- Email: [email protected]
The utility of routine postoperative radiographs after cervical spine fusion
The Spine Journal 13 (2013) 764–769
Clinical Study
The utility of routine postoperative radiographs after cervical spine fusion Bennett D. Grimm, MDa,*, Daniel P. Leas, BSb, John A. Glaser, MDa a
Department of Orthopaedic Surgery, Medical University of South Carolina, 96 Jonathan Lucas St, Ste. 708, Charleston, SC 29425, USA b School of Medicine, Medical University of South Carolina, 96 Jonathan Lucas St, Ste. 708, Charleston, SC 29425, USA Received 27 July 2011; revised 14 October 2012; accepted 13 January 2013
Abstract
BACKGROUND CONTEXT: The senior author (JAG) recently published an article questioning the utility of routine postoperative radiographs after lumbar spine fusion. That study concluded that routine postoperative radiographs in the presence of a normal physical examination rarely change the clinician’s management of these patients. Our aim was to repeat this protocol in patients after cervical spine fusion. We hypothesized that routine postoperative radiographs are unnecessary in most cases after cervical spine fusion. PURPOSE: The purpose of this study was to determine the usefulness of routine postoperative cervical spine radiographs after cervical spine fusion as to whether they help to guide clinical decision making within the first postoperative year. STUDY DESIGN: This is a retrospective chart review of 383 patients who underwent a cervical spine fusion over a 5-year period. Seven different surgeons performed the cervical spine fusions. Our review assessed a total of 1,155 postoperative clinic visits. METHODS: Each clinical postoperative visit was reviewed. The history and exam were graded as either normal or abnormal, and any plain radiographs obtained were graded similarly as either normal or abnormal. Each patient’s notes were followed up to 1 year postoperatively. Each patient had to have at least two postoperative visits with X-rays to be included in the study. We then noted any further action taken by the clinician based on the appearance of the radiograph in conjunction with the history and exam. RESULTS: In patients with normal history and exam presentations, further action was taken only 5/879 (0.57%) of the time, sometimes even in the presence of abnormal radiographs. The actions included two surgical revisions, two prolongations of cervical collar immobilization, and one patient who underwent a flexion/extension radiographic evaluation and subsequent prolonged cervical collar immobilization. There were 276 visits with abnormal history and exam; of these, 34/276 (12.3%) had abnormal X-rays. Of the clinic visits with abnormal history and exam and abnormal X-rays, 15/34 (44%) went on to revision. CONCLUSIONS: Routine postoperative radiographs after cervical spine fusion rarely appear to be of value when patients present with a normal history and exam and may expose patients to unnecessary diagnostic studies and expenses. Patients exhibiting a normal postoperative history and exam are likely to have no further additional action taken in the presence of either normal or abnormal radiographs. Ó 2013 Elsevier Inc. All rights reserved.
Keywords:
Cervical spine; X-rays; Postoperative care
Introduction FDA device/drug status: Not applicable. Author disclosures: BDG: Nothing to disclose. DPL: Nothing to disclose. JAG: Trips/Travel: SI Bone (A, Paid directly to institution/employer); Grants: SI Bone (D, Paid directly to institution/employer). The disclosure key can be found on the Table of Contents and at www. TheSpineJournalOnline.com. * Corresponding author. Department of Orthopaedic Surgery, Medical University of South Carolina, 96 Jonathan Lucas St, Suite 708, Charleston, SC 29425, USA. Tel.: (843) 792-0245; fax: (843) 792-3674. E-mail address: [email protected] (B.D. Grimm) 1529-9430/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.spinee.2013.01.018
Cervical spine fusion via an anterior, posterior, or combined approach with concomitant decompression as indicated reliably alleviates symptoms from radiculopathy/ myelopathy and can impart stability after trauma or destabilizing cervical spine lesions [1–8]. It is a standard protocol for many spine surgeons to obtain routine postoperative radiographs at distinct intervals (typically at 2 and 6 weeks, 3 months, 6 months, and 1 year
B.D. Grimm et al. / The Spine Journal 13 (2013) 764–769
postoperatively) whether or not the patient is symptomatic. The rationale for obtaining routine postoperative X-rays is threefold: they give the surgeon and patient piece of mind that the patient is healing and the surgery was a success; so that asymptomatic/symptomatic hardware migration or failure may be identified; and routine surveillance radiographs may help to protect the physician from liability via documentation of the patient’s clinical status in the medical record. The senior author (JAG) concluded in a study that routine postoperative radiographs after lumbar spine fusion almost never altered the course of treatment [9]. There has not been, to our knowledge, an investigation into the necessity of obtaining routine postoperative radiographs after cervical spine fusion for one or more levels. By assessing the utility in decision making of such routine imaging in postoperative decision making, physicians may avoid exposing the patient to unnecessary charges and radiation. Materials and methods On approval by our institutional review board, we began a retrospective chart review of 686 patients who underwent cervical fusion for any reason between January 1, 2002, and December 31, 2007 (Table 1). For inclusion into this study, patients had to be a minimum 18 years of age and have