ORIGINAL ARTICLE
Spinal Decompression in Achondroplastic Patients Using High-Speed Drill Versus Ultrasonic Bone Curette: Technical Note and Outcomes in 30 Cases Mohamad Bydon, MD,* Mohamed Macki, BS,* Risheng Xu, AM,*w Michael C. Ain, MD,* Edward S. Ahn, MD,* and George I. Jallo, MD*
Background: This manuscript describes the clinical and operative characteristics of achondroplastic children who undergo multilevel thoracolumbar decompressions using either the highspeed drill or the ultrasonic bone curette (BoneScalpel). Methods: We retrospectively reviewed 30 thoracolumbar decompressions in achondroplastic patients at a single institution between 2008 and 2013. Patients were classified into either the high-speed drill cohort or the BoneScalpel cohort, depending on which instrument was utilized to perform the decompression. A technical note on the role of the ultrasonic bone curette in decompressing stenotic achondroplastic spines is also provided. Results: In comparison with the high-speed drill cohort, the BoneScalpel cohort experienced less overall perioperative complications, including durotomy, cerebrospinal fluid leak, pseudomeningoceles, wound infection, and wound dehiscence. Although 45.0% of patients experienced a durotomy in the highspeed drill cohort, only 30.0% of patients experienced a durotomy in the BoneScalpel cohort (P = 0.694). In the high-speed drill cohort, the number of patients complaining of sensory disturbances, back pain, ataxia, incontinence, neurogenic claudication, radiculopathy, ataxia, and/or weakness decreased postoperatively. Similar results were observed in the BoneScalpel cohort. Conclusions: Although spinal decompression provides symptomatic resolution in patients with achondroplasia, intraoperative complications, in general, and durotomies, in particular, are common. Here, we report a decreased incidence in intraoperative durotomy and overall perioperative complication rates in the BoneScalpel cohort, although this did not reach the level of statistical significance. Nonetheless, the data demonFrom the *Department of Neurosurgery; and wMedical Scientist Training Program, Johns Hopkins University School of Medicine, Baltimore, MD. M.B. and M.M. contributed equally. G.I.J. receives fellowship support from Medtronic. M.C.A. is a consultant for Stryker Spine. The remaining authors declare no conflicts of interest. Reprints: George I. Jallo, MD, Division of Pediatric Neurosurgery, Johns Hopkins Hospital, 600 N Wolfe Street, Phipps 556, Baltimore, MD 21287. E-mail: [email protected]. Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Website, www. pedorthopaedics.com. Copyright r 2014 by Lippincott Williams & Wilkins
780 | www.pedorthopaedics.com
strate that the BoneScalpel is a safe and efficacious alternative to the high-speed drill in these challenging patients. Level of evidence: Level II—retrospective study. Key Words: achondroplasia, ultrasonic bone curette, pediatric, spine, surgery, outcomes (J Pediatr Orthop 2014;34:780–786)
A
chondroplasia is a genetic bone disorder that causes the most common type of dwarfism. Although the length of the vertebral column is not as severely affected as the limbs, the vertebrae undergo structural changes that may compromise the spinal cord.1 During in utero development, achondroplasia causes impaired longitudinal growth of the posterior arches, accounting for shortened pedicles and interpedicle distances. These developmental malformations result in nerve root compression and spinal stenosis.2,3 Although narrowing of the lumbar canal occurs in all achondroplastic spines, 78% subsequently develop neurological symptoms,4–6 and one third requiring surgical intervention.6–18 Due to the significantly altered spinal anatomy in achondroplastic patients, considerable challenges arise when decompressing the narrowed spinal canal.19–21 A review of the literature indicates that durotomies and cerebrospinal fluid (CSF) leaks are the most common perioperative complications.1,22–24 In this manuscript, we describe the role of an ultrasonic bone curette, the BoneScalpel, in performing a multilevel thoracolumbar decompression of the achondroplastic spine. To understand the risks and benefits of the bone curette in this challenging patient population, we review surgical outcomes in 30 decompressions in achondroplastic patients using either the conventional highspeed drill or the BoneScalpel. In addition, we provide a technical note of a thoracolumbar laminectomy in the achondroplastic spine using the ultrasonic bone curette.
METHODS Patient Population Over the 6-year period between 2008 and 2013, we collected all thoracolumbar decompressions for spinal stenosis on patients with achondroplasia at a single J Pediatr Orthop
Volume 34, Number 8, December 2014
J Pediatr Orthop
Volume 34, Number 8, December 2014
institution. In 2008, the BoneScalpel, an ultrasonic bone curette, was introduced to our institution. To explore the BoneScalpel utility in complex spine surgeries, surgeons randomly assigned the ultrasonic bone curette to achondroplastic decompressions, thereby eliminating selection biases. Only those surgeons who specialize in spinal decompressions on achondroplastic spines utilized the BoneScalpel, thereby eliminating any operator-dependent biases. We retrospectively divided the sample population into a high-speed drill cohort and BoneScalpel cohort. The cases were reviewed retrospectively using operative narratives and clinical notes. Both first-time surgeries and reoperations were identified. Intraoperative and perioperative complications were noted with a particular emphasis on intraoperative durotomy and postoperative CSF leak. All durotomies were due to dural tears, which in turn were caused by the surgical instruments: Kerrison punch or the BoneScalpel. Only symptomatic CSF leaks or incidental leaks identified on routine imaging were documented in our study.
Device Description The ultrasonic bone curette (BoneScalpel) is a handheld device designed for precise and rapid osteotomies. It consists of an electronic-control headpiece, an oscillating scalpel tip, and a jet nozzle for irrigation. With a 0.5 mm width and 20 mm length, the metal tip oscillates with a frequency of 22.5 kHz (22,500 strokes per second). When placed on bone, the delivery of ultrasonic oscillations cuts through the lamina with minimal movement of the handpiece. Upon contact with solid osseous tissue, the bone curette blade does not bend so that bone may be incised. However, soft tissue contact causes the bone curette blade to bend, protecting them from incidental incisions (eg, durotomies). The energy delivered by the oscillating tip also facilitates blood coagulation without necrosis of the surrounding tissue. The shaving tip enables undercutting of the lamina. In addition, unlike the rotational forces of the high-speed drill, the beveled design of the bone curette delivers ultrasonic pulses directly to bone, thereby sparing the dura and spinal cord. This is particularly important in the achondroplastic spinal canal where the shortened interpedicle distance compromises the space between the inner spinal column and dura. In our institutional experience, the BoneScalpel did not interfere with intraoperative neurophysiological monitoring.
Surgical Description: Thoracolumbar Decompressions After general endotracheal anesthesia and sterile preparation, the patient is placed in the prone position (See Supplemental Digital Content, http://links.lww.com/ BPO/A16). The procedure begins with a midline incision over the lumbar spine. The spinous processes and lamina are accessed by dissection through the paraspinal muscles in a subperiosteal manner. The spinalis muscles are deflected from the lamina and spinous processes to allow for better exposure of the posterior elements. The operation was then carried out in the following order: (1) inr
2014 Lippincott Williams & Wilkins
Spinal Decompression in Achondroplasia
strumentation of the posterior elements with pedicle screws; (2) laminectomy utilizing the BoneScalpel; and (3) placement of the contoured rods. Before decompression, almost all patients (28 out of 30) underwent spinal instrumented fusions to (1) correct the spinal malalignment in achondroplastic patients; and/or (2) prevent postlaminectomy kyphosis in children. Two patients had a prior instrumented fusion at the index level of decompression. With neuromonitoring, instrumented fusion is carried out with titanium pedicle screws under fluoroscopic guidance. After the spine is properly instrumented, the surgeon’s attention is turned to the decompression. Initially, interspinous ligaments are preserved to provide traction during the excision of laminar bone in an achondroplastic spine. With an incision no greater than 7 mm in width, the spinolaminar junction is cut bilaterally using the bone curette. Despite the thickened lamina in these patients, the bone curette reaches depths of up to 20 mm in a single stroke. The device continues to purchase laminar bone until the ligamentum flavum is reached. Using the Kerrison punches, the undersurface of the excised spinous process-laminar complex is then isolated from the posterior ligaments of the spinal canal. The interspinous ligament is transected above and below the spinous process, and the laminar complex is removed en bloc from the intact spine. Following the decompression, the rod is contoured, and the spinal alignment is corrected (Fig. 1). Bone graft using autograft from the laminectomy in addition to allograft is placed in the lateral gutters to reinforce the arthrodesis. The wound is closed in a multilayer manner, and a drain is placed above the fascia.
