Clinical Orthopaedics and Related Research®
Clin Orthop Relat Res (2014) 472:3166–3176 DOI 10.1007/s11999-014-3768-6
A Publication of The Association of Bone and Joint Surgeons®
CLINICAL RESEARCH
Does Fracture Affect the Healing Time or Frequency of Recurrence in a Simple Bone Cyst of the Proximal Femur? Soo Min Cha MD, Hyun Dae Shin MD, PhD, Kyung Cheon Kim MD, PhD, Jung Woo Park MD
Received: 28 December 2013 / Accepted: 17 June 2014 / Published online: 8 July 2014 Ó The Association of Bone and Joint Surgeons1 2014
Abstract Background Studies have focused on intramedullary nailing of femoral simple bone cysts but have not clarified the recurrence frequency or management of recurrent cysts. In particular, the affect of pathologic fractures on cyst healing, recurrence, and complications of treatment have not been reported. Questions/purposes We performed a retrospective comparative study to examine whether there were differences between simple bone cysts in the proximal femur nailed Each author certifies that he or she, or a member of his or her immediate family, has no funding or commercial associations (eg, consultancies, stock ownership, equity interest, patent/licensing arrangements, etc) that might pose a conflict of interest in connection with the submitted article. All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research1 editors and board members are on file with the publication and can be viewed on request. Clinical Orthopaedics and Related Research1 neither advocates nor endorses the use of any treatment, drug, or device. Readers are encouraged to always seek additional information, including FDA approval status, of any drug or device before clinical use. Each author certifies that his or her institution approved the human protocol for this investigation, that all investigations were conducted in conformity with ethical principles of research, and that informed consent for participation in the study was obtained. This work was performed at Regional Rheumatoid and Degenerative Arthritis Center, Chungnam National University Hospital, Chungnam National University School of Medicine, Daejeon, Korea. S. M. Cha, K. C. Kim, J. W. Park Regional Rheumatoid and Degenerative Arthritis Center, Chungnam National University Hospital, Daejeon, Korea H. D. Shin (&) Department of Orthopedic Surgery, Chungnam National University School of Medicine, 640, Daesa-Dong, Jung-Gu, Daejeon, Korea e-mail: [email protected]
123
after pathologic fracture and those without pathologic fracture in terms of (1) healing time, (2) frequency and timing of recurrence, and (3) complications. Methods From 1995 to 2005, 54 patients diagnosed with femoral simple bone cysts were treated and followed for a minimum of 8 years. Flexible nails were inserted in a retrograde fashion in 25 patients with fractures and 29 patients without fractures. The healing period, degree of radiographic consolidation based on the criteria of Capanna et al., recurrence frequency, and final bony abnormalities were analyzed. The mean followups were 107 months (range, 96– 124 months) and 103 months (range, 96–140 months) in the groups with and without fractures, respectively. With the numbers available, a post hoc calculation showed that this study had 80% power to detect a difference of 7 months of healing time as significant with a probability less than 0.05. Results With the numbers available, the mean healing period was not different between groups (25 versus 30 months in the groups with and without fractures, respectively; p = 0.16). Complete healing was observed at 19 versus 18 months, incomplete healing at 5 versus 8 months, and recurrence was observed in one and three patients in the groups with and without fractures, respectively. No differences were found in the distribution of healing grade based on the criteria of Capanna et al. A second surgery was performed using intramedullary nails in two patients with an open physis and compression hip screw fixation was performed in two patients with a closed physis. Finally, the recurrent cysts were classified as completely healed in three patients and incompletely healed in one. Conclusions Whether a pathologic fracture had occurred before surgical treatment, intramedullary nailing of femoral simple bone cysts resulted in reliable healing, and the frequency of recurrence did not differ. Because this was a retrospective study, the optimal treatment for recurred cysts
Volume 472, Number 10, October 2014
after intramedullary nailing should be further investigated through a comparative or prospective study. Level of Evidence Level III, therapeutic study. See the Instructions for Authors for a complete description of levels of evidence.
Introduction Simple bone cysts of the proximal femur are not uncommon in children. However, healing time and frequency of recurrence after treatment of these cysts have not been specified as clearly or sufficiently as for cysts of the proximal humerus. Numerous studies have been performed on humeral cysts [4, 7, 11, 19, 24, 25], whereas only a few have focused on femoral cysts [3, 13, 16, 20]. Several aspects, including weightbearing, the presence of a final deformity at the end of growth, limb length, and normal gait, should be considered for femoral cysts. The majority of patients report having a painful hip after pathologic fractures, and the fractures are diagnosed as displaced or nondisplaced [16]. Incidental cysts accompanied by discomfort are reported after minor trauma [7, 16]. Curettage and bone grafting are the traditional treatment methods for simple bone cysts of the proximal femur. These treatment options have a complication rate of 15%, leading to sequelae such as infection, coxa vara, epiphyseal arrest, and limb shortening [5, 9, 14, 17, 18]. Furthermore, the recurrence rate is as much as 40% [13, 14, 17, 18, 24]. Oppenheim and Galleno [17] concluded that curettage and bone grafting should not be considered the preferred treatments for simple bone cysts. The major drawbacks of other surgical techniques, such as subtotal resection and total subperiosteal resection, include physeal damage, intraoperative blood loss, intraoperative fracture, and a long period of postoperative immobilization [14, 24]. Catier et al. [3] reported successful results after flexible intramedullary nailing for treatment of simple bone cysts in the proximal aspect of the femur. Knorr et al. [11] and Santori et al. [21] reported complete healing in all of their patients treated by intramedullary nailing. The advantage of this method is the same as that described in earlier series [5, 6, 22]: continuous decompression of the cyst and a subsequent decrease in the intralesional pressure. Vigler et al. [26] used an angled blade plate with an external fixator for treatment of subtrochanteric fractures attributable to simple bone cysts, but this approach has several apparent drawbacks, including physeal injury, invasiveness, and problems with implant removal. Because of the drawbacks of the other approaches to this problem, we have considered flexible intramedullary nailing an
Intramedullary Nailing for Simple Bone Cysts
3167
attractive approach. Flexible nails offer a decompression effect for the cyst, stability around the cyst despite a pathologic fracture, and simplicity of the procedure. However, to our knowledge, no study has determined whether the efficacy of flexible nailing is comparable in patients with and without pathologic fractures at the time of presentation. Therefore we performed a retrospective comparative study to examine whether there were differences between simple bone cysts in the proximal femur nailed after pathologic fracture and those without pathologic fracture in terms of (1) healing time, (2) frequency and timing of recurrence, and (3) complications.
Patients and Methods Patient Selection Sixty patients with simple bone cysts of the proximal femur were admitted to our institute from January 1995 to December 2005. Of these 60 patients, 57 (95%) were treated with flexible intramedullary nailing. Three of these 57 patients were lost to regular followup. Thus, 54 patients (95% of those treated with flexible nails) with at least 8 years of followup were included in this study (Table 1). The first group was composed of 25 patients with simple bone cysts and pathologic fractures at initial presentation. There were 15 boys and 10 girls, with a mean age of 9.7 years. The second group was composed of 29 patients with simple bone cysts but without pathologic fractures. There were 16 boys and 13 girls, with a mean age of 10.1 years. In this group, 19 patients reported progressive hip pain and limping and the cysts of the remaining 10 patients were detected incidentally on radiographs obtained just after minor trauma. The inclusion criteria were patients (1) diagnosed with simple bone cysts by an orthopaedic oncologist and a radiologist, (2) with cysts having a modified cyst index (described below) greater than 3.5 [4], (3) undergoing treatment without bone grafting or steroid injection, (4) for whom complete medical records and radiographic data were available, (5) with skeletal immaturity, and (6) who had complete records at a minimum of 8 years. We excluded patients with the following characteristics: (1) modified cyst index [4] less than 3.5, (2) healing process underway after pathologic fracture during nonoperative treatment of a simple bone cyst, (3) recurrence after healing of a simple bone cyst treated nonoperatively or by any other treatment options, and (4) inconsistent diagnosis between the radiologist and oncologist necessitating biopsy or MRI.