at least two postoperative visits with radiographs within 12 months after the index procedure. This narrowed our cohort to 383 patients, totaling 401 procedures including 18 revision surgeries (2 on the same patient) with an average of 3 postoperative visits. We included patients who had undergone anterior cervical fusion, posterior fusion, or combined anterior/posterior fusion. Procedures were performed by a total of seven surgeons: six neurosurgeons and one spine fellowship–trained orthopedic surgeon. One hundred ninety-nine patients were men and 184 patients were women. Patient ages ranged from 18 to 84 years, and the numbers of fused levels were from 1 to 6. An anterioronly approach alone was used 350 times, 40 through a posterior approach only, and 10 via a staged anterior and posterior approach (Table 2). A senior-level orthopedic resident with the assistance of a medical student reviewed each postoperative clinic visit note. A qualifying clinic visit was defined as having a history and exam as well as an anterior-posterior and lateral X-ray of the cervical spine. The history and exam were determined to be either normal or abnormal based on the clinician’s interpretation of the patient’s symptoms. A normal history and exam (H/E NL) included only findings that were deemed appropriate during the course of recovery. An abnormal history and exam (H/E NOT) was defined as pain greater than would be expected by the clinical situation or an unresolved or new onset neurologic deficit. The radiographs were also reviewed and graded according to the clinician’s note (or the radiologist’s interpretation if the clinician did not comment on the X-ray) as normal
765
Context Postoperative radiographs are commonly obtained at follow-up visits after cervical fusion. Contribution Based on this retrospective review, the authors found that only 5 of 879 patients with normal history and physical exams at follow-up visits had abnormal X-ray findings (displacement of graft, plate, or pseudarthrosis) requiring action on the part of the treating surgeon (revision surgery, prolonged brace use). Implication The findings are interesting and suggest changes in current common practice. However, any change would have to consider gradual failure of graft and/or instrumentation that remains asymptomatic until catastrophic failure and catastrophic morbidities occur (a ‘‘threshold’’ phenomenon). What is the rate of these uncommon events? What rate is acceptable to miss? Are there particular clinical circumstances (eg, infection, tumor, deformity) with greater risk of progressive asymptomatic failure? —The Editors
(X-ray NL) or abnormal (X-ray NOT). Abnormal radiographic findings included failure of fixation, increasing listhesis/kyphosis, motion between fused segments, or any change from the baseline postoperative X-ray. We then determined whether or not the clinician took action based solely on the appearance of the routine radiographs (meaning did the clinician change management of the patient if they had a normal exam but an abnormal routine X-ray), such as additional imaging or reoperation. Chi-square analysis was used to analyze the data.
Results The clinic notes of 1,155 visits were reviewed and graded. A total of 879 visits were graded as H/E NL and 276 visits were graded as H/E NOT. Radiographs were graded as X-ray NL for 1,090 visits and as X-ray NOT in 65 cases. These data were organized into four categories (Table 3). No action was taken 100% of the time, when clinic visits Table 1 Pathologic condition leading to fusion Pathology
No. of patients
Degenerative Traumatic Other
302 70 11
766
B.D. Grimm et al. / The Spine Journal 13 (2013) 764–769
Table 2 Demographics of patients Demographics Age (y) 18–35 36–50 51–65 66–80 81þ Total number of patients Levels of fusion per procedure 1 2 3 4þ Total number of procedures Ethnicity Caucasian African-American Hispanic Asian
No. of patients 45 142 149 45 2 383 135 160 79 27 401 270 107 3 3
were graded as H/E NL and X-ray NL. In patients graded as H/E NL and X-ray NOT, no additional action was taken 26/ 31 or 84% of the time. A brief discussion of the five patients with a normal exam, but an abnormal X-ray, is warranted. Patient 1, an 80-year-old man, underwent a C4–C6 anterior decompression and fusion with a fibular allograft and plate fixation for myelopathy. At his 1-month postoperative visit, the lateral radiograph showed that the plate and screws had dissociated from the bone superiorly (Fig. 1, Left). He did not complain of any undue pain at this time and actually felt well enough to go on an extended vacation. The treating surgeon, however, felt that the instrumentation migration was unacceptable, and the patient underwent a revision procedure shortly thereafter (Fig. 1, Right). He subsequently went on to an asymptomatic fusion. Patient 2, a 53-year-old man, with a three-pack per day smoking history had a previous C1–C2 fusion (Fig. 2, Left) 1 year before presented to our institution complaining of persistent neck pain and myelopathic symptoms. He was found to have a computed tomography scan proven pseudoarthrosis and underwent a revision with C-1 lateral mass screws plus translaminar and pedicle screws at C-2 supplemented with autogenous iliac crest bone graft (Fig. 2, Middle). Approximately 2.5 months postoperatively, flexion/ extension radiographs showed a persistent pseudoarthrosis with a 10-mm atlantodens interval apparent on dynamic
Table 3 Categorized patient data Category
No. of visits
H/E H/E H/E H/E
848 31 242 34
NL and X-ray NL NL and X-ray NOT NOT and X-ray NL NOT and X-ray NOT
H/E NL, a normal history and exam; H/E NOT, an abnormal history and exam; X-ray NL, a normal X-ray; X-ray NOT, an abnormal X-ray.
