Introduction

The current trend of evolution of all spinal surgery has been toward less-invasive techniques. Stookey [1] described the clinical symptoms and anatomic location of cervical disc herniation in 1928. Subsequently, the landmark paper by Mixter and Barr [2] clearly established the relationship between herniated discs and sciatica, and provided evidence that laminectomy and disc excision could successfully relieve the pain associated with radiculopathy. Bailey and Badgley [3], Cloward [4], and Robinson and Smith [5] popularized the anterior approach with interbody fusion in the 1950s. Hirsch [6] and Robertson [7] recommended cervical discectomy without fusion. Fukushima et al. [8] introduced the ventriculofiberscope in 1973 and further enhanced the foundation for percutaneous endoscopic cervical discectomy [9].

Minimally invasive treatments aimed at removing nuclear material and lowering intradiscal pressure through devices were inserted percutaneously into intervertebral discs. A number of techniques have recently been developed that are applicable in the treatment of disc herniation. Smith et al. [10, 11] introduced chymopapain chemonucleolysis to treat herniated nucleus pulposus. Hijikata [12] described percutaneous lumbar discectomy. Ascher [13] then reported laser discectomy.

Percutaneous disc decompression, regardless of technique, has been based on the principle that a small reduction of volume in a closed hydraulic space, like an intact disc, results in a disproportionately large fall of pressure. Case et al. [14] studied showing that a large rise in pressure will regularly result from a small increase in volume, confirming the biochemical basis for the benefits obtained from interventions designed for disc decompression. Percutaneous cervical discectomy (PCD) has been developed as an effective treatment option for soft cervical disc herniation. Percutaneous nucleoplasty (PCN) is a new minimally invasive technique which uses radiofrequency energy to ablate the nucleus pulposus in a controlled manner for disc decompression.

In general, all these have shown moderate or good clinical results of PCN and PCD. To date, no study has been published in which investigators examine the therapeutic effects of PCN and PCD for the treatment of cervical disc herniation. The aim of this paper was to compare the clinical outcomes of PCN and PCD in contained cervical disc herniation.

Materials and methods

Patient population

A retrospective study on patients with contained cervical disc herniations from June 2003 to July 2005 had procedures of PCD or PCN. Patients had to satisfy specific inclusion and exclusion criteria to be enrolled. All patients had the radiographically determined contained disc herniation on magnetic resonance imaging (MRI). Inclusion criteria were contained disc hemiation (diagnosed by MRI) complaints of radicular pain with or without neck pain, and no improvement was observed for at least 6 weeks of conservative therapy (i.e., physical therapy together with use of anti-inflammatory medications and muscle relaxants at the manufacturer’s recommended therapeutic dose). Exclusion criteria were extruded disc fragment, hemorrhagic diasthesis, spondylolisthesis, spinal canal stenosis, ossification of longitudinal ligament (OPLL), previous surgery at the indicated level, and cases of myelopathy. Initially, 204 patients fulfilled the study criteria and 28 were lost in follow-up. Of the remaining 176 patients available for analysis, 81 had PCN (group 1, 43 men and 38 women with an average age 50.73 ± 11.61 years), and 95 had PCD (group 2, 49 men and 46 women with an average age 51.51 ± 11.17 years). Both groups had similar age and sex distribution, level of involvement, and the pain history as shown in Table 1, and there were no significant differences between the groups (P > 0.05).

Table 1 Comparison of the general data between PCN and PCD groups (x ± s)
Full size table

Surgical technique

Under local anesthesia, the patient was placed in a supine position with the neck extended by placing a rolled towel under the shoulders. A soft strap was placed over the forehead for stabilization. The shoulders were gently distracted downward with tape. C-arm fluoroscopy was used in anteroposterior and lateral planes to direct the placement of a spinal needle onto the disc surface. Initially, at the point of entry which is adjacent to the medial border of the right sternocleidomastoid muscle, firm pressure was digitally applied in the space between the muscle and the trachea and pointed toward the vertebral surface. The larynx and trachea were displaced medially and the carotid artery laterally. The anterior cervical spine was palpated with the fingertips, and a spinal needle was used to puncture the right-side skin of the neck and passed into the disc space. The position was confirmed by C-arm fluoroscopy.

The fiber of the Perc-D Spine Wand was inserted through the 18-gauge needle. The wand was connected to a standard Arthrocare power generator. The power of nucleoplasty ablation was 3 W with 1 s coagulation. If it incites pain syndrome, the Perc-D Spine Wand should be replaced in disc space. No pains during coagulation, had coblation of 15 s while moving the Perc-D Spine Wand on the monitor of C-arm fluoroscopy. When the Perc-D Spine Wand returned to annulus, coagulated 1 s to shrinkage of the surrounding collagen and widening channel. Redo this process for four to six times. The procedure is shown in Fig. 1.

Fig. 1
figure 1

Pre-, intro-, and post-operative lateral views of radiograph of patient operated on with PCN at C3-4

Full size image

In the PCD procedure, the instrument (rongeur or nucleotome) is then inserted into the disc and the removal of the nucleus pulposus material was estimated to be 1 g. The positioning of instruments is carefully monitored by C-arm fluoroscopy throughout the procedure. The procedure is shown in Fig. 2.

