Research design and study sample

The data of patients who underwent a BSSO according to the Hunsuck modification (occasionally combined with a Le Fort I osteotomy and/or genioplasty) were examined in this retrospective observational cohort study. A total of 80 sagittal splits in 40 patients treated between January 2013 and February 2017 were included. The study group consisted of 14 men and 6 women (mean age 26.1 years; standard deviation (SD) 9.9 years) who had been treated with BO and TC at the University Medical Center Groningen, the Netherlands. Eighteen of the study patients were treated for a class II malocclusion and two for a class III malocclusion, with distinct gap formation between the osteotomy segments. The control group, without any grafting, consisted of 7 men and 13 women (mean age 30.2 years; SD 12.3 years) treated at the Radboud University Medical Center, the Netherlands. All 20 of the control patients were treated for a class II malocclusion. Patients were excluded if one of the required CBCT scans was not available (the first from within 1 month after surgery (T1) and the second from 10–14 months after the first CBCT scan (T2)), and if the mandibular border could not be examined closely (e.g., due to a bad quality CBCT scan or scattering of the low placed miniplates). Once selected, the patients in both groups were divided into subgroups based on the amount of horizontal displacement: group I, 0.0–6.9 mm; group II, 7.0–8.9 mm; group III, 9.0–15.0 mm.

Surgical protocol

The BSSO procedure was executed according to the Hunsuck modification by a maxillofacial surgeon, frequently assisted by a resident. Both sides of the mandible were permanently fixated with miniplates (study group: 2.0-mm, KLS Martin, Tuttlingen, Germany; control group: Champy 2.0-mm plates, same manufacturer) and care was taken to avoid upward rotation of the proximal segments.

For the study group patients, BO xenograft (granules of 0.25–1 mm) was mixed evenly with the TC fibrin glue using a spatula and subsequently formed manually into two well-mixed blocks ( Fig. 1 ). The volume of BO and TC required to fill the cortical defect in the inferior border of the mandible was evaluated preoperatively, but the volume used was preferably more than the estimated gap size ( Fig. 2 ). Although the quantity depended on the size of the gap, the maximum for each side was 0.25 ml BO and 0.5 ml TC. Each BO and TC block was carefully placed into the osteotomy gap in the inferior mandibular border using a spatula, with the aim of filling this gap and restoring the contour. The mixture does not adhere well to the bone surfaces in the osteotomy gap and therefore a tight fit is important to keep the mixture in place. The mandibular inferior border was then palpated through the skin to check for any irregularity. If there was, the BO and TC block was modified and checked again by extraoral palpation.

Mixed xenograft (Bio-Oss) and fibrin glue (Tissucol) block.

Intraoperative photo of the applied xenograft–fibrin glue block (Bio-Oss and Tissucol) and the inferior mandibular border.

The transoral placement of graft material in the inferior border is done with limited direct vision. Verifying correct placement is therefore difficult, but this is important because, if the placement and the osteotomy planes are incorrect, the osteoconductive properties of the xenograft will be affected.

In all cases, wound closure was performed with a standard running resorbable suture.

Radiographic and three-dimensional analyses

Postoperative CBCT scans were obtained at T1 and T2 as part of the routine follow-up. A Planmeca Pro-Max CBCT system (Planmeca Oy, Helsinki, Finland) was used for the study group patients, with settings of 120 kVp voltage, 5 mA dose, and 5.8 s exposure time. An i-CAT CBCT scanner (Imaging Sciences International, Hatfield, PA, USA) was used for the control group patients, with settings of 120 kVp voltage, 3–8 mA dose, and 2 × 20 s exposure time. The same head positioning protocol was applied for all CBCT scans at T1 and T2, giving reproducible data.

The raw image files were segmented using ProPlan CMF v. 3.0 software (Materialise, Leuven, Belgium). The horizontal displacement between the osteotomy segments at the inferior mandibular border was measured for both groups on each three-dimensional (3D) model ( Fig. 3 ). The 3D models from T1 and T2 were then superimposed using a closest point algorithm in Geomagic studio software (3D Systems, Morrisville, NC, USA). The T2 superimposed models were imported into 3-Matic v. 11.0 (Materialise, Leuven, Belgium) to measure the height of the defect and the volume percentage between the osteotomy segments.

Horizontal displacement measured between the inferior mandibular borders of the proximal and distal segments.

The height of the defect was measured by first drawing a line that crossed the defect at the inferior border ( Fig. 4 ). A second line was then drawn down from the highest indentation point of the osteotomy gap, perpendicular to the first line that crossed the defect; this line was used to measure the height of the defect in millimetres.

Measuring the height of the defect at T2. The red line illustrates a reproducible line that crosses the defect; the perpendicular black line shows the height of the defect (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

The volume percentage of the defect (i.e., volume of the defect) was obtained by measuring the volume of the osteotomy gap at T2 and comparing this with an individualized cubic volume, representing the optimal contour volume ( Fig. 5 ). The volume at T2 was created by measuring the volume of the mandible between the osteotomy lines with an even height of 5 mm from the line that crosses the defect at the inferior border.

Cut-out volume. The grey shaded area represents the volume of the osteotomy gap and the pink shaded area represents the optimal contour volume (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

The optimal contour volume had the same borders as the volume at T2, but the width was measured from a line between the mesial and distal top width of the volume at T2.

Both volumes were calculated before dividing the volume at T2 by the optimal contour volume to obtain a percentage, which represents the degree of bone regeneration.

In addition, two maxillofacial surgeons (JJ, RS) scored the hard tissue 3D models of both groups by visual evaluation as either a noticeable defect (i.e., distinct contour change in the continuity of the inferior border) or no defect. All of the measurement process steps were executed by one researcher (HH) and were calibrated by the two surgeons (JJ, RS).

Statistical analysis

All border defects were analysed individually because of the large discrepancy between the right and left mandibular borders within the same patient. Differences in characteristics between the two groups in terms of age, horizontal displacement, and period of follow-up were compared using the Mann–Whitney test. Differences in sex distribution between the groups were compared using the χ ² test.

An independent samples t -test was used to compare the height and volume of the defect. The hard tissue models that were scored by the two surgeons were compared with χ ² tests ( P < 0.05). The results were processed using IBM SPSS Statistics version 25.0 (IBM Corp., Armonk, NY, USA).

The effect of the intervention on defect heights and volumes according to the level of horizontal displacement was assessed in the six subgroups using Cohen’s d effect size. This was done because an effect size calculation, which indicates the effect of the intervention, is possible with small groups.