lateral radiographs despite having an H/E NL. He subsequently underwent a C1–C4 posterior revision procedure with supplemental halo fixation (Fig. 2, Right). There were three other patients who had further action taken in the presence of an X-ray NOT but an H/E NL. One patient had a 3.5-mm subluxation on a lateral X-ray after a one-level anterior cervical fusion with plate fixation because of instability from trauma. This radiographic finding prompted the treating physician to order prolonged cervical collar immobilization and a flexion/extension X-ray to determine the stability of the construct. Another patient had 2 to 3 mm of plate/bone separation on a lateral X-ray after a twolevel anterior cervical decompression and fusion with plating for myelopathy, resulting in prolonged use of a cervical collar for a total of 3 months. The third patient underwent a three-level anterior instrumented fusion secondary to a destabilizing infection. This construct ultimately required two further revisions within the first 2 months after the initial procedure because of painful instability. After the third surgery, the patient’s discomfort greatly improved, but settling of his graft was noted on lateral radiographs, which led the physician to prolong cervical collar immobilization. None of these patients had any further evidence of instability within the first year of surgery after the index procedure. Overall, we found that if a patient had an H/E NL with a X-ray NOT, further action was taken in only 5/879 events (0.57%) (p5.0001). A revision procedure was performed 2/ 879 (0.23%) times in the presence of an H/E NL including in patients with abnormal radiographic findings. There were 276 visits when the patient had an H/E NOT. Thirty-four of the 276 (12.3%) visits also had an X-ray NOT. Of those patients who had both an H/E NOT and an X-ray NOT, 18 of those visits resulted in further action by the clinician. Fifteen of the 34 visits went on to revision and 3 led to the requesting of magnetic resonance images by the clinician. Therefore, these patients had a 44.1% chance of getting a revision surgery and 53% chance of any further action. We did not examine the data regarding patients with an H/E NOT in the presence an X-ray NL. Any further action taken would have been based solely on exam findings, which was not the focus of this study. Discussion Common postoperative cervical spine fusion practice includes routine follow-up visits at discreet intervals. The frequency of visits varies by individual practice and practitioner training. During many of these visits, the patient typically undergoes a focused examination as well as AP and lateral radiographs. At our institution, the charge for AP, lateral, and odontoid radiographs totals $384 (including professional fees). This can add up to more than $1,000 in radiology charges during the first postoperative year. Although this is probably not a large portion of the cost of care for a single patient, applied to all patients
B.D. Grimm et al. / The Spine Journal 13 (2013) 764–769
767
Fig. 1. (Left) Postoperative radiograph 1 month s/p C4-C6 anterior cervical decompression and fusion showing superior plate dissociation in asymptomatic patient. (Right) Lateral radiograph 2 weeks s/p revision after superior plate dissociation.
who undergo cervical fusions each year, the cost savings as a whole would certainly be significant. A recent study showed that postoperative radiographs are an insensitive tool when a pseudoarthrosis is suspected and that computed tomography scan is the test of choice [10]. Another study has demonstrated that implant failure, defined as fracture or loosening of the construct after anterior cervical fusion, may pose no danger to the patient and that immediate removal of the failed hardware is rarely necessary [11]. This study concluded that cervical spine construct failure is uncommon, and if it occurs, the patient is likely to be symptomatic. Ning et al. [12], retrospectively, reviewed
2,233 patients who had undergone cervical spine fusion with an anterior locking plate. They found a hardware complication rate of 10%; however, they stated that most of these complications were not symptomatic and should be treated conservatively. Another study looking at long-term follow-up of 71 patients who underwent anterior cervical discectomy and fusion found that although there was a high rate of graft subsidence (47.9%), only six patients (8.5%) had segmental kyphosis and seven patients (9.9%) had construct failure, none of whom, however, required revision surgery [13]. Ugokwe et al. [14] recently, retrospectively, reviewed the charts of 53 patients who underwent
Fig. 2. (Left) Preoperative radiograph showing pseudoarthrosis between C1-C2. (Middle) Postoperative revision C1-C2 fusion showing persistent instability. Note large atlantodens interval. (Right) Lateral radiograph 2 weeks after revision C1-C2 fusion with extension of fusion to C4 with instrumentation and halo placement.
768
B.D. Grimm et al. / The Spine Journal 13 (2013) 764–769
a single-level anterior cervical decompression and fusion and found that obtaining routine postoperative radiographs did not change their course of management. In light of this information, it is important to determine whether routine postoperative radiographs are necessary in the asymptomatic patient after cervical spine fusion. This review demonstrated that in the face of a normal exam and a normal radiograph, no further action was taken 100% of the time. This represented about 73% of the total number of visits. Abnormal radiographs in the presence of a normal history and exam were present 3.5% of the time (31/879) and only 0.57% (5/879) of the time did the clinician take further action based on the radiographic findings. Only two of these further actions were surgical revisions. Furthermore, the radiographs, which showed abnormalities, that led to further action, were lateral views or flexion/ extension views, suggesting that the AP view may not add value to the overall clinical picture. These findings imply that, in the presence of a normal history and exam, radiographic findings are unlikely to alter the postoperative treatment course after cervical spine fusion. The strengths of our study include the power (383) and the limited variables of the study. Our original census comprised 686 patients; however, we frequently encountered patients who only had one postoperative visit or were lost to follow-up. One of the surgeons included in the study had a cohort of 214 patients who had undergone cervical spine fusion within our designated 1-year time period; however, only 66 of these patients had more than 1 postoperative visit and, therefore, 148 patients of his were excluded from our study. This particular spine surgeon frequently