Medulloscopy-Assisted Surgery for Osteonecrosis of the Knee Following Treatment for Adolescent Leukemia: Mid-term Results

Abstract

Background

This study aimed to introduce the medulloscopy-assisted autologous bone graft for osteonecrosis of the knee following adolescent leukemia treatment and to report the mid-term clinical and radiological results.

Materials and Methods

We retrospectively reviewed records of eight consecutive knees in six patients with extensive osteonecrosis of the distal femoral condyles, from 2013 to 2015. All patients were treated with autologous bone grafts using medulloscopy. We evaluated functional and objective Knee Society Score (KSS) pre- and postoperatively. Clinical and radiological outcomes were evaluated at the last follow-up.

Results

The mean age of the patients at the time of surgery was 21.5 years. The mean follow-up period was 45 months. No additional surgery was required during the follow-up period. The mean objective KSS significantly increased from 60.6 points (range 55–65) to 90.0 points (range 70–95), postoperatively (p < 0.001). The mean functional KSS significantly improved from 48.8 points (range 45–55) to 90.0 points (range 80–100) postoperatively (p < 0.001). No patients had progression of collapse of the femoral condyles at the final follow-up. At the final follow-up, the onset of minimal osteoarthritis was seen in two knees, with a Kellgren–Lawrence grade 1.

Conclusion

Medulloscopy-assisted bone grafting can be an effective treatment option for early symptomatic osteonecrosis of knee following treatment for adolescent leukemia.

Introduction

Osteonecrosis (ON)-associated morbidity is a common musculoskeletal complication that affects 13–47% of acute lymphoblastic leukemia (ALL) survivors [1]. It has been reported that 20–25% of patients with symptomatic ON continue to have pain and gradually progressing symptoms [2, 3]. Progression of ON can result in physical impairment and the need for arthroplasty. A nationwide population-based registry showed that patients with hematologic and lymphatic malignancies had an increased risk for joint arthroplasty, compared with the general population [4]. The standardized incidence rate of arthroplasty was highest (45.5%) in ALL patients under 35 years of age [4]. Among ALL survivors, ON occurs most commonly in the hip; the knee is the second most common site of ON [5, 6]. There are many studies reporting ON of the femur head; however, research regarding ON in the distal femur is relatively rare.

ON of the distal femur is commonly associated with subchondral bone fracture and collapse, which results in secondary osteoarthritis. Conservative treatments, such as protective weight-bearing therapy and nonsteroidal anti-inflammatory drugs, eventually lead to the need for total knee arthroplasty within 6 years [7]. Knee arthroplasty is an inadequate treatment for young and active adult ALL survivors. Various surgical options, including core decompression, autologous bone grafting, and osteochondral grafting, have been performed, based on the stage of ON [7–11]. However, there is still controversy regarding which procedures are best for preserving the knee joint in ON patients [12]. Impacted morcellized bone grafting for steroid-associated ON of the knee showed favorable outcomes in midterm reports [8, 13]. However, removal of the necrotic bone in the subchondral lesion by a fluoroscopy guide is inaccurate and may cause cartilage damage [8].

Bone endoscopy (medulloscopy) has been reported as a treatment for osteomyelitis, bone grafting and bone cysts [14–17]. Many studies have described osteomyelitis or intramedullary benign bone tumors and highlighted the distinction between the nonviable sequestrum and healthy cancellous bone [15, 16, 18]. There is also a study reporting necrotic bone removal monitoring for the treatment of ON; this method is performed by alternately inserting curettage and endoscopy using a single portal in the femur head [14]. However, no studies have used medulloscopy for treating ON in the distal femur. In particular, no studies have reported the usage of safe curettage and bone grafting, with separated portals for viewing and working.

The authors have performed medulloscopy-assisted surgery using the separated portal for precise and safe curettage of the necrotic bone. The aim of this study was to retrospectively evaluate the efficacy and safety of medulloscopy-assisted autologous bone grafting for ON of the knee following treatment for adolescent leukemia.

Methods

Study Design and Ethical Considerations

We retrospectively reviewed records of 94 children (0–18 years) diagnosed with and treated for ALL at our institution from March 2010 to May 2015. The inclusion criteria were as follows: (1) ON of the distal femoral condyles following systemic treatment of adolescent acute lymphoblastic leukemia, (2) knee pain without improvement despite medication for 6 months, (3) modified Ficat and Arlet classification II and III, (4) patients underwent medulloscopy-assisted autologous bone grafting, (5) followed-up for at least 3 years with sufficient radiological and medical records. Ethical approval by the appropriate institutional review board was obtained for this study with waiver of informed consent because of the study’s retrospective nature.

Evaluations

In all patients, magnetic resonance imaging (MRI) scans were performed for the final diagnosis and assessment of the lesion. A single-photon emission computerized tomography (SPECT) scan was conducted for three patients. Each knee was evaluated functionally and objectively using the Knee Society Score (KSS) (Table 1). A KSS score ≥ 80 indicates good functional outcomes. Lesions were staged according to the Ficat and Arlet classifications, which were modified for the knee [19]. The size of lesions was assessed as a percentage of the involved femoral condyle, as previously established [8].

