- Email: [email protected]
A minimally invasive approach to defects of the pars interarticularis: Restoring function in competitive athletes
Clinical Neurology and Neurosurgery 139 (2015) 29–34
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
A minimally invasive approach to defects of the pars interarticularis: Restoring function in competitive athletes Christopher C Gillis a,∗ , Kurt Eichholz b , William J Thoman c , Richard G Fessler d a
Division of Neurosurgery, University of Nebraska Medical Center, Omaha, NE, United States Department of Neurosurgery, Vanderbilt University, Nashville, TN, United States Department of Neurological Surgery, The Ohio State University, Columbus, OH, United States d Department of Neurosurgery, Rush University Medical Center, Chicago, IL, United States b c
a r t i c l e
i n f o
Article history: Received 14 May 2015 Received in revised form 10 August 2015 Accepted 17 August 2015 Available online 28 August 2015 Keywords: Spondylolysis Lumbar spine Minimally invasive Spine Pars interarticularis Athletic spine injury
a b s t r a c t Objectives: To understand that young athletes have a higher incidence of pars interarticularis defects than the general population. This may be due to an immature spine put under higher stress loads at an early age. Traditionally, surgery was reserved for those who failed conservative therapy, and consisted of open exposure, bone grafting and placement of pedicle screws. This leads to a long recovery period and limited ability to return to competitive sport. Methods: Four collegiate and professional level athletes, three high school athletes, and one member of the National Guard presented with back pain from spondylolysis without spondylolisthesis. All underwent minimally invasive surgery (MIS) to directly repair the pars defect, for a total of sixteen pars defects repaired in eight patients. Described is an application of a MIS pars repair technique that has not previously been reported, which recreates the normal anatomy rather fusing across a motion segment. Results: Five patients were discharged the day following surgery and three were discharged on postoperative day 2. Six of the patients returned to their previous level of competitiveness. Two were unable to achieve the same level of play, both of whom failed to fuse the spondylolysis. Patients all initially reported clinical improvement postoperatively and there was overall mean improvement on patient reported outcome measures (SF36 physical and mental component scores, visual analog scale, and Oswestry disability index). Conclusion: MIS advantages include less muscle tissue disruption and restoration of the natural anatomy. This leads to a more rapid recovery, decreased perioperative pain, minimal blood loss, earlier mobilization and decreased hospital length of stay. Overall this allows the athlete to start therapy earlier and return to competition sooner and at his/her pre-operative competitive level. The described MIS repair technique outcomes are similar to those that have been reported in the literature and have allowed a high rate of return to athletics in high performing patients; critical to their quality of life. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Elite young athletes (teens to late 1920s) have a higher incidence of pars interarticularis defect than the general population. The incidence in the general population ranges from 3% to 6%, with 80% of them being asymptomatic [10,14]. In comparison, an incidence of 13.9% has been reported in young athletes, with the incidence
∗ Corresponding author. Division of Neurosurgery, University of Nebraska Medical Center, 982035 Nebraska Medical Center, Omaha, NE 68198-2035, United States. Tel.: +1 402 559 9611; fax: +1 402 559 7779. E-mail addresses: [email protected] (C.C. Gillis), [email protected] (K. Eichholz), [email protected] (W.J. Thoman), [email protected] (R.G. Fessler). http://dx.doi.org/10.1016/j.clineuro.2015.08.024 0303-8467/© 2015 Elsevier B.V. All rights reserved.
varying depending on the particular sport activity involved [18,24]. Examples of sports with a higher incidence of pars defects include gymnastics, football (specifically lineman), weight lifting, diving, and wrestling [10,23,24]. In a review of 3132 competitive athletes, Rossi and Dragoni found spondylolysis in 23% of weightlifters, 43% of divers and 30% of wrestlers [24]. In other reviews of the subject, Soler and Calderon [27] found spondylolysis in 17% of gymnasts on reviewing 3152 elite athletes, and Iwamoto et al. [12] found spondylolysis in 10.4% of 742 college football players. The exact etiology of spondylolysis has been extensively studied but remains a subject of ongoing deliberation. It is defined as a defect in the pars interarticularis, and can be either unilateral or bilateral in occurrence. It is likely an acquired defect, occurring due to acute or repetitive microtrauma injury during childhood or
30
C.C. Gillis et al. / Clinical Neurology and Neurosurgery 139 (2015) 29–34
during early adulthood. Evidence to support this includes Baker’s prospective study in the 1950s which found the incidence of a lytic pars defect to be 4.4% at 6 years of age, increasing to 6% by the time the population reached adulthood [9]. In addition, if there is any listhesis associated with the pars defect it occurs during childhood or early adulthood and stabilizes [2,9,11]. There may be a congenital weakness present in the pars of susceptible patients that predisposes them to an isthmic spondylolysis when associated with certain activities [9,33]. Alternatively, it may be that an immature spine, placed under supranormal stress at an early age is more prone to develop this type of injury. High-impact activities requiring alternating flexion and extension with repetition may lead to a fatigue or stress fracture of the pars interarticularis. The levels most susceptible to pars defects are L5 (85–95%) and L4 (5–15%) [28]. Recent advances in minimal access techniques have allowed many procedures which traditionally have been performed through an open, muscle dissecting approach to be performed through a muscle dilating approach. Minimally invasive spine (MIS) decompression has been shown in comparison with open laminectomy, to decrease hospital length of stay, postoperative pain score, narcotic requirement, and allow faster mobilization and postoperative recovery time. If this is also true in the spondylolysis scenario, it could translate to faster return to competitive sport in this specific patient population [17]. MIS has also been shown to be more cost-effective than traditional open procedures [1]. In this paper, we describe a novel, minimally invasive procedure which allows direct repair of the spondylolysis, without requiring fusion across a motion segment. This procedure is based on traditional pedicle screw insertion, which is a technique familiar to all spine surgeons. In this way, the patient receives not only the above-mentioned benefits of a minimally invasive approach, but also avoids the deleterious effects of fusion across a lumbar motion segment.
Fig. 1. T2-weighted sagittal cut magnetic resonance image (MRI) of L-spine showing relatively normal disk anatomy at L4-5 and L5-S1 in patient with L5 pars defect.
2. Methods We developed a minimally invasive surgery (MIS) for pars interarticularis defect repair using the Dynesys® cable system (Zimmer Holdings, Warsaw, Indiana). Four collegiate and professional level athletes (2 football, 1 volleyball, 1 hockey), three high school athletes (1 volleyball, 1 track, and 1 football), and one member of the National Guard, presenting with symptomatic spondylolysis without spondylolisthesis, underwent bilateral pars defect repair (total of 16 pars defects). Ages of the athletes at the time of surgery ranged from 16 to 23 years of age. The member of the National Guard was 46 years of age. All patients had failed medical management and were unable to participate in their high impact activities secondary to ongoing pain. Six of the patients had spondylolysis at L5, and two had pars defects at L3. Seven of the patients reported back pain exacerbated by extension. Four patients reported bilateral leg symptoms, which was positional. Disc pathology as the cause of pain was ruled out via MRI (see example in Fig. 1) preoperative injections into the pars defects provided relief of symptoms, confirming the diagnosis. Preoperative imaging consisted of MRI as well as CT and plain X-rays with flexion and extension (see Figs. 2 and 3). The patients were not required to wear postoperative external lumbar orthotic devices. Routine postoperative care included physical therapy, which started within 4–6 weeks of surgery. All patients had postoperative CT and serial plain films (see case examples in Figs. 4–6). Postoperative fusion was assessed by the senior author based on the presence of bony growth across the defect on X-ray and/or CT and lack of motion on flexion-extension X-rays. Patient reported outcome measures were assessed preoperatively and then postoperatively at last follow-up. Patient follow-up was performed by the senior author.