123
3168
Clinical Orthopaedics and Related Research1
Cha et al.
Table 1. Demographic data Variable
Group 1
Group 2
p value
Number of patients
25
29
Age at treatment (years) (range)
10 ± 1.9 (6–14)
10 ± 2.2 (7–16)
Gender, males, number
15
16
0.72
Total followup (months) (range)
107 ± 9.6 (96–124)
103 ± 8.4 (96–140)
0.99
Neck
4
7
0.46
Intertrochanter
8
9
0.94
Subtrochanter
13
13
0.6
5.34 ± 0.81 (3.2–6.2)
4.81 ± 0.83 (3.5–6.0)
0.02
Active
6
11
0.27
Latent
19
18
Time to surgery (days)
1.9 ± 1.0
7.8 ± 9.9
0.007
Number of inserted nails
4.1 ± 0.3
4.1 ± 0.3
0.88
0.5
Site of lesion
Modified cyst index (range) Cyst activity
Surgical Technique and Postoperative Management In the group with fractures, displaced fractures were reduced first with the patients on the traction table and nails were inserted. The length of the nails chosen was checked using the image intensifier by placing the nail on the anterior surface of the thigh. The diameter of the nails was selected such that the sum of inserted nails would occupy approximately 80% of the medullary canal. Four or five nails were inserted. All the nails were titanium alloy and ranged from 2 mm to 3 mm. No open or percutaneous biopsy was performed before nailing, and no curettage was performed in this series. For retrograde insertion of nails, two straight incisions were made 3 cm above the distal epiphysis. In the group without fractures, all procedures were identical, except that the ordinary operation table was used instead of the fracture table. Postoperative mobilization depended on whether there was a pathologic fracture and on the stability of the fixation. Patients with fractures of the femur were allowed nonweightbearing, crutch-assisted ambulation after a mean of 5 postoperative days (range, 3–7 days). For patients with poor stability owing to larger cysts or extensive cortical thinning, nonweightbearing, crutch-assisted walking was allowed for 3 weeks, and weightbearing ambulation was gradually permitted after monitoring via simple followup radiographs. Patients with stable nailed fractures were allowed to bear weight as tolerated. All patients in both groups were allowed to resume daily activities after discharge. The distribution of cysts was four in the neck, eight in the intertrochanter, and 13 in the subtrochanter area for the patients with fractures. Six patients had active cysts. For the patients without fractures, the distribution of cysts was seven in the neck, nine in the intertrochanter, and 13 in the subtrochanter area. Eleven patients had active cysts. Mean
123
followups were 107 months (range, 96–124 months) and 103 months (range, 96–140 months) in the groups with and without fractures, respectively. The two groups were similar with the exception of time to surgery after diagnosis (1.9 days versus 7.8 days for the groups with and without fractures, respectively; p = 0.007; Table 1).
Modified Cyst Index We used the modified cyst index [4] to evaluate the cyst area. This is a newly developed method of determining the cyst index based on picture archiving and communication system (PACS) software (m-view 5.4; Marosis Technologies Inc, Seoul, Korea). The area of simple bone cysts was calculated automatically on PACS software and then the area was divided by the square of the diameter of the femoral diaphysis. Two orthopaedic surgeons (SMC, JWP) measured the index on the AP and lateral radiographs and defined the mean value as the modified cyst index. Similar to the cutoff value of the conventional cyst index [10, 12] for high risk of pathologic fracture in femoral simple bone cysts, we used modified cyst index values greater than 3.5 as the criterion for surgery in the group without fractures. Bland and Altman plots and repeatability coefficients [1] were used to measure interobserver and intraobserver repeatability of evaluations of the modified cyst index. The 95% limits of agreement represented a visual interpretation of how well the measurements of the two orthopaedic surgeons agreed. By definition, the measurement of error was smaller than the repeatability coefficient for 95% of the observations. The interobserver and intraobserver repeatability coefficients of the modified cyst index were 0.77 and 0.75,
Group 2
Group 2
Group 2
2
3
4
* Group 1 = group with fracture; Group 2 = group without fracture.
Complete
Complete Incomplete Incomplete Recurrence Complete Incomplete
Complete
3169
Incomplete Incomplete Incomplete
Complete
Complete
Incomplete Group 1 1
Incomplete
6 5 4 3 2 1
Initial group*
Radiographic assessments were performed every 3 months until 2 years postoperatively and every 6 months thereafter. According to the classification system of Capanna et al. [2], a cyst was classified as completely healed when it was filled with bone and had healed and the cortical margins had thickened. A cyst was classified as incompletely healed when it was well consolidated with bone; that is, most of the cyst was filled with bone and had healed and the cortical margins had thickened, but there were still small, residual areas of radiolucency. A cyst was classified as a recurrence when it had healed initially and been filled with bone, but large areas of radiolucency and cortical thinning subsequently developed. A cyst was classified as having no response when there was no sign of consolidation or healing. Recurrence and no response evaluations represented treatment failures. Two orthopaedic surgeons (SMC, JWP) other than the operating surgeon evaluated the radiographs. The ability of the reviewers to classify the cyst using the four criteria of Capanna et al. [2] was expressed using the kappa value. According to Fleiss and Cohen [8], a kappa value greater than 0.75 is regarded as excellent, 0.40 to 0.75 is fair to good, and less than 0.40 is poor. The reviewers’ ability to classify the cyst according to the criteria of Capanna et al. [2] at final followup as determined on simple AP and lateral radiographs was excellent (kappa values of 0.83 and 0.84, respectively). The healing time was defined as
Table 3. Annual change of cyst grade in four patients with a recurred cyst
Radiographic Classification and Measurement of the Cyst
Patient
respectively on AP radiographs and 0.79 and 0.81 respectively on lateral radiographs, indicating high reproducibility.
Annual cyst status (postoperative years) according to the classification of Capanna et al. [2]
7
* Fisher’s exact test.
Complete
1* 0.65*
Incomplete
2 2
Incomplete
0.62*
1 3
Deformity Limb length discrepancy at final followup
Incomplete
0.55*
3
Recurrence
8
1
Incomplete
5
Recurrence
Incomplete
Incomplete healing
Incomplete
0.38*
Complete
18
Incomplete
19
Complete
9
Complete healing
8
Capanna et al. criteria at final followup
Incomplete
0.31
Incomplete
30 ± 11.3 (14–54)
Incomplete
28 ± 9.1 (16–44)
Incomplete
Nail removal, postoperative (months) (range)
Incomplete
0.16
Incomplete
30 ± 11.2 (12–54)
Recurrence
25 ± 9.6 (14–44)
Incomplete
Healing time (months) (range)
Incomplete
p value
Recurrence
Group 2
Incomplete
Group 1
10
Outcomes
Incomplete
Table 2. Radiologic and clinical outcomes
Incomplete
Intramedullary Nailing for Simple Bone Cysts
Incomplete
Volume 472, Number 10, October 2014
123
3170
Clinical Orthopaedics and Related Research1
Cha et al.
Fig. 1A–F A 6-year-old boy presented with a painful hip. (A) His initial radiograph shows a right pathologic fracture and simple bone cysts. (B) Thirteen months after flexible nailing, incomplete healing of the cyst was observed. (C) The nails were removed at 20 months. (D) Four years after the initial diagnosis, refracture occurred. (E)
Reoperation using flexible nails was performed. (F) A 1.5-cm limb length discrepancy was present at final followup 10 years after the first diagnosis; this patient had incomplete healing according to the classification of Capanna et al. [2].
the postoperative period when the cysts had complete healing or incomplete healing, as determined at regular followups.
abnormalities including avascular necrosis, and angulation were evaluated at final followup and were regarded as long-term complications. Limb length discrepancy observed on the final orthoroentgenogram was defined as a difference greater than 1 cm in length.