Table 1.

Patient clinical characteristics

Case	Patient	Sex	Age (years)	Side	Location of lesion	Ficat Arlet classification	Size of lesion (necrotic angle)	follow-up (months)	Underlying disease	Preoperative		Follow-upb
										KSSa	Functional KSSa	KSSa	Functional KSSa
1	1	M	20	Rt	Medial	2	275°	55	ALL (B cell)	65	55	95	90
2	2	M	19	Rt	Medial	3	263°	50	ALL (B cell)	55	45	70	80
3		M	20	Lt	Medial/Lateral	3	271°	48	ALL (B cell)	60	45	90	100
4	3	F	24	Lt	Lateral	2	190°	44	ALL (B cell)	65	55	95	100
5	4	M	21	Rt	Medial	3	266°	44	ALL (T cell)	65	45	95	100
6		M	22	Lt	Medial/Lateral	3	287°	43	ALL (T cell)	55	45	90	80
7	5	F	25	Lt	Medial	2	174°	40	ALL (B cell)	65	55	90	90
8	6	M	21	Rt	medial	3	277°	37	ALL (B cell)	55	45	95	80

ALL acute lymphoblastic leukemia

^aKnee Society Score

^bAt last follow-up

Surgical Technique

All procedures were performed by the same surgeon. Patients were placed in the supine position on the radiolucent table. A stab incision was made on the contralateral side of the metaphyseal area; this was done to establish the viewing portal percutaneously. A percutaneous sleeve assembly was inserted through a stab incision and through the soft tissue, until it reached the bone (DePuy Synthes Cannulated Screw System, PA, USA). A 2.8-mm guide wire was inserted through the proximal end of the affected condyle under an image intensifier. A 7.3-mm cannulated drill bit was used to drill the cortex and make a hole for the arthroscope. A 70° angled sheathed arthroscope was inserted through the hole. Another skin incision was made on the affected side, just proximal to the epicondyle, for the bone graft (Fig. 1a). After the removal of the periosteum, a 2 × 2 cm cortical window was made. Once the viewing and working portals were established, irrigation was performed through the viewing portal to create a wet medulloscopy state. Under medulloscopy, the necrotic bone demonstrates a hard and dense texture that is distinct from the viable cancellous bone (Fig. 1b). The curette and powered burr were used to remove the necrotic bone until the viable cancellous bone was exposed (Fig. 1c). Fluoroscopy was applied to prevent chondral damage. Autologous cancellous bone was harvested from the ipsilateral tibia or iliac crest, and impacted layer-by-layer under medulloscopy monitoring (Fig. 1d). Healthy cancellous bone from the metaphyseal area was also used to fill the remaining cavity. The cortical window was replaced and the periosteum was sutured.

Fig. 1.

a Percutaneous viewing and open working portals to approach the osteonecrotic lesion. b Curettage of the necrotic bone under medulloscopy guide. c After curettage, medulloscopy can be used to check for the typical honeycomb texture of the viable cancellous bone (arrow). d The autologous bone was impacted under medulloscopy monitoring

Postoperative Course and Follow-up

Full range of motion of the knee and toe-touch weight bearing, with the use of double crutches, were allowed for the first 6 weeks. Partial weight bearing, with the use of a single crutch, was instructed during the next 6 weeks. The surgeon followed up all patients for the assessment of clinical and radiological results every 3 months at outpatient department. Clinical and radiological success were defined as previously documented [8]. Radiological failure was determined as the onset or the progression of collapse or osteoarthritis. Clinical success was defined as a minimal objective score of 80 points and no conversion to total knee arthroplasty.

Statistical analyses were performed using SPSS® software (Version 24; SPSS Inc., Chicago, IL, USA). Analyses of continuous data were performed using the paired t test. p values less than 0.05 were considered to indicate significant differences.

Results

Clinical Results

A total of 12 patients had MRI-confirmed ON of the distal femoral condyles. Eight consecutive knees in six patients were enrolled. The mean age at ALL diagnosis was 14.4 years (range 13–16 years). The mean duration of knee pain was 17 months (range 12–21 months). All patients had knee pain for > 6 months, despite protected weight bearing using a cane and the use of non-steroidal anti-inflammatory drugs. The mean age of the six patients at the time of surgery was 21.5 years (range 19–25 years). The mean follow-up period from surgery was 45 months (range 37–55 months). Details of ON lesions and involvement are shown in Table 1. All patients underwent one operation and no additional surgery was required during the follow-up period. There was no perforation of the cortex of the condyle during surgery. The mean operation time was 66 min (range 48–95 min), while the mean duration of hospital stay was 7 days (range 4–11 days). The mean objective KSS increased from 60.6 points (range 55–65) to 90.0 points (range 70–95), postoperatively (p < 0.01). The mean functional KSS improved from 48.8 points (range: 45–55) to 90.0 points (range 80–100), postoperatively (p < 0.01). One patient (case number 2) complained of recurrent knee pain (KSS score 70), but did not show radiological progression at 18 months. The patient was able to control the pain with intermittent medication and they maintained a functional KSS score of 80, until the last follow-up. The clinical success rate was 88% (seven of eight knees).