Fig. 2. Pre-operative computed tomography scan (CT) demonstrating defects in the (A) right pars and (B) left pars.
Fig. 3. Pre-operative CT coronal cut image showing bilateral pars defects.
C.C. Gillis et al. / Clinical Neurology and Neurosurgery 139 (2015) 29–34
31
Fig. 4. Post-operative upright X-rays (A) anterior–posterior, and (B) lateral views.
Fig. 7. Example of serial muscle dilation used to approach the pars defect and the placement of working channel with retractors affixed to the operating room table.
Fig. 5. Post-operative lateral upright X-rays in (A) flexion, and (B) extension.
3. Surgical technique The patient is positioned prone on a radiolucent frame or with chest rolls on a Jackson table. The level is confirmed by anterior–posterior and lateral fluoroscopy. Bilateral stab incisions 3 cm paramedian to the midline are performed. Kirschner (K)-wires are inserted under fluoroscopic guidance, directly overlying the pars defects with caution not to enter into the foramina. A scalpel is used to incise the fascia along the plane of the K-wires. A series of muscle dilators are then used to retract the paraspinous musculature over the defects. Working channels (Quadrant®, Medtronic, Minneapolis, MN is preferred by the senior author) are then placed over the pars defect with fluoroscopic confirmation (see Fig. 7). The defects are visually identified under loupe magnification with
Fig. 6. One-year post-operative CT showing: (A) left sided pedicle screw in the vertebral body and healing of the L5 pars between the superior and inferior articulating process, and (B) bilateral pedicle screws in place with bilateral bony healing visible medial to the screws.
headlight illumination through the working channels. The defects are debrided of all fibrous tissue using a combination of curettes and drill. The pars are drilled out until cancellous bone is identified. Using a curette and fluoroscopy, the defects are inspected to ensure adequate cancellous bone is exposed. A “sandwich” of corticocancellous bone inside a sponge of bone morphogenic protein (BMP) (Infuse, Medtronic Sofamor Danek, Memphis, TN) is placed into each defect. While preserving the facet capsule, the working channels are then readjusted to visualize the traditional pedicle screw entry points. Under both direct visualization and with fluoroscopy, pedicle screws are placed at the junction of the transverse process and facet of the vertebral body containing the lysis. Next, the tip of a disposable ventriculo-peritoneal shunt passer is bent into a “U” shaped curve. It is then percutaneously passed from one screw head underneath the fascia and inferior to the spinous process to the base of the tube on the contralateral side. This is done without having to modify the already created MIS incisions. The shunt passer is removed from the sheath and the Dynesys® cord (see Fig. 8) is introduced into the sheath (the cord is radiolucent and cannot be visualized on fluoroscopy or radiographs). The sheath and cord are then pulled back in the opposite motion bringing the cord with them to the other screw head (see Figs. 9 and 10). Spacers are applied to the cord to hold it in place, and the tips of the cord are threaded through the head of the screws. The cord system is then placed under tension to re-approximate the pars defect (already earlier filled with BMP and bone) together. The pars defect is visually inspected and fluoroscopy is obtained to verify alignment and reduction. The wound is irrigated and examined for hemostasis. At the end of the procedure, the working channel is removed and the wound closed with vicryl sutures in the fascia and in a subcuticular layer. Skin glue (Dermabond, Ethicon, Cincinnati, Ohio) is placed over both incisions.
32
C.C. Gillis et al. / Clinical Neurology and Neurosurgery 139 (2015) 29–34
Table 1 General patient demographics. Patient
Sport
Level
Length of stay
Fusion
1 2 3 4 5 6 7
Professional football College football College volleyball Professional hockey HS football HS track HS volleyball
L5 L5 L3 L5 L5 L3 L5
1d 2d 1d 1d 1d 2d 1d
Yes Yes No* Yes Yes Yes Yes*
8
National guard
L5
2d
Yes
Fig. 8. Photograph of the Dynesys® radiolucent cord and specialized pedicle screw, with a head for the cord to pull through, a set screw goes into the top of the ring to hold the cable in place.
4. Results
Complications/reoperations
Screw revision for radiculopathy Lumbar fusion performed 4 years later
Recurrence of symptoms at 18 months with pseudoarthrosis
Return to sport at previous level Yes Yes No Yes Yes Yes No Yes
Standard follow-up for patients was set at 6 weeks, 3 months, 6 months and 1 year. The patients all started physical therapy within 6 weeks of surgery (usually at 4 weeks). Patients started an exercise program 1–3 months after surgery and were able to return to their high-impact activity at 6 months. Three out of four collegiate and professional level athlete patients returned to the premorbid competition level of sport without missing any time from their regular season. The athlete who was fused 4 years after initial surgery was unable to return to sport. For the high school athletes and National Guard patient they were all able to return to high impact activity within 6 months to 1 year postoperatively. However, one high school athlete began to have pain again at 18 months postoperatively. A CT scan was performed, indicating fusion across the pars defect was incomplete. This athlete has since been unable to return to sport at his previous competitive level. The final outcome measures were administered between oneyear and fifty-four months postoperatively, depending on length of patient follow-up (see Table 2). Two patients were lost to long term follow-up and were not included in calculations performed on outcomes data. The mean change in SF-36v2 physical component score (PCS) and mental component score (MCS) from baseline to final
Sixteen pars defect repairs in eight patients were performed in total with this new technique. Six patients had repair of their L5 pars defects while two had L3 pars defects repaired (see Table 1 for summary). The average time of surgery was 127 min and blood loss was less than 50 cc. Five patients were discharged the following day, and three were discharged after two days. No complications were observed intraoperatively. One patient returned for repeat surgery to alleviate ongoing post-operative radiculopathy; a screw revision, for medial breech, was performed ten days after the initial surgery. This pain completely resolved within 3 months. For another patient, a fusion was performed at the same level by a different surgeon, at four years postoperatively. During the surgery, the Dynesys® cable was observed to be broken by the secondary surgeon.
Fig. 9. Illustrations of the Dynesys® cord placed under tension, allowing the surgeon to pull the pars defect together. The hashed black line represents the position of the cord, which is radiolucent.
Fig. 10. Schematic drawing of pars repair. The red line represents the pars defect. Yellow dots represent the screw head. The green line represents the cable placing tension on the posterior elements and re-approximating the pars defect (For interpretation of the color information in this figure legend, the reader is referred to the web version of the article.).