RESULTS Between 2008 and 2013, 30 operations were carried out for the management of symptomatic spinal stenosis in achondroplastic patients. All patients underwent decompression of the thoracolumbar spine with either a highspeed drill (Table 1) or the BoneScalpel (Table 2). The high-speed drill was used in 20 cases (66.7%), whereas the BoneScalpel was used in 10 cases (33.3%). The cases included 15 (50.0%) revision surgeries and 15 (50.0%) firsttime operations. Of the 15 reoperations, 1 was observed in the BoneScalpel cohort. Of the study cohort of 30 patients, 70.0% were male (n = 21). Mean age (± SD) at the time of surgery was 15.0 ± 4.79 in the high-speed drill cohort and 14.10 ± 5.63 in the BoneScalpel cohort (P = 0.650). Between the 2 cohorts, patients did not statistically differ in terms of follow-up time from index surgery and length of hospital stay. Mean follow-up time was 21.90 ± 8.53 and 20.70 ± 6.85 months in the high-speed drill and BoneScalpel cohorts, respectively (P = 0.702). Mean length of hospital stay in the high-speed drill cohort (6.5 ± 1.99 d) did not statistically differ from the BoneScalpel cohort (6.50 ± 2.64 d) (P = 1.000). However, the median number of levels decompressed in the highspeed drill cohort, 3, was statistically higher than the median 5.5 levels decompressed in the BoneScalpel cohort www.pedorthopaedics.com |
781
Bydon et al
J Pediatr Orthop
Volume 34, Number 8, December 2014
FIGURE 1. Images of an achondroplastic child with stenosis of the thoracolumbar spine (T10-L4). A, Preoperative T2-weighted magnetic resonance imaging (MRI), axial image. B, Preoperative T2-weighted MRI, sagittal image. C, Preoperative, lateral x-ray. D, Postoperative, lateral x-ray with instrumentation from T10 to L4.
(P = 0.005), whereas a median of 6 levels were fused in both the high-speed drill and BoneScalpel cohorts (P = 0.687). An exact McNemar test was used to compare the number of patients with preoperative and postoperative symptoms in the high-speed drill and BoneScalpel cohorts (Table 3). In the high-speed drill group, the number of symptomatic patients postoperatively decreased in all measured findings. The number of patients with sensory disturbances decreased by 77.7%, back pain by 85.71%, incontinence by 85.71%, neurogenic claudication by 100.00%, radiculopathy by 66.67%, ataxia by 25.00%, and weakness by 75.00%. Differences in sensory disturbances, back pain, incontinence, and neurogenic claudication reached statistical significance. In the BoneScalpel cohort, all patients complaining of preoperative neurogenic claudication and weakness experienced complete resolution of their symptoms. The number of patients complaining of sensory disturbances decreased by 25.00%, back pain by 66.67%, incontinence by 50.00%, and ataxia by 60.00%. Improvement in neurogenic claudication and weakness reached statistical significance. Following Cobb angle measurements, the spinal alignment corrected by a mean of 11.88 degrees (P = 0.002). Ten patients in the high-speed drill cohort experienced at least 1 perioperative complication, whereas 3 patients in the BoneScalpel cohort experienced clinically insignificant durotomies (without CSF leaks). Individual complications in the high-speed drill cohort and the BoneScalpel cohort were compared with Fisher exact test (Table 4). During the 10 cases in which the BoneScalpel was used, the incidence of durotomies decreased by 15.00% when compared with the high-speed drill cases (P = 0.694). Of the 9 durotomies (45.0%) in the highspeed drill cohort, postoperative cerebrospinal fluid (CSF) leaks occurred in 2 patients (10.0%), one of whom developed a pseudomeningocele and seroma (5.0%). Other complications in the high-speed drill cohort in-
782 | www.pedorthopaedics.com
cluded wound infection (10.0%) and wound dehiscence (5.0%). The cases with pseudomeningoceles and wound dehiscence required surgical management with a complex plastics closure. Taken together, the BoneScalpel cohort experienced less intraoperative and perioperative complications, although this did not reach statistical significance (P = 0.440).
DISCUSSION Symptomatic spinal stenosis is a common diagnosis in achondroplastic children.2–4,6,7,12,20,21,25–32 Management of spinal stenosis in achondroplastic children requires surgical intervention with a multilevel decompression. In a retrospective review of 18 achondroplastic children undergoing spinal decompressions, Baca et al12 assigned each patient a symptomatic score (SS) based on the presence of lower extremity weakness, paresthesias, reflexes, incontinence, and ataxia. Mean preoperative SS values (4.0 ± 0.9) significantly differed from postoperative SS values (1.6 ± 1.7). Although patients experience postoperative symptomatic relief, perioperative complications in achondroplastic patients underscore the technical difficulties of operating on an achondroplasic spine.1,26,27,29,31 Although the overall complication rate after surgical decompressions in the general population is approximately 22%, this rate almost triples to 61% in achondroplastic patients.15,30 Lutter et al1 reports the most common complications encountered as dural tears (37%), followed by neurological complications (23%). Surgical-site infections (9%), deep vein thrombosis (3%), pulmonary complications (3%), and gastrointestinal complications (3%) were also reported. These values corroborate our complication rates in the conventional high-speed drill cohort with durotomies at 45.0% and wound infections at 10.0%. Wang et al24 cited an even higher durotomy rate at 55%. These durotomy rates in achondroplastic patients are strikingly high in comparison r
2014 Lippincott Williams & Wilkins
r
14
21
30
22
15
15/M
13/M
8/M
20/F
14/M
17/M
22/F
3
4
5
2014 Lippincott Williams & Wilkins
6
7
8
9
10 7/M
11 13/M
12 13/M
13 16/F
17
19
17
15 12/M
16 23/M
Paresthesia
Ataxia
Radiculopathy
Asymptomatic
Urinary incontinence Ataxia
Asymptomatic
Asymptomatic
Asymptomatic
Paresthesias
Low back pain
Asymptomatic
Ataxia
Asymptomatic
Asymptomatic
Asymptomatic
8
8
5
7
5
7
4
6
13
7
8
4
5
7
6
5
x
x
x
x
x
x
x
x x
x
Postoperative LOS Postoperative Wound Wound Condition (d) Durotomy CSF Leak Pseudomeningocele Infection Dehiscence
Volume 34, Number 8, December 2014
14 14/M
28
30
12
18
42
18
Surgery
