Cranioplasty is required to reconstruct cranial defects for patients undergoing decompressive craniectomy (DC) to treat refractory intracranial hypertension due to traumatic brain injury, cerebral infarction, intracranial hemorrhage, and various causes of brain edema, or craniectomy for compound comminuted depressed and/or open/contaminated skull fractures (1–5). Cranioplasty offers not only cerebral protection and cosmetic repair, but also restores the intracranial and atmospheric pressure balance, improving the flow dynamics of cerebrospinal fluid (CSF) (6–13). Although there is no clear consensus on which material is the most appropriate, several materials are used for cranioplasty, including autologous bone, polymethyl methacrylate (PMMA), hydroxyapatite cement, polyetheretherketone (PEEK), and titanium (8, 14–17). Recent advances in three-dimensional (3D) printing technology and medical imaging have enabled the production of custom-made prefabricated patient-specific synthetic implants, offering a precise fit (8, 18–21). Despite these advances and the technical simplicity of the procedure, cranioplasty is associated with a high rate of complications, such as infection, bone resorption, postoperative hemorrhage, seizure, and hydrocephalus, which can increase the morbidity and mortality (6, 8, 9, 14, 22). Surgical site infection (SSI) is a devastating complication ranges from 12.3 to 29.7% (1, 6, 22–24) that can lead to cranioplasty failure, additional surgery, and neurologic deterioration. The identification of possible predictive factors of SSI may help surgeons decide which materials to use, recognize at-risk patients, and guide prophylactic care. The aim of the present study was to identify predictive factors of SSI following cranioplasty by reviewing procedures performed over a 10-year period, including 172 cases, 48 of which involved the use of 3D printed patient-specific implants.

In the present study, we identified significant differences in pre- and postoperative fluid collections between the non-SSI and SSI groups (Table 1), and found that the presence of fluid collections on CT images demonstrated a significant correlation with the event-free survival rate for SSI (Figures 4C,D). Furthermore, univariate analysis revealed that fluid collections were a significant predictor of SSI (Table 2). In multivariate analysis, however, the presence of fluid collections on CT scans after cranioplasty was the only factor found to correlate with the occurrence of SSI (Table 2). Fluid collections post-DC have a variety of sources, including: subgaleal/epidural fluid resulting from CSF leakage due to loose approximation of the dura or closure with artificial dural implants (28), accumulation of exudate from the dissected subgaleal region and muscle (29), and subdural hygroma or external hydrocephalus due to altered CSF hydrodynamics (30). This may be related to dead space that forms when the swelling in the brain has receded after the resolution of the edema (7). Most fluid collections regress spontaneously over time, and previous studies have suggested that cranioplasty may actually improve cerebral blood flow and CSF hydrodynamics, resulting in the resolution of fluid collections due to external hydrocephalus (31–33). However, one previous study suggested that cranioplasty might increase the risk of hydrocephalus when performed <90 days after initial craniectomy (34). Based on the results of the present study, one way to prevent the development of SSIs following cranioplasty is to ensure that there are no fluid collections post-cranioplasty, even if there were fluid collections before the operation. If fluid collections are identified on CT images prior to the cranioplasty, the surgical evacuation of fluid collections during cranioplasty (Figure 2) or concurrent VPS placement in cases of external hydrocephalus may be considered. The patient who underwent sugical evacuation of fluid collection did not develop surgical site infection until his last follow-up (13 months) (Figure 2).

Hydrocephalus has been reported to occur in 10–40% of patients who have undergone DC (10, 35). Permanent methods of CSF diversion, such as VPSs, are required if hydrocephalus persists even after intracerebral pressure management with external ventricular drainage during the acute phase. However, there are varying results regarding when VPS placement should be performed, and there is no clear consensus on the risk of complications, particularly for the development of SSIs with staged or concurrent cranioplasty and VPS placement. Previous studies have reported that concurrent VPS placement and cranioplasty resulted in an increased rate of SSIs compared with staged operations (36, 37). Contrarily, other studies have concluded that the rate of SSIs did not differ significantly between concurrent and staged surgeries for VPS placement and cranioplasty (9, 13, 14). Postoperative hydrocephalus requiring VPS placement was observed in 25% of the cases in the present study. Of the 43 patients who underwent VPS placement, nine underwent placement prior to cranioplasty, and 34 underwent concurrent cranioplasty and VPS placement. We evaluated the event-free survival among patients who had undergone VPS placement prior to cranioplasty, concurrent VPS placement and cranioplasty, and no VPS placement. We identified no significant differences in the event-free survival rates for SSIs among these groups (Figure 4B). From the standpoint that concurrent VPS placement and cranioplasty increases the occurrence of SSIs, the cause of this is thought to be the negative gradient force induced by the over drainage of CSF through the VPS (21, 38). The negative gradient force causes not only a more depressed brain with postoperative dead space, but also pulls at the skin, leading to the exposure of graft materials (39). In the present study only five out of 34 patients who underwent concurrent VPS and cranioplasty had external hydrocephalus. Although the case was too small to analyze the statistical significance separately, four out of five patients improved external hydrocephalus and did not develop SSI during the follow-up period, but one patient did not improve external hydrocephalus and developed SSI. Although VPS placement was not a statistically significant variable that increased or decreased the risk of SSI in the present study, we still suggest that patients with persistent hydrocephalus, especially those with external hydrocephalus, undergo concurrent VPS placement, and cranioplasty. However, we also suggest that the pressure should be slowly adjusted via a programmable valve, to prevent excessive drainage of the CSF.