C.C. Gillis et al. / Clinical Neurology and Neurosurgery 139 (2015) 29–34 Table 2 Mean change in patient reported outcomes as recorded at last patient follow-up. Score (n = 6)
Mean change
SF36 physical component (PCS) SF 36 mental component (MCS) Oswestry disability index Visual analog score back Visual analog score leg
+12.2 +2.9 −29 −3 −1.5
outcomes time point showed an increase in 12.2 and 2.9 points, respectively. PCS improved more than the ten point standard deviation of the metric. The mean change in Oswestry Disability Index (ODI) was a decrease by 29 points. This corresponds to a change from a moderately disabled/borderline severely disabled preoperative score at mean 40 points, to minimally disabled postoperatively at a mean 11 points. The mean change in VAS back and leg scores was a decrease of 3 and 1.5 points, respectively. 5. Discussion The higher incidence of isthmic spondylolysis in young athletes may be due to an immature spine or a genetic predisposition to pars weakness combined with repetitive stresses. It has been shown that some sports have a higher incidence, especially those that tend to have situations requiring repeated loading in flexion and hyperextension [24]. Two of our patients were football players that played on the defensive line. These athletes are required to take a three-point stance, which puts their lumbar spine in a flexed (reduced lordosis) position. As the whistle blows, they are required to quickly transition with power into a hyperextended position as they block the opposition [9]. These athletes are large individuals and this blocking technique can cause forces greater than 8600 N at the L4-5 motion segment [3]. A third athlete in our group was a hockey goalie. Although this patient may not have been exposed to high levels of force, goalies spend a large amount of time in a crouching position while goaltending. This situation may relate to a study looking at the isthmic spondylolysis in an Inuit population that revealed a frequency as high as 60%. It was postulated that it may be due to a genetic predisposition in addition to the fact that Inuit spends long periods of time in a crouching position while skinning whale blubber [29]. Two of our patients were volleyball players. In this sport, athletes hyperextend their lower back as they go up to spike the ball in order to create power. In this case the athlete goes from an extended position to a flexed position as they follow through which follows with the theory of repetitive flexion and extension mechanism causing stress fractures in the pars. Traditionally, symptomatic pars defects were treated with rest, bracing, and physical therapy. This would include activity modifications as part of the therapy. Ninety percent of patients returned to their previous level of activity within 6 months [30]. Long term follow up on young athletes also showed approximately 90% good to excellent outcome 11 years out [16]. For patients performing at the highest level of competition, however, there is a great deal of pressure on return to play and conservative management can be time consuming for the athletes [30]. Surgery was reserved for patients who had failed at least 6 months of conservative therapy. Originally the surgery of choice was a posterior or posterolateral fusion with loss of motion across the fusion. The first motion sparing pars repair, introduced by Buck in 1970, consists of a screw fixation of the actual pars itself [6]. The Scott’s wiring [20] technique and the Morscher screw and hook technique [19] were introduced in the mid-eighties. The Buck screw technique requires a full exposure of the pars defect along with the entire lamina of the lumbar vertebra. The exposure must be long enough caudally in order to visualize the screw entry point and to keep the appropriate angle as the screw
33
goes through the pars defect [6]. An advantage to the Buck screw is that it addresses the pars defect directly. The disadvantage is that it required a wide exposure and is technically challenging in order to get the screw through the pars without breaking the walls. The construct relies completely on the screw and there are reported cases of screw breakage [8]. The Scott’s wiring technique involves exposure of the lamina and transverse processes of the affected level. A right angle instrument is passed underneath the transverse processes to allow for a loop of wire to be passed. One loop of wire is passed underneath each transverse process. The ends of the loops are then brought together by passing one of the wire loops through the interspinous ligament just caudal the spinous process of the affected vertebra. The bone graft is then placed into the defect and layered on top of the defect [20,22]. This method is considered technically less challenging than the Buck Screw technique but requires an even more aggressive lateral exposure in order to visualize the transverse processes. A modified Scott’s technique is similar to the procedure described above, but the transverse processes do not need to be fully exposed since pedicle screws are used to anchor the wire. An open technique utilizing similar instrumentation employed in this manuscript has been described by Lin [13]. A variation of the Morscher screw and hook technique involves a screw-rod-hook fixation for pars defect [19,31]. Again this requires a full open exposure. Once the defect is exposed and decorticated, bilateral pedicle screws are placed. The lamina is cleared caudally in order to place hooks underneath it bilaterally. The hooks are then connected to the pedicle screws with small rods. This technique provides a very strong construct but, again, requires a wide exposure. Additionally, there have been reports of hardware failure with the Morscher screw and hook construct [8]. Biomechanical testing showed that all of these described techniques resulted in satisfactory outcomes [7]. These open exposures affect the paraspinous musculature and lead to increased scar formation and muscle atrophy [5]. Healing for these procedures requires 6–12 months with greater than 90% of patients able to return to high impact activities but not all at the same competitive level [13,22,23]. Due to the concerns with open techniques, minimally invasive approaches to the Buck screw and the hook, rod and screw techniques have been proposed. Brennan et al. [4] reported a successful attempt using image guidance to repair bilateral pars defects. A true muscle dilation MIS modification of the Buck technique was described by Widi et al. [32] and most recently, Snyder et al. [26] reported a series of the Buck procedure through a small midline incision in 16 patients with symptom improvement in 94% of patients and they had 8 athletes in their series that were able to return to play at last follow-up. Noggle et al. [21] presented a series of five patients who had undergone pars repair with pedicle screws and hooks under a minimally invasive exposure. Sairyo et al. [25] performed a minimally invasive pars repair with a pedicle screw and hook-rod system in two patients. In all of these cases, patients did well without the morbidity of larger open cases. In our technique, we used a BMP ‘sandwich’ after denuding the pars defect of scar tissue to promote fusion. The use of BMP in the pars could be a concern if heterotopic ossification occurred and led to narrowing of the foramina, but this was not seen in any of our postoperative imaging and there were no cases of delayed radicular symptoms. Consideration can give to using other graft replacement materials in the pars defect in lieu of BMP. In comparing recently reported literature, our technique resulted in outcomes in operative time at 127 min and blood loss of less than 50 ml which was similar to that reported by Noggle et al. in their MIS approach. Our clinical outcomes are similar to that reported by Snyder et al. with pseudoarthrosis in one patient and one patient requiring screw revision for radiculopathy. We did
34
C.C. Gillis et al. / Clinical Neurology and Neurosurgery 139 (2015) 29–34
have a hardware failure in one patient where it was later observed that the cable had broken, leading to a delayed non-union. Our patients had their SF36v2 PCS scores improve significantly as well as their ODI. Two patients were lost to long term follow-up but had initially clinically improved after their surgeries, last assessed at 3 months after surgery, they were left out of patient reported outcome measures. 6. Conclusions Conservative management is successful in the majority of patients with spondylolysis, but when these measures fail, surgery becomes an option. We describe a novel MIS technique closest in relation to an open technique described by Lin [13] and his technique was a derivation of the modified Scott technique. This technique allows restoration of the natural anatomy with preservation of motion. The technical basis of the procedure is pedicle screw insertion through traditional trajectory, which is familiar to all spine surgeons and thus technically accessible. This technique was used in professional and collegiate athletes with success and allowed them to return to preoperative competitive level in high-impact activities, and with improvement in patient reported outcome measures. Disclosure Dr. Fessler receives royalties from Depuy, Medtronic and Stryker. He has an ownership interest in In Queue Innovations. There is no relationship with Zimmer, who created the system described—and had no effect on the instrumentation or development of technique described in this manuscript. There are no other financial disclosures and none of the other authors have any industry relationships. No funding was received in preparation of this manuscript. References [1] R.T. Allen, S.R. Garfin, The economics of minimally invasive spine surgery: the value perspective, Spine (Phil Pa 1976) 35 (26 Suppl.) (2010) S0375–S382. [2] W.J. Beutler, B.E. Fredrickson, A. Murtland, C.A. Sweeney, W.D. Grant, D. Baker, The natural history of spondylolysis and spondylolisthesis: 45-year follow-up evaluation, Spine (Phila Pa 1976) 28 (10) (2003) 1027–1035. [3] C.M. Bono, Low-back pain in athletes, J. Bone Joint Surg. Am. 86-A (2) (2004) 382–396. [4] R.P. Brennan, P.Y. Smucker, E.M. Horn, Minimally invasive image-guided direct repair of bilateral L-5 pars interarticularis defects, Neurosurg. Focus 25 (2) (2008) E13. [5] L.E.F.R. Bresnahan, R.G. Fessler, Paraspinal muscle changes on MRI following posterior lumbar spine surgery, in: AANS Annual Meeting, 2009, p. 1171. [6] J.E. Buck, Direct repair of the defect in spondylolisthesis. Preliminary report, J. Bone Joint Surg. Br. 52 (3) (1970) 432–437. [7] M. Deguchi, A.J. Rapoff, T.A. Zdeblick, Biomechanical comparison of spondylolysis fixation techniques, Spine (Phila Pa 1976) 24 (4) (1999) 328–333. [8] V. Dreyzin, S.I. Esses, A comparative analysis of spondylolysis repair, Spine (Phil Pa 1976) 19 (17) (1994) 1909–1914, discussion 1915.
