s experience of the resolution of cervical radiculopathy after dorsal forami-notomy and ventral discectomy. J Clin Neurosci 18:1090–1092, 2011

21. Silveri CP, Simpson JM, Simeone FA, Balderston RA: Cervi-cal disk disease and the keyhole foraminotomy: proven effi-cacy at extended long-term follow up. Orthopedics 20:687–692, 1997

22. Wang TY, Lubelski D, Abdullah KG, Steinmetz MP, Benzel EC, Mroz TE: Rates of anterior cervical discectomy and fu-sion after initial posterior cervical foraminotomy. Spine J [epub ahead of print], 2013

23. Witzmann A, Hejazi N, Krasznai L: Posterior cervical forami-notomy. A follow-up study of 67 surgically treated patients with compressive radiculopathy. Neurosurg Rev 23:213–217, 2000

24. Woertgen C, Holzschuh M, Rothoerl RD, Haeusler E, Brawa-nski A: Prognostic factors of posterior cervical disc surgery: a prospective, consecutive study of 54 patients. Neurosurgery 40:724–729, 1997

25. Woertgen C, Rothoerl RD, Henkel J, Brawanski A: Long term outcome after cervical foraminotomy. J Clin Neurosci 7:312–315, 2000

26. Zeidman SM, Ducker TB: Posterior cervical laminoforami-notomy for radiculopathy: review of 172 cases. Neurosurgery 33:356–362, 1993

Manuscript submitted December 6, 2013.Accepted July 7, 2014.Please include this information when citing this paper: published

online August 15, 2014; DOI: 10.3171/2014.7.SPINE131110.Address correspondence to: Ali Bydon, M.D., Department of

Neurosurgery, The Johns Hopkins Hospital, 600 N. Wolfe St., Meyer 5-109, Baltimore, MD 21287. email: abydon1@jhmi.edu.

Unauthenticated | Downloaded 11/14/20 12:45 PM UTC