L4-S1 decompression T4-L4 instrumentation T10-L2 decompression T8-L5 instrumentation Urinary incontinence, neurogenic Reop, thoracolumbar claudication decompression T10-L4 instrumented fusion Ataxia, paresthesia Reop, lumbosacral decompression Neurogenic claudication L2-L4 decompression T10-L4 instrumentation Low back pain, radiculopathy Reop, T6-T9, T12-L4 decompression T4-L4 instrumented fusion Lower extremity numbness Reop, T10-L2 decompression T10-L4 instrumented fusion Low back pain, headaches T12-L5 decompression T8-L4 instrumented fusion Back pain, leg numbness Reop, T12-L5 decompression T9-L4 instrumented fusion Back pain, weakness L1-L4 decompression T10-L4 instrumentation Neurogenic claudication, urinary Reop, L4-S1 incontinence decompression Ataxia, urinary incontinence Reop, L4-S1 decompression T6-L5 instrumented fusion Back pain, radiculopathy, numbness, Reop, L4-S1 urinary incontinence decompression T12-S1 instrumented fusion Radiculopathy, paresthesia, sleep Reop, L1-L5 apnea, urinary and fecal incontinence decompression T11-L4 instrumented fusion Ataxia, neurogenic claudication, Reop, L4-S1 urinary incontinence decompression T9-L5 instrumented fusion Paresthesia, weakness Reop, L4-S1 decompression L2-sacral instrumented fusion
Hypoesthesia and paresthesia, back pain, neurogenic claudication Hyporeflexia, weakness
Preoperative Indications
Complications
27
10/M
2
39
14/M
Follow-up (mo)
1
Age (y)/ Sex
TABLE 1. Achondroplastic Patients who Underwent Lumbar Decompressions With the High-Speed Drill (2008-2013)
J Pediatr Orthop Spinal Decompression in Achondroplasia
www.pedorthopaedics.com |
783
784 | www.pedorthopaedics.com
19
25
13
18 15/M
19 24/M
20 10/M
Surgery
Reop, L5-S1 decompression T8-T10 instrumented fusion Weakness and paresthesia in the lower Reop, L5 -S2 extremities, urinary incontinence decompression L3-S2 instrumented fusion Back pain Reop, L4-S1 decompression L4-S2 instrumented fusion Neurogenic claudication L2-L5 decompression T10-L4 instrumented fusion
Ataxia, paresthesia
Preoperative Indications
Asymptomatic
Asymptomatic
Leg weakness
Asymptomatic
5
7
6
7
x
x
x
x
r
17
15
12/F
9/F
20/F
17/M
6
7
8
9
10 14/M
Neurogenic claudication, paresthesias, weakness Low back pain, neurogenic claudication Neurogenic claudication, paresthesia Neurogenic claudication, incontinence Ataxia, low back pain, paresthesias
Surgery T11-L2 decompression T7-L4 instrumented fusion T11-L4 decompression T9-L4 instrumented fusion T12-L5 decompression T9-L4 instrumented fusion Reop, T10-L4 decompression T10-L4 instrumented fusion T12-L4 decompression T10-L4 instrumented fusion T12-L4 decompression T10-L4 instrumented fusion T10-L3 decompression T7-L4 instrumented fusion T11-L5 decompression T10-L4 instrumented fusion L1-L5 decompression T10-L4 instrumented fusion L1-L5 decompression T12-L3 instrumented fusion Asymptomatic
Asymptomatic
Low back pain
Asymptomatic
Asymptomatic
Ataxia, paresthesias
4
6
6
5
7
5
4
13
8
Numbness Paresthesias, bowel incontinence Ataxia
7
Asymptomatic
x
x
x
Complications Postoperative Condition LOS Postopeartive Wound Wound at Last Follow-up (d) Durotomy CSF Leak Pseudomeningocele Infection Dehiscence
CSF indicates cerebrospinal fluid; LOS (d), length of stay (days); Reop, reoperation from a prior decompression; x, corresponds to the complication which the patient experienced.
12
23
30
26
8/F
12
30
5
13/M
3
24
Ataxia, neurogenic claudication Low back pain, paresthesia, ataxia Weakness, ataxia, bowel incontinence Irritability, cognitive delay, ataxia Neurogenic claudication
Preoperative Indications
7/M
16/F
2
18
Follow-up (mo)
J Pediatr Orthop
4
25/M
1
Age (y)/ Sex
TABLE 2. Achondroplastic Patients Who Underwent Spinal Decompressions With the BoneScalpel (2008-2013)
x
Postoperative LOS Postoperative Wound Wound Condition (d) Durotomy CSF Leak Pseudomeningocele Infection Dehiscence
Complications
CSF indicates cerebrospinal fluid; LOS (d), length of stay (days); Reop, reoperation from a prior decompression; x, corresponds to the complication which the patient experienced.
12
Follow-up (mo)
17 20/F
Age (y)/ Sex
TABLE 1. (continued)
Bydon et al Volume 34, Number 8, December 2014
2014 Lippincott Williams & Wilkins
J Pediatr Orthop
Volume 34, Number 8, December 2014
Spinal Decompression in Achondroplasia
TABLE 3. Perioperative Findings in Patients in the High-Speed Drill Cohort Versus BoneScalpel Cohort High-Speed Drill Cohort
Findings Sensory disturbancesw Back pain Incontinence Neurogenic claudication Radiculopathy Ataxia Weakness
Preoperative Indications (n = 20)
Postoperative Conditions (n = 20)
9/20 (45.0%)
BoneScalpel Cohort
P*
Preoperative Indications (n = 10)
Postoperative Conditions (n = 10)
P*
2/20 (10.0%)
0.015
4/10 (40.0%)
3/10 (30.0%)
1.000
7/20 (35.0%) 7/20 (35.0%) 6/20 (30.0%)
1/20 (5.0%) 1/20 (5.0%) 0/20 (0.0%)
0.031 0.031 0.031
3/10 (30.0%) 2/10 (20.0%) 6/10 (60.0%)
1/10 (10.0%) 1/10 (10.0%) 0/10 (0.0%)
0.625 1.000 0.031
3/20 (15.0%) 4/20 (20.0%) 4/20 (20.0%)
1/20 (5.0%) 3/20 (15.0%) 1/20 (5.0%)
0.500 1.000 0.250
0/10 (0.0%) 5/10 (50.0%) 2/10 (20.0%)
0/10 (0.0%) 2/10 (20.0%) 0/10 (0.0%)
— 0.375 0.500
*P-values are calculated from an exact McNemar test; statistical significance is set at Pr0.05. wSensory disturbances describe numbness and/or paresthesias. Statistically significant values are in bold.
with the average rate of 5% in the general population undergoing spinal decompressions.15 The dural tears may be attributed to the higher number of levels requiring decompression in an achondroplastic spine.15 Moreover, the spinal canal of an achondroplastic patient has not only an unusually thin dura but also one half to one third of the diameter in comparison with the general population.28 Finally, the epidural fat that protects the neural components in the canal is also decreased in these patients.33 To overcome these limitations in achondroplastic patients, we have adopted the bone curette for spinal osteotomies. Although the safety profile between the ultrasonic bone curette is similar to the high-speed drill in the adult (nonachondroplastic) population, we wondered about the application of the curette in our sample population.34 We experienced a lower durotomy rate of 30.0% (3/10) in the BoneScalpel cohort compared with 45.0% in the high-speed drill cohort (P = 0.694). Two of 3 durotomies in the BoneScalpel cohort (patients 7 and 8) were caused by the Kerrison punches, whereas the ultrasonic curette was responsible for only 1 durotomy. The 3 durotomies in the BoneScalpel cohort measured, on average, less 1 cm in length. In patient 2 and 8, the dural tears occurred at the L3-L4 interspace and L1 level, respectively. Patient 7 experienced a durotomy at the nerve
TABLE 4. Complications in the High-Speed Drill Cohort Versus BoneScalpel Cohort Complications Durotomy (intraoperative) CSF leak (postoperative) Pseudomeningoceles Wound infection Wound dehiscence Total no. patients with Z1 complications
High-Speed Drill
BoneScalpel
P*
9/20 (45.0%)
3/10 (30.0%)
0.694
0/10 0/10 0/10 0/10 3/10
0.540 1.000 0.540 1.000 0.440
2/20 1/20 2/20 1/20 10/20
(10.0%) (5.0%) (10.0%) (5.0%) (50.00%)
(0.0%) (0.0%) (0.0%) (0.0%) (30.0%)
*The P-values for complications are derived from Fisher exact test; statistical significance is set at P < 0.05. CSF indicates cerebrospinal fluid.
