Outcome in patient-specific PEEK cranioplasty: A two-center cohort study of 40 implants

Abstract

Objective

The best material choice for cranioplasty following craniectomy remains a subject to discussion. Complication rates after cranioplasty tend to be high. Computer-assisted 3-dimensional modelling of polyetheretherketone (PEEK) was recently introduced for cranial reconstruction. The aim of this study was to evaluate patient- and surgery-related characteristics and risk factors that predispose patients to cranioplasty complications.

Material and methods

This retrospective study included a total of 40 cranial PEEK implants in 38 patients, performed at two reference centers in the Netherlands from 2011 to 2014. Complications were registered and patient- and surgery-related data were carefully analysed.

Results

The overall complication rate of PEEK cranioplasty was 28%. Complications included infection (13 %), postoperative haematoma (10 %), cerebrospinal fluid leak (2.5 %) and wound-related problems (2.5 %). All postoperative infections required removal of the implant. Nonetheless removed implants could be successfully re-used after re-sterilization.

Conclusion

Although overall complication rates after PEEK cranioplasty remain high, outcomes are satisfactory, as our results compare favourably to recent literature reports on cranial vault reconstruction.

Introduction

Cranioplasty aims to repair a defect in the cranium and is one of the oldest neurosurgical procedures. Archeological evidence dates back to 3000 BC and suggests that the Incas performed skull reconstruction using gold plates (Rifkinson-Mann, 1988). In the 16th century Fallopius also recommended repair with gold plates (Sanan and Haines, 1997) and one century later, in 1668, the Dutch surgeon van Meekeren reported on the repair of a cranial defect in a Russian soldier with bone derived from a canine skull (Sanan and Haines, 1997).

Cranioplasty provides protection to the underlying brain and is performed for both functional and aesthetic reasons. It aspires to neurologic recovery, as described with reconstruction for the sinking scalp flap or syndrome of the trephined (Dujovny et al., 1997a, Dujovny et al., 1999, Goldstein et al., 2013, Kuo et al., 2004, Stula, 1982, Winkler et al., 2000). Disadvantages to delayed cranioplasty involve a temporarily unprotected brain as well as an aesthetic deformity (Kshettry et al., 2012). Timing seems to be important in the neurological outcome of patients but also in avoiding complications (Yadla et al., 2011). Cranioplasty is most commonly performed after previous craniectomy for traumatic brain injury, stroke, after intracranial tumour surgery and intracranial infections (Diedler et al., 2009, Honeybul and Ho, 2013, Kakar et al., 2009, Mak et al., 2013, Morley et al., 2002, Perez-Bovet et al., 2012, Schirmer et al., 2007, Schwab et al., 1998, Vahedi et al., 2007, Whitfield et al., 2001, Winter et al., 2005).

Material choice for cranioplasty is still controversial, which brings complexity to this seemingly straightforward procedure (Klinger et al., 2014, Sahoo et al., 2010, Walcott et al., 2013, Yadla et al., 2011). Harvest sites for autologous bone grafts include iliac crest, rib, sternum, scapula and the skull (Shah et al., 2014). At present, autologous bone flap replacement using the previously removed bone flap is the most common practice. Autologous bone does not exert immune rejection and is effective as a substrate for bone ingrowth and revascularization. Besides this autologous bone reconstruction has relatively low costs (Grant et al., 2004). However, there is a risk of infection, resorption and in this case its strength gradually reduces. This has led to a search for synthetic materials (Cheng et al., 2014, Goldstein et al., 2013, Matsuno et al., 2006, Schoekler and Trummer, 2014, Walcott et al., 2013, Yadla et al., 2011). At present, there are primarily 3 classes of allografts: metal, ceramic and polymer (Bonda et al., 2015). Titanium is the only metal still in use, it is a biocompatible material with a low infection rate (Lethaus et al., 2012). Nonetheless titanium has certain disadvantages: the material is expensive and leads to artifacts on imaging (Hill et al., 2012, Matsuno et al., 2006). Furthermore, it is a very strong material that shows no deflection in cases of traumatic stress and consequently it has no protective energy-absorbing properties (Lethaus et al., 2012). Hydroxyapatite is a ceramic, which is known to be a good scaffolding material for bony ingrowth (Bonda et al., 2015). Unfortunately, it is rather limited for use in larger defects because of its brittleness and low tensile strength (Ducic, 2002, Dujovny et al., 1997b). Polymethyl methacrylate (PMMA), a polymer, has been widely used because of its low cost, radiolucency and lack of thermoconduction. Nonetheless it is associated with complications such as infection, fragmentation and a lack of incorporation (Blum et al., 1997, Matsuno et al., 2006).

Computer-assisted design (CAD) and computer-assisted manufacturing (CAM) has been used to make titanium, hydroxyapatite and PMMA implants. Prefabrication of a patient-specific implant (PSI) reduces operation time and produces superb cosmetic results (Bonda et al., 2015). Recently, computer-assisted 3-dimensional modelling of polyetheretherketone (PEEK), another polymer, has been successfully introduced for cranial reconstruction (Hanasono et al., 2009, Kurtz and Devine, 2007). It is a strong and highly thermoplastic material. It resembles titanium in its perfect intraoperative fitting and its resistance to aggressive sterilization procedures (heat and ionizing radiation). The elasticity and energy-absorbing properties of PEEK match closer to bone than the mechanical properties of titanium. And in contrast to titanium, PEEK is a radiolucent non-magnetic material, facilitating postoperative imaging (Kurtz and Devine, 2007, Lethaus et al., 2011, Lethaus et al., 2012, Lethaus et al., 2014, Shah et al., 2014). PEEK has a few disadvantages, but it has no bioactive potential and the costs related to the manufacturing of a PEEK PSI are high (Lethaus et al., 2014).

The aim of this study is to evaluate patient- and surgery-related characteristics and risk factors that predispose patients to an increased risk of complications after PEEK cranioplasty.