[9] R.J. Ferguson, J.H. McMaster, C.L. Stanitski, Low back pain in college football linemen, J. Sports Med. 2 (2) (1974) 63–69. [10] B.E. Fredrickson, D. Baker, W.J. McHolick, H.A. Yuan, J.P. Lubicky, The natural history of spondylolysis and spondylolisthesis, J. Bone Joint Surg. Am. 66 (5) (1984) 699–707. [11] T. Ikata, R. Miyake, S. Katoh, T. Morita, M. Murase, Pathogenesis of sportsrelated spondylolisthesis in adolescents. Radiographic and magnetic resonance imaging study, Am. J. Sports Med. 24 (1) (1996) 94–98. [12] J. Iwamoto, H. Abe, Y. Tsukimura, K. Wakano, Relationship between radiographic abnormalities of lumbar spine and incidence of low back pain in high school and college football players: a prospective study, Am. J. Sports Med. 32 (3) (2004) 781–786. [13] P. Lin, Surgical repair of pars interarticularis defect, 12 year experience, Spine J. 7 (1S) (2007) 163S. [14] G. Logroscino, O. Mazza, G. Aulisa, L. Pitta, E. Pola, L. Aulisa, Spondylolysis and spondylolisthesis in the pediatric and adolescent population, Childs Nerv. Syst. 17 (11) (2001) 644–655. [16] S.F. Miller, J. Congeni, K. Swanson, Long-term functional and anatomical follow up of early detected spondylolysis in young athletes, Am. J. Sports Med. 32 (4) (2004) 928–933. [17] R.J. Mobbs, J. Li, P. Sivabalan, D. Raley, P.J. Rao, Outcomes after decompressive laminectomy for lumbar spinal stenosis: comparison between minimally invasive unilateral laminectomy for bilateral decompression and open laminectomy, J. Neurosurg. Spine 21 (2014) 179–186. [18] M. Muschik, H. Hahnel, P.N. Robinson, C. Perka, C. Muschik, Competitive sports and the progression of spondylolisthesis, J. Pediatr. Orthop. 16 (3) (1996) 364–369. [19] E. Morscher, B. Gerber, J. Fasel, Surgical treatment of spondylolisthesis by bone grafting and direct stabilization of spondylolysis by means of a hook screw, Arch. Orthop. Trauma Surg. 103 (3) (1984) 175–178. [20] R.O. Nicol, J.H. Scott, spondylolysis. Lytic, Repair by wiring, Spine (Phila Pa 1976) 11 (10) (1986) 1027–1030. [21] J.C. Noggle, D.M. Sciubba, A.F. Samdani, D.G. Anderson, R.R. Betz, J. Asghar, Minimally invasive direct repair of lumbar spondylolysis with a pedicle screw and hook construct, Neurosurg. Focus 25 (2) (2008) E15. [22] S. Nozawa, K. Shimizu, K. Miyamoto, M. Tanaka, Repair of pars interarticularis defect by segmental wire fixation in young athletes with spondylolysis, Am. J. Sports Med. 31 (3) (2003) 359–364. [23] C.A. Reitman, S.I. Esses, Direct repair of spondylolytic defects in young competitive athletes, Spine J. 2 (2) (2002) 142–144. [24] F.D.S. Rossi, The prevalence of spondylolysis and spondylolisthesis in symptomatic elite athletes: radiographic findings, Radiography 7 (2002). [25] K. Sairyo, T. Sakai, N. Yasui, Minimally invasive technique for direct repair of pars interarticularis defects in adults using a percutaneous pedicle screw and hook-rod system, J. Neurosurg. Spine 10 (5) (2009) 492–495. [26] L.A. Snyder, H. Shufflebarger, M.F. O’Brien, H. Thind, N. Theodore, U.K. Kakarla, Spondlyosis outcomes in adolescents after direct screw repair of the pars interarticularis, J. Neurosurg. Spine 21 (3) (2014) 329–333. [27] T. Soler, C. Calderon, The prevalence of spondylolysis in the Spanish elite athlete, Am. J. Sports Med. 28 (1) (2000) 57–62. [28] C.J. Standaert, S.A. Herring, B. Halpern, O. King, Spondylolysis, Phys. Med. Rehabil. Clin. N. Am. 11 (4) (2000) 785–803. [29] T.D. Stewart, The age incidence of neural-arch defects in Alaskan natives, considered from the standpoint of etiology, J. Bone Joint Surg. Am. 35-A (4) (1953) 937–950. [30] J. Sys, J. Michielsen, P. Bracke, M. Martens, J. Verstreken, Nonoperative treatment of active spondylolysis in elite athletes with normal X-ray findings: literature review and results of conservative treatment, Eur. Spine J. 10 (6) (2001) 498–504. [31] Y. Tokuhashi, H. Matsuzaki, Repair of defects in spondylolysis by segmental pedicular screw hook fixation. A preliminary report, Spine (Phila Pa 1976) 21 (17) (1996) 2041–2045. [32] G.A. Widi, S.K. Williams, A.D. Levi, Minimally invasive direct repair of bilateral pars defects in athletes, Case Rep. Med. 2013 (2013) 5. [33] L.L. Wiltse, Etiology of spondylolisthesis, Clin. Orthop. 10 (1957) 48–60.