r
2014 Lippincott Williams & Wilkins
root. All 3 defects were corrected with Prolene sutures and DuraSeal. None of the patients experienced a postoperative CSF leak. Because 2 of the 3 patients who experienced wound infection and dehiscence in the high-speed drill cohort also had CSF leaks, the wound complications may be a sequela of the CSF leak. Per continuum, by lowering the durotomy and subsequent CSF leak rates, the BoneScalpel may also decrease the rate of wound infections and dehiscence. These improved outcomes in the BoneScalpel cohort may be attributed to the instrument’s ability to distinguish between bone and soft tissue. The oscillations produced by the scalpel tip purchase bone, but soft-tissue structures (eg, nerves, dura) are spared via reflexive vibrations. Also, as the bone curette does not cut cotton, cottonoid applications may be used to protect adjacent soft-tissue structures in an already condensed space within the spinal column. Finally, because the speed of osteotomies with the bone curette allows for a shorter length of operation, the decreased time of surgical-site exposure may account for this lower rate of surgical-site infections and dehiscence. However, according to our experience, the utility of the BoneScalpel was limited in patients undergoing reoperations. In the setting of postoperative scarring from a previous operation, the laminar bone is difficult to isolate from the surrounding fibrous tissue. Because the curette cuts bone, but not soft tissue, achieving wide decompressions in these patients was slightly onerous. Nevertheless, achondroplastic children routinely experience symptomatic improvement postoperatively. In a retrospective review of decompressions in the thoracolumbar spine of achondroplastic children, Pyeritz et al29 reported symptomatic relief in 91% of patients with a mean follow-up of 8 years. Consistent with Pyeritz and colleague’s finding, all of our patients reported symptomatic improvement of their sensory disturbances, back pain, incontinence, neurogenic claudication, radiculopathy, ataxia, and weakness in both the high-speed drill cohort and BoneScalpel cohort. As with all retrospective analysis, our clinical study had inherent limitations. Prospective, randomized-controlled www.pedorthopaedics.com |
785
J Pediatr Orthop
Bydon et al
trials may assist in delineating the benefits of the ultrasonic bone curette. Moreover, additional studies with larger cohorts may increase the statistical power of our analysis.
CONCLUSIONS Spinal stenosis represents a clinically significant complication of patients with achondroplasia. Neurological symptoms in the thoracolumbar spine can range anywhere from back pain to sensory disturbances. The management of these symptoms usually includes spinal decompression followed by instrumented fusions if necessary. However, multilevel laminectomies pose unique challenges in the achondroplstic spine, with intraoperative durotomy being one of the highest complications noted. Our review of 30 cases revealed a decreased number of durotomies as well as decreased total number of complications in the BoneScalpel cohort versus the high-speed drill cohort. The data in this manuscript suggest that the ultrasonic bone curette is a safe and efficacious alternative to the high-speed drill in these challenging patients. REFERENCES 1. Lutter LD, Longstein JE, Winter RB, et al. Anatomy of the achondroplastic lumbar canal. Clin Orthop Relat Res. 1977;126:139–142. 2. Nelson MA. Spinal stenosis in achondroplasia. Proc R Soc Med. 1972;65:1028–1029. 3. Nelson MA. Kyphosis and lumbar stenosis in achondroplasia. Basic Life Sci. 1988;48:305–311. 4. Hall JG. The natural history of achondroplasia. Basic Life Sci. 1988;48:3–9. 5. Hunter AG, Bankier A, Rogers JG, et al. Medical complications of achondroplasia: a multicentre patient review. J Med Genet. 1998; 35:705–712. 6. Schkrohowsky JG, Hoernschemeyer DG, Carson BS, et al. Early presentation of spinal stenosis in achondroplasia. J Pediatr Orthop. 2007;27:119–122. 7. Morgan DF, Young RF. Spinal neurological complications of achondroplasia. Results of surgical treatment. J Neurosurg. 1980; 52:463–472. 8. Misra SN, Morgan HW. Thoracolumbar spinal deformity in achondroplasia. Neurosurg Focus. 2003;14:e4. 9. Vleggeert-Lankamp C, Peul W. Surgical decompression of thoracic spinal stenosis in achondroplasia: indication and outcome. J Neurosurg Spine. 2012;17:164–172. 10. Modi HN, Suh SW, Hong JY, et al. Magnetic resonance imaging study determining cord level and occupancy at thoracolumbar junction in achondroplasia—a prospective study. Ind J Orthop. 2011;45:63–68. 11. Carlisle ES, Ting BL, Abdullah MA, et al. Laminectomy in patients with achondroplasia: the impact of time to surgery on long-term function. Spine. 2011;36:886–892. 12. Baca KE, Abdullah MA, Ting BL, et al. Surgical decompression for lumbar stenosis in pediatric achondroplasia. J Pediatr Orthop. 2010;30:449–454. 13. King JA, Vachhrajani S, Drake JM, et al. Neurosurgical implications of achondroplasia. J Neurosurg Pediatr. 2009;4:297–306.
786 | www.pedorthopaedics.com
Volume 34, Number 8, December 2014
14. Jha RM, Klimo P, Smith ER. Foramen magnum stenosis from overgrowth of the opisthion in a child with achondroplasia. J Neurosurg Pediatr. 2008;2:136–138. 15. Ain MC, Chang TL, Schkrohowsky JG, et al. Rates of perioperative complications associated with laminectomies in patients with achondroplasia. J Bone Joint Surg Am. 2008;90:295–298. 16. Jeong ST, Song HR, Keny SM, et al. MRI study of the lumbar spine in achondroplasia. A morphometric analysis for the evaluation of stenosis of the canal. J Bone Joint Surg Br. 2006;88:1192–1196. 17. Fortuna A, Ferrante L, Acqui M, et al. Narrowing of thoraco-lumbar spinal canal in achondroplasia. J Neurosurg Sci. 1989;33:185–196. 18. Hidaka K, Chiba Y, Takeda H. Clinical application of ultrasonic osteotome for the spinous process-splitting laminoplasty. Spinal Surg. 1998;12:19–24. 19. Sun W, Zhang JG, Qiu GX, et al. [Posterior osteotomy and decompression for spinal deformity in patients with achondroplasia]. Zhonghua yi xue za zhi. 2010;90:3068–3071. 20. Sciubba DM, Noggle JC, Marupudi NI, et al. Spinal stenosis surgery in pediatric patients with achondroplasia. J Neurosurg. 2007; 106(suppl 5):372–378. 21. Ryken TC, Menezes AH. Cervicomedullary compression in achondroplasia. J Neurosurg. 1994;81:43–48. 22. Katz J, Temple AP. Surgery for achondroplasia. Perioperative challenges of dwarfism. AORN J. 1987;46:96–105. 23. Ganz JC. Lumbar spinal stenosis: postoperative results in terms of preoperative posture-related pain. J Neurosurg. 1990;72:71–74. 24. Wang JC, Bohlman HH, Riew KD. Dural tears secondary to operations on the lumbar spine. Management and results after a two-year-minimum follow-up of eighty-eight patients. J Bone Joint Surg Am. 1998;80:1728–1732. 25. Aryanpur J, Hurko O, Francomano C, et al. Craniocervical decompression for cervicomedullary compression in pediatric patients with achondroplasia. J Neurosurg. 1990;73:375–382. 26. Farmer KW, Brinkley MF, Skolasky RL, et al. Lumbar fusion in achondroplasia: does fusion to the sacrum affect function? J Pediat Orthop. 2009;29:476–480. 27. Gordon N. The neurological complications of achondroplasia. Brain Dev. 2000;22:3–7. 28. Nakagawa H, Kim SD, Mizuno J, et al. Technical advantages of an ultrasonic bone curette in spinal surgery. J Neurosurg Spine. 2005; 2:431–435. 29. Pyeritz RE, Sack GH Jr, Udvarhelyi GB. Thoracolumbosacral laminectomy in achondroplasia: long-term results in 22 patients. Am J Med Genet. 1987;28:433–444. 30. Silvers HR, Lewis PJ, Asch HL. Decompressive lumbar laminectomy for spinal stenosis. J Neurosurg. 1993;78:695–701. 31. Yasuoka S, Peterson HA, MacCarty CS. Incidence of spinal column deformity after multilevel laminectomy in children and adults. J Neurosurg. 1982;57:441–445. 32. Ito K, Ishizaka S, Sasaki T, et al. Safe and minimally invasive laminoplastic laminotomy using an ultrasonic bone curette for spinal surgery: technical note. Surg Neurol. 2009;72:470–475; discussion 5. 33. Hancock DO, Philips DG. Spinal compression in achondroplasia. Paraplegia. 1965;3:23–33. 34. Bydon M, Xu R, Papademetriou K, et al. Safety of spinal decompression using an ultrasonic bone curette compared with a high-speed drill: outcomes in 337 patients. J Neurosurg Spine. 2013;18:627–633.
