Key words
patellar instability - blood supply - imaging - remodeling - graft healing - ligament reconstruction
Introduction
The medial patellofemoral ligament (MPFL) is the primary mechanism supporting medial
patellofemoral joint stability. When the continuity of the MPFL is disrupted by
patellar dislocation, the stability of the PF joint is reduced. Reconstruction by
guiding a graft into the bone tunnel is widely used as a treatment to improve
stability. In order to obtain good stability, postoperative biosorption of the graft
to the bone is important. On the other hand, the resumption of blood flow is an
important factor in the reconstruction process of the graft, which can develop into
necrosis early after reconstruction [1]
[2]. Contrast-enhanced magnetic resonance
imaging (MRI) and magnetic resonance angiography (MRA) have been used to evaluate
blood flow to the tendon graft [3]
[4]
[5].
Contrast-enhanced MRI images are taken in the venous phase, when the transplanted
tendon is well-constructed, making it difficult to assess the blood flow path.
However, MRA has revealed blood flow changes associated with graft reconstruction by
visualizing blood flow to the bone tunnel wall and graft after anterior cruciate
ligament (ACL) 1 bundle reconstruction, using a semitendinosus tendon [6]
[7]
[8]. Therefore, we consider MRA
to be a helpful tool to elucidate the remodeling process of the graft.
We hypothesized that inflow vessels to the tendon graft after MPFL reconstruction are
revealed applying the technique of MRA. Fujii et al. [9] have reported chronological radiographic
evaluation after MPFL reconstruction. If chronological blood flow change to the bone
tunnel wall and reconstructed ligament after MPFL reconstruction is clarified in
vivo, it will help to elucidate the remodeling process. Against this background, we
performed MRA after MPFL reconstruction in order to confirm whether blood flow could
be evaluated. We also examined the changes in hemodynamics over time using clinical
and radiological evaluations.
Materials and methods
Subjects
Ten cases and 11 knees with recurrent patellar dislocation that underwent MPFL
reconstruction using the semitendinosus tendon from July 2013 to September 2016
were included in this study. Habitual patellar dislocations were excluded. The
mean age of the patients was 26.4 years with 8 female cases involving 9 knees
and 2 male cases involving 2 knees. The mean height and body mass index were
160.5 cm and 21.6 kg/m2, respectively. The
patients’ demographic data are presented in [Table 1].
Table 1 Patients’ demographic data.
|
Case no.
|
Age (years)
|
Sex
|
Laterality
|
Duration to 1st MRA (months)
|
Duration to 2nd MRA (montlhs)
|
Follow-up (months)
|
|
1
|
43
|
Male
|
R
|
3.1
|
11.8
|
89
|
|
2
|
13
|
Female
|
R
|
3.3
|
14.4
|
95
|
|
3
|
45
|
Female
|
L
|
2.6
|
13.6
|
89
|
|
4
|
37
|
Female
|
L
|
2.2
|
11.7
|
83
|
|
5
|
11
|
Female
|
L
|
3.3
|
13.9
|
80
|
|
6
|
24
|
Female
|
L
|
2.7
|
14.1
|
73
|
|
7
|
23
|
Male
|
L
|
2.0
|
15.1
|
71
|
|
8
|
21
|
Female
|
L
|
2.0
|
11.6
|
71
|
|
9
|
22
|
Female
|
R
|
3.0
|
16.9
|
59
|
|
10
|
14
|
Female
|
L
|
2.9
|
13.1
|
59
|
|
11
|
37
|
Female
|
R
|
2.2
|
10.7
|
51
|
|
Average
|
26.4±12.2
|
–
|
–
|
2.7±0.5
|
13.4±1.8
|
74.5±14.2
|
Ethics approval for this study was obtained from the Ethical Review Board of our
hospital (ERB-C-268-3). All procedures were conducted in accordance with the
ethical standards of the current ethical regulations for research [10] and the Helsinki Declaration of 1975,
as revised in 2000.