discharged patients from his clinic after two postoperative clinic visits seeing them after 2 and 6 weeks with X-rays at 6 weeks. After which time, if the patients were doing as expected and the X-rays appeared within normal limits, they were discharged from clinic with follow-up on an as-needed basis with the presumption that if they had an impending construct failure, they would be symptomatic and seek treatment. There are several limitations of our study. First, the subjectivity of the clinician’s perception of the patients’ well-being, multiplied by the subjectivity of the reviewer’s interpretation of the clinician’s note and the accuracy of the clinician’s note, leaves open the possibility for a substantial amount of bias. Certainly, the treating physician may consciously or subconsciously minimize or dismiss a patient’s complaint of neck or arm discomfort if he or she views a normal postoperative X-ray before meeting the patient at that clinic visit. Conversely, he or she may exaggerate a patient’s pain in the face of an abnormal radiograph, thereby introducing his or her own bias into the medical record. Because there was no standardization protocol, the definition of what constitutes a normal or abnormal X-ray may vary among clinicians and, therefore, bias the results. Moreover, the chart reviewers had to interpret the clinician’s notes and make a judgment from the content of those notes as to whether a physical exam finding or complaint of
pain was normal or not, which may further confound the results of our study. In addition, there was no standardization of postoperative protocols, thus adding problems inherent in this type of retrospective, multisurgeon study. The vast majority of patients had a 2- to 3-week postoperative visit and a 6-week visit, but after the 6-week period, follow-up intervals became quite variable. Follow-up for more than 1 year may improve the study as well. By not obtaining routine surveillance radiographs after cervical spine fusion, the treating physician may expose the patient to the possibility of having an asymptomatic failure of the construct, potentially even a catastrophic one. Although we could not find any reports of asymptomatic catastrophic failures after cervical fusion, there are reports in the literature describing late asymptomatic screw migration from a cervical plate into the esophagus status after ACDF [15]. Weighing the financial and emotional cost of missing an asymptomatic, catastrophic failure of a cervical fusion versus the cost of obtaining routine, surveillance postoperative radiographs to screen for such an event is unknowable from this study. This study also does not address the medicolegal implications of not obtaining such X-rays, which could theoretically compromise a surgeon’s defense in malpractice litigation if there were a catastrophic failure, which may have been prevented by obtaining routine postoperative radiographs. Despite the limitations of our study, our data are similar to previous studies, which show that routine postoperative radiographs rarely change clinical management [9,11,14,16].
Conclusions As the health-care costs continue to rise, it will become ever more imperative to control such costs and reduce unnecessary medical expenditures. Overall, we believe that we have identified a potential expense that could be reduced in the form of routine postoperative radiographs after cervical spine fusion. Our data demonstrate that routine radiographs rarely change the physicians’ management after cervical fusions and that those decisions are more influenced by the patients’ history and physical exam. We suggest that an immediate postoperative radiograph should be taken and that any future X-rays should be obtained based on the patient’s exam and treating physician’s judgment. It appears that in the face of a normal clinical exam, radiographs rarely alter treatment regardless of findings.
Acknowledgments The authors thank William T. Barfield, PhD, and James F. Mooney, MD, of Medical University of South Carolina.
B.D. Grimm et al. / The Spine Journal 13 (2013) 764–769
References [1] Abumi K, Kaneda K. Pedicle screw fixation for nontraumatic lesions of the cervical spine. Spine 1997;22:1853–63. [2] Abumi K, Kaneda K, Shono Y, Fujiya M. One-stage posterior decompression and reconstruction of the cervical spine by using pedicle screw fixation systems. J Neurosurg 1999;90(1 Suppl):19–26. [3] Bailey RW, Badgley CE. Stabilization of the cervical spine by anterior fusion. J Bone Joint Surg Am 1960;42:565–94. [4] Brown JA, Havel P, Ebraheim N, et al. Cervical stabilization by plate and bone fusion. Spine 1988;13:236–40. [5] Cloward RB. The anterior approach for removal of ruptured cervical disks. J Neurosurg 1958;15:602–17. [6] Gok B, Sciubba DM, McLoughlin GS, et al. Surgical treatment of cervical spondylotic myelopathy with anterior compression: a review of 67 cases. J Neurosurg Spine 2008;9:152–7. [7] Macdonald RL, Fehlings MG, Tator CH, et al. Multilevel anterior cervical corpectomy and fibular allograft fusion for cervical myelopathy. J Neurosurg 1997;86:990–9. [8] Mayr MT, Subach BR, Comey CH, et al. Cervical spinal stenosis: outcome after anterior corpectomy, allograft reconstruction, and instrumentation. J Neurosurg 2002;96:10–6. [9] Romero NC, Glaser J, Walton Z. Are routine radiographs needed in the first year after lumbar spinal fusions? Spine 2009;34: 1578–80.
769
[10] Epstein NE, Silvergleide RS. Documenting fusion following anterior cervical surgery: a comparison of roentgenogram versus two dimensional computed tomographic findings. J Spinal Disord Tech 2003;16:243–7. [11] Lowery GL, McDonough RF. The significance of hardware failure in anterior cervical plate fixation. Patients with 2- to 7-year follow-up. Spine 1998;23:181–6. [12] Ning X, Wen Y, Xiao-Jian Y, et al. Anterior cervical locking platerelated complications; prevention and treatment recommendations. Int Orthop 2008;32:649–55. [13] Yue WM, Brodner W, Highland TR. Long-term results after anterior cervical discectomy and fusion with allograft and plating: a 5- to 11-year radiologic and clinical follow-up study. Spine 2005;30: 2138–44. [14] Ugokwe KT, Kalfas IH, Mroz TE, Steinmetz M. Review of the utility of obtaining repeated postoperative radiographs following singlelevel anterior cervical decompression, fusion, and plate placement. J Neurosurg Spine 2008;9:175–9. [15] Pompili A, Canitano S, Caroli F, et al. Asymptomatic esophageal perforation caused by late screw migration after anterior cervical plating: report of a case and review of relevant literature. Spine 2002;27:E499–502. [16] Yamashita T, Steinmetz M, Lieberman I, et al. The utility of repeated postoperative radiographs after lumbar instrumented fusion for degenerative lumbar spine. Spine 2011;36:1955–60.