r
2014 Lippincott Williams & Wilkins
Spinal Decompression in Achondroplastic Patients Using High-Speed Drill Versus Ultrasonic Bone Curette: Technical Note and Outcomes in 30 Cases Mohamad Bydon, MD,* Mohamed Macki, BS,* Risheng Xu, AM,*w Michael C. Ain, MD,* Edward S. Ahn, MD,* and George I. Jallo, MD*
Background: This manuscript describes the clinical and operative characteristics of achondroplastic children who undergo multilevel thoracolumbar decompressions using either the highspeed drill or the ultrasonic bone curette (BoneScalpel). Methods: We retrospectively reviewed 30 thoracolumbar decompressions in achondroplastic patients at a single institution between 2008 and 2013. Patients were classified into either the high-speed drill cohort or the BoneScalpel cohort, depending on which instrument was utilized to perform the decompression. A technical note on the role of the ultrasonic bone curette in decompressing stenotic achondroplastic spines is also provided. Results: In comparison with the high-speed drill cohort, the BoneScalpel cohort experienced less overall perioperative complications, including durotomy, cerebrospinal fluid leak, pseudomeningoceles, wound infection, and wound dehiscence. Although 45.0% of patients experienced a durotomy in the highspeed drill cohort, only 30.0% of patients experienced a durotomy in the BoneScalpel cohort (P = 0.694). In the high-speed drill cohort, the number of patients complaining of sensory disturbances, back pain, ataxia, incontinence, neurogenic claudication, radiculopathy, ataxia, and/or weakness decreased postoperatively. Similar results were observed in the BoneScalpel cohort. Conclusions: Although spinal decompression provides symptomatic resolution in patients with achondroplasia, intraoperative complications, in general, and durotomies, in particular, are common. Here, we report a decreased incidence in intraoperative durotomy and overall perioperative complication rates in the BoneScalpel cohort, although this did not reach the level of statistical significance. Nonetheless, the data demonFrom the *Department of Neurosurgery; and wMedical Scientist Training Program, Johns Hopkins University School of Medicine, Baltimore, MD. M.B. and M.M. contributed equally. G.I.J. receives fellowship support from Medtronic. M.C.A. is a consultant for Stryker Spine. The remaining authors declare no conflicts of interest. Reprints: George I. Jallo, MD, Division of Pediatric Neurosurgery, Johns Hopkins Hospital, 600 N Wolfe Street, Phipps 556, Baltimore, MD 21287. E-mail: [email protected]. Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Website, www. pedorthopaedics.com. Copyright r 2014 by Lippincott Williams & Wilkins
780 | www.pedorthopaedics.com
strate that the BoneScalpel is a safe and efficacious alternative to the high-speed drill in these challenging patients. Level of evidence: Level II—retrospective study. Key Words: achondroplasia, ultrasonic bone curette, pediatric, spine, surgery, outcomes (J Pediatr Orthop 2014;34:780–786)
A
chondroplasia is a genetic bone disorder that causes the most common type of dwarfism. Although the length of the vertebral column is not as severely affected as the limbs, the vertebrae undergo structural changes that may compromise the spinal cord.1 During in utero development, achondroplasia causes impaired longitudinal growth of the posterior arches, accounting for shortened pedicles and interpedicle distances. These developmental malformations result in nerve root compression and spinal stenosis.2,3 Although narrowing of the lumbar canal occurs in all achondroplastic spines, 78% subsequently develop neurological symptoms,4–6 and one third requiring surgical intervention.6–18 Due to the significantly altered spinal anatomy in achondroplastic patients, considerable challenges arise when decompressing the narrowed spinal canal.19–21 A review of the literature indicates that durotomies and cerebrospinal fluid (CSF) leaks are the most common perioperative complications.1,22–24 In this manuscript, we describe the role of an ultrasonic bone curette, the BoneScalpel, in performing a multilevel thoracolumbar decompression of the achondroplastic spine. To understand the risks and benefits of the bone curette in this challenging patient population, we review surgical outcomes in 30 decompressions in achondroplastic patients using either the conventional highspeed drill or the BoneScalpel. In addition, we provide a technical note of a thoracolumbar laminectomy in the achondroplastic spine using the ultrasonic bone curette.
METHODS Patient Population Over the 6-year period between 2008 and 2013, we collected all thoracolumbar decompressions for spinal stenosis on patients with achondroplasia at a single J Pediatr Orthop
Volume 34, Number 8, December 2014
J Pediatr Orthop
Volume 34, Number 8, December 2014
institution. In 2008, the BoneScalpel, an ultrasonic bone curette, was introduced to our institution. To explore the BoneScalpel utility in complex spine surgeries, surgeons randomly assigned the ultrasonic bone curette to achondroplastic decompressions, thereby eliminating selection biases. Only those surgeons who specialize in spinal decompressions on achondroplastic spines utilized the BoneScalpel, thereby eliminating any operator-dependent biases. We retrospectively divided the sample population into a high-speed drill cohort and BoneScalpel cohort. The cases were reviewed retrospectively using operative narratives and clinical notes. Both first-time surgeries and reoperations were identified. Intraoperative and perioperative complications were noted with a particular emphasis on intraoperative durotomy and postoperative CSF leak. All durotomies were due to dural tears, which in turn were caused by the surgical instruments: Kerrison punch or the BoneScalpel. Only symptomatic CSF leaks or incidental leaks identified on routine imaging were documented in our study.
Device Description The ultrasonic bone curette (BoneScalpel) is a handheld device designed for precise and rapid osteotomies. It consists of an electronic-control headpiece, an oscillating scalpel tip, and a jet nozzle for irrigation. With a 0.5 mm width and 20 mm length, the metal tip oscillates with a frequency of 22.5 kHz (22,500 strokes per second). When placed on bone, the delivery of ultrasonic oscillations cuts through the lamina with minimal movement of the handpiece. Upon contact with solid osseous tissue, the bone curette blade does not bend so that bone may be incised. However, soft tissue contact causes the bone curette blade to bend, protecting them from incidental incisions (eg, durotomies). The energy delivered by the oscillating tip also facilitates blood coagulation without necrosis of the surrounding tissue. The shaving tip enables undercutting of the lamina. In addition, unlike the rotational forces of the high-speed drill, the beveled design of the bone curette delivers ultrasonic pulses directly to bone, thereby sparing the dura and spinal cord. This is particularly important in the achondroplastic spinal canal where the shortened interpedicle distance compromises the space between the inner spinal column and dura. In our institutional experience, the BoneScalpel did not interfere with intraoperative neurophysiological monitoring.
Surgical Description: Thoracolumbar Decompressions After general endotracheal anesthesia and sterile preparation, the patient is placed in the prone position (See Supplemental Digital Content, http://links.lww.com/ BPO/A16). The procedure begins with a midline incision over the lumbar spine. The spinous processes and lamina are accessed by dissection through the paraspinal muscles in a subperiosteal manner. The spinalis muscles are deflected from the lamina and spinous processes to allow for better exposure of the posterior elements. The operation was then carried out in the following order: (1) inr
2014 Lippincott Williams & Wilkins
Spinal Decompression in Achondroplasia
strumentation of the posterior elements with pedicle screws; (2) laminectomy utilizing the BoneScalpel; and (3) placement of the contoured rods. Before decompression, almost all patients (28 out of 30) underwent spinal instrumented fusions to (1) correct the spinal malalignment in achondroplastic patients; and/or (2) prevent postlaminectomy kyphosis in children. Two patients had a prior instrumented fusion at the index level of decompression. With neuromonitoring, instrumented fusion is carried out with titanium pedicle screws under fluoroscopic guidance. After the spine is properly instrumented, the surgeon’s attention is turned to the decompression. Initially, interspinous ligaments are preserved to provide traction during the excision of laminar bone in an achondroplastic spine. With an incision no greater than 7 mm in width, the spinolaminar junction is cut bilaterally using the bone curette. Despite the thickened lamina in these patients, the bone curette reaches depths of up to 20 mm in a single stroke. The device continues to purchase laminar bone until the ligamentum flavum is reached. Using the Kerrison punches, the undersurface of the excised spinous process-laminar complex is then isolated from the posterior ligaments of the spinal canal. The interspinous ligament is transected above and below the spinous process, and the laminar complex is removed en bloc from the intact spine. Following the decompression, the rod is contoured, and the spinal alignment is corrected (Fig. 1). Bone graft using autograft from the laminectomy in addition to allograft is placed in the lateral gutters to reinforce the arthrodesis. The wound is closed in a multilayer manner, and a drain is placed above the fascia.