Fig. 2
figure 2

Pre-, intro-, and post-operative lateral views of radiograph of patient operated on with PCD at C5-6

Full size image

The skin incision (2–3 mm) was then closed with Steri-Strips and the patients were discharged home within a day after the procedure. The operation time of procedures were recorded and compared between PCD and PCN. Patients received periprocedural and after-operation oral antibiotics every 8 h for 48 h. Postoperatively, patients were allowed unlimited walking, standing, and sitting. Return to sedentary or simple work was permitted at 3–4 days following the surgery.

Radiographic assessment

The radiographic assessment of the segment stability was determined by two independent radiologists, who were blinded to the assigned treatment group of the patients. The radiologists studied anteroposterior, lateral, and flexion–extension lateral radiographs. The radiographical stability has been considered <11° of angular displacement (AD), and <3 mm of horizontal displacement (HD) had to be present at the site of the procedure segment when comparing flexion–extension lateral radiographic views. AD ≥11° or HD ≥3 mm was considered radiographically instable.

Statistical analyses

Statistical analyses were performed independently by a non-clinical research assistant and an outside party to ensure objectivity, using SPSS version 11.5. The clinical results of the effect of treatment were evaluated by the Macnab standard [15] as shown in Table 2, and using visual analog pain scale (VAS) for cervical radiculopathy. The Student’s t test with a two-tailed paired comparison was used to compare the means between visits and to compare the success based on demographic variables. Results were considered statistically significant if P ≤ 0.05 for continuous variables.

Table 2 Clinical results of the Macnab standard
Full size table

Results

Puncture of the needle into the disc space was accurately performed under C-arm fluoroscopy guidance in all cases and there were no intraoperative deaths in our study. At the end, 176 cases had follow-up and 28 were lost with follow-up rate 88.0% (81/92) in the PCN group and 84.8% (95/112) in the PCD group. The follow-up time ranged from 16 to 48 months (average 29 months), and averages of 28.86 ± 4.52 months on PCN and 28.42 ± 3.21 months on PCD (t = −0.24, P = 0.81, >0.05). The averages of operation time of PCN 4.67 ± 1.16 min and PCD 11.95 ± 1.80 min. There was a significant difference in the operation time between PCN and PCD (P < 0.01).

The pain index improved from 7.12 ± 1.13 to 2.74 ± 0.89 (t = 27.03, P = 0.0000, <0.001) in PCN patients and improved from 7.18 ± 1.09 to 2.71 ± 0.91 (t = 29.57, P = 0.0000, <0.001) in PCD patients.

The clinical results assessed by Macnab standard of PCN was excellent in 31 cases, good 32 cases, fair 13 cases, and poor 5 cases; and that for PCD 33, 42, 12, and 7 cases, respectively, and 1 in discitis. The good and excellent was 78.4% (77.8% in PCN and 79.5% in PCD, P > 0.05).

There was one case of PCN that had the partial Perc-D Spine Wand broken in the disc space and which cannot been moved by the percutaneous cervical discectomy and remained there itself (Fig. 3). The clinical outcome of this case was good and there was no complicated occurrence. There was one case that had discitis in this study after PCD (Fig. 4). The patient presented with neck pain and associated radicular pain and numbness in the left upper-limb after 8 days of PCD. The history was negative for fever, night pain or constitutional symptoms. Cervical X-ray film showing narrowed C5-6 intervertebral space and erosive in the 5-cervical vertebral body wherein 3 months after operation shows narrowed disc space and sclerotic remodeling.

Fig. 3
figure 3

Shown is the case that had Perc-D SpineWand broken in C4-5 space when procedure. The partial Perc-D SpineWand to bend (a; b—an enlargement in the radiograph showed it) and broken in C4-5 space cannot be moved by the percutaneous cervical discectomy (c) and remain there itself (d). The clinical outcome of this case is good with no complications

Full size image
Fig. 4
figure 4

The C5-6 discitis case after PCD. Cervical X-ray film showing narrowed C5-6 intervertebral space and erosive in the 5-cervical vertebral body (a); 3 months after operation shows the disc space narrowed and sclerotic remodeling in the 5-cervical vertebral body (b)

Full size image

The procedure segment stability of pre- and the post-procedures are illustrated in Table 2. There were no instability cases after procedure of PCN and PCD. There were no significant difference stability in the preoperatively and postoperatively between PCN and PCD (P > 0.05).