Clinical Study
The utility of routine postoperative radiographs after cervical spine fusion Bennett D. Grimm, MDa,*, Daniel P. Leas, BSb, John A. Glaser, MDa a
Department of Orthopaedic Surgery, Medical University of South Carolina, 96 Jonathan Lucas St, Ste. 708, Charleston, SC 29425, USA b School of Medicine, Medical University of South Carolina, 96 Jonathan Lucas St, Ste. 708, Charleston, SC 29425, USA Received 27 July 2011; revised 14 October 2012; accepted 13 January 2013
Abstract
BACKGROUND CONTEXT: The senior author (JAG) recently published an article questioning the utility of routine postoperative radiographs after lumbar spine fusion. That study concluded that routine postoperative radiographs in the presence of a normal physical examination rarely change the clinician’s management of these patients. Our aim was to repeat this protocol in patients after cervical spine fusion. We hypothesized that routine postoperative radiographs are unnecessary in most cases after cervical spine fusion. PURPOSE: The purpose of this study was to determine the usefulness of routine postoperative cervical spine radiographs after cervical spine fusion as to whether they help to guide clinical decision making within the first postoperative year. STUDY DESIGN: This is a retrospective chart review of 383 patients who underwent a cervical spine fusion over a 5-year period. Seven different surgeons performed the cervical spine fusions. Our review assessed a total of 1,155 postoperative clinic visits. METHODS: Each clinical postoperative visit was reviewed. The history and exam were graded as either normal or abnormal, and any plain radiographs obtained were graded similarly as either normal or abnormal. Each patient’s notes were followed up to 1 year postoperatively. Each patient had to have at least two postoperative visits with X-rays to be included in the study. We then noted any further action taken by the clinician based on the appearance of the radiograph in conjunction with the history and exam. RESULTS: In patients with normal history and exam presentations, further action was taken only 5/879 (0.57%) of the time, sometimes even in the presence of abnormal radiographs. The actions included two surgical revisions, two prolongations of cervical collar immobilization, and one patient who underwent a flexion/extension radiographic evaluation and subsequent prolonged cervical collar immobilization. There were 276 visits with abnormal history and exam; of these, 34/276 (12.3%) had abnormal X-rays. Of the clinic visits with abnormal history and exam and abnormal X-rays, 15/34 (44%) went on to revision. CONCLUSIONS: Routine postoperative radiographs after cervical spine fusion rarely appear to be of value when patients present with a normal history and exam and may expose patients to unnecessary diagnostic studies and expenses. Patients exhibiting a normal postoperative history and exam are likely to have no further additional action taken in the presence of either normal or abnormal radiographs. Ó 2013 Elsevier Inc. All rights reserved.
Keywords:
Cervical spine; X-rays; Postoperative care
Introduction FDA device/drug status: Not applicable. Author disclosures: BDG: Nothing to disclose. DPL: Nothing to disclose. JAG: Trips/Travel: SI Bone (A, Paid directly to institution/employer); Grants: SI Bone (D, Paid directly to institution/employer). The disclosure key can be found on the Table of Contents and at www. TheSpineJournalOnline.com. * Corresponding author. Department of Orthopaedic Surgery, Medical University of South Carolina, 96 Jonathan Lucas St, Suite 708, Charleston, SC 29425, USA. Tel.: (843) 792-0245; fax: (843) 792-3674. E-mail address: [email protected] (B.D. Grimm) 1529-9430/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.spinee.2013.01.018
Cervical spine fusion via an anterior, posterior, or combined approach with concomitant decompression as indicated reliably alleviates symptoms from radiculopathy/ myelopathy and can impart stability after trauma or destabilizing cervical spine lesions [1–8]. It is a standard protocol for many spine surgeons to obtain routine postoperative radiographs at distinct intervals (typically at 2 and 6 weeks, 3 months, 6 months, and 1 year
B.D. Grimm et al. / The Spine Journal 13 (2013) 764–769
postoperatively) whether or not the patient is symptomatic. The rationale for obtaining routine postoperative X-rays is threefold: they give the surgeon and patient piece of mind that the patient is healing and the surgery was a success; so that asymptomatic/symptomatic hardware migration or failure may be identified; and routine surveillance radiographs may help to protect the physician from liability via documentation of the patient’s clinical status in the medical record. The senior author (JAG) concluded in a study that routine postoperative radiographs after lumbar spine fusion almost never altered the course of treatment [9]. There has not been, to our knowledge, an investigation into the necessity of obtaining routine postoperative radiographs after cervical spine fusion for one or more levels. By assessing the utility in decision making of such routine imaging in postoperative decision making, physicians may avoid exposing the patient to unnecessary charges and radiation. Materials and methods On approval by our institutional review board, we began a retrospective chart review of 686 patients who underwent cervical fusion for any reason between January 1, 2002, and December 31, 2007 (Table 1). For inclusion into this study, patients had to be a minimum 18 years of age and have at least two postoperative visits with radiographs within 12 months