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
A minimally invasive approach to defects of the pars interarticularis: Restoring function in competitive athletes Christopher C Gillis a,∗ , Kurt Eichholz b , William J Thoman c , Richard G Fessler d a
Division of Neurosurgery, University of Nebraska Medical Center, Omaha, NE, United States Department of Neurosurgery, Vanderbilt University, Nashville, TN, United States Department of Neurological Surgery, The Ohio State University, Columbus, OH, United States d Department of Neurosurgery, Rush University Medical Center, Chicago, IL, United States b c
a r t i c l e
i n f o
Article history: Received 14 May 2015 Received in revised form 10 August 2015 Accepted 17 August 2015 Available online 28 August 2015 Keywords: Spondylolysis Lumbar spine Minimally invasive Spine Pars interarticularis Athletic spine injury
a b s t r a c t Objectives: To understand that young athletes have a higher incidence of pars interarticularis defects than the general population. This may be due to an immature spine put under higher stress loads at an early age. Traditionally, surgery was reserved for those who failed conservative therapy, and consisted of open exposure, bone grafting and placement of pedicle screws. This leads to a long recovery period and limited ability to return to competitive sport. Methods: Four collegiate and professional level athletes, three high school athletes, and one member of the National Guard presented with back pain from spondylolysis without spondylolisthesis. All underwent minimally invasive surgery (MIS) to directly repair the pars defect, for a total of sixteen pars defects repaired in eight patients. Described is an application of a MIS pars repair technique that has not previously been reported, which recreates the normal anatomy rather fusing across a motion segment. Results: Five patients were discharged the day following surgery and three were discharged on postoperative day 2. Six of the patients returned to their previous level of competitiveness. Two were unable to achieve the same level of play, both of whom failed to fuse the spondylolysis. Patients all initially reported clinical improvement postoperatively and there was overall mean improvement on patient reported outcome measures (SF36 physical and mental component scores, visual analog scale, and Oswestry disability index). Conclusion: MIS advantages include less muscle tissue disruption and restoration of the natural anatomy. This leads to a more rapid recovery, decreased perioperative pain, minimal blood loss, earlier mobilization and decreased hospital length of stay. Overall this allows the athlete to start therapy earlier and return to competition sooner and at his/her pre-operative competitive level. The described MIS repair technique outcomes are similar to those that have been reported in the literature and have allowed a high rate of return to athletics in high performing patients; critical to their quality of life. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Elite young athletes (teens to late 1920s) have a higher incidence of pars interarticularis defect than the general population. The incidence in the general population ranges from 3% to 6%, with 80% of them being asymptomatic [10,14]. In comparison, an incidence of 13.9% has been reported in young athletes, with the incidence
∗ Corresponding author. Division of Neurosurgery, University of Nebraska Medical Center, 982035 Nebraska Medical Center, Omaha, NE 68198-2035, United States. Tel.: +1 402 559 9611; fax: +1 402 559 7779. E-mail addresses: [email protected] (C.C. Gillis), [email protected] (K. Eichholz), [email protected] (W.J. Thoman), [email protected] (R.G. Fessler). http://dx.doi.org/10.1016/j.clineuro.2015.08.024 0303-8467/© 2015 Elsevier B.V. All rights reserved.
varying depending on the particular sport activity involved [18,24]. Examples of sports with a higher incidence of pars defects include gymnastics, football (specifically lineman), weight lifting, diving, and wrestling [10,23,24]. In a review of 3132 competitive athletes, Rossi and Dragoni found spondylolysis in 23% of weightlifters, 43% of divers and 30% of wrestlers [24]. In other reviews of the subject, Soler and Calderon [27] found spondylolysis in 17% of gymnasts on reviewing 3152 elite athletes, and Iwamoto et al. [12] found spondylolysis in 10.4% of 742 college football players. The exact etiology of spondylolysis has been extensively studied but remains a subject of ongoing deliberation. It is defined as a defect in the pars interarticularis, and can be either unilateral or bilateral in occurrence. It is likely an acquired defect, occurring due to acute or repetitive microtrauma injury during childhood or
30
C.C. Gillis et al. / Clinical Neurology and Neurosurgery 139 (2015) 29–34
during early adulthood. Evidence to support this includes Baker’s prospective study in the 1950s which found the incidence of a lytic pars defect to be 4.4% at 6 years of age, increasing to 6% by the time the population reached adulthood [9]. In addition, if there is any listhesis associated with the pars defect it occurs during childhood or early adulthood and stabilizes [2,9,11]. There may be a congenital weakness present in the pars of susceptible patients that predisposes them to an isthmic spondylolysis when associated with certain activities [9,33]. Alternatively, it may be that an immature spine, placed under supranormal stress at an early age is more prone to develop this type of injury. High-impact activities requiring alternating flexion and extension with repetition may lead to a fatigue or stress fracture of the pars interarticularis. The levels most susceptible to pars defects are L5 (85–95%) and L4 (5–15%) [28]. Recent advances in minimal access techniques have allowed many procedures which traditionally have been performed through an open, muscle dissecting approach to be performed through a muscle dilating approach. Minimally invasive spine (MIS) decompression has been shown in comparison with open laminectomy, to decrease hospital length of stay, postoperative pain score, narcotic requirement, and allow faster mobilization and postoperative recovery time. If this is also true in the spondylolysis scenario, it could translate to faster return to competitive sport in this specific patient population [17]. MIS has also been shown to be more cost-effective than traditional open procedures [1]. In this paper, we describe a novel, minimally invasive procedure which allows direct repair of the spondylolysis, without requiring fusion across a motion segment. This procedure is based on traditional pedicle screw insertion, which is a technique familiar to all spine surgeons. In this way, the patient receives not only the above-mentioned benefits of a minimally invasive approach, but also avoids the deleterious effects of fusion across a lumbar motion segment.
Fig. 1. T2-weighted sagittal cut magnetic resonance image (MRI) of L-spine showing relatively normal disk anatomy at L4-5 and L5-S1 in patient with L5 pars defect.
2. Methods We developed a minimally invasive surgery (MIS) for pars interarticularis defect repair using the Dynesys® cable system (Zimmer Holdings, Warsaw, Indiana). Four collegiate and professional level athletes (2 football, 1 volleyball, 1 hockey), three high school athletes (1 volleyball, 1 track, and 1 football), and one member of the National Guard, presenting with symptomatic spondylolysis without spondylolisthesis, underwent bilateral pars defect repair (total of 16 pars defects). Ages of the athletes at the time of surgery ranged from 16 to 23 years of age. The member of the National Guard was 46 years of age. All patients had failed medical management and were unable to participate in their high impact activities secondary to ongoing pain. Six of the patients had spondylolysis at L5, and two had pars defects at L3. Seven of the patients reported back pain exacerbated by extension. Four patients reported bilateral leg symptoms, which was positional. Disc pathology as the cause of pain was ruled out via MRI (see example in Fig. 1) preoperative injections into the pars defects provided relief of symptoms, confirming the diagnosis. Preoperative imaging consisted of MRI as well as CT and plain X-rays with flexion and extension (see Figs. 2 and 3). The patients were not required to wear postoperative external lumbar orthotic devices. Routine postoperative care included physical therapy, which started within 4–6 weeks of surgery. All patients had postoperative CT and serial plain films (see case examples in Figs. 4–6). Postoperative fusion was assessed by the senior author based on the presence of bony growth across the defect on X-ray and/or CT and lack of motion on flexion-extension X-rays. Patient reported outcome measures were assessed preoperatively and then postoperatively at last follow-up. Patient follow-up was performed by the senior author.