Discussion

In recent years, the general trend in spinal surgery has been one of reductionism and minimalization. Treatment of cervical disk hernia has been performed under conservative therapy and sometimes under surgical intervention, such as anterior fixation of the cervical spine. At those times, if the patient failed, conservative treatments of open surgery were the only available option. Conventional open cervical discectomy, with or without bony fusion, is considered the standard treatment for cervical disc protrusion [16]. However, some patients were not available for surgery because of their poor general conditions and some refused surgery even though they were good candidates. In an attempt to avoid prolonged suffering from unsuccessful conservative treatment and to minimize the chance of morbidity often associated with open surgery, many surgeons sought alternative ways of decompressing a pathological disc. In our series, we selected patients contained disc herniation complaints of radicular pain with or without neck pain, and no improvement for at least 6 weeks of conservative therapy (i.e., physical therapy together with use of anti-inflammatory medications and muscle relaxants at the manufacturer’s recommended therapeutic dose). The pain index improved from 7.12 ± 1.13 to 2.74 ± 0.89 in PCN patients and that for PCD from 7.18 ± 1.09 to 2.71 ± 0.91. Clinical results of good and excellent was 78.4% (77.8% in PCN and 79.5% in PCD, P > 0.05). Therefore, myelopathy is considered outside the indication of PCN and PCD for cervical disk hernia.

Internal disc disruption and disc herniations are the common causes of extremity pain which may become chronic if not diagnosed and treated. Annular tears lead to migration of the nuclear material and derange of internal architecture. In the chronically damaged intervertebral disc, leakage of nuclear material from annular tears can initiate, promote, and continue the inflammatory process and delay or stop recovery of vital remaining intradiscal tissue. Inflammatory chemicals from the response to a damaged disc may activate or injure the dorsal root ganglion. Smaller protrusions are postulated to cause referred extremity pain due to neural inflammation and axial pain due to a combination of a sensitized outer annulus and increased outer annular tension [17]. Cervical radiculopathy is due to an abnormal process that involves the nerve root. It is caused mostly by cervical disk herniation or cervical spondylosis. The symptoms of cervical radiculopathy are neck and brachial-radicular pain, with or without motor weakness or paresthesia, in 80–100% of the patients [18]. Partial removal of the nucleus has been shown to decompress herniated discs, relieving pressure on nerve roots, and in some cases, offering relief from disc pain. In this study, puncturing needle into the disc space was accurately performed in all cases and no intraoperative deaths were observed in our study. The operation time of PCN averaged 4.67 ± 1.16 min and that PCD 11.95 ± 1.80 min (P < 0.01), and also the clinical outcome was very good (Figs. 1, 2).

In 1989, Hijikata [12] reported on his series of microdiscectomy patients and reinforced the clinical perception that successful outcome of disc surgery was independent of the amount of disc material excised. Increasing nuclear pressures by injecting the disc with fluid will be reflected as a proportional increase in outer annular pressure when a radial annular tear extends to the outer annulus [19, 20]. This increased pressure is probably analogous to compressing a well-hydrated nucleus and transmitting horizontal forces through the herniated nuclear material to the outer annulus. Castro et al. [21] showed that for each gram of nucleus removed, the disc space narrows to 1.42 mm and the disc bulge decreases to 0.45 mm. The advantage of PCN is that it reduces the volume and pressure of the affected disk without damaging other spinal structures. Ablation of a relatively small volume of the nucleus pulposus results in a significant reduction in intradisk pressure. This pressure was markedly reduced in the younger, healthy disc cadaver. There was an inverse correlation between the degree of disc degeneration and the change in intradiscal pressure in the Chen’s study of human cadavers [22]. Histologic examination revealed no evidence of direct mechanical or thermal damage to the surrounding tissues in human cadavers [23], while giving these radial thermal penetrations, high temperatures and lethal thermal doses in small regions outside of the nucleus, or within the bone endplates in human cadavers [24]. In this study, coblation for 15 s and coagulation for 1 s and redoing this process for four to six times were carried out in PCN. While, 1 g is removed by the PCD procedures; there were no instability change of the operation segment (Table 3).

Table 3 Comparison of the stability of segment between pre- and post-operation (x ± s)
Full size table

Percutaneous disc decompression has produced less number of complications [25–27]. There was one case that had the Perc-D SpineWand broken in the C4-5 space during PCN procedure (Fig. 3). The partial Perc-D Spine Wand broken in C4-5 space could not be moved by the percutaneous cervical discectomy and remain there itself. The clinical outcome of this case was good, and there were no complication occurrence. There was one case that had discitis in this study after PCD (Fig. 4). The patient presented with neck pain, associated radicular pain, and numbness in the left upper-limb after 8 days of PCD. The history was negative for fever and night pain symptoms. Cervical X-ray film showing narrowed C5-6 intervertebral space and erosive in the 5-cervical vertebral body wherein 3 months postoperatively shows narrowed disc space and sclerotic remodeling. There were some potential complications of PCN and PCD that include infections, bleeding, nerve damage, worsened pain, failure of technique, and recurrence of herniation. With the approach from the anterior neck to disk space, it is important to monitor the distance from the tip of the needle to the spinal canal. Therefore, needle monitoring is essential during this procedure. X-ray fluoroscopy confirmed the correct position of the needle tip during puncture of the needle on axial images, permitting accurate nucleoplasty of the intervertebral disk. The probe tip can be damaged or broken if it is forced against the vertebral endplates.

Conclusions

PCN and PCD treatments of contained cervical disc herniation show good outcomes and no differences on the stability of cervical spine were observed. Both treatments are safe, minimally invasive, and no difference in clinical outcomes.