The present study evaluated the risk of SSI following cranioplasty among different graft materials, including 48 custom-made 3D printed implants. Traditionally, the material of choice for cranioplasty has been the patient's own preserved bone flap, due to a decreased risk of excessive immune response to foreign materials, and its ability to undergo bony regrowth and revascularization (18, 40). However, previous studies have demonstrated that ABF is correlated with the development of SSIs as well as bone flap resorption, resulting in graft failure (41, 42). This may be due to the denaturation of ABF, depending on the storage method (24). There are still no standard guidelines for the sterilization and preservation of ABFs. The largest study concerning predictors of infection after craniolasty by Morton et al. assessed the predictive value of intraoperative bone flap cultures, which are not performed in our center. The authors suggested that intraoperative bone cultures in the absence of infection should be discontinued since the culture results were not a reliable predictor of postcranioplasty infection in their analysis (43). The most commonly used techniques to preserve ABFs are cryopreservation and subcutaneous implantation of the flap into an abdominal pocket (44, 45). Our center has traditionally performed cryopreservation, via direct freezing at −80°C; however, recently, ABFs have not been as widely used for cranioplasty, and 3D printed implants have been used for 27.9% of the total number of cranioplasties performed at our center. In the present study, the most commonly used materials for cranioplasty were ABF (61.6%), 3D printed implants (titanium, n = 48) (27.9%), titanium mesh (7.6%), and PMMA (2.9%), was and we found no significant differences between the graft materials for the event-free survival rates for SSIs following cranioplasty (Figure 4A). Additionally, graft materials were not found to correlate with the occurrence of SSI following cranioplasty (Table 2). This finding was consistent with that of meta-analyses conducted by Yadla et al. (40) and Punchak et al. (46). Various synthetic materials have been used for cranioplasty, including PMMA, ceramics, hydroxyapatite, PEEK, and titanium. Each material has advantages and disadvantages, and as of yet there is no consensus on which one is most ideal for cranioplasty. Titanium provides a strong and non-corrosive material that can be manually shaped during operation; however, thermal conductivity, radiopacity, risk of metal hypersensitivity, and abrasiveness to overlying soft tissues are disadvantages of this material (8, 47). One study suggested that cranioplasty using titanium can greatly increase the implant exposure rate (2). Benefits of PEEK include chemical inertness, robustness, comfort, radiolucency, and thermal non-conductivity (20); however, extrusion has been reported due to the incorporation of bone defects. Synthetic materials are considered to have a higher risk of infection following cranioplasty, but the present study revealed that graft materials were not predictive factors for SSI, although we did demonstrate the superiority of 3D printed implants in providing a precise fit and satisfactory esthetic results. The 3D printed implant used in the early days required multiple screws to secure it to the bone (Figures 3A,B), but as 3D printing technology advanced, not only did the implant fit perfectly, but a screw fixing part and a tenting part to prevent epidural hemorrhage were created and provided (Figures 3C,D). In the present study, we identified the safety of patient-specific 3D printed implants; however, prefabricated implants are still expensive, and long-term complications remain to be investigated. Further investigations with continuous follow-up are necessary to confirm the long-term safety of 3D printed implants in the setting of cranioplasty.

The present study investigated predictive factors that may help identify patients at risk of SSI following cranioplasty and provide guidelines associated with the procedure. The presence of fluid collections on CT scans before and after cranioplasty showed a significant association with the event-free survival rate for SSI. In the univariate analysis, the presence of fluid collections before and after cranioplasty was also a significant predictor for SSI. In the multivariate analysis, however, only the presence of fluid collections on CT scans after cranioplasty was significantly associated with the occurrence of SSIs. Surgery-specific risk factors, including graft materials and VPS placement, did not demonstrate a significant correlation with SSI. The surgical evacuation of fluid collections during cranioplasty or concurrent VPS placement may be considered in cases of external hydrocephalus as a way to reduce fluid collections present prior to cranioplasty. Further prospectively designed studies with long-term follow-term are needed to confirm our conclusions.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Korea University Ansan Hospital Institutional Review Board. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

S-DK: conceptualization, writing—review and editing, and resources. MK: visualization and roles/writing—original draft. H-BL: data curation and formal analysis. S-KH: project administration investigation. D-JL: supervision and validation. All authors contributed to the article and approved the submitted version.

This work was supported by Korea University (Grant Numbers K2101781 and K2118811) and Korea University Ansan Hospital (Grant Number O2105791).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.