MPFL reconstruction
MPFL reconstruction was performed by modifying the method reported by Toritsuka
et al. [11] and by Nakagawa et al. [12]. This procedure can reconstruct MPFL
anatomically, albeit at the risk of a patellar fracture. All operations were
performed by 2 experienced orthopedic surgeons, who worked at same time, under
the supervision of the director of Sports Orthopaedics.
The semitendinosus tendon was harvested according to the method of Hara et al.
[13]. A femoral tunnel with a depth of
30 mm was created with reference to the osseous landmarks distal to the
adductor tubercle (proximally and posteriorly from the centre of the medial
femoral epicondyle) toward the lateral femoral bone cortex, as reported by
Nomura et al. [14]. Two patellar bone
tunnels with a depth of 15 mm each and 4.5-mm in diameter were created
with a reamer using 2 K-wires as guidewires inserted 1/2 and 1/3
of the distance from the proximal edge of the patella. The bone tunnel wall and
the graft were in close contact with each other, and the fit good in all cases.
The tendinous region of the excised and semitendinosus muscle tendon was sutured
to both ends of the #2 FiberWire (Arthrex, Naples, FL) to make a double-bundle
tendon graft. The FiberWire was passed through the patellar bone tunnel, and
both ends of the tendon graft were then drawn into the tunnel and fixed to the
lateral cortex of the patella using an EndoButton (Smith & Nephew,
London, England). The patella was fixed to the center of the femoral trochlea,
while the axial position was confirmed using imaging. For fixation on the
femoral side, the ToggleLoc Fixation Device (Zimmer Biomet, Warsaw, IN) was
used, according to the method described by Nakagawa et al.
[12]. Lateral retinacular release was not
performed.
Postoperative management
From the day after MPFL reconstruction surgery, passive- and active-assisted
range of knee motion was initiated. Weight bearing was gradually increased to
full at 3 weeks postoperatively. Running was allowed at 3 months, followed by a
return to previous sporting activity(ies) at 6 months [9]
[15]
[16].
Magnetic resonance angiography
Imaging was performed with a 3.0 T magnetic resonance imager (Gyroscan
Achieva; Philips Medical Systems, Best, Netherlands) with an 8-channel knee coil
[6]. First, we obtained the
conventional images. Then, an intravenous infusion line was established on the
dorsal side of the patient’s hand and the patient was placed in the
supine position with the knee fully extended in the neutral position. A
fat-suppressed 3D-gradient echo T1-wighted image was used to obtain
contrast-enhanced MRA images. Contrast-enhanced MRA was performed after
intravenous injection of Gd-DTPA (0.1 mmol/kg body weight). The
imaging parameter was as follows: repetition time
(T
R
)=7 msec, echo time
(T
E
)=4.4 msec, flip angle (FA)=12,
field of view=150 mm, slice thickness=2.4 mm,
gap between slices=–1.2 mm, with a 256×190
matrix. Contrast-enhanced MRA was performed every 27 sec. Imaging was
confined to oblique and sagittal sections to align the popliteal artery. Total
imaging time was 2 min and 34 sec. Maximum intensity projection
images were obtained in each of the sagittal and transverse projections selected
to encompass the nutrient blood from the popliteal artery to the tendon graft
and bone tunnel wall. MRA was performed after approximately 3 months and 12
months after surgery.
Evaluation
All patients were evaluated before, 3, and 12 months after MPFL reconstruction.
Clinical data, including the incidence of recurrent subluxation and dislocation,
patellar apprehension, and the Knee Society, Lysholm, and Kujala scores were
measured preoperatively and 12 months after surgery [17]. Plain radiography of the knee,
including conventional anteroposterior and lateral views, and the flexed
patellar axis was performed 3 and 12 months after reconstruction. Trochlear
dysplasia, according to the Dejour classification, was measured on lateral
radiographs [18]. Computed tomography or
magnetic resonance imaging (MRI) was utilized to measure the tibial
tubercle-trochlear groove (TT-TG) distance [19]. The evaluation was performed by the knee surgery specialist.
