Predictors of Outcomes in 900 Alveolar Bone Grafts

Abstract

Background:

Significant discrepancies exist in the reported variables influencing alveolar bone graft outcomes. The purpose of this study was to evaluate graft success and identify outcome predictors in a large patient cohort using an objective cone beam computed tomography assessment tool.

Methods:

Consecutive patients with cleft lip/palate who underwent alveolar bone grafting by 1 surgeon were included. Predictor variables were age at graft, oronasal fistula, canine position, concurrent premaxillary osteotomy, size of cleft, presence of bony palatal bridge, history of failed graft, location of primary repair, and surgeon experience. The outcome variable was graft success, determined using a cone beam computed tomography assessment tool and defined as a score of 3 or 4 (out of 4) in the following domains: vertical bone level, labiopalatal thickness, and nasal piriform symmetry.

Results:

The sample included 900 alveolar cleft sites (median graft age, 9.9 years). The success rate was 94.6%. Presence of an erupted canine, large cleft defect, or premaxillary osteotomy were independent predictors of graft failure; presence of a bony palatal bridge was associated with graft success (P < 0.05).

Conclusions:

Presence of an erupted canine, large bony defect, or premaxillary osteotomy increase the risk of failure, and a bony palatal bridge portends success. Age 12 years or older, visible oronasal fistula, history of failed graft, primary cleft repaired at outside institution, and surgeon experience were associated with higher graft failure, but were not independent predictors when controlling for covariates. Surgeons should be aware that these factors in combination increase the odds of graft failure.

CLINICAL QUESTION/LEVEL OF EVIDENCE:

Risk, III.

Reported success rates for alveolar bone grafting in patients with cleft lip and palate vary widely, ranging from 32% to 95%, with most studies documenting greater than 80% success.¹ A multitude of factors have been reported to be associated with bone graft outcomes, including patient age,^2–5 cleft type,^6–8 canine position,^9–13 oral hygiene,^2,6,14 periodontal health of cleft-adjacent teeth,^15,16 alveolar cleft size,^7,16–19 presence of an oronasal fistula,⁵ presurgical orthodontic treatment,^20,21 history of multiple revisions,⁵ and surgeon experience.^7,22,23 Most studies, however, are limited by small sample size, short follow-up, and subjective or inadequate methods of outcome assessment.

The majority of reports use periapical radiographs to document success of alveolar cleft grafting. These two-dimensional (2D) images overestimate bone volume and cannot be used to evaluate bone in the horizontal dimension.²⁴ Cone beam computed tomography (CBCT) scans permit superior assessment of three-dimensional (3D) bony anatomy. Padwa and colleagues²⁵ developed an objective radiographic outcome assessment tool using CBCT scans and demonstrated its reliability in the largest cohort of patients reported.

The purpose of this study was to use the objective radiographic assessment tool to evaluate outcomes of alveolar bone grafting in a large single-center sample. We hypothesized that there are variables that affect alveolar graft results. Specific aims were to determine the success rate of alveolar bone grafting and to identify variables associated with bone graft outcomes.

PATIENTS AND METHODS

Study Design and Patients

This was a retrospective cohort study of consecutive patients who underwent cancellous iliac crest bone graft to an alveolar cleft by 1 surgeon from 2005 through 2021.²⁶ To be included, patients had to have cleft lip and alveolus with or without cleft of the secondary palate and CBCT scan obtained more than 4 months after alveolar bone grafting. Patients were excluded if the CBCT was inadequate or if they had a corticocancellous block graft.

This study was approved by the institutional review board of the Center for Applied Investigation (protocol P00033122) with a waiver of informed consent. All research activities were conducted as per the Declaration of Helsinki.

Variables

Predictor variables were sex, cleft type, syndromic diagnosis, location of primary cleft repair (our institution, elsewhere in the United States, or international), use of a presurgical orthopedic appliance before labial repair (yes, no, unknown), history of a failed bone graft (yes, no), age at graft, presence of an erupted lateral incisor on the major segment or premaxilla (yes, no), impacted teeth in the cleft site (maintained or removed), extraction of erupted cleft-adjacent teeth (8 weeks before or at time of graft), premaxillary osteotomy (at time of labial or palatal repair or bone graft), pregraft maxillary expansion or orthodontic treatment (yes, no), presence of visible oronasal fistula (yes, no), and surgeon experience (first quartile of grafts performed; yes, no).

Preoperative radiographs (panorex or CBCT scan) were used to judge canine position at the time of graft (unerupted or erupted) (Fig. 1). Bony cleft defect size was determined by measuring the largest mesial–distal distance in the axial plane on available pregraft CBCT scans (Fig. 2). A bony palatal bridge in the axial plane was noted as present or absent in the alveolar clefts with pregraft CBCT scans (Fig. 3).