Fig. 2. Pre-operative computed tomography scan (CT) demonstrating defects in the (A) right pars and (B) left pars.
Fig. 3. Pre-operative CT coronal cut image showing bilateral pars defects.
C.C. Gillis et al. / Clinical Neurology and Neurosurgery 139 (2015) 29–34
31
Fig. 4. Post-operative upright X-rays (A) anterior–posterior, and (B) lateral views.
Fig. 7. Example of serial muscle dilation used to approach the pars defect and the placement of working channel with retractors affixed to the operating room table.
Fig. 5. Post-operative lateral upright X-rays in (A) flexion, and (B) extension.
3. Surgical technique The patient is positioned prone on a radiolucent frame or with chest rolls on a Jackson table. The level is confirmed by anterior–posterior and lateral fluoroscopy. Bilateral stab incisions 3 cm paramedian to the midline are performed. Kirschner (K)-wires are inserted under fluoroscopic guidance, directly overlying the pars defects with caution not to enter into the foramina. A scalpel is used to incise the fascia along the plane of the K-wires. A series of muscle dilators are then used to retract the paraspinous musculature over the defects. Working channels (Quadrant®, Medtronic, Minneapolis, MN is preferred by the senior author) are then placed over the pars defect with fluoroscopic confirmation (see Fig. 7). The defects are visually identified under loupe magnification with
Fig. 6. One-year post-operative CT showing: (A) left sided pedicle screw in the vertebral body and healing of the L5 pars between the superior and inferior articulating process, and (B) bilateral pedicle screws in place with bilateral bony healing visible medial to the screws.
headlight illumination through the working channels. The defects are debrided of all fibrous tissue using a combination of curettes and drill. The pars are drilled out until cancellous bone is identified. Using a curette and fluoroscopy, the defects are inspected to ensure adequate cancellous bone is exposed. A “sandwich” of corticocancellous bone inside a sponge of bone morphogenic protein (BMP) (Infuse, Medtronic Sofamor Danek, Memphis, TN) is placed into each defect. While preserving the facet capsule, the working channels are then readjusted to visualize the traditional pedicle screw entry points. Under both direct visualization and with fluoroscopy, pedicle screws are placed at the junction of the transverse process and facet of the vertebral body containing the lysis. Next, the tip of a disposable ventriculo-peritoneal shunt passer is bent into a “U” shaped curve. It is then percutaneously passed from one screw head underneath the fascia and inferior to the spinous process to the base of the tube on the contralateral side. This is done without having to modify the already created MIS incisions. The shunt passer is removed from the sheath and the Dynesys® cord (see Fig. 8) is introduced into the sheath (the cord is radiolucent and cannot be visualized on fluoroscopy or radiographs). The sheath and cord are then pulled back in the opposite motion bringing the cord with them to the other screw head (see Figs. 9 and 10). Spacers are applied to the cord to hold it in place, and the tips of the cord are threaded through the head of the screws. The cord system is then placed under tension to re-approximate the pars defect (already earlier filled with BMP and bone) together. The pars defect is visually inspected and fluoroscopy is obtained to verify alignment and reduction. The wound is irrigated and examined for hemostasis. At the end of the procedure, the working channel is removed and the wound closed with vicryl sutures in the fascia and in a subcuticular layer. Skin glue (Dermabond, Ethicon, Cincinnati, Ohio) is placed over both incisions.
32
C.C. Gillis et al. / Clinical Neurology and Neurosurgery 139 (2015) 29–34
Table 1 General patient demographics. Patient
Sport
Level
Length of stay
Fusion
1 2 3 4 5 6 7
Professional football College football College volleyball Professional hockey HS football HS track HS volleyball
L5 L5 L3 L5 L5 L3 L5
1d 2d 1d 1d 1d 2d 1d
Yes Yes No* Yes Yes Yes Yes*
8
National guard
L5
2d
Yes
Fig. 8. Photograph of the Dynesys® radiolucent cord and specialized pedicle screw, with a head for the cord to pull through, a set screw goes into the top of the ring to hold the cable in place.
4. Results
Complications/reoperations
Screw revision for radiculopathy Lumbar fusion performed 4 years later
Recurrence of symptoms at 18 months with pseudoarthrosis
Return to sport at previous level Yes Yes No Yes Yes Yes No Yes
Standard follow-up for patients was set at 6 weeks, 3 months, 6 months and 1 year. The patients all started physical therapy within 6 weeks of surgery (usually at 4 weeks). Patients started an exercise program 1–3 months after surgery and were able to return to their high-impact activity at 6 months. Three out of four collegiate and professional level athlete patients returned to the premorbid competition level of sport without missing any time from their regular season. The athlete who was fused 4 years after initial surgery was unable to return to sport. For the high school athletes and National Guard patient they were all able to return to high impact activity within 6 months to 1 year postoperatively. However, one high school athlete began to have pain again at 18 months postoperatively. A CT scan was performed, indicating fusion across the pars defect was incomplete. This athlete has since been unable to return to sport at his previous competitive level. The final outcome measures were administered between oneyear and fifty-four months postoperatively, depending on length of patient follow-up (see Table 2). Two patients were lost to long term follow-up and were not included in calculations performed on outcomes data. The mean change in SF-36v2 physical component score (PCS) and mental component score (MCS) from baseline to final
Sixteen pars defect repairs in eight patients were performed in total with this new technique. Six patients had repair of their L5 pars defects while two had L3 pars defects repaired (see Table 1 for summary). The average time of surgery was 127 min and blood loss was less than 50 cc. Five patients were discharged the following day, and three were discharged after two days. No complications were observed intraoperatively. One patient returned for repeat surgery to alleviate ongoing post-operative radiculopathy; a screw revision, for medial breech, was performed ten days after the initial surgery. This pain completely resolved within 3 months. For another patient, a fusion was performed at the same level by a different surgeon, at four years postoperatively. During the surgery, the Dynesys® cable was observed to be broken by the secondary surgeon.
Fig. 9. Illustrations of the Dynesys® cord placed under tension, allowing the surgeon to pull the pars defect together. The hashed black line represents the position of the cord, which is radiolucent.
Fig. 10. Schematic drawing of pars repair. The red line represents the pars defect. Yellow dots represent the screw head. The green line represents the cable placing tension on the posterior elements and re-approximating the pars defect (For interpretation of the color information in this figure legend, the reader is referred to the web version of the article.).
