The arthrotomy is extended proximally and distally at an angle of 45° up to the border of the vastus lateralis [3]. This technique does not require any changes to the rehabilitation protocol following the revision procedure and allows immediate active mobilization.

This extended approach may be considered when a quadriceps snip is not sufficient. It consists in continuing the quadriceps snip arthrotomy through the vastus lateralis tendon and the lateral retinaculum up to the superior portion of the iliotibial band, thus preserving the vascularization of the lateral geniculate artery. This approach entails a postoperative restriction of knee flexion for 4–6 weeks and may induce a knee extension lag. For that reason, it is only used in very specific cases where lengthening of the extensor mechanism is required.

An 8–10 cm-long osteotomy is performed on the medial side, leaving a bone bridge on the lateral cortex and fixing the osteotomy posteriorly with K-wires and a screw. To carry out this osteotomy, the anteromedial incision must be lengthened exposing 10–15 cm of the proximal tibia. A saw must be used from medial to lateral, creating a 10-cm-long, 2-cm-wide, and 1-cm-thick bone fragment. The lateral part of the osteotomy is completed with a chisel. The fragment is elevated subperiostically, taking care to spare the periosteum and leaving the muscle attached to the fragment. Flexion, extension, and weight bearing are allowed from the second week post-op. This technique has been associated with a large number of complications, including fractures of the bone fragment, delayed healing, and loosening of the fixation devices. In contrast to the V-Y quadriceps turndown, tibial tuberosity osteotomy results in low postoperative patient satisfaction and a lower postoperative range of motion. However, the V-Y quadriceps turndown tends to give rise to a significant knee extension lag [4].

These spacers may be produced by a manufacturer or be custom-made by the surgeon intraoperatively using molds of variable sizes or shaping them using the explanted components as a model. These spacers have demonstrated infection resolution rates of 80–100% in different series. Complications such as spacer subluxation or breakage are uncommon. Although placement of these spacers tends to be straightforward, care must be exercised to ensure correct orientation of the femoral component so as to minimize the incidence of spacer breakage.

This kind of spacer consists of a handmade cement femoral component and a stemmed all-poly tibial component. The infection control rate for this type of spacer has been reported to be in excess of 90% [11].

This category includes a wide range of options that go from using the re-sterilized retrieved femoral component with a new polyethylene insert [12], both fixed with antibiotic-loaded cement, to using the PROSTALAC system [13], which consists of a femoral component with a stainless steel articulating surface and a posterior-stabilized all-poly tibial component. The PROSTALAC system exhibits infection control rates, ranging between 88% and 95% in the different series. Disadvantages of this kind of spacer include its relatively poorer infection control potential and its higher cost.

Few studies exist that compare the different kinds of dynamic spacers. In a comparison of cement-on-cement vs. metal-on-polyethylene spacers, Jämsen et al. [14] obtained a higher Knee Society Score (KSS) and a shorter first-stage operative time in patients with a metal-on-polyethylene spacer. Nonetheless, no significant differences were observed following the second surgical stage between the groups as regards knee range of motion.

Insall [3] reported that a period of approximately 6 weeks of intravenous (IV) antibiotic treatment must precede the second surgical stage. However, the length of this treatment period has been called into question as it may increase the cost of hospitalization, exacerbate the toxic effect of antibiotics, and aggravate bacterial resistance to commonly used antibiotics. As periarticular tissue receives little vascular supply in the presence of periprosthetic infection, delivery of systemic IV antibiotics to the joint may prove challenging. The use of spacers fixed with antibiotic-loaded cement tends to be effective as the antibiotic is released at the infection site and bactericidal activity lasts several months. Several studies on the effect of short-course (2–3 weeks) intravenous antibiotic therapy [15, 16] have shown infection resolution rates similar (>90%) to protocols based on long-term IV antibiotic therapy. Few studies compare both regimens (short vs. long) of systemic antibiotic treatment in revision TKA. Hsieh et al. [17] compared a 1-week vs. a 6-week regimen IV treatment in hip revision surgery and found no significant differences regarding infection control in both groups. However, length of hospital stay, cost, and nephrotoxicity were higher in the longer regimen.