after the index procedure. This narrowed our cohort to 383 patients, totaling 401 procedures including 18 revision surgeries (2 on the same patient) with an average of 3 postoperative visits. We included patients who had undergone anterior cervical fusion, posterior fusion, or combined anterior/posterior fusion. Procedures were performed by a total of seven surgeons: six neurosurgeons and one spine fellowship–trained orthopedic surgeon. One hundred ninety-nine patients were men and 184 patients were women. Patient ages ranged from 18 to 84 years, and the numbers of fused levels were from 1 to 6. An anterioronly approach alone was used 350 times, 40 through a posterior approach only, and 10 via a staged anterior and posterior approach (Table 2). A senior-level orthopedic resident with the assistance of a medical student reviewed each postoperative clinic visit note. A qualifying clinic visit was defined as having a history and exam as well as an anterior-posterior and lateral X-ray of the cervical spine. The history and exam were determined to be either normal or abnormal based on the clinician’s interpretation of the patient’s symptoms. A normal history and exam (H/E NL) included only findings that were deemed appropriate during the course of recovery. An abnormal history and exam (H/E NOT) was defined as pain greater than would be expected by the clinical situation or an unresolved or new onset neurologic deficit. The radiographs were also reviewed and graded according to the clinician’s note (or the radiologist’s interpretation if the clinician did not comment on the X-ray) as normal
765
Context Postoperative radiographs are commonly obtained at follow-up visits after cervical fusion. Contribution Based on this retrospective review, the authors found that only 5 of 879 patients with normal history and physical exams at follow-up visits had abnormal X-ray findings (displacement of graft, plate, or pseudarthrosis) requiring action on the part of the treating surgeon (revision surgery, prolonged brace use). Implication The findings are interesting and suggest changes in current common practice. However, any change would have to consider gradual failure of graft and/or instrumentation that remains asymptomatic until catastrophic failure and catastrophic morbidities occur (a ‘‘threshold’’ phenomenon). What is the rate of these uncommon events? What rate is acceptable to miss? Are there particular clinical circumstances (eg, infection, tumor, deformity) with greater risk of progressive asymptomatic failure? —The Editors
(X-ray NL) or abnormal (X-ray NOT). Abnormal radiographic findings included failure of fixation, increasing listhesis/kyphosis, motion between fused segments, or any change from the baseline postoperative X-ray. We then determined whether or not the clinician took action based solely on the appearance of the routine radiographs (meaning did the clinician change management of the patient if they had a normal exam but an abnormal routine X-ray), such as additional imaging or reoperation. Chi-square analysis was used to analyze the data.
Results The clinic notes of 1,155 visits were reviewed and graded. A total of 879 visits were graded as H/E NL and 276 visits were graded as H/E NOT. Radiographs were graded as X-ray NL for 1,090 visits and as X-ray NOT in 65 cases. These data were organized into four categories (Table 3). No action was taken 100% of the time, when clinic visits Table 1 Pathologic condition leading to fusion Pathology
No. of patients
Degenerative Traumatic Other
302 70 11
766
B.D. Grimm et al. / The Spine Journal 13 (2013) 764–769
Table 2 Demographics of patients Demographics Age (y) 18–35 36–50 51–65 66–80 81þ Total number of patients Levels of fusion per procedure 1 2 3 4þ Total number of procedures Ethnicity Caucasian African-American Hispanic Asian
No. of patients 45 142 149 45 2 383 135 160 79 27 401 270 107 3 3
were graded as H/E NL and X-ray NL. In patients graded as H/E NL and X-ray NOT, no additional action was taken 26/ 31 or 84% of the time. A brief discussion of the five patients with a normal exam, but an abnormal X-ray, is warranted. Patient 1, an 80-year-old man, underwent a C4–C6 anterior decompression and fusion with a fibular allograft and plate fixation for myelopathy. At his 1-month postoperative visit, the lateral radiograph showed that the plate and screws had dissociated from the bone superiorly (Fig. 1, Left). He did not complain of any undue pain at this time and actually felt well enough to go on an extended vacation. The treating surgeon, however, felt that the instrumentation migration was unacceptable, and the patient underwent a revision procedure shortly thereafter (Fig. 1, Right). He subsequently went on to an asymptomatic fusion. Patient 2, a 53-year-old man, with a three-pack per day smoking history had a previous C1–C2 fusion (Fig. 2, Left) 1 year before presented to our institution complaining of persistent neck pain and myelopathic symptoms. He was found to have a computed tomography scan proven pseudoarthrosis and underwent a revision with C-1 lateral mass screws plus translaminar and pedicle screws at C-2 supplemented with autogenous iliac crest bone graft (Fig. 2, Middle). Approximately 2.5 months postoperatively, flexion/ extension radiographs showed a persistent pseudoarthrosis with a 10-mm atlantodens interval apparent on dynamic
Table 3 Categorized patient data Category
No. of visits
H/E H/E H/E H/E
848 31 242 34
NL and X-ray NL NL and X-ray NOT NOT and X-ray NL NOT and X-ray NOT
H/E NL, a normal history and exam; H/E NOT, an abnormal history and exam; X-ray NL, a normal X-ray; X-ray NOT, an abnormal X-ray.