C.C. Gillis et al. / Clinical Neurology and Neurosurgery 139 (2015) 29–34 Table 2 Mean change in patient reported outcomes as recorded at last patient follow-up. Score (n = 6)
Mean change
SF36 physical component (PCS) SF 36 mental component (MCS) Oswestry disability index Visual analog score back Visual analog score leg
+12.2 +2.9 −29 −3 −1.5
outcomes time point showed an increase in 12.2 and 2.9 points, respectively. PCS improved more than the ten point standard deviation of the metric. The mean change in Oswestry Disability Index (ODI) was a decrease by 29 points. This corresponds to a change from a moderately disabled/borderline severely disabled preoperative score at mean 40 points, to minimally disabled postoperatively at a mean 11 points. The mean change in VAS back and leg scores was a decrease of 3 and 1.5 points, respectively. 5. Discussion The higher incidence of isthmic spondylolysis in young athletes may be due to an immature spine or a genetic predisposition to pars weakness combined with repetitive stresses. It has been shown that some sports have a higher incidence, especially those that tend to have situations requiring repeated loading in flexion and hyperextension [24]. Two of our patients were football players that played on the defensive line. These athletes are required to take a three-point stance, which puts their lumbar spine in a flexed (reduced lordosis) position. As the whistle blows, they are required to quickly transition with power into a hyperextended position as they block the opposition [9]. These athletes are large individuals and this blocking technique can cause forces greater than 8600 N at the L4-5 motion segment [3]. A third athlete in our group was a hockey goalie. Although this patient may not have been exposed to high levels of force, goalies spend a large amount of time in a crouching position while goaltending. This situation may relate to a study looking at the isthmic spondylolysis in an Inuit population that revealed a frequency as high as 60%. It was postulated that it may be due to a genetic predisposition in addition to the fact that Inuit spends long periods of time in a crouching position while skinning whale blubber [29]. Two of our patients were volleyball players. In this sport, athletes hyperextend their lower back as they go up to spike the ball in order to create power. In this case the athlete goes from an extended position to a flexed position as they follow through which follows with the theory of repetitive flexion and extension mechanism causing stress fractures in the pars. Traditionally, symptomatic pars defects were treated with rest, bracing, and physical therapy. This would include activity modifications as part of the therapy. Ninety percent of patients returned to their previous level of activity within 6 months [30]. Long term follow up on young athletes also showed approximately 90% good to excellent outcome 11 years out [16]. For patients performing at the highest level of competition, however, there is a great deal of pressure on return to play and conservative management can be time consuming for the athletes [30]. Surgery was reserved for patients who had failed at least 6 months of conservative therapy. Originally the surgery of choice was a posterior or posterolateral fusion with loss of motion across the fusion. The first motion sparing pars repair, introduced by Buck in 1970, consists of a screw fixation of the actual pars itself [6]. The Scott’s wiring [20] technique and the Morscher screw and hook technique [19] were introduced in the mid-eighties. The Buck screw technique requires a full exposure of the pars defect along with the entire lamina of the lumbar vertebra. The exposure must be long enough caudally in order to visualize the screw entry point and to keep the appropriate angle as the screw
33
goes through the pars defect [6]. An advantage to the Buck screw is that it addresses the pars defect directly. The disadvantage is that it required a wide exposure and is technically challenging in order to get the screw through the pars without breaking the walls. The construct relies completely on the screw and there are reported cases of screw breakage [8]. The Scott’s wiring technique involves exposure of the lamina and transverse processes of the affected level. A right angle instrument is passed underneath the transverse processes to allow for a loop of wire to be passed. One loop of wire is passed underneath each transverse process. The ends of the loops are then brought together by passing one of the wire loops through the interspinous ligament just caudal the spinous process of the affected vertebra. The bone graft is then placed into the defect and layered on top of the defect [20,22]. This method is considered technically less challenging than the Buck Screw technique but requires an even more aggressive lateral exposure in order to visualize the transverse processes. A modified Scott’s technique is similar to the procedure described above, but the transverse processes do not need to be fully exposed since pedicle screws are used to anchor the wire. An open technique utilizing similar instrumentation employed in this manuscript has been described by Lin [13]. A variation of the Morscher screw and hook technique involves a screw-rod-hook fixation for pars defect [19,31]. Again this requires a full open exposure. Once the defect is exposed and decorticated, bilateral pedicle screws are placed. The lamina is cleared caudally in order to place hooks underneath it bilaterally. The hooks are then connected to the pedicle screws with small rods. This technique provides a very strong construct but, again, requires a wide exposure. Additionally, there have been reports of hardware failure with the Morscher screw and hook construct [8]. Biomechanical testing showed that all of these described techniques resulted in satisfactory outcomes [7]. These open exposures affect the paraspinous musculature and lead to increased scar formation and muscle atrophy [5]. Healing for these procedures requires 6–12 months with greater than 90% of patients able to return to high impact activities but not all at the same competitive level [13,22,23]. Due to the concerns with open techniques, minimally invasive approaches to the Buck screw and the hook, rod and screw techniques have been proposed. Brennan et al. [4] reported a successful attempt using image guidance to repair bilateral pars defects. A true muscle dilation MIS modification of the Buck technique was described by Widi et al. [32] and most recently, Snyder et al. [26] reported a series of the Buck procedure through a small midline incision in 16 patients with symptom improvement in 94% of patients and they had 8 athletes in their series that were able to return to play at last follow-up. Noggle et al. [21] presented a series of five patients who had undergone pars repair with pedicle screws and hooks under a minimally invasive exposure. Sairyo et al. [25] performed a minimally invasive pars repair with a pedicle screw and hook-rod system in two patients. In all of these cases, patients did well without the morbidity of larger open cases. In our technique, we used a BMP ‘sandwich’ after denuding the pars defect of scar tissue to promote fusion. The use of BMP in the pars could be a concern if heterotopic ossification occurred and led to narrowing of the foramina, but this was not seen in any of our postoperative imaging and there were no cases of delayed radicular symptoms. Consideration can give to using other graft replacement materials in the pars defect in lieu of BMP. In comparing recently reported literature, our technique resulted in outcomes in operative time at 127 min and blood loss of less than 50 ml which was similar to that reported by Noggle et al. in their MIS approach. Our clinical outcomes are similar to that reported by Snyder et al. with pseudoarthrosis in one patient and one patient requiring screw revision for radiculopathy. We did
34
C.C. Gillis et al. / Clinical Neurology and Neurosurgery 139 (2015) 29–34
have a hardware failure in one patient where it was later observed that the cable had broken, leading to a delayed non-union. Our patients had their SF36v2 PCS scores improve significantly as well as their ODI. Two patients were lost to long term follow-up but had initially clinically improved after their surgeries, last assessed at 3 months after surgery, they were left out of patient reported outcome measures. 6. Conclusions Conservative management is successful in the majority of patients with spondylolysis, but when these measures fail, surgery becomes an option. We describe a novel MIS technique closest in relation to an open technique described by Lin [13] and his technique was a derivation of the modified Scott technique. This technique allows restoration of the natural anatomy with preservation of motion. The technical basis of the procedure is pedicle screw insertion through traditional trajectory, which is familiar to all spine surgeons and thus technically accessible. This technique was used in professional and collegiate athletes with success and allowed them to return to preoperative competitive level in high-impact activities, and with improvement in patient reported outcome measures. Disclosure Dr. Fessler receives royalties from Depuy, Medtronic and Stryker. He has an ownership interest in In Queue Innovations. There is no relationship with Zimmer, who created the system described—and had no effect on the instrumentation or development of technique described in this manuscript. There are no other financial disclosures and none of the other authors have any industry relationships. No funding was received in preparation of this manuscript. References [1] R.T. Allen, S.R. Garfin, The economics of minimally invasive spine surgery: the value perspective, Spine (Phil Pa 1976) 35 (26 Suppl.) (2010) S0375–S382. [2] W.J. Beutler, B.E. Fredrickson, A. Murtland, C.A. Sweeney, W.D. Grant, D. Baker, The natural history of spondylolysis and spondylolisthesis: 45-year follow-up evaluation, Spine (Phila Pa 1976) 28 (10) (2003) 1027–1035. [3] C.M. Bono, Low-back pain in athletes, J. Bone Joint Surg. Am. 86-A (2) (2004) 382–396. [4] R.P. Brennan, P.Y. Smucker, E.M. Horn, Minimally invasive image-guided direct repair of bilateral L-5 pars interarticularis defects, Neurosurg. Focus 25 (2) (2008) E13. [5] L.E.F.R. Bresnahan, R.G. Fessler, Paraspinal muscle changes on MRI following posterior lumbar spine surgery, in: AANS Annual Meeting, 2009, p. 1171. [6] J.E. Buck, Direct repair of the defect in spondylolisthesis. Preliminary report, J. Bone Joint Surg. Br. 52 (3) (1970) 432–437. [7] M. Deguchi, A.J. Rapoff, T.A. Zdeblick, Biomechanical comparison of spondylolysis fixation techniques, Spine (Phila Pa 1976) 24 (4) (1999) 328–333. [8] V. Dreyzin, S.I. Esses, A comparative analysis of spondylolysis repair, Spine (Phil Pa 1976) 19 (17) (1994) 1909–1914, discussion 1915.