Complications Surgery-related complications were defined as physeal damage attributable to hardware, infection, refracture during the inserted nail state, and problems related to the protruding nail ends at the insertion site. Additionally, long-term complications included joint contracture, limb length discrepancy, and gait abnormality [7, 23]. Deformities including coxa vara, coxa brevia, femoral head
123
Statistical Analysis Preoperative demographic data, distribution of cyst location, activity, and modified cyst index were analyzed and compared using a t-test or chi-square test. Final values were compared between the two groups using paired Student’s t-tests. For analyses involving smaller quantities,
Volume 472, Number 10, October 2014
Intramedullary Nailing for Simple Bone Cysts
3171
Fig. 2A–E A 7-year-old boy presented with a simple bone cyst. (A) Flexible nails were inserted to treat a subtrochanteric simple bone cyst in the left femur. (B) Twenty-two months after flexible nailing, the nails were removed. (C) Seven years after the initial diagnosis, recurrence and fracture occurred during daily activities. (D)
Compression hip screw and autogenous iliac bone grafting were performed. (E) Three years after screw fixation, he had complete healing according to the classification of Capanna et al. [2], and the limb length discrepancy was insignificant.
such as for patients experiencing complications, Fisher’s exact test was used. Data were analyzed using SPSS1 for Windows1 (SPSS Inc, Chicago, IL, USA). A post hoc calculation showed that this study had 80% power to detect a difference of 7 months of healing time as significant at a probability less than 0.05.
The two groups were not different with respect to frequency of recurrence (one patient versus three patients with recurrences in the groups with and without fractures, respectively; p = 0.62; Table 2). All recurrences followed a pathologic fracture. The one patient in the group without fractures had progressive hip pain throughout 3 weeks, and a fracture was diagnosed after minor trauma. The other three patients had an acutely painful hip and subsequently were diagnosed with recurrence and refracture. Three of the recurrences occurred after removal of intramedullary nails with incomplete healing and one with complete healing 84 months after the initial surgery. The recurrences occurred 4 years after initial diagnosis in the patient in the group with a fracture and at 2, 4, and 7 years after initial diagnosis in the three patients in the group without fractures (Table 3). For these four patients, the mean healing
Results With the numbers available, there was no difference in healing period between the two groups (25 versus 30 months in the groups with and without fractures, respectively; p = 0.16; Table 2). Distribution of the criteria of Capanna et al. [2] at final followup was not different between groups.
123
Clinical Orthopaedics and Related Research1
2 cm shortening
None
between
* Group 1 = group with fracture; Group 2 = group without fracture; S = subtrochanter; I = intertrochanter.
No
No Complete
Complete
demographic
variables
and
Demographic variables
Recurrence (4 patients)
No recurrence (50 patients)
p value
0.07
Sex
120
124 5 Nailing
Compression 6 hip screw 84
54
5.05
20
4.12
5.65 14 7
9
Group 2 M 4
I Group 2 M 3
S
13
3.84
18
1.5 cm overgrowth No Incomplete
Complete 128
122 7 Nailing
Compression 8 hip screw 30
50 18
16
7.8
3.64
4.95
3.94 18
10 6
16 Group 2 M
I
Group 1 M
2
S
Table 5. Relationship recurrence
Male
1
Criteria of Deformity Capanna et al. at final [2] at final followup followup Healing time Total after second followup operation (months) (months) Time at Second diagnosis operation of recurrence method (postoperative months) Modified Healing cyst period index at (months) recurrence Patient Initial Sex Site Age at Age at Initial group* diagnosis recurrence modified (years) (years) cyst index
Table 4. Data for four patients with a recurred cyst
123
Coxa brevia None
Cha et al.
Limb length discrepancy at final followup
3172
4
27
Female Age (years)
0 10 ± 4.51
23 10 ± 1.84
0.85
Modified cyst index
4.60 ± 0.86
5.04 ± 0.87
0.34
time was 18 months, and the mean time of recurrence was 55 months postoperatively. A second surgery was performed using flexible intramedullary nails in two patients with an open physis (Fig. 1), and compression hip screw fixation with autogenous iliac bone grafting was performed in two patients with a closed physis (Fig. 2). The mean healing period after the second surgery was 7 months and the recurrent cysts were classified as completely healed in three patients and incompletely healed in one patient. The total followup from the time of initial diagnosis of these four patients was 124 months (Table 4). None of the main demographic variables, such as sex, age, or modified cyst index, affected the incidence of recurrence (Table 5). One patient in each group had coxa vara at final followup. No discrepancies greater than 2 cm were found and no patient had a Trendelenburg sign or any other gait abnormality. Shortening less than 2 cm was observed in two patients in each group. One patient with recurrence in the fracture group had lengthening of 1.5 cm, and one of the patients with less than 2-cm shortening in the group without fractures had a recurrent cyst. Other complications, such as physeal damage attributable to the hardware, infection, or problems related to the protruding nail ends at the insertion site, were not encountered.
Discussion Femoral simple bone cysts accompanied by fractures require fracture reduction and stability, but few studies [3, 13, 20] have focused on femoral cysts that have not yet fractured. Studies of intramedullary nailing of femoral simple bone cysts do not clarify whether there are differences in treatment efficacy with this approach based on whether a fracture has or has not occurred before treatment [7, 16, 20]. We therefore performed a retrospective comparative study to examine whether there were differences between simple bone cysts in the proximal femur nailed after pathologic fracture and those without pathologic fracture in terms of (1) healing time, (2) frequency and timing of recurrence, and (3) complications.
Volume 472, Number 10, October 2014
Intramedullary Nailing for Simple Bone Cysts
3173
Fig. 3A–B (A) Flexible nails were anchored in the neck area in a patient from the group with fractures. (B) In this patient in the group without fractures, the neck cyst was fixed with nailing.
Fig. 4 A 6-year-old boy with a fracture had a cyst that expanded almost to the epiphysis. Two flexible nails were inserted that penetrated the epiphyseal plate for stability.
This study had some limitations. Despite having 25 patients in the group with fractures and 29 patients in the group without fractures, few patients in either group had a recurrence, complicating statistical comparisons of its incidence. In addition, risk factors and predictive factors for recurrence were not identified. Four patients underwent a second surgery for recurrence with fixation with compression hip screws or intramedullary nailing. Stable fixation for a fractured cyst helped to guide selection of the method. However, we could not compare the usefulness of compression hip screws and nailing. All patients with recurrences presented with fractures. Therefore, in this study we could not ascertain the proper guidelines for treatment of cyst recurrence without fracture. Studies have confirmed the usefulness of this method based on a sufficient number of patients, high rate of response for treatment, and long followup [7, 19, 20]. de Sanctis and Andreacchio [7] reported that no difference in healing time was noted between the simple bone cysts with pathologic fractures and those without them. The relatively small number of patients in their study precluded identifying any statistical differences in healing periods or outcomes between the groups of patients. Roposch et al. [20] reported that 11 patients with simple bone cysts of the proximal femur were treated by intramedullary nails. The mean healing time was 48 months for five patients with
123
3174
Cha et al.
Clinical Orthopaedics and Related Research1
Fig. 5A–C (A) A 13-year-old boy had a recurrence 4 years after diagnosis. (B) A reoperation using flexible nails was performed, but the nails perforated the weak portion of the trochanteric area. (C) At final followup, a 2-cm discrepancy was observed, but it was asymptomatic.
