Validation of a Novel Clinical Score: The Rostock Functional and Cosmetic Cranioplasty Score

Introduction

Cranioplasty (CP), the surgical repair of a skull defect, is as old as the trephination itself. One of the oldest documented CPs was performed from the Paracas Culture in Peru in 2000 BC using a thin plate of gold to cover a frontal defect. ^1,2 Other materials also had been used, according to the social status of the patient, such as shells and gourds. In the 16th century, Gabriele Fallopius first described the procedure of CP and its indication of use: after a cranial fracture, repair should be performed with a thin gold plate if the underlying dura was damaged. ¹ Later on, xenografts (bone and horn) were used to fill defects, which were derived from dog, ape, goose, rabbit, calf, and eagle. ³

One of the first cranial reconstructions using an autologous bone graft was described by Müller and König in 1890. ^4,5 Synthetic materials were tested largely as bone substitutes for CP ^3,6 to prevent complications associated with the use of autologous grafts (aseptic bone necrosis, infection, reduced strength of the graft) and in cases in which the initial bone graft could not be reused such as in cases of an infiltrating tumor. Besides focusing on the materials used in CP, with a sinking mortality rate, more attention was paid toward the cosmetic results after CP.

CP aims to restore the integrity of the skull not only for safety or cosmetic reasons, but also to improve the neurological condition of patients with the so-called syndrome of the threphined. ^7,8 CP can restore the physiological cortical blood flow and cerebrospinal fluid (CSF) circulation. ⁹ During the last decades, the number of CP procedures has been rising continuously, mainly as a consequence of a rising number of decompressive craniectomies for patients with space-occupying ischemic strokes. ^{10 –15} Decompressive craniectomy may also be indicated in patients with malignant brain edema of various etiologies such as traumatic brain injury (TBI), intracerebral bleeding (ICB), and after aneurysmal subarachnoid hemorrhages (aSAH). ^{16 –21} Other reasons for the initial removal of the skull bone are osseous tumors or meningiomas with bone infiltration. Subsequently, the number of patients in all these circumstances with a good clinical outcome is rising, ¹¹ with the need for CP as a second operative procedure.

Beside cosmetic reasons, the functional aspect after CP is of special interest. Besides scarring and wound healing issues, the main attention is drawn toward the temporal muscle and its role in CP for several reasons. During the initial operation, the muscle is often damaged while detaching it from the temporal bone. ²² The dissection of the muscle initially and again during the CP procedure often leads to scarring, muscle retraction, and atrophy, partly caused by impaired vascular supply and denervation. The muscle loss and functional impairment can be related to masticatory difficulties because of an impaired ability of mouth opening and jaw movement, and cosmetic impairment per temporal hollowing. ^{23 –25}

One major drawback of nearly all recently published studies had been the fact that there is no existing clinical scoring system to rate the results after CP and make them comparable. ^{22,26 –28} In particular, there is no score to our knowledge that combines functional and aesthetic aspects after CP, both being equally of utmost significance. Therefore, the aim of our study was to design and validate a clinical score that is able to rate both functional and cosmetic results after CP and that makes clinical results comparable after different techniques of CP. In addition, besides being simple and accurate, obtaining the score should be independent of the possible interaction with the patient, making it eligible also for patients who lost their ability of proper communication because of aphasia after a cerebral infarction of the dominant hemisphere.

Methods

In this observational single-center study, we tested and validated our Rostock Functional and Cosmetic Cranioplasty (RFCC)-Score in terms of objectivity and reliability. Institutional Review Board approval was obtained before the start of this study (Reg.nr A2016-0018). Written consent was obtained from all patients.

Results

A total of 27 patients met the inclusion criteria, with a median age of 52 ± 11.51 years and a slightly female dominance (1.25:1) (Table 1). The prevalent reason for the initial operation was a cerebral infarction with consecutive space-occupying edema (44.4%), followed by aSAH (18.5%) and TBI (18.5%). Two patients were included after an initial tumor removal (meningioma and an osteolytic bone cyst) with an extended temporal trepanation (7.4%); three patients had ICB and the need of a decompressive craniectomy (11.1%). A total of 38 CP operations were performed on the 27 patients included with a mean number of 1.52 implants per patient, meaning that 37% of the enrolled patients had a complication with the need of a secondary CP.

The main reason for a re-operation was superficial or deep infection of the surgical site (64.3%). An aseptic bone necrosis was recorded in two cases (14.3%), which led to discarding the bone and implanting a Computer Aided Design/Computer-Aided Manufacturing (CAD/CAM) CP, being in accordance with other studies. ³⁷ In concordance with our institutional policy, primary implant for CP should always be the autologous bone (68.4%). Mean follow-up period after the last CP was 56.3 ± 46.20 months (range 9–142 months).

