Costa et al. [37] reviewed 23 level I and II studies and found no significant difference in clinical scores between MIS and conventional techniques. In Li et al.’s [39] meta-analysis MIS-TKA has shown significantly improved results in objective and subjective outcome scores, VAS, ROM, knee flexion, flexion 90 day and straight leg rising day, all of which resulted in patients achieving faster recovery. In another study [40] results were analysed for measurements available at 6 weeks, 3 months and 6 months or more. The objective score at 3 months was just marginally significant in favour of the MIS group. In the same study the (VAS) was significantly improved in the MIS group.

ROM is an important parameter of functional outcome. In control cohort studies, the MIS approach was found to be superior to the conventional approach in terms of ROM [41, 42]. Alcelic et al. [40] evaluated 507 MIS versus 513 conventional TKAs in a metanalysis. Knee flexion was significantly greater by 9.9° on average in the MIS group at 1 week postoperatively but not at 3 months.

Postoperative pain is an important parameter in terms of patient satisfaction and also greatly affects postoperative rehabilitation. Recently published meta-analyses report better results in VAS scores and postoperative pain [39, 40]. In our RCT [23] we report that during the first post-operative week pain was apparently greater in the minimally invasive group. In our opinion, pain management is a serious confounding factor in many published studies, and reduced pain levels should not be presented as a benefit of minimally-invasive TKA (Fig. 11.5).

The difficulty of consistently showing a definite clinical advantage of MIS techniques has led investigators to try to measure clinical outcome by other means. Bonutti et al. [43] used the contralateral knee as a control group and found that peak extensor muscle strength was significantly and consistently higher in the MIS group. Most patients preferred the knee treated with the MIS approach. Interestingly, these patients had different objective Knee Society Scores but similar Knee Function Scores. The authors conclude that this measure may not be sensitive enough to detect the improved outcomes of MIS techniques which are detectable with more specific measures, such as isokinetic muscle strength testing, ambulation and straight leg-raising time.

Costa et al. [37] concludes, in a systematic review, that the only significant difference observed was in the recovery of quadriceps muscle function (shorter in patients who had undergone a minimally invasive approach).

Alcelic et al. [40] also found a significant reduction in number of days to SLR in the MIS group.

MIS techniques seem to offer the advantage of less total blood loss and early postoperative decrease of haemoglobin [39, 40].

Despite initial reservations, the fear of implant malposition does not seem to be confirmed. Most of the meta-analyses [37, 39, 44] did not find inferior radiological results in terms of component orientation and outliers with MIS techniques. Bonutti et al. [45] reviewed their first 1,000 MIS TKAs and their most important radiologic finding were 3 impending component failures, 2 tibial and 1 femoral. They raise the concern of potential tibial component loosening related to decreased exposure and possibly poor cement pressurization. In another study by there were more cement voids and more retained cement debris in the minimally invasive cohort. In our study a high incidence of wrong implant placement was found and we argue that femoral component malrotation is not detected by conventional radiography [23].

Costa et al. [37], after reviewing a total of 23 level I or II studies, conclude that there were no differences in complication rates between patients with various minimally invasive approaches compared with a standard approach. The minimally invasive lateral approach had more complications than the other minimally invasive approaches. Alcelic et al. [40], in a systematic review and metanalysis, analyzed four RCTs and quasi RCTs with a total of 296 MIS TKAs for complications. They did not find an increased risk of wound healing problems but MIS techniques showed a significantly increased risk of developing intraoperative complications with a risk ratio (RR) of 7.6. The most common complication was quadriceps tendon laceration (31 cases) and anterior femoral notching (5 cases). Other reported complications were patellar tendon rupture or partial avulsion (four cases), laceration of the popliteus (two cases), palsy of the deep peroneal nerve (two cases), one medial epicondylar avulsion fracture, one femoral condylar fracture, one tibial plateau fracture, one supracondylar fracture and the inability to displace the patella or achieve adequate exposure in four cases in the MIS group requiring conversion to the standard approach. This finding, in our opinion, reflects the technical difficulties that surgeons encounter when performing MIS techniques.

On the contrary, Li et al. [39] discovered significant differences in superficial wound healing problems and a risk of increased skin necrosis with MIS techniques but not in superficial wound infection, deep infection, DVT, fractures, femoral notching, peroneal nerve palsy, stiffness requiring manipulation, polities tendon injury, or knee instability. However, we do not feel that superficial wound problems, which are usually easily treated, should be regarded as severe complications.

MIS techniques are associated with longer operative and tourniquet times, especially during the learning curve period [40], but there are studies indicating that these parameters reduce to levels comparable with conventional techniques as experience is accumulated by the surgical team [24].