pathologic fractures and 31 months for six patients without pathologic fractures. However critical statistical comparison was limited owing to the small number of patients. Nevertheless, they assumed that, despite the lack of statistical significance, a longer healing time was needed in patients who did not have a pathologic fracture [19]. They pointed out that recurrent cysts grew outward from an area not covered by nails. However, these results differ from those of our study, as three subtrochanteric and one intertrochanteric cyst recurred. Flexible nailing yielded stability, decompression, and covered the entire cystic area, but the cysts recurred as pathologic fractures. An additional difference in our study was shortened nails attributable to bone growth. Roposch et al. [19] exchanged nails in nine patients, but we removed the nails as soon as possible when the cyst showed at least incomplete healing according to the classification of Capanna et al. [2]. Namely, when the residual area of radiolucency was
123
smaller than 30% of the initial size, the nails were removed. Thus, all nails were removed 14 to 54 months postoperatively. Norman-Taylor et al. [16] reported that, to reduce recurrence, a Neer healing grade of 3 or 4 [15] is necessary during the first year of diagnosis and throughout followup. Norman-Taylor et al. [16] evaluated the difference in refracture after healing by nonsurgical treatment depending on displacement at the initial presentation. They concluded that displaced fractures were able to achieve union and healing in 1 year, but major complications such as coxa vara and avascular necrosis of the femoral head were frequent. However, undisplaced fractures needed a longer period of healing, additional steroid injections were needed to maintain the grade, and recurrence and refractures were more frequent, leading to coxa vara and short limb length. Their study [16] was meaningful for its comparison of the natural healing potential of a fractured cyst of the proximal
Volume 472, Number 10, October 2014
femur, even without statistically significant evidence. However, their results may differ from ours because we used flexible nailing. The healing potential, time, and pattern may be variable according to the decompression and fixation of the cysts. Teoh et al. [25] found age of 5 years or older; right-sided, large, multilocular cysts; and unimpacted fracture to be predictive factors for recurrence of simple bone cysts of the proximal humerus. Recurrence was not related to sex, cyst location, or cyst activity level. However, with the numbers available, in our study, age, sex, cyst size (modified cyst index), and the presence of fracture all were unrelated to the recurrence of the cyst. In the group without fractures, three patients were diagnosed with recurrence; however, the incidence was not different from that of the group with fractures. In our study, three patients with fractures and seven without fractures had a neck cyst with proper cancellous bone stock for stable anchoring with the nails (Fig. 3). In one patient in the group with fractures, in whom the cyst had expanded nearly to the physis, the nails were fixed through the epiphyseal plate (Fig. 4). Roposch et al. [20] reported that, with large cysts and extensive cortical weakening, nails may exit from the bone. In such cases, avascular necrosis of the femoral head must be considered a potential complication. Such a perforation of a nail through the weakened cortex was observed in one of our patients who experienced recurrence, but the extrusion was through the weak greater trochanter and was not a substantial problem at final followup (Fig. 5). The strength of our study was its comparative nature based on a large number of patients in both included groups through long-term followups. To our knowledge, previous reports [7, 19, 20] of femoral bone cysts were case series and the number of patients who underwent surgery before pathologic fracture was relatively small, making statistical comparisons difficult. Second, we performed surgery using a single method without bone grafting or any other additional fixation. Few patients were lost to followup, so the recurrence frequency in each group could be meaningful. Additionally, the change of the degree of healing was evaluated at the time of recurrence and after the secondary fixation. Whether a pathologic fracture had occurred before surgical treatment, intramedullary nailing of femoral simple bone cysts resulted in reliable healing and the frequency of recurrence did not differ. However, because this was a retrospective study, the optimal treatment for a recurred cyst after intramedullary nailing should be investigated further through a comparative or prospective study. Acknowledgments We thank Nam-Hee Kim PhD (YN Company, Seoul, Korea) for assistance with the statistical analysis.
Intramedullary Nailing for Simple Bone Cysts
3175
References 1. Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat Methods Med Res. 1999;8:135–160. 2. Capanna R, Albisinni U, Caroli GC, Campanacci M. Contrast examination as a prognostic factor in the treatment of solitary bone cyst by cortisone injection. Skeletal Radiol. 1984;12:97– 102. 3. Catier P, Bracq H, Canciani JP, Allouis M, Babut JM. [The treatment of upper femoral unicameral bone cysts in children by Ender’s nailing technique] [in French]. Rev Chir Orthop Reparatrice Appar Mot. 1981;67:147–149. 4. Cha SM, Shin HD, Kim KC, Kang DH. Flexible intramedullary nailing in simple bone cysts of the proximal humerus: prospective study for high-risk cases of pathologic fracture. J Pediatr Orthop B. 2013;22:475–480. 5. Chigira M, Watanabe H, Arita S, Udagawa E. [Simple bone cyst: pathophysiology and treatment] [in Japanese]. Nihon Seikeigeka Gakkai Zasshi. 1983;57:759–766. 6. Cohen J. Unicameral bone cysts: a current synthesis of reported cases. Orthop Clin North Am. 1977;8:715–736. 7. de Sanctis N, Andreacchio A. Elastic stable intramedullary nailing is the best treatment of unicameral bone cysts of the long bones in children? Prospective long-term follow-up study. J Pediatr Orthop. 2006;26:520–525. 8. Fleiss JL, Cohen J. The equivalence of weighted kappa and the intraclass correlation coefficient as measures of reliability. Educ Psychol Meas. 1973;33:613–639. 9. Gentile JV, Weinert CR, Schlechter JA. Treatment of unicameral bone cysts in pediatric patients with an injectable regenerative graft: a preliminary report. J Pediatr Orthop. 2013;33:254–261. 10. Kaelin AJ, MacEwen GD. Unicameral bone cysts: natural history and the risk of fracture. Int Orthop. 1989;13:275–282. 11. Knorr P, Schmittenbecher PP, Dietz HG. [Treatment of pathological fractures of long tubular bones in childhood using elastic stable intramedullary nailing] [in German]. Unfallchirurg. 1996;99:410–414. 12. Lee JH, Reinus WR, Wilson AJ. Quantitative analysis of the plain radiographic appearance of unicameral bone cysts. Invest Radiol. 1999;34:28–37. 13. Malkawi H, Shannak A, Amir S. Surgical treatment of pathological subtrochanteric fractures due to benign lesions in children and adolescents. J Pediatr Orthop. 1984;4:63–69. 14. McKay DW, Nason SS. Treatment of unicameral bone cysts by subtotal resection without grafts. J Bone Joint Surg Am. 1977;59:515–519. 15. Neer CS 2nd, Francis KC, Marcove RC, Terz J, Carbonara PN. Treatment of unicameral bone cyst: a follow-up study of one hundred seventy-five cases. J Bone Joint Surg Am. 1966;48:731–745. 16. Norman-Taylor FH, Hashemi-Nejad A, Gillingham BL, Stevens D, Cole WG. Risk of refracture through unicameral bone cysts of the proximal femur. J Pediatr Orthop. 2002;22:249–254. 17. Oppenheim WL, Galleno H. Operative treatment versus steroid injection in the management of unicameral bone cyst. J Pediatr Orthop. 1984;4:1–7. 18. Peltier LF, Jones RH. Treatment of unicameral bone cysts by curettage and packing with plaster-of-Paris pellets. Clin Orthop Relat Res. 2004;422:145–147. 19. Roposch A, Saraph V, Linhart WE. Flexible intramedullary nailing for the treatment of unicameral bone cysts in long bones. J Bone Joint Surg Am. 2000;82:1447–1453. 20. Roposch A, Saraph V, Linhart WE. Treatment of femoral neck and trochanteric simple bone cysts. Arch Orthop Trauma Surg. 2004;124:437–442.
123
3176
Cha et al.
21. Santori FS, Ghera S, Castelli V, Tollis A. Dynamic endomedullary nailing in the treatment of extensive bone cysts in young patients: a pathogenetic interpretation. Ital J Orthop Traumatol. 1986;12:411–417. 22. Shinozaki T, Arita S, Watanabe H, Chigira M. Simple bone cysts treated by multiple drill-holes: 23 cysts followed 2–10 years. Acta Orthop Scand. 1996;67:288–290. 23. Stanton RP, Abdel-Mot’al MM. Growth arrest resulting from unicameral bone cyst. J Pediatr Orthop. 1998;18:198–201.
123
Clinical Orthopaedics and Related Research1 24. Sturz H, Zenker H, Buckl H. Total subperiosteal resection treatment of solitary bone cysts of the humerus. Arch Orthop Trauma Surg. 1979;93:231–239. 25. Teoh KH, Watts AC, Chee YH, Reid R, Porter DE. Predictive factors for recurrence of simple bone cyst of the proximal humerus. J Orthop Surg (Hong Kong). 2010;18:215–219. 26. Vigler M, Weigl D, Schwarz M, Ben-Itzhak I, Salai M, Bar-On E. Subtrochanteric femoral fractures due to simple bone cysts in children. J Pediatr Orthop B. 2006;15:439–442.
