The goal in treating comminuted patellar fractures is to restore the extensor mechanism, to anatomically reduce the articular surface, and to provide a stable construct to allow early rehabilitation [2, 18, 19]. In order to achieve this goal, fixation constructs should be versatile, stable, and strong enough to allow early mobilization. While various techniques of internal fixation have been recommended [3–8], a gold standard of treatment has not been established. In addition, many severely comminuted patellar fractures are commonly treated with partial or total patellectomy, which results in devastating outcomes [20]. These facts indicate that management of comminuted patella fractures remains to be a challenge.
Surgical fixation of articular fragments in the comminuted patellar fractures is often too complex and difficult because of inherent weakness of the bone and relatively small fracture fragments. In the treatment of such fractures, the modified tension band wiring fixation alone is often unable to achieve effective fixation. Yang et al. [3] suggested the management of displaced comminuted patellar fracture with titanium cable cerclage. In this approach, the articular fragments were not fixed directly, and the reduction was maintained only by the cerclage. Failure may develop during the rehabilitation due to a lack of direct fixation to the articular fragments. The modified tension band technique can be used in conjunction with the cerclage to enhance the stability of the fixation. However, circumferential cerclage may also result in multiple segmentation of the wire. We applied a reduction technique for comminuted patellar fractures, which was based upon suture reduction concept, this suture reduction technique has been applied to some comminuted fractures with small fragments, but there is no more application report in the preservation of comminuted patellar fractures.
The current study presented the clinical outcomes of SR combined with the modified K-wire tension band in management of comminuted patellar fractures (Fig. 2 and Fig. 3). Herein, the surgical time and blood loss in the SR group were less than those in the CR group, which could be explained by the following points. First of all, during the SR process, we mainly sutured the periosteum to initially restore the integrity of the patella, without excessive dissection of the fracture, therefore reducing tissue bleeding. Secondly, preoperative attention to the characteristics of fracture and intraoperative protection of periosteum integrity were conducive to the implementation of SR technique and shorten the operation time. Finally, SR did not rely too much on some reduction tools, such as reduction forceps, K-wires, etc. and the process was relatively simple. In addition, the difference in surgical time between the two groups was not statistically significant, which may be related to the small sample size. In our series, fracture union was achieved in SR group at an average of 10.1 weeks. Full knee flexion/extension was achieved in SR group except 2 cases with limited knee flexion. Clinical results were comparative with the results obtained by other scholars[5, 7, 8, 21–23]. In those studies, the majority of patients achieved union with full ROM. Compared with the results of those studies and CR group, our SR technique took less time and less blood loss, which was consistent with the current concept of minimally invasive surgery. Functional results of SR group and other case series using different fixation techniques were summarized in Table 3.
Table 3
Different studies reporting on functional outcomes of comminuted patella fractures
| Singer et al.5 | Suh et al.7 | Cho et al.8 | Gao et al.21 | Sun et al.22 | Wang et al.23 | SR Group |
Case (n) | 9 | 13 | 30 | 16 | 38 | 25 | 35 |
Fracture type ( AO/OTA classification) | 34-C3 | 12 cases: 34-C3; One case: 34-C1 | 25 cases 34-C3; 5 cases 34-C2 | No mention | 28 cases: 34-C3; 10 cases: 34-C2 | 6 cases: 34-C3; 19 cases: 34-C2 | 34-C3 |
Closed fractures (n) | 9 | 11 | 27 | 16 | 38 | 25 | 35 |
Fixation method | Low profile mesh plate | Headless compression screws with additional separate vertical wiring | Miniplate augmented tension-band wiring (TBW) | A miniature plate with a tension band wire | Modified cerclage wiring | Modified anterior ellipsoidal cap titanium cable tension band | suture reduction combined with the modified Kirschner-wire tension band |
follow-up (months) | 19.6 (12–33) | 16 (12–25) | 20 (12–28) | 15.6 (12–20) | 16.1 (6–36) | 25 (17–39) | 35.7 (6–64) |
Surgical time (minutes) | 69 (55–90 ) | No mention | 71.1 (60–79) | No mention | 66.4 (55–80) | No mention | 63.6 (45–105) |
Postoperative complications | One case superficial wound issues; One case deep venous thrombosis (DVT) | Four cases thigh muscle wasting; Three cases mild anterior knee pain; Six cases removal of hardware | One case postoperative infection; Four cases implant removal; Four cases flexion contracture; Three cases extension lag | No | No | Two cases minor complications (soft tissue irritation, cellulitis) | One case limited flexion; One case knee stiffness; Two cases anterior knee pain; Two cases K-wires migration |
Range of motion (ROM) | Eight cases: full range of knee flexion/extension; One case:Extension:-10 degrees, flexion: 90 degrees | 134.2 degrees (120–145) | 120 degrees (110–130) | All cases: full range of knee flexion/extension | 130 degrees (110–140) | No mention | 122 degrees (60–135) |
Knee function score | Lysholm score: 89.1(82–95); Böstman scale: 27.2(22–30). | Lysholm score: 94.4 (84–100); Böstman scale: 28.7 (25–30) | Lysholm score: 94.3 (82–100); Böstman scale:28.6(26–30) | Lysholm score: 91.6(84–97); Böstman scale: 26.4(22–30) | Böstman scale: 28.7 (20–30) | Böstman scale: 27.3 (23–30) | Lysholm score: 91.8(65–100); Böstman scale: 27.6(13–30). |
Symptomatic implants are one of the reasons that many surgeons have abandoned the K-wire tension band construct for patella procedures[9, 24–26]. Hoshino et al.[9] reported symptomatic hardware removal in 36.8% of patients treated by K-wires and cerclage. In another retrospective study, rate of implant prominence and subsequent removal was 13%. Most irritation was usually associated with the proximal end of the K-wire when the wire initially fails to be imbedded deeply under the soft tissue or was displaced due to loss of anchoring [27]. In the current study, owing to our efforts to reduce implant prominence on the anterior cortex, three cases of hardware complications or removal were reported. For example, the development of one-layered soft-tissue flaps along the surgical incision may reduce this low rate of implant prominence.
The limitations of our study were the small sample size, retrospective nature, and the lack of long-term follow-up. A biomechanical analysis would also be beneficial. Additional prospective and biomechanics studies should be conducted to confirm these outcomes in the future.