With the increase of total knee arthroplasty surgery year by year, the incidence of prosthetic joint infection is also increasing. When a joint prosthesis becomes infected, a poorly controlled infection can lead not only to failure of the joint arthroplasty, but in severe cases even to amputation [40, 41]. Two stage revision surgery is the most commonly used method in clinical practice and it can offers the best results in terms of treatment. In the two stage revision surgery, the commonly used spacers include dynamic spacers and static spacers and each has its own characteristics. The structure of the dynamic spacer is match to the anatomy of the knee joint, thus it can reducing adhesions and scarring of the soft tissues surrounding the knee joint. The dynamic spacer can also improve the recovery of knee function after revision surgery and reduce the incidence of some complications. However, there are also some reports suggested that dynamic spacers are less effective in controlling prosthetic joint infection. In contrast, static spacers are less prone to dislocation during fixation and it can also provide good joint stability. In addition, the static spacer can provide a high concentration of antibiotics for better infection control while maintaining limb length. Of course, there is still controversy in clinical and related research regarding the difference between dynamic and static spacers in controlling prosthetic infection and improving the prognosis of revision surgery. In the current meta-analysis, we collected relevant clinically controlled studies and performed a meta-analysis of these. Our aim is to further investigate the differences between dynamic spacers and static spacers in terms of therapeutic effect and impact on knee function.
A total of 14 articles were included in this meta-analysis, and we grouped one primary and four secondary outcome indicators according to the indicators in each article. Finally, we did the corresponding meta-analysis for each of the five outcome indicators. The rate of infection control in prosthetic joint infection is the most important indicator of the effectiveness of revision surgery treatment. Based on the final results we can find that the treatment effect between dynamic spacers and static spacers is the same in terms of the main outcome indicator of prosthetic infection control, with no significant difference between the two groups (p = 0.179 > 0.05). This suggests us that dynamic spacers are no less effective than static spacers for prosthetic infection control and it can also achieve the same clinical outcomes. This result is consistent with the results of many current clinical studies and further validates the efficacy of joint spacers [42–44]. According to relevant literature reports, the concentration of antibiotics and the duration of anti-infection therapy are important factors in the outcome of two-stage revision surgery for prosthetic infections. The use of adequate antibiotic concentrations and duration of anti-infection therapy not only results in more satisfactory infection control, but also better reduces the recurrence of prosthetic infections [45, 46]. In addition, revision surgery at an early stage of prosthetic infection can also provide better control of prosthetic infection, and early anti-infection treatment is one of the keys to treatment and reduces the risk of surgery [47, 48]. Therefore, not only does the correct use of spacers will influence the outcome of treatment, but timely and adequate anti-infective treatment is also crucial to the success of treatment.
Of the four secondary outcome indicators, KSSs scores, KSS functional scores and ROM were used to evaluate the impact on knee function after patients received spacer implantion, all as an indicator of post-operative recovery and efficacy. In the results of our meta-analysis, the results for all three indicators, KSSs scores, KSS functional scores and ROM, were statistically significant (p < 0.05). This indicates a significant difference between the results of the dynamic spacer group and the static spacer group in these three indicators. In terms of improved prognosis for knee function, patients with dynamic spacers had significantly better functional scores and knee mobility than those with static spacers. These results once again verified the superiority of the dynamic spacers over the static spacers, in comparison, the dynamic spacer could better improve the patient's motion function and range of motion [49–51]. Bone loss is a common complication after spacer implantation and the patient's bone loss is a judgment indicator to evaluate the impact of the spacer on the patient's bone health. Therefore, bone loss was also one of the reference indicators in our current meta-analysis. Based on the results of the meta-analysis, it is known that the patients with dynamic spacers experienced significantly less bone loss than those with static spacers (p < 0.05). There was a significant difference in the results between the two groups. This is also in line with some current findings that patients using static spacers are more likely to experience bone loss [52]. Combining these results, we can see that the dynamic spacer can achieve similar results to the static spacer in terms of infection control of the prosthesis. The performance of dynamic spacers is better than that of static spacers in improving the prognosis of knee function and preventing bone loss. For patients requiring prosthetic revision surgery, the use of dynamic spacers may provide a better prognosis and recovery of joint function.
Of course, there are some limitations to our current meta-analysis. Firstly, although we included a total of 14 relevant literatures, the total number of patients studied was only 799, which may not be a large enough sample size. Perhaps we need more clinical studies with larger samples to further confirm our results. Secondly, the articles we included were not clinical randomised controlled studies, but rather retrospective clinical studies and prospective trials. The use of blinding was also not reported for the assessment of outcomes and scores related to knee function, which lacks a certain degree of concealment. In addition, there is a lack of uniformity in the assessment and the associated results may be influenced by subjective factors. Thirdly, the possible influence of relevant factors on the treatment outcomes (e.g. gender, age, height, weight, etc.) was not adequately considered in determining patient groupings.