Clin Orthop Relat Res (2014) 472:3166–3176 DOI 10.1007/s11999-014-3768-6
A Publication of The Association of Bone and Joint Surgeons®
CLINICAL RESEARCH
Does Fracture Affect the Healing Time or Frequency of Recurrence in a Simple Bone Cyst of the Proximal Femur? Soo Min Cha MD, Hyun Dae Shin MD, PhD, Kyung Cheon Kim MD, PhD, Jung Woo Park MD
Received: 28 December 2013 / Accepted: 17 June 2014 / Published online: 8 July 2014 Ó The Association of Bone and Joint Surgeons1 2014
Abstract Background Studies have focused on intramedullary nailing of femoral simple bone cysts but have not clarified the recurrence frequency or management of recurrent cysts. In particular, the affect of pathologic fractures on cyst healing, recurrence, and complications of treatment have not been reported. Questions/purposes We performed a retrospective comparative study to examine whether there were differences between simple bone cysts in the proximal femur nailed Each author certifies that he or she, or a member of his or her immediate family, has no funding or commercial associations (eg, consultancies, stock ownership, equity interest, patent/licensing arrangements, etc) that might pose a conflict of interest in connection with the submitted article. All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research1 editors and board members are on file with the publication and can be viewed on request. Clinical Orthopaedics and Related Research1 neither advocates nor endorses the use of any treatment, drug, or device. Readers are encouraged to always seek additional information, including FDA approval status, of any drug or device before clinical use. Each author certifies that his or her institution approved the human protocol for this investigation, that all investigations were conducted in conformity with ethical principles of research, and that informed consent for participation in the study was obtained. This work was performed at Regional Rheumatoid and Degenerative Arthritis Center, Chungnam National University Hospital, Chungnam National University School of Medicine, Daejeon, Korea. S. M. Cha, K. C. Kim, J. W. Park Regional Rheumatoid and Degenerative Arthritis Center, Chungnam National University Hospital, Daejeon, Korea H. D. Shin (&) Department of Orthopedic Surgery, Chungnam National University School of Medicine, 640, Daesa-Dong, Jung-Gu, Daejeon, Korea e-mail: [email protected]
123
after pathologic fracture and those without pathologic fracture in terms of (1) healing time, (2) frequency and timing of recurrence, and (3) complications. Methods From 1995 to 2005, 54 patients diagnosed with femoral simple bone cysts were treated and followed for a minimum of 8 years. Flexible nails were inserted in a retrograde fashion in 25 patients with fractures and 29 patients without fractures. The healing period, degree of radiographic consolidation based on the criteria of Capanna et al., recurrence frequency, and final bony abnormalities were analyzed. The mean followups were 107 months (range, 96– 124 months) and 103 months (range, 96–140 months) in the groups with and without fractures, respectively. With the numbers available, a post hoc calculation showed that this study had 80% power to detect a difference of 7 months of healing time as significant with a probability less than 0.05. Results With the numbers available, the mean healing period was not different between groups (25 versus 30 months in the groups with and without fractures, respectively; p = 0.16). Complete healing was observed at 19 versus 18 months, incomplete healing at 5 versus 8 months, and recurrence was observed in one and three patients in the groups with and without fractures, respectively. No differences were found in the distribution of healing grade based on the criteria of Capanna et al. A second surgery was performed using intramedullary nails in two patients with an open physis and compression hip screw fixation was performed in two patients with a closed physis. Finally, the recurrent cysts were classified as completely healed in three patients and incompletely healed in one. Conclusions Whether a pathologic fracture had occurred before surgical treatment, intramedullary nailing of femoral simple bone cysts resulted in reliable healing, and the frequency of recurrence did not differ. Because this was a retrospective study, the optimal treatment for recurred cysts
Volume 472, Number 10, October 2014
after intramedullary nailing should be further investigated through a comparative or prospective study. Level of Evidence Level III, therapeutic study. See the Instructions for Authors for a complete description of levels of evidence.
Introduction Simple bone cysts of the proximal femur are not uncommon in children. However, healing time and frequency of recurrence after treatment of these cysts have not been specified as clearly or sufficiently as for cysts of the proximal humerus. Numerous studies have been performed on humeral cysts [4, 7, 11, 19, 24, 25], whereas only a few have focused on femoral cysts [3, 13, 16, 20]. Several aspects, including weightbearing, the presence of a final deformity at the end of growth, limb length, and normal gait, should be considered for femoral cysts. The majority of patients report having a painful hip after pathologic fractures, and the fractures are diagnosed as displaced or nondisplaced [16]. Incidental cysts accompanied by discomfort are reported after minor trauma [7, 16]. Curettage and bone grafting are the traditional treatment methods for simple bone cysts of the proximal femur. These treatment options have a complication rate of 15%, leading to sequelae such as infection, coxa vara, epiphyseal arrest, and limb shortening [5, 9, 14, 17, 18]. Furthermore, the recurrence rate is as much as 40% [13, 14, 17, 18, 24]. Oppenheim and Galleno [17] concluded that curettage and bone grafting should not be considered the preferred treatments for simple bone cysts. The major drawbacks of other surgical techniques, such as subtotal resection and total subperiosteal resection, include physeal damage, intraoperative blood loss, intraoperative fracture, and a long period of postoperative immobilization [14, 24]. Catier et al. [3] reported successful results after flexible intramedullary nailing for treatment of simple bone cysts in the proximal aspect of the femur. Knorr et al. [11] and Santori et al. [21] reported complete healing in all of their patients treated by intramedullary nailing. The advantage of this method is the same as that described in earlier series [5, 6, 22]: continuous decompression of the cyst and a subsequent decrease in the intralesional pressure. Vigler et al. [26] used an angled blade plate with an external fixator for treatment of subtrochanteric fractures attributable to simple bone cysts, but this approach has several apparent drawbacks, including physeal injury, invasiveness, and problems with implant removal. Because of the drawbacks of the other approaches to this problem, we have considered flexible intramedullary nailing an
Intramedullary Nailing for Simple Bone Cysts
3167
attractive approach. Flexible nails offer a decompression effect for the cyst, stability around the cyst despite a pathologic fracture, and simplicity of the procedure. However, to our knowledge, no study has determined whether the efficacy of flexible nailing is comparable in patients with and without pathologic fractures at the time of presentation. Therefore we performed a retrospective comparative study to examine whether there were differences between simple bone cysts in the proximal femur nailed after pathologic fracture and those without pathologic fracture in terms of (1) healing time, (2) frequency and timing of recurrence, and (3) complications.
Patients and Methods Patient Selection Sixty patients with simple bone cysts of the proximal femur were admitted to our institute from January 1995 to December 2005. Of these 60 patients, 57 (95%) were treated with flexible intramedullary nailing. Three of these 57 patients were lost to regular followup. Thus, 54 patients (95% of those treated with flexible nails) with at least 8 years of followup were included in this study (Table 1). The first group was composed of 25 patients with simple bone cysts and pathologic fractures at initial presentation. There were 15 boys and 10 girls, with a mean age of 9.7 years. The second group was composed of 29 patients with simple bone cysts but without pathologic fractures. There were 16 boys and 13 girls, with a mean age of 10.1 years. In this group, 19 patients reported progressive hip pain and limping and the cysts of the remaining 10 patients were detected incidentally on radiographs obtained just after minor trauma. The inclusion criteria were patients (1) diagnosed with simple bone cysts by an orthopaedic oncologist and a radiologist, (2) with cysts having a modified cyst index (described below) greater than 3.5 [4], (3) undergoing treatment without bone grafting or steroid injection, (4) for whom complete medical records and radiographic data were available, (5) with skeletal immaturity, and (6) who had complete records at a minimum of 8 years. We excluded patients with the following characteristics: (1) modified cyst index [4] less than 3.5, (2) healing process underway after pathologic fracture during nonoperative treatment of a simple bone cyst, (3) recurrence after healing of a simple bone cyst treated nonoperatively or by any other treatment options, and (4) inconsistent diagnosis between the radiologist and oncologist necessitating biopsy or MRI.