RESULTS Between 2008 and 2013, 30 operations were carried out for the management of symptomatic spinal stenosis in achondroplastic patients. All patients underwent decompression of the thoracolumbar spine with either a highspeed drill (Table 1) or the BoneScalpel (Table 2). The high-speed drill was used in 20 cases (66.7%), whereas the BoneScalpel was used in 10 cases (33.3%). The cases included 15 (50.0%) revision surgeries and 15 (50.0%) firsttime operations. Of the 15 reoperations, 1 was observed in the BoneScalpel cohort. Of the study cohort of 30 patients, 70.0% were male (n = 21). Mean age (± SD) at the time of surgery was 15.0 ± 4.79 in the high-speed drill cohort and 14.10 ± 5.63 in the BoneScalpel cohort (P = 0.650). Between the 2 cohorts, patients did not statistically differ in terms of follow-up time from index surgery and length of hospital stay. Mean follow-up time was 21.90 ± 8.53 and 20.70 ± 6.85 months in the high-speed drill and BoneScalpel cohorts, respectively (P = 0.702). Mean length of hospital stay in the high-speed drill cohort (6.5 ± 1.99 d) did not statistically differ from the BoneScalpel cohort (6.50 ± 2.64 d) (P = 1.000). However, the median number of levels decompressed in the highspeed drill cohort, 3, was statistically higher than the median 5.5 levels decompressed in the BoneScalpel cohort www.pedorthopaedics.com |
781
Bydon et al
J Pediatr Orthop
Volume 34, Number 8, December 2014
FIGURE 1. Images of an achondroplastic child with stenosis of the thoracolumbar spine (T10-L4). A, Preoperative T2-weighted magnetic resonance imaging (MRI), axial image. B, Preoperative T2-weighted MRI, sagittal image. C, Preoperative, lateral x-ray. D, Postoperative, lateral x-ray with instrumentation from T10 to L4.
(P = 0.005), whereas a median of 6 levels were fused in both the high-speed drill and BoneScalpel cohorts (P = 0.687). An exact McNemar test was used to compare the number of patients with preoperative and postoperative symptoms in the high-speed drill and BoneScalpel cohorts (Table 3). In the high-speed drill group, the number of symptomatic patients postoperatively decreased in all measured findings. The number of patients with sensory disturbances decreased by 77.7%, back pain by 85.71%, incontinence by 85.71%, neurogenic claudication by 100.00%, radiculopathy by 66.67%, ataxia by 25.00%, and weakness by 75.00%. Differences in sensory disturbances, back pain, incontinence, and neurogenic claudication reached statistical significance. In the BoneScalpel cohort, all patients complaining of preoperative neurogenic claudication and weakness experienced complete resolution of their symptoms. The number of patients complaining of sensory disturbances decreased by 25.00%, back pain by 66.67%, incontinence by 50.00%, and ataxia by 60.00%. Improvement in neurogenic claudication and weakness reached statistical significance. Following Cobb angle measurements, the spinal alignment corrected by a mean of 11.88 degrees (P = 0.002). Ten patients in the high-speed drill cohort experienced at least 1 perioperative complication, whereas 3 patients in the BoneScalpel cohort experienced clinically insignificant durotomies (without CSF leaks). Individual complications in the high-speed drill cohort and the BoneScalpel cohort were compared with Fisher exact test (Table 4). During the 10 cases in which the BoneScalpel was used, the incidence of durotomies decreased by 15.00% when compared with the high-speed drill cases (P = 0.694). Of the 9 durotomies (45.0%) in the highspeed drill cohort, postoperative cerebrospinal fluid (CSF) leaks occurred in 2 patients (10.0%), one of whom developed a pseudomeningocele and seroma (5.0%). Other complications in the high-speed drill cohort in-
782 | www.pedorthopaedics.com
cluded wound infection (10.0%) and wound dehiscence (5.0%). The cases with pseudomeningoceles and wound dehiscence required surgical management with a complex plastics closure. Taken together, the BoneScalpel cohort experienced less intraoperative and perioperative complications, although this did not reach statistical significance (P = 0.440).
DISCUSSION Symptomatic spinal stenosis is a common diagnosis in achondroplastic children.2–4,6,7,12,20,21,25–32 Management of spinal stenosis in achondroplastic children requires surgical intervention with a multilevel decompression. In a retrospective review of 18 achondroplastic children undergoing spinal decompressions, Baca et al12 assigned each patient a symptomatic score (SS) based on the presence of lower extremity weakness, paresthesias, reflexes, incontinence, and ataxia. Mean preoperative SS values (4.0 ± 0.9) significantly differed from postoperative SS values (1.6 ± 1.7). Although patients experience postoperative symptomatic relief, perioperative complications in achondroplastic patients underscore the technical difficulties of operating on an achondroplasic spine.1,26,27,29,31 Although the overall complication rate after surgical decompressions in the general population is approximately 22%, this rate almost triples to 61% in achondroplastic patients.15,30 Lutter et al1 reports the most common complications encountered as dural tears (37%), followed by neurological complications (23%). Surgical-site infections (9%), deep vein thrombosis (3%), pulmonary complications (3%), and gastrointestinal complications (3%) were also reported. These values corroborate our complication rates in the conventional high-speed drill cohort with durotomies at 45.0% and wound infections at 10.0%. Wang et al24 cited an even higher durotomy rate at 55%. These durotomy rates in achondroplastic patients are strikingly high in comparison r
2014 Lippincott Williams & Wilkins
r
14
21
30
22
15
15/M
13/M
8/M
20/F
14/M
17/M
22/F
3
4
5
2014 Lippincott Williams & Wilkins
6
7
8
9
10 7/M
11 13/M
12 13/M
13 16/F
17
19
17
15 12/M
16 23/M
Paresthesia
Ataxia
Radiculopathy
Asymptomatic
Urinary incontinence Ataxia
Asymptomatic
Asymptomatic
Asymptomatic
Paresthesias
Low back pain
Asymptomatic
Ataxia
Asymptomatic
Asymptomatic
Asymptomatic
8
8
5
7
5
7
4
6
13
7
8
4
5
7
6
5
x
x
x
x
x
x
x
x x
x
Postoperative LOS Postoperative Wound Wound Condition (d) Durotomy CSF Leak Pseudomeningocele Infection Dehiscence
Volume 34, Number 8, December 2014
14 14/M
28
30
12
18
42
18
Surgery
