Abstract
The restoration of mandibular bone defects with non-vascularized bone grafts depends on the diagnosis, anatomical site, extent of the defect, and the patient’s age, as well as the surgeon’s experience. The aim of this study was to perform a systematic literature review on mandibular reconstruction for segmental mandibular bone defects using non-vascularized bone grafts to answer the following question: Is there scientific evidence to support the use of this technique? The initial literature search in PubMed, Scopus, and Cochrane databases identified 862 articles. Of these, 25 were included in the final review. These articles encompassed 926 procedures with non-vascularized bone grafts; 76.1% were from the iliac crest. Benign tumours were the major cause of these defects (56.8%), and 44.7% of defects were located in the lateral mandibular area. Although this technique showed a high occurrence of complications (290 in 873 patients, some with more than one complication), these did not account for treatment failure. The restoration of bone defects due to malignant tumours treated with radiation therapy had lower success rates, and these appear to be a contraindication for the technique. Although standardized randomized controlled clinical studies are needed to obtain better clinical evidence for treatment choices in general, the use of non-vascularized bone grafts for mandibular reconstruction showed an 87.6% success rate in this review.
Mandibular bone defects due to resections for a pathology or trauma result in a loss of bone continuity and create a ‘socially mutilated’ patient due to the significant loss of function and aesthetics. The treatment of these defects should preferably be performed by means of immediate bone reconstruction, with the aim of maintaining facial contours, aesthetics, and function, with correct maxillomandibular positioning. The restoration of mandibular segmental defects can be done with either non-vascularized or vascularized grafts.
Although vascularized grafts are the best reconstruction technique because of the graft blood supply and soft tissue support, they do have limitations and do not assure the best procedure results. High costs, a longer surgical time, the need for two specialized surgical teams, morbidity, complications, and a longer hospital stay are some of the disadvantages presented by this technique.
Reconstruction using non-vascularized grafts is preferable in cases of minor defects, preserved mandibular continuity, or benign tumours. Some available donor sites are the iliac crest, ribs, tibia, fibula, calvaria, and sternum. The iliac crest is most often used due to the ease of access and availability of cortical and medullary bone; also, when used properly, such grafts can facilitate dental implant placement and prosthetic rehabilitation.
However, there is no consensus on the success and complication rates with the use of non-vascularized graft reconstruction: the success rate varies between 46% and 100% and the complication rate between 20% and 35%. This wide variety of results can be explained by the various existing variables in the planning and surgery itself that can lead to procedure failure. Currently, mandibular reconstruction with non-vascularized grafts relies on the surgeon’s experience and familiarity with the technique.
The aim of this study was to perform a systematic review of the literature on mandibular reconstruction for segmental bone defects using non-vascularized bone grafts to answer the following question: Is there scientific evidence to support the use of this technique?
Materials and methods
Search strategy and selection criteria
An initial literature search in PubMed (MEDLINE), Scopus, and Cochrane Library databases was performed by scanning the subject of interest through four lines of search elements: (1) ‘mandibular defect’ AND ‘non-vascularized bone graft’, (2) ‘mandibular defect’ AND ‘nonvascularized bone graft’, (3) ‘mandibular defect’ AND ‘non vascularized bone graft’, (4) ‘mandibular defect’ AND ‘free bone graft’.
For the initial selection, three independent reviewers (LBM, PHAC, and OLCJr) analysed the title and/or abstract against the following inclusion criteria: studies on humans, specific studies using non-vascularized grafts for segmental mandibular defect restoration, studies reported in English, and a case series, retrospective study, or prospective study design; there was no time restriction with regard to publication date.
After the initial selection, the three examiners read the full-text articles using the same selection criteria to determine the final inclusion or exclusion of studies. Disagreements between reviewers were settled by further discussion. All selection criteria were established prior to the start of the study.
Data collection
Data from selected studies were collected by the examiners, and the following variables were included: type of study, donor site, mandibular involvement according to the HCL classification proposed by Jewer et al., size and cause of the defect, postoperative complications, and the success rate. Data were analysed by descriptive statistics.
Quality evaluation
The evaluation of methodological quality was performed using the PRISMA statement criteria to verify the strength of the scientific evidence available in the current literature for clinical decision-making. The following criteria were used to classify the potential risk of bias of each study, as applied in previous reviews : random selection in the population (sample), definition of inclusion/exclusion criteria, report of losses to follow-up (monitoring, follow-up), validated measurements obtained, and statistical analysis. Studies meeting all of these criteria were classified as having a low risk of bias; those that did not meet one of the criteria were classified as having a moderate risk of bias and those that did not meet two or more criteria were classified as presenting a high risk of bias.
Results
The electronic search was conducted on 5 January 2015 and 862 articles were identified. Thirty-nine articles were classified as relevant after reading the title and/or abstract. One full-text article could not be obtained ; the full texts of the remaining 38 studies were evaluated against the previously established inclusion criteria. Thirteen articles did not meet one or more criteria for inclusion and were excluded from the study. At the end of the selection process, 25 articles were included. A flowchart of the selection and inclusion process is given in Fig. 1 .