123
3168
Clinical Orthopaedics and Related Research1
Cha et al.
Table 1. Demographic data Variable
Group 1
Group 2
p value
Number of patients
25
29
Age at treatment (years) (range)
10 ± 1.9 (6–14)
10 ± 2.2 (7–16)
Gender, males, number
15
16
0.72
Total followup (months) (range)
107 ± 9.6 (96–124)
103 ± 8.4 (96–140)
0.99
Neck
4
7
0.46
Intertrochanter
8
9
0.94
Subtrochanter
13
13
0.6
5.34 ± 0.81 (3.2–6.2)
4.81 ± 0.83 (3.5–6.0)
0.02
Active
6
11
0.27
Latent
19
18
Time to surgery (days)
1.9 ± 1.0
7.8 ± 9.9
0.007
Number of inserted nails
4.1 ± 0.3
4.1 ± 0.3
0.88
0.5
Site of lesion
Modified cyst index (range) Cyst activity
Surgical Technique and Postoperative Management In the group with fractures, displaced fractures were reduced first with the patients on the traction table and nails were inserted. The length of the nails chosen was checked using the image intensifier by placing the nail on the anterior surface of the thigh. The diameter of the nails was selected such that the sum of inserted nails would occupy approximately 80% of the medullary canal. Four or five nails were inserted. All the nails were titanium alloy and ranged from 2 mm to 3 mm. No open or percutaneous biopsy was performed before nailing, and no curettage was performed in this series. For retrograde insertion of nails, two straight incisions were made 3 cm above the distal epiphysis. In the group without fractures, all procedures were identical, except that the ordinary operation table was used instead of the fracture table. Postoperative mobilization depended on whether there was a pathologic fracture and on the stability of the fixation. Patients with fractures of the femur were allowed nonweightbearing, crutch-assisted ambulation after a mean of 5 postoperative days (range, 3–7 days). For patients with poor stability owing to larger cysts or extensive cortical thinning, nonweightbearing, crutch-assisted walking was allowed for 3 weeks, and weightbearing ambulation was gradually permitted after monitoring via simple followup radiographs. Patients with stable nailed fractures were allowed to bear weight as tolerated. All patients in both groups were allowed to resume daily activities after discharge. The distribution of cysts was four in the neck, eight in the intertrochanter, and 13 in the subtrochanter area for the patients with fractures. Six patients had active cysts. For the patients without fractures, the distribution of cysts was seven in the neck, nine in the intertrochanter, and 13 in the subtrochanter area. Eleven patients had active cysts. Mean
123
followups were 107 months (range, 96–124 months) and 103 months (range, 96–140 months) in the groups with and without fractures, respectively. The two groups were similar with the exception of time to surgery after diagnosis (1.9 days versus 7.8 days for the groups with and without fractures, respectively; p = 0.007; Table 1).
Modified Cyst Index We used the modified cyst index [4] to evaluate the cyst area. This is a newly developed method of determining the cyst index based on picture archiving and communication system (PACS) software (m-view 5.4; Marosis Technologies Inc, Seoul, Korea). The area of simple bone cysts was calculated automatically on PACS software and then the area was divided by the square of the diameter of the femoral diaphysis. Two orthopaedic surgeons (SMC, JWP) measured the index on the AP and lateral radiographs and defined the mean value as the modified cyst index. Similar to the cutoff value of the conventional cyst index [10, 12] for high risk of pathologic fracture in femoral simple bone cysts, we used modified cyst index values greater than 3.5 as the criterion for surgery in the group without fractures. Bland and Altman plots and repeatability coefficients [1] were used to measure interobserver and intraobserver repeatability of evaluations of the modified cyst index. The 95% limits of agreement represented a visual interpretation of how well the measurements of the two orthopaedic surgeons agreed. By definition, the measurement of error was smaller than the repeatability coefficient for 95% of the observations. The interobserver and intraobserver repeatability coefficients of the modified cyst index were 0.77 and 0.75,
Group 2
Group 2
Group 2
2
3
4
* Group 1 = group with fracture; Group 2 = group without fracture.
Complete
Complete Incomplete Incomplete Recurrence Complete Incomplete
Complete
3169
Incomplete Incomplete Incomplete
Complete
Complete
Incomplete Group 1 1
Incomplete
6 5 4 3 2 1
Initial group*
Radiographic assessments were performed every 3 months until 2 years postoperatively and every 6 months thereafter. According to the classification system of Capanna et al. [2], a cyst was classified as completely healed when it was filled with bone and had healed and the cortical margins had thickened. A cyst was classified as incompletely healed when it was well consolidated with bone; that is, most of the cyst was filled with bone and had healed and the cortical margins had thickened, but there were still small, residual areas of radiolucency. A cyst was classified as a recurrence when it had healed initially and been filled with bone, but large areas of radiolucency and cortical thinning subsequently developed. A cyst was classified as having no response when there was no sign of consolidation or healing. Recurrence and no response evaluations represented treatment failures. Two orthopaedic surgeons (SMC, JWP) other than the operating surgeon evaluated the radiographs. The ability of the reviewers to classify the cyst using the four criteria of Capanna et al. [2] was expressed using the kappa value. According to Fleiss and Cohen [8], a kappa value greater than 0.75 is regarded as excellent, 0.40 to 0.75 is fair to good, and less than 0.40 is poor. The reviewers’ ability to classify the cyst according to the criteria of Capanna et al. [2] at final followup as determined on simple AP and lateral radiographs was excellent (kappa values of 0.83 and 0.84, respectively). The healing time was defined as
Table 3. Annual change of cyst grade in four patients with a recurred cyst
Radiographic Classification and Measurement of the Cyst
Patient
respectively on AP radiographs and 0.79 and 0.81 respectively on lateral radiographs, indicating high reproducibility.
Annual cyst status (postoperative years) according to the classification of Capanna et al. [2]
7
* Fisher’s exact test.
Complete
1* 0.65*
Incomplete
2 2
Incomplete
0.62*
1 3
Deformity Limb length discrepancy at final followup
Incomplete
0.55*
3
Recurrence
8
1
Incomplete
5
Recurrence
Incomplete
Incomplete healing
Incomplete
0.38*
Complete
18
Incomplete
19
Complete
9
Complete healing
8
Capanna et al. criteria at final followup
Incomplete
0.31
Incomplete
30 ± 11.3 (14–54)
Incomplete
28 ± 9.1 (16–44)
Incomplete
Nail removal, postoperative (months) (range)
Incomplete
0.16
Incomplete
30 ± 11.2 (12–54)
Recurrence
25 ± 9.6 (14–44)
Incomplete
Healing time (months) (range)
Incomplete
p value
Recurrence
Group 2
Incomplete
Group 1
10
Outcomes
Incomplete
Table 2. Radiologic and clinical outcomes
Incomplete
Intramedullary Nailing for Simple Bone Cysts
Incomplete
Volume 472, Number 10, October 2014
123
3170
Clinical Orthopaedics and Related Research1
Cha et al.
Fig. 1A–F A 6-year-old boy presented with a painful hip. (A) His initial radiograph shows a right pathologic fracture and simple bone cysts. (B) Thirteen months after flexible nailing, incomplete healing of the cyst was observed. (C) The nails were removed at 20 months. (D) Four years after the initial diagnosis, refracture occurred. (E)
Reoperation using flexible nails was performed. (F) A 1.5-cm limb length discrepancy was present at final followup 10 years after the first diagnosis; this patient had incomplete healing according to the classification of Capanna et al. [2].
the postoperative period when the cysts had complete healing or incomplete healing, as determined at regular followups.
abnormalities including avascular necrosis, and angulation were evaluated at final followup and were regarded as long-term complications. Limb length discrepancy observed on the final orthoroentgenogram was defined as a difference greater than 1 cm in length.