L4-S1 decompression T4-L4 instrumentation T10-L2 decompression T8-L5 instrumentation Urinary incontinence, neurogenic Reop, thoracolumbar claudication decompression T10-L4 instrumented fusion Ataxia, paresthesia Reop, lumbosacral decompression Neurogenic claudication L2-L4 decompression T10-L4 instrumentation Low back pain, radiculopathy Reop, T6-T9, T12-L4 decompression T4-L4 instrumented fusion Lower extremity numbness Reop, T10-L2 decompression T10-L4 instrumented fusion Low back pain, headaches T12-L5 decompression T8-L4 instrumented fusion Back pain, leg numbness Reop, T12-L5 decompression T9-L4 instrumented fusion Back pain, weakness L1-L4 decompression T10-L4 instrumentation Neurogenic claudication, urinary Reop, L4-S1 incontinence decompression Ataxia, urinary incontinence Reop, L4-S1 decompression T6-L5 instrumented fusion Back pain, radiculopathy, numbness, Reop, L4-S1 urinary incontinence decompression T12-S1 instrumented fusion Radiculopathy, paresthesia, sleep Reop, L1-L5 apnea, urinary and fecal incontinence decompression T11-L4 instrumented fusion Ataxia, neurogenic claudication, Reop, L4-S1 urinary incontinence decompression T9-L5 instrumented fusion Paresthesia, weakness Reop, L4-S1 decompression L2-sacral instrumented fusion
Hypoesthesia and paresthesia, back pain, neurogenic claudication Hyporeflexia, weakness
Preoperative Indications
Complications
27
10/M
2
39
14/M
Follow-up (mo)
1
Age (y)/ Sex
TABLE 1. Achondroplastic Patients who Underwent Lumbar Decompressions With the High-Speed Drill (2008-2013)
J Pediatr Orthop Spinal Decompression in Achondroplasia
www.pedorthopaedics.com |
783
784 | www.pedorthopaedics.com
19
25
13
18 15/M
19 24/M
20 10/M
Surgery
Reop, L5-S1 decompression T8-T10 instrumented fusion Weakness and paresthesia in the lower Reop, L5 -S2 extremities, urinary incontinence decompression L3-S2 instrumented fusion Back pain Reop, L4-S1 decompression L4-S2 instrumented fusion Neurogenic claudication L2-L5 decompression T10-L4 instrumented fusion
Ataxia, paresthesia
Preoperative Indications
Asymptomatic
Asymptomatic
Leg weakness
Asymptomatic
5
7
6
7
x
x
x
x
r
17
15
12/F
9/F
20/F
17/M
6
7
8
9
10 14/M
Neurogenic claudication, paresthesias, weakness Low back pain, neurogenic claudication Neurogenic claudication, paresthesia Neurogenic claudication, incontinence Ataxia, low back pain, paresthesias
Surgery T11-L2 decompression T7-L4 instrumented fusion T11-L4 decompression T9-L4 instrumented fusion T12-L5 decompression T9-L4 instrumented fusion Reop, T10-L4 decompression T10-L4 instrumented fusion T12-L4 decompression T10-L4 instrumented fusion T12-L4 decompression T10-L4 instrumented fusion T10-L3 decompression T7-L4 instrumented fusion T11-L5 decompression T10-L4 instrumented fusion L1-L5 decompression T10-L4 instrumented fusion L1-L5 decompression T12-L3 instrumented fusion Asymptomatic
Asymptomatic
Low back pain
Asymptomatic
Asymptomatic
Ataxia, paresthesias
4
6
6
5
7
5
4
13
8
Numbness Paresthesias, bowel incontinence Ataxia
7
Asymptomatic
x
x
x
Complications Postoperative Condition LOS Postopeartive Wound Wound at Last Follow-up (d) Durotomy CSF Leak Pseudomeningocele Infection Dehiscence
CSF indicates cerebrospinal fluid; LOS (d), length of stay (days); Reop, reoperation from a prior decompression; x, corresponds to the complication which the patient experienced.
12
23
30
26
8/F
12
30
5
13/M
3
24
Ataxia, neurogenic claudication Low back pain, paresthesia, ataxia Weakness, ataxia, bowel incontinence Irritability, cognitive delay, ataxia Neurogenic claudication
Preoperative Indications
7/M
16/F
2
18
Follow-up (mo)
J Pediatr Orthop
4
25/M
1
Age (y)/ Sex
TABLE 2. Achondroplastic Patients Who Underwent Spinal Decompressions With the BoneScalpel (2008-2013)
x
Postoperative LOS Postoperative Wound Wound Condition (d) Durotomy CSF Leak Pseudomeningocele Infection Dehiscence
Complications
CSF indicates cerebrospinal fluid; LOS (d), length of stay (days); Reop, reoperation from a prior decompression; x, corresponds to the complication which the patient experienced.
12
Follow-up (mo)
17 20/F
Age (y)/ Sex
TABLE 1. (continued)
Bydon et al Volume 34, Number 8, December 2014
2014 Lippincott Williams & Wilkins
J Pediatr Orthop
Volume 34, Number 8, December 2014
Spinal Decompression in Achondroplasia
TABLE 3. Perioperative Findings in Patients in the High-Speed Drill Cohort Versus BoneScalpel Cohort High-Speed Drill Cohort
Findings Sensory disturbancesw Back pain Incontinence Neurogenic claudication Radiculopathy Ataxia Weakness
Preoperative Indications (n = 20)
Postoperative Conditions (n = 20)
9/20 (45.0%)
BoneScalpel Cohort
P*
Preoperative Indications (n = 10)
Postoperative Conditions (n = 10)
P*
2/20 (10.0%)
0.015
4/10 (40.0%)
3/10 (30.0%)
1.000
7/20 (35.0%) 7/20 (35.0%) 6/20 (30.0%)
1/20 (5.0%) 1/20 (5.0%) 0/20 (0.0%)
0.031 0.031 0.031
3/10 (30.0%) 2/10 (20.0%) 6/10 (60.0%)
1/10 (10.0%) 1/10 (10.0%) 0/10 (0.0%)
0.625 1.000 0.031
3/20 (15.0%) 4/20 (20.0%) 4/20 (20.0%)
1/20 (5.0%) 3/20 (15.0%) 1/20 (5.0%)
0.500 1.000 0.250
0/10 (0.0%) 5/10 (50.0%) 2/10 (20.0%)
0/10 (0.0%) 2/10 (20.0%) 0/10 (0.0%)
— 0.375 0.500
*P-values are calculated from an exact McNemar test; statistical significance is set at Pr0.05. wSensory disturbances describe numbness and/or paresthesias. Statistically significant values are in bold.
with the average rate of 5% in the general population undergoing spinal decompressions.15 The dural tears may be attributed to the higher number of levels requiring decompression in an achondroplastic spine.15 Moreover, the spinal canal of an achondroplastic patient has not only an unusually thin dura but also one half to one third of the diameter in comparison with the general population.28 Finally, the epidural fat that protects the neural components in the canal is also decreased in these patients.33 To overcome these limitations in achondroplastic patients, we have adopted the bone curette for spinal osteotomies. Although the safety profile between the ultrasonic bone curette is similar to the high-speed drill in the adult (nonachondroplastic) population, we wondered about the application of the curette in our sample population.34 We experienced a lower durotomy rate of 30.0% (3/10) in the BoneScalpel cohort compared with 45.0% in the high-speed drill cohort (P = 0.694). Two of 3 durotomies in the BoneScalpel cohort (patients 7 and 8) were caused by the Kerrison punches, whereas the ultrasonic curette was responsible for only 1 durotomy. The 3 durotomies in the BoneScalpel cohort measured, on average, less 1 cm in length. In patient 2 and 8, the dural tears occurred at the L3-L4 interspace and L1 level, respectively. Patient 7 experienced a durotomy at the nerve
TABLE 4. Complications in the High-Speed Drill Cohort Versus BoneScalpel Cohort Complications Durotomy (intraoperative) CSF leak (postoperative) Pseudomeningoceles Wound infection Wound dehiscence Total no. patients with Z1 complications
High-Speed Drill
BoneScalpel
P*
9/20 (45.0%)
3/10 (30.0%)
0.694
0/10 0/10 0/10 0/10 3/10
0.540 1.000 0.540 1.000 0.440
2/20 1/20 2/20 1/20 10/20
(10.0%) (5.0%) (10.0%) (5.0%) (50.00%)
(0.0%) (0.0%) (0.0%) (0.0%) (30.0%)
*The P-values for complications are derived from Fisher exact test; statistical significance is set at P < 0.05. CSF indicates cerebrospinal fluid.