[9] R.J. Ferguson, J.H. McMaster, C.L. Stanitski, Low back pain in college football linemen, J. Sports Med. 2 (2) (1974) 63–69. [10] B.E. Fredrickson, D. Baker, W.J. McHolick, H.A. Yuan, J.P. Lubicky, The natural history of spondylolysis and spondylolisthesis, J. Bone Joint Surg. Am. 66 (5) (1984) 699–707. [11] T. Ikata, R. Miyake, S. Katoh, T. Morita, M. Murase, Pathogenesis of sportsrelated spondylolisthesis in adolescents. Radiographic and magnetic resonance imaging study, Am. J. Sports Med. 24 (1) (1996) 94–98. [12] J. Iwamoto, H. Abe, Y. Tsukimura, K. Wakano, Relationship between radiographic abnormalities of lumbar spine and incidence of low back pain in high school and college football players: a prospective study, Am. J. Sports Med. 32 (3) (2004) 781–786. [13] P. Lin, Surgical repair of pars interarticularis defect, 12 year experience, Spine J. 7 (1S) (2007) 163S. [14] G. Logroscino, O. Mazza, G. Aulisa, L. Pitta, E. Pola, L. Aulisa, Spondylolysis and spondylolisthesis in the pediatric and adolescent population, Childs Nerv. Syst. 17 (11) (2001) 644–655. [16] S.F. Miller, J. Congeni, K. Swanson, Long-term functional and anatomical follow up of early detected spondylolysis in young athletes, Am. J. Sports Med. 32 (4) (2004) 928–933. [17] R.J. Mobbs, J. Li, P. Sivabalan, D. Raley, P.J. Rao, Outcomes after decompressive laminectomy for lumbar spinal stenosis: comparison between minimally invasive unilateral laminectomy for bilateral decompression and open laminectomy, J. Neurosurg. Spine 21 (2014) 179–186. [18] M. Muschik, H. Hahnel, P.N. Robinson, C. Perka, C. Muschik, Competitive sports and the progression of spondylolisthesis, J. Pediatr. Orthop. 16 (3) (1996) 364–369. [19] E. Morscher, B. Gerber, J. Fasel, Surgical treatment of spondylolisthesis by bone grafting and direct stabilization of spondylolysis by means of a hook screw, Arch. Orthop. Trauma Surg. 103 (3) (1984) 175–178. [20] R.O. Nicol, J.H. Scott, spondylolysis. Lytic, Repair by wiring, Spine (Phila Pa 1976) 11 (10) (1986) 1027–1030. [21] J.C. Noggle, D.M. Sciubba, A.F. Samdani, D.G. Anderson, R.R. Betz, J. Asghar, Minimally invasive direct repair of lumbar spondylolysis with a pedicle screw and hook construct, Neurosurg. Focus 25 (2) (2008) E15. [22] S. Nozawa, K. Shimizu, K. Miyamoto, M. Tanaka, Repair of pars interarticularis defect by segmental wire fixation in young athletes with spondylolysis, Am. J. Sports Med. 31 (3) (2003) 359–364. [23] C.A. Reitman, S.I. Esses, Direct repair of spondylolytic defects in young competitive athletes, Spine J. 2 (2) (2002) 142–144. [24] F.D.S. Rossi, The prevalence of spondylolysis and spondylolisthesis in symptomatic elite athletes: radiographic findings, Radiography 7 (2002). [25] K. Sairyo, T. Sakai, N. Yasui, Minimally invasive technique for direct repair of pars interarticularis defects in adults using a percutaneous pedicle screw and hook-rod system, J. Neurosurg. Spine 10 (5) (2009) 492–495. [26] L.A. Snyder, H. Shufflebarger, M.F. O’Brien, H. Thind, N. Theodore, U.K. Kakarla, Spondlyosis outcomes in adolescents after direct screw repair of the pars interarticularis, J. Neurosurg. Spine 21 (3) (2014) 329–333. [27] T. Soler, C. Calderon, The prevalence of spondylolysis in the Spanish elite athlete, Am. J. Sports Med. 28 (1) (2000) 57–62. [28] C.J. Standaert, S.A. Herring, B. Halpern, O. King, Spondylolysis, Phys. Med. Rehabil. Clin. N. Am. 11 (4) (2000) 785–803. [29] T.D. Stewart, The age incidence of neural-arch defects in Alaskan natives, considered from the standpoint of etiology, J. Bone Joint Surg. Am. 35-A (4) (1953) 937–950. [30] J. Sys, J. Michielsen, P. Bracke, M. Martens, J. Verstreken, Nonoperative treatment of active spondylolysis in elite athletes with normal X-ray findings: literature review and results of conservative treatment, Eur. Spine J. 10 (6) (2001) 498–504. [31] Y. Tokuhashi, H. Matsuzaki, Repair of defects in spondylolysis by segmental pedicular screw hook fixation. A preliminary report, Spine (Phila Pa 1976) 21 (17) (1996) 2041–2045. [32] G.A. Widi, S.K. Williams, A.D. Levi, Minimally invasive direct repair of bilateral pars defects in athletes, Case Rep. Med. 2013 (2013) 5. [33] L.L. Wiltse, Etiology of spondylolisthesis, Clin. Orthop. 10 (1957) 48–60.