Statistical analysis
Data are expressed as the mean±standard deviation. Clinical scores
(Kujala score, Knee Society score, and Lysholm score) and radiographic data
(tilting angle, lateral tilt ratio, and congruence angle) were analyzed using
the paired t test. In all analyses, p<0.05 was considered
statistically significant. The 95% confidence interval (CI) was used to
assess the demographic characteristics. Cohen’s d was used to
estimate effect sizes.
Results
The mean time of MRA imaging in the early postoperative period was 2.7 months, and
the mean 12-month postoperative timing was 13.4 months ([Table 1]).
Patient radiographic data are shown in [Table
2]. The parameters about patellofemoral joint alignment improved
postoperatively. All of the clinical scores for the full study population improved
from baseline to 12 months post-operatively ([Table
3]). Cohen’s d of tilting angle, lateral shift ratio,
congruence angle, Kujala score, Knee Society score, and Lysholm score are 0.95,
0.74, 0.86, 2.57, 3.70, and 2.29, respectively.
Table 2 Radiographic data.
|
Pre-operative mean±SD, 95%CI
|
Post-operative mean±SD, 95%CI
|
p value,Cohen’s d
|
|
Femorotibial angle, degrees
|
175.45±2.54, 173.75–177.16
|
–
|
–
|
|
Caton-Deschamps index
|
1.19±0.23, 1.03–1.35
|
–
|
–
|
|
Sulcus angle, degrees
|
148.27±9.68, 141.77–154.77
|
–
|
–
|
|
Dejour classification, number of knees,
N/A/B/C/D
|
2/6/1/2/0
|
–
|
–
|
|
TT-TG distance, mm
|
13.97±4.48, 10.96–16.98
|
–
|
–
|
|
Tilting angle, degrees
|
25.36±11.36, 17.73–33.0,
|
16.82±5.60, 13.06–20.58
|
<0.01,0.95
|
|
Lateral shift ratio, %
|
43.54±24.96, 26.78–60.31
|
28.35±15.21, 18.14–38.57
|
<0.01,0.74
|
|
Congruence angle, degrees
|
22.55±25.46, 17.73–33.00
|
2.82±20.33, –0.84–16.48
|
<0.01,0.86
|
The paired t-test was used for statistical analysis. SD, standard deviation;
CI, confidence interval; TT-TG, tibial tubercle-trochlear groove.
Table 3 Pre- and post-operative clinical scores.
|
Pre-operative mean±SD, 95%CI
|
Post-operative mean±SD, 95%CI
|
p value, Cohen’s d
|
|
Kujala score
|
61.27±20.05, 47.81–74.74
|
98.36±3.88, 95.76–100.97
|
<0.01,2.57
|
|
Knee Society score
|
71.82±9.10, 65.71–77.93
|
96.82±2.96, 94.83–98.81
|
<0.01,3.70
|
|
Lysholm score
|
77.09±13.19, 68.23–85.96
|
99.00±3.00, 96.99–101.02
|
<0.01,2.29
|
The paired t-test was used for statistical analysis. SD, standard deviation;
CI, confidence interval.
Early postoperative MRA showed an inflow vessel from the medial superior genicular
artery into the femoral bone tunnel wall ([Fig.
1a]). Blood flow to the patellar bone tunnel wall was from the articular
branch of the descending genicular artery and the medial superior genicular artery,
proximally, and from the medial inferior genicular artery, distally. In addition, a
contrast effect was observed around the tendon graft between the bone tunnel ([Fig. 2a–c], [Fig. 3a–c]). Approximately 1 year
after surgery, the contrast effects around the tendon graft and at the bone tunnel
wall were diminished ([Fig. 1b]) In the
sagittal section, the bone tunnel wall of both the femur and patella was contrasted
as a ring ([Fig. 2d–f], [Fig. 3d–f]). This contrast effect was
seen from the early postoperative period and tended to decrease up to 1 year after
surgery. Similar results were obtained in all cases.