Of the articles included in the final review, 21 were retrospective studies, two were prospective, and two were case series. In the quality evaluation, three articles showed a low risk of bias, seven showed a moderate risk of bias, and 15 showed a high risk of bias ( Table 1 ).
| Year | Author and reference | Random selection in population | Defined inclusion/exclusion criteria | Report loss to follow-up | Validated measurements | Statistical analysis | Estimated potential risk of bias |
|---|---|---|---|---|---|---|---|
| 2014 | Akbay and Aydogan | Yes | No | No | Yes | Yes | High |
| 2014 | Ndukwe et al. | Yes | No | No | Yes | No | High |
| 2013 | Ahmad and Choudhary | Yes | No | No | No | No | High |
| 2013 | Bai et al. | Yes | Yes | Yes | Yes | No | Moderate |
| 2013 | Magesh et al. | Yes | Yes | No | Yes | Yes | Moderate |
| 2012 | Guerrier et al. | Yes | Yes | Yes | Yes | Yes | Low |
| 2012 | Okoje et al. | Yes | Yes | No | Yes | Yes | Moderate |
| 2011 | Gadre et al. | Yes | Yes | Yes | Yes | No | Moderate |
| 2011 | Handschel et al. | Yes | No | No | Yes | Yes | High |
| 2011 | Rana et al. | Yes | Yes | Yes | Yes | Yes | Low |
| 2010 | Maurer et al. | Yes | No | No | Yes | Yes | High |
| 2010 | Mooren et al. | Yes | Yes | Yes | No | No | High |
| 2010 | Ogundale et al. | Yes | No | No | No | No | High |
| 2009 | Van Germet et al. | Yes | No | Yes | Yes | Yes | Moderate |
| 2008 | Chiapasco et al. | Yes | Yes | Yes | Yes | Yes | Low |
| 2003 | Jin et al. | Yes | Yes | No | Yes | Yes | Moderate |
| 1999 | Foster et al. | Yes | No | No | Yes | Yes | High |
| 1999 | Schliephake et al. | Yes | No | Yes | Yes | Yes | Moderate |
| 1997 | Pogrel et al. | Yes | No | No | Yes | Yes | High |
| 1996 | Holtz | Yes | No | Yes | Yes | No | High |
| 1996 | Shirota et al. | No | No | No | No | No | High |
| 1994 | Cheung et al. | Yes | No | No | Yes | Yes | High |
| 1992 | El-Sheikh et al. | Yes | No | No | Yes | No | High |
| 1990 | Tidstrom and Keller | No | No | No | Yes | No | High |
| 1980 | Giordano et al. | Yes | No | Yes | Yes | No | High |
Table 2 shows the data collected and the results of all of the studies included in the review. In all, 926 non-vascularized mandibular reconstruction grafts were performed. The most frequent donor site was the iliac crest ( n = 705, 76.1%), followed by the ribs ( n = 150, 16.2%) and fibula ( n = 52, 5.6%); calvaria, sternum, and scapula grafts, as well as combinations of grafts, were also used.
| Year | Author and reference | Type of study | Graft performed | Number of patients | Defect classification (HCL) | Defect size, cm | Aetiology | Complications | Success rate (%) |
|---|---|---|---|---|---|---|---|---|---|
| 2014 | Akbay and Aydogan | Retrospective | 9 iliac crest 1 scapula 1 fibula |
11 | 7 L 2 C 2 LCL |
1.5–14.0 Mean 5.0 |
2 benign tumours 9 gunshot injuries |
2 infections | – |
| 2014 | Ndukwe et al. | Retrospective | 17 rib 8 iliac crest |
25 | 10 LCL 8 LC 4 L 2 H 1 C |
– | 24 benign tumours 1 gunshot injury |
8 infections 3 graft loss 2 dehiscence |
88.0 |
| 2013 | Ahmad and Choudhary | Retrospective | 50 rib | 50 | – | – | 50 benign tumours | 2 infections 1 graft loss |
90.0 |
| 2013 | Bai et al. | Retrospective | 7 particulate iliac crest | 7 | 3 L 3 LC 1 C |
6.0–13.0 Mean 9.1 |
7 benign tumours | No complications | 100.0 |
| 2013 | Magesh et al. | Prospective | 13 iliac crest | 13 | – | – | 13 benign tumours | 2 infections 2 dehiscence |
100.0 |
| 2012 | Guerrier et al. | Retrospective | 35 iliac crest | 35 | 14 L 9 LCL 7 LC 5 C |
16 grafts <5.0 19 grafts >5.0 |
35 gunshot injuries | 3 total graft loss 4 partial graft loss |
80.0 |
