In this prospective clinical study, iodine-supported implants were found to be useful for the prevention and treatment of postoperative infections in compromised hosts.

Several factors may contribute to implant failure: soft tissue defects for tumor cases and poor soft tissue condition in revision cases leading to instability [3, 4] bone quality and cement technique that may contribute to loosening and prosthetic or peri-prosthetic fracture [3]; chemotherapy and radiotherapy, large bone defects and prostheses, increasing surgery time (> 2.5 h), increased patient body mass index, lower preoperative hemoglobin or albumin levels, and postoperative hematoma contribute to a higher risk of infection [3, 5].

In this study, four patients (9%) had hip dislocation as type 1 failures. In all the patients, iodine-coated implants were used for therapeutic purposes. Garbuz et al. [6] reported that the dislocation rate in revision THA using a 32-mm head was 8.7%, and hip dislocation was independent of the iodine coating.

Aseptic loosening (type 2) accounted for 3% of cases in this study. Tumors accounted for the majority of cases. The incidence of aseptic loosening failure of tumor prostheses in the literature is 4.7–10% [3]. In comparison, the aseptic loosening rate of iodine-coated implants (3.0%) was satisfactory.

In this study, the structural failure (type 3) rate was 9.0%, including an implant fracture rate of 6.1%. Spinal instrument fractures were most common (4.4%). The mechanical failure of spinal instruments is reported to be 2.0–10% [7, 8]. On the other hand, the mechanical failure of a magaprosthesis is reported to be 11.7% [9]. In the study by Qu et al. [10], the rate of fixation failure for tumor-bearing bone was 7.4%. The implant fracture rate in the current study was 1.5%, which is better than that reported in these studies. Peri-implant fractures are also associated with structural failures. Previous reports have reported periprosthetic fractures after THA and TKA in 0–18% and 0.3–2.5% of patients, respectively [11, 12]. In this study, there was only one case of periprosthetic fracture, which was very rare. In contrast, most peri-implant fractures were recycled bone fractures in patients with tumors. In a previous report, Paholpak et al. [13] reported that fractures of frozen bone using liquid nitrogen occurred in 8% of patients. In comparison, 2.1% of patients in our study had such fractures which indicates a very low rate. We believe that the small number of structural failures is because iodine-coated implants have high bone affinity and good fixation.

Infection (type 4) is the worst and most frequent cause of failure [9, 14,15,16]. In this study, the final infection rate using iodine-coated implants was 6.8% (3.7% for prevention and 15.3% for treatment) in immunocompromised hosts and in postoperatively infected cases. The incidence of SSI using spinal instruments is reported to be 4.4–21.9% in the literature [15, 16]. Pala et al. [9] reported an infection rate of 6.9% when silver-coated titanium implants were used to prevent infection, which was similar to the infection rate when iodine coating was used. On the other hand, according to Fiore et al. [17], when a silver-coated maegaprosthesis was used for infection prevention and treatment, the overall infection rate was 17.6% (primary: 9.2%, revision: 13.7%). The results of this study indicate that iodine-coated implants are superior to non-coated implants. Because many of the infected cases in this study were refractory bacterial infections (osteomyelitis, pyogenic spondylitis, infected THA, and infected TKA), re-infections were likely to occur. According to previous reports, staphylococci are the most common cause of postoperative infections, especially MRSA and MRSE [18]. In this study, MRSA was also detected as the most common pathogen. In revision surgery, two-stage revision surgery is usually recommended [5, 19]; however, in this study, there was no significant difference in the re-infection rate between one-stage and two-stage replacement (14.6% vs. 16.0%, respectively). These rates were lower than those reported by Nucci et al. [19] (45.5% vs. 27.3%, respectively). Pala et al. [9] reported a mean time to the onset of infection of 25 months. In the current study, the mean time to the onset of infection was 9.1 months, which tended to be short. Because the patients in this study were immunocompromised hosts and patients with postoperative infections, this may have resulted in a reduction in the time of infection onset. All 45 infected cases were eventually cured, but limb amputation was performed in three cases (6.6%, 3/45). In previous reports, post-infectious amputations ranged from 23.5 to 87% [20]. In comparison, treatment with iodine-coated implants greatly reduced amputation. Furthermore, a one-stage replacement surgery for postoperative spinal infection did not result in infection. This result has the potential to alter conventional wisdom.

Tumor progression (type 5) failure occurred in 4.3–4.8% of cases in the literature [3, 14]. In the current study, type 5 failure occurred in 26/297 (8.7%) patients, which tended to be slightly higher than previously reported. This is because many metastatic tumors were treated.

The present study had several limitations. First, there was heterogeneity in the primary diseases. Second, various types of implants were used, such as plates, prostheses, and spinal instruments. Third, this study was a comparison with historical controls and was not a randomized controlled trial. However, the positive effect of iodine-supported implants in preventing and treating postoperative infection is encouraging and warrants further study.