Fig. 1 (a) An oblique, sagittal magnetic resonance angiography
image of the knee taken 2 months after left medial patellofemoral ligament
reconstruction. The articular branch of the descending artery and the medial
superior genicular artery extended to the proximal patellar bone tunnel in
all patients and the medial inferior genicular artery extended to the distal
patellar bone tunnel. The medial superior genicular artery also extended to
the femoral bone tunnel.; (b) The same slice as in [Fig. 1a], 12 months after MPFL
reconstruction is shown.; All of the contrast effects seen at 2 months
postoperatively were attenuated.; Dotted line circle: patellar bone tunnels;
Solid line circle: femoral bone tunnel
Fig. 2 Magnetic resonance imaging of case No. 4;
a–c 2 months after left MPFL reconstruction.;
d–f 12 months after left MPFL reconstruction.;
a, d Axial image of the left knee.; b, e
Sagittal section of line 1 shows the femoral bone tunnel (square) and the
tendon graft (circle).; c, f Sagittal section of line 2 shows
the patellar bone tunnels (square).
Fig. 3 Magnetic resonance imaging of case No.10;
a–c 3 months after left MPFL reconstruction.;
d–f 12 months after left MPFL reconstruction.;
a, d Axial image of the left knee; b, e
Sagittal section of line 1 shows the femoral bone tunnel (square) and the
tendon graft (circle).; c, f Sagittal section of line 2 shows
the patellar bone tunnels (square).
Two knees showed no signs of trochlear dysplasia, whereas 6 exhibited type A, one
exhibited type B, 2 exhibited type C, and none exhibited type D trochlear dysplasia
([Table 2]). Before the operation, the
tilting angle, the lateral shift ratio, and the congruence angle were
25.4°±11.4°, 43.5±25.0%, and
22.5°±25.5°, respectively. On plain radiographs obtained 12
months post-operatively, the tilting angle, the lateral shift ratio, and the
congruence angle were 16.8°±5.6°,
28.4±15.2%, and 2.8°±20.3°, respectively.
The tilting and congruence angles and the lateral shift ratio at 12 months
post-operatively significantly increased relative to pre-operatively
(p=0.01, p<0.01, p<0.01, respectively).
None of these patients showed recurrence of subluxation or dislocation of the
patella. They all returned to their original sports level. The average preoperative
Tegner activity score [20] was 3.0±1.5
points.
Discussion
The aim of this study was to evaluate blood flow after
MPFL reconstruction applying the technique of MRA. The main findings of this study
demonstrate that an inflow vessel from the medial superior genicular artery into the
femoral bone tunnel wall and blood flow to the patellar bone tunnel wall was from
the articular branch of the descending genicular artery and the medial superior
genicular artery, proximally, and from the medial inferior genicular artery,
distally. Moreover, contrast effects of bone tunnel walls were seen from the early
postoperative period and tended to decrease up to 1 year after surgery.
One of
the representative surgeries to place the graft in the bone tunnel is ACL
reconstruction using the hamstring tendon. Remodeling of the graft and strong union
of the bone-tendon junction are important factors that determine the outcome after
ACL reconstruction [21]
[22]. It has been reported that the tendon graft
undergoes the processes of necrosis, angiogenesis, cell proliferation, and collagen
remodeling that histologically approximate the normal ACL [23]. However, the histological changes of the
transplanted tendon after MPFL reconstruction remain unknown. On the other hand, the
hemodynamics of the surrounding bone tunnel wall and the graft after ACL
reconstruction have been analyzed using MRA; Arai et al.