| 2012 | Okoje et al. | Retrospective | 47 iliac crest | 47 | 21 L 12 LC 8 H 6 C |
– | 42 benign tumours 3 malignant tumours 1 trauma 1 gunshot injury |
12 infections 4 dehiscence |
89.4 |
| 2011 | Gadre et al. | Retrospective | 68 iliac crest 16 fibula 3 rib |
87 | 47 L 27 LC 11 HC 2 H |
– | 54 benign tumours 33 malignant tumours |
10 graft loss 8 dehiscence 7 infections 3 recurrence |
88.5 |
| 2011 | Handschel et al. | Retrospective | 84 iliac crest | 84 | 55 L 17 LC 12 C |
Mean 4.9 ± 2.1 | 57 malignant tumours 23 benign tumours 1 osteomyelitis 1 trauma 2 other causes |
27 infections 6 dehiscence 5 fixation failure 7 other complications |
75.0 |
| 2011 | Rana et al. | Retrospective | 80 iliac crest 39 rib 31 fibula 16 sternum |
166 | – | – | – | 30 graft loss 18 infections |
81.9 |
| 2010 | Maurer et al. | Retrospective | 26 iliac crest 4 fibula |
30 | – | – | – | – | – |
| 2010 | Mooren et al. | Retrospective | 20 iliac crest associated with particulate iliac crest | 20 | – | 2.0–12.0 | 10 malignant tumours 7 benign tumours 3 trauma |
3 dehiscence 3 infections |
85.0 |
| 2010 | Ogundale et al. | Retrospective | 37 iliac crest | 37 | – | – | 36 benign tumours 1 gunshot injury |
10 infections | 91.9 |
| 2009 | Van Germet et al. | Retrospective | 74 iliac crest | 74 | 42 L 29 LC 2 C 1 LCL |
– | 29 benign tumours 26 malignant tumours 8 osteomyelitis 7 osteoradionecrosis |
27 infections | 75.7 |
| 2008 | Chiapasco et al. | Retrospective | 14 iliac crest 2 calvaria |
16 | 5 L 5 LC 3 H 2 C 1 HC |
– | 14 benign tumours 2 malignant tumours |
1 fixation failure 1 dehiscence |
93.7 |
| 2003 | Jin et al. | Retrospective | 15 iliac crest | 15 | 5 LC 4 L 4 H 1 C 1 LCL |
– | 15 benign tumours | 4 infections 1 fixation failure |
– |
| 1999 | Foster et al. | Retrospective | 22 iliac crest 4 particulate iliac crest |
– | – | 2.5–11.0 Mean 8.1 |
21 benign tumours 3 malignant tumours 2 osteoradionecrosis |
8 graft loss | 69.2 |
| 1999 | Schliephake et al. | Retrospective | 23 iliac crest | 23 | – | 2.0–12.0 Mean 6.7 |
20 malignant tumours 3 benign tumours |
– | – |
| 1997 | Pogrel et al. | Retrospective | 22 iliac crest 4 particulate iliac crest 3 rib |
29 | – | 3.0–14.0 Mean 8.3 |
20 benign tumours 5 malignant tumours 3 infections 1 trauma |
8 graft loss 3 infections |
75.9 |
| 1996 | Holtz | Retrospective | 10 iliac crest | 10 | 4 H 2 LC 2 L 1 C 1 LCL |
4.0–14.0 Mean 9.0 |
– | No complications | 100.0 |
| 1996 | Shirota et al. | Prospective | 7 particulate iliac crest 3 iliac crest |
10 | 4 LC 4 L 2 C |
– | 9 malignant tumours 1 benign tumour |
No complications | 100.0 |
| 1994 | Cheung et al. | Retrospective | 20 particulate iliac crest 2 particulate iliac crest associated with rib |
22 | 12 LC 4 L 3 C 1 LCL 1 H 1 HC |
3.5–12.5 Mean 7.5 |
7 benign tumours 7 malignant tumours 3 osteoradionecrosis 1 osteomyelitis 4 reoperations |
7 plate removal 3 infections 3 dehiscence 2 positive margin 1 recurrence 2 other complications |
72.7 |
| 1992 | El-Sheikh et al. | Retrospective | 38 rib | 38 | 14 H 8 LC 8 L 6 C 2 LCL |
– | 33 benign tumours 2 malignant tumours 3 other causes |
15 aesthetic loss 7 graft loss 5 fixation removal 1 recurrence |
94.7 |
| 1990 | Tidstrom and Keller | Case series | 34 iliac crest | 34 | 10 LC 10 LCL 6 L 5 H 2 C 1 HC |
– | 17 infection/trauma 11 malignant tumours 6 benign tumours |
4 infections 1 graft loss 1 dehiscence 1 fixation failure |
100.0 |
| 1980 | Giordano et al. | Case series | 17 particulate iliac crest | 17 | 12 L 4 C 1 LCL |
2.0–12.0 Mean 4.2 |
9 gunshot injuries 8 trauma |
1 infection 1 plate removal |
88.2 |
Stay updated, free dental videos. Join our Telegram channel
VIDEdental - Online dental courses