lateral radiographs despite having an H/E NL. He subsequently underwent a C1–C4 posterior revision procedure with supplemental halo fixation (Fig. 2, Right). There were three other patients who had further action taken in the presence of an X-ray NOT but an H/E NL. One patient had a 3.5-mm subluxation on a lateral X-ray after a one-level anterior cervical fusion with plate fixation because of instability from trauma. This radiographic finding prompted the treating physician to order prolonged cervical collar immobilization and a flexion/extension X-ray to determine the stability of the construct. Another patient had 2 to 3 mm of plate/bone separation on a lateral X-ray after a twolevel anterior cervical decompression and fusion with plating for myelopathy, resulting in prolonged use of a cervical collar for a total of 3 months. The third patient underwent a three-level anterior instrumented fusion secondary to a destabilizing infection. This construct ultimately required two further revisions within the first 2 months after the initial procedure because of painful instability. After the third surgery, the patient’s discomfort greatly improved, but settling of his graft was noted on lateral radiographs, which led the physician to prolong cervical collar immobilization. None of these patients had any further evidence of instability within the first year of surgery after the index procedure. Overall, we found that if a patient had an H/E NL with a X-ray NOT, further action was taken in only 5/879 events (0.57%) (p5.0001). A revision procedure was performed 2/ 879 (0.23%) times in the presence of an H/E NL including in patients with abnormal radiographic findings. There were 276 visits when the patient had an H/E NOT. Thirty-four of the 276 (12.3%) visits also had an X-ray NOT. Of those patients who had both an H/E NOT and an X-ray NOT, 18 of those visits resulted in further action by the clinician. Fifteen of the 34 visits went on to revision and 3 led to the requesting of magnetic resonance images by the clinician. Therefore, these patients had a 44.1% chance of getting a revision surgery and 53% chance of any further action. We did not examine the data regarding patients with an H/E NOT in the presence an X-ray NL. Any further action taken would have been based solely on exam findings, which was not the focus of this study. Discussion Common postoperative cervical spine fusion practice includes routine follow-up visits at discreet intervals. The frequency of visits varies by individual practice and practitioner training. During many of these visits, the patient typically undergoes a focused examination as well as AP and lateral radiographs. At our institution, the charge for AP, lateral, and odontoid radiographs totals $384 (including professional fees). This can add up to more than $1,000 in radiology charges during the first postoperative year. Although this is probably not a large portion of the cost of care for a single patient, applied to all patients
B.D. Grimm et al. / The Spine Journal 13 (2013) 764–769
767
Fig. 1. (Left) Postoperative radiograph 1 month s/p C4-C6 anterior cervical decompression and fusion showing superior plate dissociation in asymptomatic patient. (Right) Lateral radiograph 2 weeks s/p revision after superior plate dissociation.
who undergo cervical fusions each year, the cost savings as a whole would certainly be significant. A recent study showed that postoperative radiographs are an insensitive tool when a pseudoarthrosis is suspected and that computed tomography scan is the test of choice [10]. Another study has demonstrated that implant failure, defined as fracture or loosening of the construct after anterior cervical fusion, may pose no danger to the patient and that immediate removal of the failed hardware is rarely necessary [11]. This study concluded that cervical spine construct failure is uncommon, and if it occurs, the patient is likely to be symptomatic. Ning et al. [12], retrospectively, reviewed
2,233 patients who had undergone cervical spine fusion with an anterior locking plate. They found a hardware complication rate of 10%; however, they stated that most of these complications were not symptomatic and should be treated conservatively. Another study looking at long-term follow-up of 71 patients who underwent anterior cervical discectomy and fusion found that although there was a high rate of graft subsidence (47.9%), only six patients (8.5%) had segmental kyphosis and seven patients (9.9%) had construct failure, none of whom, however, required revision surgery [13]. Ugokwe et al. [14] recently, retrospectively, reviewed the charts of 53 patients who underwent
Fig. 2. (Left) Preoperative radiograph showing pseudoarthrosis between C1-C2. (Middle) Postoperative revision C1-C2 fusion showing persistent instability. Note large atlantodens interval. (Right) Lateral radiograph 2 weeks after revision C1-C2 fusion with extension of fusion to C4 with instrumentation and halo placement.
768
B.D. Grimm et al. / The Spine Journal 13 (2013) 764–769
a single-level anterior cervical decompression and fusion and found that obtaining routine postoperative radiographs did not change their course of management. In light of this information, it is important to determine whether routine postoperative radiographs are necessary in the asymptomatic patient after cervical spine fusion. This review demonstrated that in the face of a normal exam and a normal radiograph, no further action was taken 100% of the time. This represented about 73% of the total number of visits. Abnormal radiographs in the presence of a normal history and exam were present 3.5% of the time (31/879) and only 0.57% (5/879) of the time did the clinician take further action based on the radiographic findings. Only two of these further actions were surgical revisions. Furthermore, the radiographs, which showed abnormalities, that led to further action, were lateral views or flexion/ extension views, suggesting that the AP view may not add value to the overall clinical picture. These findings imply that, in the presence of a normal history and exam, radiographic findings are unlikely to alter the postoperative treatment course after cervical spine fusion. The strengths of our study include the power (383) and the limited variables of the study. Our original census comprised 686 patients; however, we frequently encountered patients who only had one postoperative visit or were lost to follow-up. One of the surgeons included in the study had a cohort of 214 patients who had undergone cervical spine fusion within our designated 1-year time period; however, only 66 of these patients had more than 1 postoperative visit and, therefore, 148 patients of his were excluded from our study. This particular spine surgeon frequently