[6]
[7]
successfully visualized the hemodynamics after ACL reconstruction using MRA, and
Kanamura et al. [8] subsequently reported the
changes over time. In this study, a similar technique was used after MPFL
reconstruction. As a result, contrast effects of the medial superior genicular
artery on the femoral bone tunnel and the articular branch of the descending
genicular artery and medial superior genicular artery on the proximal patellar bone
tunnel were confirmed. In the distal patellar bone tunnel, a contrast effect from
the medial inferior genicular artery was observed. The current findings showed that
MRA after MPFL reconstruction exhibited a contrast effect around the graft and that
blood flow to the bone tunnel wall was observed at about 3 months after surgery. In
this study, blood flow to the bone tunnel wall was restored after MPFL
reconstruction as well as after ACL reconstruction, suggesting that the biological
union between the bone tunnel wall and the graft had progressed.
Also, at 12
months postoperatively, these blood flows were reduced but still visible, suggesting
that remodeling is ongoing. Fujii et al. studied the changes in PF joint conformity
over time and reported that it changes up to 3 months postoperatively, but
stabilizes thereafter [9]. Based on the above,
we concluded that this study demonstrated that ligamentization of the graft
progressed even if the conformity of the PF joint did not change ([Fig. 4]).
Fig. 4 Chronological radiographic changes of the patellar axis after
MPFL reconstruction; The tilting and congruence angles improved immediately
after the operation (b) compared with before the operation
(a). The tilting and congruence angles 3 months after the operation
(c) were decreased relative to those recorded immediately after
the operation (b). The tilting and congruence angles were not
different 3 months and one year after the operation (d).
The rehabilitation protocol after MPFL reconstruction varies in the
literature, and systematic reviews show that most patients return to sports 3 to 6
months after surgery [16]. Similarly, our
postoperative protocol is to start athletic rehabilitation from 2 to 3 months after
surgery and to return to sports at 6 months after surgery. Based upon the findings
of this study, we believe that the postoperative period should be limited to linear
athletic rehabilitation, avoiding rotational movements that place a heavy load on
the graft, as is the case after ACL reconstruction.
In addition, a contrast
effect was seen around the graft between the bone tunnels approximate 3 months after
surgery, indicating that the graft was covered with tissue with good blood flow
outside the bone tunnel. In ACL reconstruction, the central portion of the graft is
located intra-articularly and has no surrounding soft tissue to contact. Thus, the
only way to restore blood flow to the central portion of the transplanted tendon is
through the tendon-bone junction on the femoral or tibial side. In contrast, the
MPFL is located outside the joint cavity, and the central portion of the graft in
MPFL reconstruction is surrounded by soft tissue. Therefore, we considered that a
good contrast effect around the graft tendon was achieved. The direct influx of
blood flow from the soft tissues around the graft may have facilitated remodeling of
the graft between the bone tunnels.
The results of the present study indicate
that hemodynamics of the graft may change up to 3 months post-operatively. In
post-MPFL reconstruction rehabilitation programs, it is important to avoid knee
rotation, such as turns and side steps, to prevent excessive lateral stress on the
patella up to 3 months postoperatively.
There are several limitations to this
study. Firstly, the sample size was too small to evaluate statistically and the
gender ratio is also not uniform. Furthermore, the evaluations were qualitative, and
the signal values could not be evaluated quantitatively. Finally, we evaluated the
subjects only during the early stage (approximate 3 months postoperatively) and 12
months postoperatively, and the changes during these periods are unknown.
In
conclusion, we evaluated blood flow to the graft after MPFL reconstruction and
demonstrated that MRA can be used to detect blood flow and that contrast effects can
be obtained in the bone tunnel wall and around the graft at approximate 3 months
postoperatively. We further found that these contrast effects decrease at 12 months
postoperatively. The results of this study may be useful in elucidating the
reconstruction and maturation process of the graft after MPFL
reconstruction.