Radiological Results

Six knees had only distal femoral involvement, while two had both femoral and tibial involvement. Seven patients presented bilateral involvement of ON. Four knees had only medial condylar ON, one had only lateral condylar ON, and three knees had both condylar lesions. The mean lesion size on preoperative MRI was 250.4° (range 174°–287°). Two lesions were medium-sized, with a combined necrotic angle ranging from 151 to 249° (medium size); the remaining lesions were all > 250° (Fig. 2).

Fig. 2.

a Preoperative anteroposterior radiogram presented the subchondral collapse of the medial femoral condyle (Ficat and Arlet stage 3). b Preoperative MRI showed a separation of cartilage from the subchondral bone (arrowhead) c Follow-up MRI conducted 1 year after medulloscopy-assisted bone grafting showed recovery of the lesion. d A simple radiogram performed 4 years after surgery showed no progression of the subchondral collapse (black arrowhead)

No patients had progression of preoperative collapse of the femoral condyles at the final follow-up. At the final follow-up, the onset of minimal osteoarthritis was seen in two knees (case numbers 2 and 6); these knees demonstrated a Kellgren–Lawrence grade of 1. The radiographic success rate was 75% (six of eight knees). Separation of the cartilage from the subchondral bone was seen in one patient (case number 2). The follow-up MRI showed recovery of the cartilage lesion 1 year after the operation (Fig. 2).

Discussion

This study suggested a new treatment option, using medulloscopy, for extensive ON lesions of the knee. Medulloscopy-assisted autogenous bone grafting could be a feasible treatment option for preserving the native knee joints of ON patients. All patients had subjective improvement after surgery. Considering 45.5% as the highest incidence rate for arthroplasty in ALL patients under 35 years of age, finding appropriate treatments for ON in young adults is crucial. Although this study is a small series, it is significant as it reports the treatment results of a homogeneous cohort, which includes young ALL survivors with symptomatic ON of the distal femur.

Impaction bone grafting of the femoral condyle has been suggested as relatively simple and effective technique in younger patient with early stage ON [8]. However, removal of the necrotic bone through the small cortical window is challenging due to the limited field of vision. Fluoroscopy guidance from multiple directions takes time and requires radiation exposure. A previous study using fluoroscopy reported perforations of the cortex of the condyle in 22% of knees (two of nine knees) [8]. The extent of bone necrosis was predetermined according to MRI. However, the exact extent of bone necrosis is difficult to identify during surgery; furthermore, it is difficult to confirm that it has been removed properly. The efficacy of medulloscopy to differentiate between healthy bone and pathologic lesions has been described several times in previous studies, in regard to osteomyelitis and bone tumor surgery [14–18]. The authors can use medulloscopy to distinguish the typical honeycomb structure of healthy cancellous bone and the dense structure of necrotic bone (Fig. 1c).

The authors applied a separated viewing portal to monitor the procedures simultaneously. Most studies to date, have used a single portal for viewing and working [14–17]. In single portal surgery, the scope and instrument should be used alternately. Therefore, the single portal approach has some drawbacks; it is associated with a long operation time, maintaining the scope orientation is difficult, and simultaneous monitoring cannot be done. In this series, the biportal approach enables adequate curettage and thorough bone grafting, with a relatively short operation time (66 min). The mean operation time for open bone grafting with fluoroscopy was 89 min in another study [8]. The viewing portal on the contralateral side of the metaphyseal area can be established easily using a percutaneous approach and a currently used cannulated screw system.

Limitations

This study has some limitations. The small number of patients included in the study may not be enough to confirm the efficacy of our novel technique. However, ALL with ON is a condition with low incidence and other studies, to date, included a small number of patients within a heterogeneous cohort. The present homogeneous cohort, which is focused on a specific location (knee joint) and treatment option (medullloscopy-assisted bone grafting), might be sufficient to suggest our technique as an effective modality for ON. Regarding the natural history of ON, the promising results of our midterm follow-up study should be confirmed by long-term evaluations with more patients.

Conclusion

Medulloscopy-assisted bone grafting can be an effective treatment option for early symptomatic ON of the knee joint.

Author contributions

ESC conception of the study, acquisition and analysis of data, drafting the article. HDS interpretation of data, revising it critically.

Funding

None.

Compliance with Ethical Standards

Conflict of interest

The authors declare that they have no conflict of interest. The manuscript has been read and approved by all the authors. The manuscript represents honest work.

Ethical standard statement

This article does not contain any studies with human or animal subjects performed by the any of the authors.

Informed consent

For this type of study informed consent is not required.