discharged patients from his clinic after two postoperative clinic visits seeing them after 2 and 6 weeks with X-rays at 6 weeks. After which time, if the patients were doing as expected and the X-rays appeared within normal limits, they were discharged from clinic with follow-up on an as-needed basis with the presumption that if they had an impending construct failure, they would be symptomatic and seek treatment. There are several limitations of our study. First, the subjectivity of the clinician’s perception of the patients’ well-being, multiplied by the subjectivity of the reviewer’s interpretation of the clinician’s note and the accuracy of the clinician’s note, leaves open the possibility for a substantial amount of bias. Certainly, the treating physician may consciously or subconsciously minimize or dismiss a patient’s complaint of neck or arm discomfort if he or she views a normal postoperative X-ray before meeting the patient at that clinic visit. Conversely, he or she may exaggerate a patient’s pain in the face of an abnormal radiograph, thereby introducing his or her own bias into the medical record. Because there was no standardization protocol, the definition of what constitutes a normal or abnormal X-ray may vary among clinicians and, therefore, bias the results. Moreover, the chart reviewers had to interpret the clinician’s notes and make a judgment from the content of those notes as to whether a physical exam finding or complaint of
pain was normal or not, which may further confound the results of our study. In addition, there was no standardization of postoperative protocols, thus adding problems inherent in this type of retrospective, multisurgeon study. The vast majority of patients had a 2- to 3-week postoperative visit and a 6-week visit, but after the 6-week period, follow-up intervals became quite variable. Follow-up for more than 1 year may improve the study as well. By not obtaining routine surveillance radiographs after cervical spine fusion, the treating physician may expose the patient to the possibility of having an asymptomatic failure of the construct, potentially even a catastrophic one. Although we could not find any reports of asymptomatic catastrophic failures after cervical fusion, there are reports in the literature describing late asymptomatic screw migration from a cervical plate into the esophagus status after ACDF [15]. Weighing the financial and emotional cost of missing an asymptomatic, catastrophic failure of a cervical fusion versus the cost of obtaining routine, surveillance postoperative radiographs to screen for such an event is unknowable from this study. This study also does not address the medicolegal implications of not obtaining such X-rays, which could theoretically compromise a surgeon’s defense in malpractice litigation if there were a catastrophic failure, which may have been prevented by obtaining routine postoperative radiographs. Despite the limitations of our study, our data are similar to previous studies, which show that routine postoperative radiographs rarely change clinical management [9,11,14,16].
Conclusions As the health-care costs continue to rise, it will become ever more imperative to control such costs and reduce unnecessary medical expenditures. Overall, we believe that we have identified a potential expense that could be reduced in the form of routine postoperative radiographs after cervical spine fusion. Our data demonstrate that routine radiographs rarely change the physicians’ management after cervical fusions and that those decisions are more influenced by the patients’ history and physical exam. We suggest that an immediate postoperative radiograph should be taken and that any future X-rays should be obtained based on the patient’s exam and treating physician’s judgment. It appears that in the face of a normal clinical exam, radiographs rarely alter treatment regardless of findings.
Acknowledgments The authors thank William T. Barfield, PhD, and James F. Mooney, MD, of Medical University of South Carolina.
B.D. Grimm et al. / The Spine Journal 13 (2013) 764–769
References [1] Abumi K, Kaneda K. Pedicle screw fixation for nontraumatic lesions of the cervical spine. Spine 1997;22:1853–63. [2] Abumi K, Kaneda K, Shono Y, Fujiya M. One-stage posterior decompression and reconstruction of the cervical spine by using pedicle screw fixation systems. J Neurosurg 1999;90(1 Suppl):19–26. [3] Bailey RW, Badgley CE. Stabilization of the cervical spine by anterior fusion. J Bone Joint Surg Am 1960;42:565–94. [4] Brown JA, Havel P, Ebraheim N, et al. Cervical stabilization by plate and bone fusion. Spine 1988;13:236–40. [5] Cloward RB. The anterior approach for removal of ruptured cervical disks. J Neurosurg 1958;15:602–17. [6] Gok B, Sciubba DM, McLoughlin GS, et al. Surgical treatment of cervical spondylotic myelopathy with anterior compression: a review of 67 cases. J Neurosurg Spine 2008;9:152–7. [7] Macdonald RL, Fehlings MG, Tator CH, et al. Multilevel anterior cervical corpectomy and fibular allograft fusion for cervical myelopathy. J Neurosurg 1997;86:990–9. [8] Mayr MT, Subach BR, Comey CH, et al. Cervical spinal stenosis: outcome after anterior corpectomy, allograft reconstruction, and instrumentation. J Neurosurg 2002;96:10–6. [9] Romero NC, Glaser J, Walton Z. Are routine radiographs needed in the first year after lumbar spinal fusions? Spine 2009;34: 1578–80.
769
[10] Epstein NE, Silvergleide RS. Documenting fusion following anterior cervical surgery: a comparison of roentgenogram versus two dimensional computed tomographic findings. J Spinal Disord Tech 2003;16:243–7. [11] Lowery GL, McDonough RF. The significance of hardware failure in anterior cervical plate fixation. Patients with 2- to 7-year follow-up. Spine 1998;23:181–6. [12] Ning X, Wen Y, Xiao-Jian Y, et al. Anterior cervical locking platerelated complications; prevention and treatment recommendations. Int Orthop 2008;32:649–55. [13] Yue WM, Brodner W, Highland TR. Long-term results after anterior cervical discectomy and fusion with allograft and plating: a 5- to 11-year radiologic and clinical follow-up study. Spine 2005;30: 2138–44. [14] Ugokwe KT, Kalfas IH, Mroz TE, Steinmetz M. Review of the utility of obtaining repeated postoperative radiographs following singlelevel anterior cervical decompression, fusion, and plate placement. J Neurosurg Spine 2008;9:175–9. [15] Pompili A, Canitano S, Caroli F, et al. Asymptomatic esophageal perforation caused by late screw migration after anterior cervical plating: report of a case and review of relevant literature. Spine 2002;27:E499–502. [16] Yamashita T, Steinmetz M, Lieberman I, et al. The utility of repeated postoperative radiographs after lumbar instrumented fusion for degenerative lumbar spine. Spine 2011;36:1955–60.