Complications Surgery-related complications were defined as physeal damage attributable to hardware, infection, refracture during the inserted nail state, and problems related to the protruding nail ends at the insertion site. Additionally, long-term complications included joint contracture, limb length discrepancy, and gait abnormality [7, 23]. Deformities including coxa vara, coxa brevia, femoral head
123
Statistical Analysis Preoperative demographic data, distribution of cyst location, activity, and modified cyst index were analyzed and compared using a t-test or chi-square test. Final values were compared between the two groups using paired Student’s t-tests. For analyses involving smaller quantities,
Volume 472, Number 10, October 2014
Intramedullary Nailing for Simple Bone Cysts
3171
Fig. 2A–E A 7-year-old boy presented with a simple bone cyst. (A) Flexible nails were inserted to treat a subtrochanteric simple bone cyst in the left femur. (B) Twenty-two months after flexible nailing, the nails were removed. (C) Seven years after the initial diagnosis, recurrence and fracture occurred during daily activities. (D)
Compression hip screw and autogenous iliac bone grafting were performed. (E) Three years after screw fixation, he had complete healing according to the classification of Capanna et al. [2], and the limb length discrepancy was insignificant.
such as for patients experiencing complications, Fisher’s exact test was used. Data were analyzed using SPSS1 for Windows1 (SPSS Inc, Chicago, IL, USA). A post hoc calculation showed that this study had 80% power to detect a difference of 7 months of healing time as significant at a probability less than 0.05.
The two groups were not different with respect to frequency of recurrence (one patient versus three patients with recurrences in the groups with and without fractures, respectively; p = 0.62; Table 2). All recurrences followed a pathologic fracture. The one patient in the group without fractures had progressive hip pain throughout 3 weeks, and a fracture was diagnosed after minor trauma. The other three patients had an acutely painful hip and subsequently were diagnosed with recurrence and refracture. Three of the recurrences occurred after removal of intramedullary nails with incomplete healing and one with complete healing 84 months after the initial surgery. The recurrences occurred 4 years after initial diagnosis in the patient in the group with a fracture and at 2, 4, and 7 years after initial diagnosis in the three patients in the group without fractures (Table 3). For these four patients, the mean healing
Results With the numbers available, there was no difference in healing period between the two groups (25 versus 30 months in the groups with and without fractures, respectively; p = 0.16; Table 2). Distribution of the criteria of Capanna et al. [2] at final followup was not different between groups.
123
Clinical Orthopaedics and Related Research1
2 cm shortening
None
between
* Group 1 = group with fracture; Group 2 = group without fracture; S = subtrochanter; I = intertrochanter.
No
No Complete
Complete
demographic
variables
and
Demographic variables
Recurrence (4 patients)
No recurrence (50 patients)
p value
0.07
Sex
120
124 5 Nailing
Compression 6 hip screw 84
54
5.05
20
4.12
5.65 14 7
9
Group 2 M 4
I Group 2 M 3
S
13
3.84
18
1.5 cm overgrowth No Incomplete
Complete 128
122 7 Nailing
Compression 8 hip screw 30
50 18
16
7.8
3.64
4.95
3.94 18
10 6
16 Group 2 M
I
Group 1 M
2
S
Table 5. Relationship recurrence
Male
1
Criteria of Deformity Capanna et al. at final [2] at final followup followup Healing time Total after second followup operation (months) (months) Time at Second diagnosis operation of recurrence method (postoperative months) Modified Healing cyst period index at (months) recurrence Patient Initial Sex Site Age at Age at Initial group* diagnosis recurrence modified (years) (years) cyst index
Table 4. Data for four patients with a recurred cyst
123
Coxa brevia None
Cha et al.
Limb length discrepancy at final followup
3172
4
27
Female Age (years)
0 10 ± 4.51
23 10 ± 1.84
0.85
Modified cyst index
4.60 ± 0.86
5.04 ± 0.87
0.34
time was 18 months, and the mean time of recurrence was 55 months postoperatively. A second surgery was performed using flexible intramedullary nails in two patients with an open physis (Fig. 1), and compression hip screw fixation with autogenous iliac bone grafting was performed in two patients with a closed physis (Fig. 2). The mean healing period after the second surgery was 7 months and the recurrent cysts were classified as completely healed in three patients and incompletely healed in one patient. The total followup from the time of initial diagnosis of these four patients was 124 months (Table 4). None of the main demographic variables, such as sex, age, or modified cyst index, affected the incidence of recurrence (Table 5). One patient in each group had coxa vara at final followup. No discrepancies greater than 2 cm were found and no patient had a Trendelenburg sign or any other gait abnormality. Shortening less than 2 cm was observed in two patients in each group. One patient with recurrence in the fracture group had lengthening of 1.5 cm, and one of the patients with less than 2-cm shortening in the group without fractures had a recurrent cyst. Other complications, such as physeal damage attributable to the hardware, infection, or problems related to the protruding nail ends at the insertion site, were not encountered.
Discussion Femoral simple bone cysts accompanied by fractures require fracture reduction and stability, but few studies [3, 13, 20] have focused on femoral cysts that have not yet fractured. Studies of intramedullary nailing of femoral simple bone cysts do not clarify whether there are differences in treatment efficacy with this approach based on whether a fracture has or has not occurred before treatment [7, 16, 20]. We therefore performed a retrospective comparative study to examine whether there were differences between simple bone cysts in the proximal femur nailed after pathologic fracture and those without pathologic fracture in terms of (1) healing time, (2) frequency and timing of recurrence, and (3) complications.
Volume 472, Number 10, October 2014
Intramedullary Nailing for Simple Bone Cysts
3173
Fig. 3A–B (A) Flexible nails were anchored in the neck area in a patient from the group with fractures. (B) In this patient in the group without fractures, the neck cyst was fixed with nailing.
Fig. 4 A 6-year-old boy with a fracture had a cyst that expanded almost to the epiphysis. Two flexible nails were inserted that penetrated the epiphyseal plate for stability.
This study had some limitations. Despite having 25 patients in the group with fractures and 29 patients in the group without fractures, few patients in either group had a recurrence, complicating statistical comparisons of its incidence. In addition, risk factors and predictive factors for recurrence were not identified. Four patients underwent a second surgery for recurrence with fixation with compression hip screws or intramedullary nailing. Stable fixation for a fractured cyst helped to guide selection of the method. However, we could not compare the usefulness of compression hip screws and nailing. All patients with recurrences presented with fractures. Therefore, in this study we could not ascertain the proper guidelines for treatment of cyst recurrence without fracture. Studies have confirmed the usefulness of this method based on a sufficient number of patients, high rate of response for treatment, and long followup [7, 19, 20]. de Sanctis and Andreacchio [7] reported that no difference in healing time was noted between the simple bone cysts with pathologic fractures and those without them. The relatively small number of patients in their study precluded identifying any statistical differences in healing periods or outcomes between the groups of patients. Roposch et al. [20] reported that 11 patients with simple bone cysts of the proximal femur were treated by intramedullary nails. The mean healing time was 48 months for five patients with
123
3174
Cha et al.
Clinical Orthopaedics and Related Research1
Fig. 5A–C (A) A 13-year-old boy had a recurrence 4 years after diagnosis. (B) A reoperation using flexible nails was performed, but the nails perforated the weak portion of the trochanteric area. (C) At final followup, a 2-cm discrepancy was observed, but it was asymptomatic.