The clinical examination of the patients showed a slight impairment of the facial function evaluated by the House–Brackmann facial nerve grading system (Table 2), with a median grading of 2 and five patients with a moderate dysfunction (grade 3). Higher grades of dysfunction were not observed in our cohort. The mouth opening was recorded in 18.5% of all cases as impaired; median interincisal distance was 47 mm with a cut-off value of 38 mm as a sign of impaired temporalis muscle function. In contrast to these findings, sideways jaw movement was found in 29.6% cases within the physiological range of ≥9 mm.

The patient-based JFLS-8 showed that 62.9% of all patients enrolled had no (score 0) or a little impairment of their orofacial function (score 1); on the other hand, 25.9% had severe limitations (score 3). For a detailed analysis of our patient cohort, all participants completed a screening for signs of depression, maybe affecting the other patient-based scores (CES-D); 51.9% showed no signs of a depression, 25.9% were at risk of depression, and 22.2% showed signs of a severe depression. EQ-5L-5D index values showed a mean of 0.81 ± 0.10, and the VAS a mean of 60 ± 29.3. Compared with the general German population value sets, the VAS of our cohort is clearly beyond the “reference value” of a mean VAS of 85.1. ³⁸

Looking at the RFCC-Score results obtained by the neurosurgeon, in 15 patients the post-operative result was rated good (55.6%); 12 patients showed an acceptable result (44.4%). No poor post-operative result was rated in our cohort by the neurosurgeon. In contrast, only six patients rated their post-operative result as acceptable (22.2%), and 21 patients as a good result (77.8%); also, no poor result was rated by the patients.

For validation of the RFCC-Score, we analyzed the different clinical measurements and scores and their impact on each other (Table 3). Interestingly, the number of CP operations performed for each patient influenced significantly the individual mouth opening (p = 0.001), having no impact on the other scores. The House-Brackmann grading was correlated positively with the RFCC-Score obtained from the surgeon, but not from the patient (p = 0.004 [correlation coefficient 0.54] and p = 0.458 [correlation coefficient 0.15], respectively). In addition, the facial nerve function significantly influenced both quality of life scores EQ-5D-5L and VAS (p = 0.017 and p = 0.035, respectively). The mouth opening had no influence on these two scores (p = 0.825 and p = 0.779, respectively); the jaw sideways movement significantly influenced only the VAS (p = 0.019) with no impact on the EQ-5D-5L (p = 0.106).

The JFLS-8 was not correlated with the RFCC-Score obtained by the surgeon or the patient (p = 0.625 and p = 0.286, respectively—data not shown), Neither did the measured mouth opening nor the jaw movement (p = 0.739 and p = 0.188, respectively) influence our scoring system.

The RFCC-Scores, obtained from the patient and the neurosurgeon, showed a high correlation (p = 0.018) (Table 4). Looking at the agreement of the two scores, Cohen ƙ analysis showed a poor agreement with a coefficient of 0.239 (quadratic weighted 0.373). These observations were supported by the Bland-Altmann plot (Fig. 1), which showed identical scorings in 37% of all obtained RFCC-Scorings. There were 11 patients (40.7%) who rated the post-operative results better than the surgeon (negative difference); six patients (22.2%) scored the results worse (positive difference).

OPEN IN VIEWER

FIG. 1.

Plot of the differences between the Rostock Functional and Cosmetic Cranioplasty (RFCC)-Scores obtained from the patient and the neurosurgeon per patient. No observed differences were displayed as zero on the y axis; 95% limits of agreement were displayed as reference lines (2.76 and −3.80, respectively).

Discussion

The number of decompressive craniectomies is rising constantly and thereby the number of CP procedures. In addition, the temporal and pterional approach are the most commonly used in the field of neurosurgery. The common feature of all these procedures is the dissection or mobilization of the temporal muscle, which later on needs to be refixated. Associated problems are temporomandibular disorders, temporal hollowing with aesthetic impairment, and post-operative pain. ^{23 –25} These issues are yet underrepresented in the existing literature regarding clinical outcome reports, beside their rising importance for the patient and quality of life after CP.

In the present study, we designed and validated a novel clinical score, reflecting the functional and aesthetic results after CP—the RFCC-Score. The elected cohort consisted of typical patients who underwent a decompressive craniectomy, followed by CP (81.5%). Extended temporal trephination and CP represented 18.5% of cases (patients after standard temporal or pterional approaches were not included to test our score) within a patient cohort, where the temporal muscle had been totally mobilized from the temporal bone, carrying the biggest risk of a functional impairment and temporal hollowing developing. Mean follow-up period was 56.3 month, ensuring a long-term cosmetic and functional result. Figure 2 is showing a typical patient enrolled in our study at a follow-up visit 13 months after CP.