r
2014 Lippincott Williams & Wilkins
root. All 3 defects were corrected with Prolene sutures and DuraSeal. None of the patients experienced a postoperative CSF leak. Because 2 of the 3 patients who experienced wound infection and dehiscence in the high-speed drill cohort also had CSF leaks, the wound complications may be a sequela of the CSF leak. Per continuum, by lowering the durotomy and subsequent CSF leak rates, the BoneScalpel may also decrease the rate of wound infections and dehiscence. These improved outcomes in the BoneScalpel cohort may be attributed to the instrument’s ability to distinguish between bone and soft tissue. The oscillations produced by the scalpel tip purchase bone, but soft-tissue structures (eg, nerves, dura) are spared via reflexive vibrations. Also, as the bone curette does not cut cotton, cottonoid applications may be used to protect adjacent soft-tissue structures in an already condensed space within the spinal column. Finally, because the speed of osteotomies with the bone curette allows for a shorter length of operation, the decreased time of surgical-site exposure may account for this lower rate of surgical-site infections and dehiscence. However, according to our experience, the utility of the BoneScalpel was limited in patients undergoing reoperations. In the setting of postoperative scarring from a previous operation, the laminar bone is difficult to isolate from the surrounding fibrous tissue. Because the curette cuts bone, but not soft tissue, achieving wide decompressions in these patients was slightly onerous. Nevertheless, achondroplastic children routinely experience symptomatic improvement postoperatively. In a retrospective review of decompressions in the thoracolumbar spine of achondroplastic children, Pyeritz et al29 reported symptomatic relief in 91% of patients with a mean follow-up of 8 years. Consistent with Pyeritz and colleague’s finding, all of our patients reported symptomatic improvement of their sensory disturbances, back pain, incontinence, neurogenic claudication, radiculopathy, ataxia, and weakness in both the high-speed drill cohort and BoneScalpel cohort. As with all retrospective analysis, our clinical study had inherent limitations. Prospective, randomized-controlled www.pedorthopaedics.com |
785
J Pediatr Orthop
Bydon et al
trials may assist in delineating the benefits of the ultrasonic bone curette. Moreover, additional studies with larger cohorts may increase the statistical power of our analysis.
CONCLUSIONS Spinal stenosis represents a clinically significant complication of patients with achondroplasia. Neurological symptoms in the thoracolumbar spine can range anywhere from back pain to sensory disturbances. The management of these symptoms usually includes spinal decompression followed by instrumented fusions if necessary. However, multilevel laminectomies pose unique challenges in the achondroplstic spine, with intraoperative durotomy being one of the highest complications noted. Our review of 30 cases revealed a decreased number of durotomies as well as decreased total number of complications in the BoneScalpel cohort versus the high-speed drill cohort. The data in this manuscript suggest that the ultrasonic bone curette is a safe and efficacious alternative to the high-speed drill in these challenging patients. REFERENCES 1. Lutter LD, Longstein JE, Winter RB, et al. Anatomy of the achondroplastic lumbar canal. Clin Orthop Relat Res. 1977;126:139–142. 2. Nelson MA. Spinal stenosis in achondroplasia. Proc R Soc Med. 1972;65:1028–1029. 3. Nelson MA. Kyphosis and lumbar stenosis in achondroplasia. Basic Life Sci. 1988;48:305–311. 4. Hall JG. The natural history of achondroplasia. Basic Life Sci. 1988;48:3–9. 5. Hunter AG, Bankier A, Rogers JG, et al. Medical complications of achondroplasia: a multicentre patient review. J Med Genet. 1998; 35:705–712. 6. Schkrohowsky JG, Hoernschemeyer DG, Carson BS, et al. Early presentation of spinal stenosis in achondroplasia. J Pediatr Orthop. 2007;27:119–122. 7. Morgan DF, Young RF. Spinal neurological complications of achondroplasia. Results of surgical treatment. J Neurosurg. 1980; 52:463–472. 8. Misra SN, Morgan HW. Thoracolumbar spinal deformity in achondroplasia. Neurosurg Focus. 2003;14:e4. 9. Vleggeert-Lankamp C, Peul W. Surgical decompression of thoracic spinal stenosis in achondroplasia: indication and outcome. J Neurosurg Spine. 2012;17:164–172. 10. Modi HN, Suh SW, Hong JY, et al. Magnetic resonance imaging study determining cord level and occupancy at thoracolumbar junction in achondroplasia—a prospective study. Ind J Orthop. 2011;45:63–68. 11. Carlisle ES, Ting BL, Abdullah MA, et al. Laminectomy in patients with achondroplasia: the impact of time to surgery on long-term function. Spine. 2011;36:886–892. 12. Baca KE, Abdullah MA, Ting BL, et al. Surgical decompression for lumbar stenosis in pediatric achondroplasia. J Pediatr Orthop. 2010;30:449–454. 13. King JA, Vachhrajani S, Drake JM, et al. Neurosurgical implications of achondroplasia. J Neurosurg Pediatr. 2009;4:297–306.
786 | www.pedorthopaedics.com
Volume 34, Number 8, December 2014
14. Jha RM, Klimo P, Smith ER. Foramen magnum stenosis from overgrowth of the opisthion in a child with achondroplasia. J Neurosurg Pediatr. 2008;2:136–138. 15. Ain MC, Chang TL, Schkrohowsky JG, et al. Rates of perioperative complications associated with laminectomies in patients with achondroplasia. J Bone Joint Surg Am. 2008;90:295–298. 16. Jeong ST, Song HR, Keny SM, et al. MRI study of the lumbar spine in achondroplasia. A morphometric analysis for the evaluation of stenosis of the canal. J Bone Joint Surg Br. 2006;88:1192–1196. 17. Fortuna A, Ferrante L, Acqui M, et al. Narrowing of thoraco-lumbar spinal canal in achondroplasia. J Neurosurg Sci. 1989;33:185–196. 18. Hidaka K, Chiba Y, Takeda H. Clinical application of ultrasonic osteotome for the spinous process-splitting laminoplasty. Spinal Surg. 1998;12:19–24. 19. Sun W, Zhang JG, Qiu GX, et al. [Posterior osteotomy and decompression for spinal deformity in patients with achondroplasia]. Zhonghua yi xue za zhi. 2010;90:3068–3071. 20. Sciubba DM, Noggle JC, Marupudi NI, et al. Spinal stenosis surgery in pediatric patients with achondroplasia. J Neurosurg. 2007; 106(suppl 5):372–378. 21. Ryken TC, Menezes AH. Cervicomedullary compression in achondroplasia. J Neurosurg. 1994;81:43–48. 22. Katz J, Temple AP. Surgery for achondroplasia. Perioperative challenges of dwarfism. AORN J. 1987;46:96–105. 23. Ganz JC. Lumbar spinal stenosis: postoperative results in terms of preoperative posture-related pain. J Neurosurg. 1990;72:71–74. 24. Wang JC, Bohlman HH, Riew KD. Dural tears secondary to operations on the lumbar spine. Management and results after a two-year-minimum follow-up of eighty-eight patients. J Bone Joint Surg Am. 1998;80:1728–1732. 25. Aryanpur J, Hurko O, Francomano C, et al. Craniocervical decompression for cervicomedullary compression in pediatric patients with achondroplasia. J Neurosurg. 1990;73:375–382. 26. Farmer KW, Brinkley MF, Skolasky RL, et al. Lumbar fusion in achondroplasia: does fusion to the sacrum affect function? J Pediat Orthop. 2009;29:476–480. 27. Gordon N. The neurological complications of achondroplasia. Brain Dev. 2000;22:3–7. 28. Nakagawa H, Kim SD, Mizuno J, et al. Technical advantages of an ultrasonic bone curette in spinal surgery. J Neurosurg Spine. 2005; 2:431–435. 29. Pyeritz RE, Sack GH Jr, Udvarhelyi GB. Thoracolumbosacral laminectomy in achondroplasia: long-term results in 22 patients. Am J Med Genet. 1987;28:433–444. 30. Silvers HR, Lewis PJ, Asch HL. Decompressive lumbar laminectomy for spinal stenosis. J Neurosurg. 1993;78:695–701. 31. Yasuoka S, Peterson HA, MacCarty CS. Incidence of spinal column deformity after multilevel laminectomy in children and adults. J Neurosurg. 1982;57:441–445. 32. Ito K, Ishizaka S, Sasaki T, et al. Safe and minimally invasive laminoplastic laminotomy using an ultrasonic bone curette for spinal surgery: technical note. Surg Neurol. 2009;72:470–475; discussion 5. 33. Hancock DO, Philips DG. Spinal compression in achondroplasia. Paraplegia. 1965;3:23–33. 34. Bydon M, Xu R, Papademetriou K, et al. Safety of spinal decompression using an ultrasonic bone curette compared with a high-speed drill: outcomes in 337 patients. J Neurosurg Spine. 2013;18:627–633.
r
2014 Lippincott Williams & Wilkins