pathologic fractures and 31 months for six patients without pathologic fractures. However critical statistical comparison was limited owing to the small number of patients. Nevertheless, they assumed that, despite the lack of statistical significance, a longer healing time was needed in patients who did not have a pathologic fracture [19]. They pointed out that recurrent cysts grew outward from an area not covered by nails. However, these results differ from those of our study, as three subtrochanteric and one intertrochanteric cyst recurred. Flexible nailing yielded stability, decompression, and covered the entire cystic area, but the cysts recurred as pathologic fractures. An additional difference in our study was shortened nails attributable to bone growth. Roposch et al. [19] exchanged nails in nine patients, but we removed the nails as soon as possible when the cyst showed at least incomplete healing according to the classification of Capanna et al. [2]. Namely, when the residual area of radiolucency was
123
smaller than 30% of the initial size, the nails were removed. Thus, all nails were removed 14 to 54 months postoperatively. Norman-Taylor et al. [16] reported that, to reduce recurrence, a Neer healing grade of 3 or 4 [15] is necessary during the first year of diagnosis and throughout followup. Norman-Taylor et al. [16] evaluated the difference in refracture after healing by nonsurgical treatment depending on displacement at the initial presentation. They concluded that displaced fractures were able to achieve union and healing in 1 year, but major complications such as coxa vara and avascular necrosis of the femoral head were frequent. However, undisplaced fractures needed a longer period of healing, additional steroid injections were needed to maintain the grade, and recurrence and refractures were more frequent, leading to coxa vara and short limb length. Their study [16] was meaningful for its comparison of the natural healing potential of a fractured cyst of the proximal
Volume 472, Number 10, October 2014
femur, even without statistically significant evidence. However, their results may differ from ours because we used flexible nailing. The healing potential, time, and pattern may be variable according to the decompression and fixation of the cysts. Teoh et al. [25] found age of 5 years or older; right-sided, large, multilocular cysts; and unimpacted fracture to be predictive factors for recurrence of simple bone cysts of the proximal humerus. Recurrence was not related to sex, cyst location, or cyst activity level. However, with the numbers available, in our study, age, sex, cyst size (modified cyst index), and the presence of fracture all were unrelated to the recurrence of the cyst. In the group without fractures, three patients were diagnosed with recurrence; however, the incidence was not different from that of the group with fractures. In our study, three patients with fractures and seven without fractures had a neck cyst with proper cancellous bone stock for stable anchoring with the nails (Fig. 3). In one patient in the group with fractures, in whom the cyst had expanded nearly to the physis, the nails were fixed through the epiphyseal plate (Fig. 4). Roposch et al. [20] reported that, with large cysts and extensive cortical weakening, nails may exit from the bone. In such cases, avascular necrosis of the femoral head must be considered a potential complication. Such a perforation of a nail through the weakened cortex was observed in one of our patients who experienced recurrence, but the extrusion was through the weak greater trochanter and was not a substantial problem at final followup (Fig. 5). The strength of our study was its comparative nature based on a large number of patients in both included groups through long-term followups. To our knowledge, previous reports [7, 19, 20] of femoral bone cysts were case series and the number of patients who underwent surgery before pathologic fracture was relatively small, making statistical comparisons difficult. Second, we performed surgery using a single method without bone grafting or any other additional fixation. Few patients were lost to followup, so the recurrence frequency in each group could be meaningful. Additionally, the change of the degree of healing was evaluated at the time of recurrence and after the secondary fixation. Whether a pathologic fracture had occurred before surgical treatment, intramedullary nailing of femoral simple bone cysts resulted in reliable healing and the frequency of recurrence did not differ. However, because this was a retrospective study, the optimal treatment for a recurred cyst after intramedullary nailing should be investigated further through a comparative or prospective study. Acknowledgments We thank Nam-Hee Kim PhD (YN Company, Seoul, Korea) for assistance with the statistical analysis.
Intramedullary Nailing for Simple Bone Cysts
3175
References 1. Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat Methods Med Res. 1999;8:135–160. 2. Capanna R, Albisinni U, Caroli GC, Campanacci M. Contrast examination as a prognostic factor in the treatment of solitary bone cyst by cortisone injection. Skeletal Radiol. 1984;12:97– 102. 3. Catier P, Bracq H, Canciani JP, Allouis M, Babut JM. [The treatment of upper femoral unicameral bone cysts in children by Ender’s nailing technique] [in French]. Rev Chir Orthop Reparatrice Appar Mot. 1981;67:147–149. 4. Cha SM, Shin HD, Kim KC, Kang DH. Flexible intramedullary nailing in simple bone cysts of the proximal humerus: prospective study for high-risk cases of pathologic fracture. J Pediatr Orthop B. 2013;22:475–480. 5. Chigira M, Watanabe H, Arita S, Udagawa E. [Simple bone cyst: pathophysiology and treatment] [in Japanese]. Nihon Seikeigeka Gakkai Zasshi. 1983;57:759–766. 6. Cohen J. Unicameral bone cysts: a current synthesis of reported cases. Orthop Clin North Am. 1977;8:715–736. 7. de Sanctis N, Andreacchio A. Elastic stable intramedullary nailing is the best treatment of unicameral bone cysts of the long bones in children? Prospective long-term follow-up study. J Pediatr Orthop. 2006;26:520–525. 8. Fleiss JL, Cohen J. The equivalence of weighted kappa and the intraclass correlation coefficient as measures of reliability. Educ Psychol Meas. 1973;33:613–639. 9. Gentile JV, Weinert CR, Schlechter JA. Treatment of unicameral bone cysts in pediatric patients with an injectable regenerative graft: a preliminary report. J Pediatr Orthop. 2013;33:254–261. 10. Kaelin AJ, MacEwen GD. Unicameral bone cysts: natural history and the risk of fracture. Int Orthop. 1989;13:275–282. 11. Knorr P, Schmittenbecher PP, Dietz HG. [Treatment of pathological fractures of long tubular bones in childhood using elastic stable intramedullary nailing] [in German]. Unfallchirurg. 1996;99:410–414. 12. Lee JH, Reinus WR, Wilson AJ. Quantitative analysis of the plain radiographic appearance of unicameral bone cysts. Invest Radiol. 1999;34:28–37. 13. Malkawi H, Shannak A, Amir S. Surgical treatment of pathological subtrochanteric fractures due to benign lesions in children and adolescents. J Pediatr Orthop. 1984;4:63–69. 14. McKay DW, Nason SS. Treatment of unicameral bone cysts by subtotal resection without grafts. J Bone Joint Surg Am. 1977;59:515–519. 15. Neer CS 2nd, Francis KC, Marcove RC, Terz J, Carbonara PN. Treatment of unicameral bone cyst: a follow-up study of one hundred seventy-five cases. J Bone Joint Surg Am. 1966;48:731–745. 16. Norman-Taylor FH, Hashemi-Nejad A, Gillingham BL, Stevens D, Cole WG. Risk of refracture through unicameral bone cysts of the proximal femur. J Pediatr Orthop. 2002;22:249–254. 17. Oppenheim WL, Galleno H. Operative treatment versus steroid injection in the management of unicameral bone cyst. J Pediatr Orthop. 1984;4:1–7. 18. Peltier LF, Jones RH. Treatment of unicameral bone cysts by curettage and packing with plaster-of-Paris pellets. Clin Orthop Relat Res. 2004;422:145–147. 19. Roposch A, Saraph V, Linhart WE. Flexible intramedullary nailing for the treatment of unicameral bone cysts in long bones. J Bone Joint Surg Am. 2000;82:1447–1453. 20. Roposch A, Saraph V, Linhart WE. Treatment of femoral neck and trochanteric simple bone cysts. Arch Orthop Trauma Surg. 2004;124:437–442.
123
3176
Cha et al.
21. Santori FS, Ghera S, Castelli V, Tollis A. Dynamic endomedullary nailing in the treatment of extensive bone cysts in young patients: a pathogenetic interpretation. Ital J Orthop Traumatol. 1986;12:411–417. 22. Shinozaki T, Arita S, Watanabe H, Chigira M. Simple bone cysts treated by multiple drill-holes: 23 cysts followed 2–10 years. Acta Orthop Scand. 1996;67:288–290. 23. Stanton RP, Abdel-Mot’al MM. Growth arrest resulting from unicameral bone cyst. J Pediatr Orthop. 1998;18:198–201.
123
Clinical Orthopaedics and Related Research1 24. Sturz H, Zenker H, Buckl H. Total subperiosteal resection treatment of solitary bone cysts of the humerus. Arch Orthop Trauma Surg. 1979;93:231–239. 25. Teoh KH, Watts AC, Chee YH, Reid R, Porter DE. Predictive factors for recurrence of simple bone cyst of the proximal humerus. J Orthop Surg (Hong Kong). 2010;18:215–219. 26. Vigler M, Weigl D, Schwarz M, Ben-Itzhak I, Salai M, Bar-On E. Subtrochanteric femoral fractures due to simple bone cysts in children. J Pediatr Orthop B. 2006;15:439–442.