OPEN IN VIEWER

FIG. 2.

Computed tomography (CT) scans and clinical picture of a typical patient enrolled in our study. (A) Post-operative CT scan after cranioplasty (CP) with the autologous bone. An initial decompressive craniectomy was necessitated after a space-occupying cerebral infarction of the middle cerebral artery. (B, C) The autologous implant had to be removed after a deep wound infection. After completion of wound healing and antibiotic therapy, a second CP was performed using an Computer Aided Design/Computer-Aided Manufacturing (CAD/CAM) implant (OssDsign,^® Upsala, Sweden). (D) The clinical picture is demonstrating a typical ipsilateral temporal hollowing after right-sided second CP (asterisk). A fissuring is clearly visible between the implant and the rim of the skull defect (arrows).

The jaw-specific function measurements showed an impaired sideways movement in 30% of cases, but 63% of all enrolled patients complained of no or just little impairment of their orofacial function (JFLS-8). The mouth opening, on the other hand, was impaired in 19% of our cohort. Correlation analysis showed a stronger interconnection of the jaw sideways movement toward the JFLS-8 than toward the mouth opening (p = 0.188 and p = 0.739, respectively), besides both not being statistically significant. These findings were supported by the results from our RFCC-Score, being more correlated with the jaw sideways movement than with the mouth opening (RFCC surgeon p = 0.283 and p = 0.564, respectively). One possible explanation is the selected cutoff value for a physiological mouth opening of 38 mm, diverging results from other studies and a great interpersonal variety. ³⁰ Interestingly, the mouth opening was strongly influencing the disposition for depression (p = 0.058).

The House-Brackmann score was significantly correlated with the RFCC-Score obtained from the surgeon (p = 0.004), but not the one obtained by the patient (p = 0.458). The facial nerve function, on the other hand, strongly correlated with the quality of life of the patients, assessed by the EQ-5D-5L and VAS, but did not influence the other measurements or scores. Therefore, the surgeon RFCC-Score reflects more the orofacial aspect, which is related to the quality of life of the patients after CP, rather than the single muscle function. This thesis is supported by a missing correlation between the RFCC-Scores obtained from both the surgeon and the patient and the JFLS-8. An individual's perception of pain after CP, for example, is often underrepresented and underestimated in post-operative assessments. ³⁹ Rocha-Filho and colleagues ²⁵ reported in 2007 that 47.9% of patients after pterional craniotomy for aneurysm complained of headache during dental evaluation, 28.2% reported pain during normal jaw movement. ^29,30 These functional aspects will always influence the individual experience and rating of his or her own clinical state. Therefore, it is important to involve these aspects into a scoring system without losing the advantage of an objective observer.

The difference between the RFCC-Scores obtained by the patient and the surgeon was independent of a disposition toward depression and the quality of life assessed by the patient, revealing that our score is objective and independent of the perception of patients potentially influenced by a depression. Individuals with a higher RFCC-Score obtained by themselves and a corresponding positive difference toward the surgeons scoring had 19.8 points at the CES-D, compared with 14.3 points mean of the patients without a difference or a negative difference towards the surgeon scorings. These results are showing clearly that patients with a disposition for depression tend to rate their functional and aesthetic results after CP inferior compared with the more objective results gained from the surgeon.

There was little agreement between the two RFCC-Scores, revealed by the plot and the Cohen ƙ coefficient. These findings are in concordance with the correlation analysis and what we expected. The diverging results and the influence of the individual's emotional condition and percipience on the patient scorings makes our RFCC-Score superior to patient-based scoring systems. One additional and very important benefit is the independence of our score from the patient's ability to speak. Just a minimum of interaction is required, such as the ability to open the mouth and to move the jaw.

Today, there is still a lack of good and satisfactory clinical follow-up data after CP procedures, not only concerning the general outcome, but also looking at complications. Putting us into a position in which we can compare different surgical techniques and used materials for CP is an importance aim of two registers started in the United Kingdom and Germany. ^40,41 From our point of view, it is inevitable, especially within the setting of a prospective registry, to gain valid but compressed data that can reflect the functional state and quality of life of the patient.

In contrast to other assessments, our score is able to represent these individual aspects in a bigger dimension than others, being more influenced by the functional and cosmetic aspects in total, rather than the sheer muscle function. One drawback of our study is the limited number of patients being enrolled. A larger study is therefore already being conducted, testing our RFCC-Score in general and especially for interobserver variability. Another restriction of our scoring system has to be considered: the obtained score result is dynamic over time. During the follow-up period, the result will change because of an ongoing muscle atrophy or an aseptic bone necrosis with an increased gapping between the bone fragment and the skull. Therefore, the score result as a tool for the assessment of the post-operative functional and aesthetic outcome after CP should always be rated considering its dependence on the moment of examination.