Fig. 1.

Panoramic radiographs show left alveolar cleft with unerupted canine (above) and erupted canine (below).

Fig. 2.

Size of alveolar cleft defect measured as largest mesial–distal distance in axial plane.

Fig. 3.

Bony palatal bridge in a patient with right unilateral complete cleft lip and palate.

Image Analysis

CBCT images were acquired with a standard protocol on either an i-CAT 3D (Imaging Sciences International Inc.) or Planmeca ProMax 3D Max (Planmeca Oy) imaging system. One-millimeter-slice images were assessed in the coronal and axial planes using MedView PACS Viewer software by 2 independent observers. Vertical bone level (Fig. 4); labiopalatal (horizontal) thickness at the cervical, middle, and apical thirds (Fig. 5); and nasal piriform symmetry (Fig. 6) were measured on an ordinal scale from 1 to 4 (1, failure; 2, fair result; 3, good result; 4, excellent result). Graft success was defined as a score of 3 or 4 in all domains.²⁵

Fig. 4.

Vertical bone level of erupted cleft-adjacent mesial and distal teeth. Scores wer obtained by determining the distance from the cemento-enamel junction (CEJ) to the marginal bone level in the coronal plane, with 1 = CEJ to marginal bone 75% or greater root length (least successful); 2 = CEJ to marginal bone 50% or greater to less than 75% root length; 3 = CEJ to marginal bone 25% or greater to less than 50% root length; and 4 = CEJ to marginal bone less than 25% root length (most successful).

Fig. 5.

Labiopalatal (horizontal) thickness in axial plane was scored by comparing bone thickness with root width of cleft-adjacent teeth at cervical, middle, and apical thirds of cleft-adjacent teeth, with (above, left) 1 = labiopalatal thickness less than 50% root width of cleft-adjacent teeth (least successful); (above, right) 2 = labiopalatal thickness 50% or greater root width of cleft-adjacent teeth; (below, left) 3 = labiopalatal thickness 75% or greater root width of cleft-adjacent teeth; and (below, right) 4 = labiopalatal thickness 100% or greater root width of cleft-adjacent teeth (most successful).

Fig. 6.

Piriform symmetry was scored by comparing the height of the nasal floor on both sides in the coronal plane. For unilateral clefts, grafted and unaffected sides were compared, and for bilateral clefts, 2 grafted sides were compared for symmetry. Scores were as follows: (above, left) 1 = 6 mm or greater difference (least successful); (above, right) 2 = 3 or greater and less than 6 mm difference; (below, left) 3 = 1 or greater and less than 3 mm difference; and (below, right) 4 = less than 1 mm difference (most successful).

Statistical Analysis

Analyses were conducted using SPSS version 27.0 (IBM Corp.). Frequency distributions, odds ratios with 95% confidence intervals, medians, and interquartile ranges were calculated as appropriate. Univariate logistic regressions were performed using demographic characteristics, past and present surgical and dental history, and cleft and dental characteristics as predictors of alveolar bone graft failure. A stepwise multivariable logistic regression model was then fit to determine the key predictors of graft success or failure using variables that had a 2-tailed P value less than 0.20 on univariate analyses. A receiver operating characteristic (ROC) curve assessed performance of this model, and a P value less than 0.05 was considered statistically significant for all analyses. Interrater and intrarater reliability was assessed for ordinal outcome variables using the linear weighted-kappa statistic. All reliability statistics were greater than 0.85, indicating good to excellent intrarater and interrater agreement.²⁵

RESULTS

Sample Characteristics

A total of 894 patients had alveolar bone grafting to 1116 cleft sites by 1 surgeon from 2005 through 2021; 172 patients did not meet inclusion criteria, leaving 722 patients with 900 alveolar cleft sites for analysis (59.3% male, 60.4% unilateral, 88.4% with cleft lip and palate) (Table 1). There were no significant differences between the study cohort and those lost to follow-up. The median age at alveolar bone graft was 9.9 years (range, 5.8 to 31.3 years). Postoperative CBCTs were obtained a median of 8.8 months after grafting (range, 4.2 to 191.6 months).

Table 1.

Demographic Characteristics

Demographic Characteristics	Alveolar Bone Graft, n = 900
Male, no. (%)	534 (59.3)
Cleft type, no.(%)
Unilateral cleft lip and alveolus ± cleft palate	544 (60.4)
Bilateral cleft lip and alveolus ± cleft palate	356 (39.6)
Cleft lip and palate	796 (88.4)
Cleft lip and alveolus with intact palate	104 (11.6)
Age at bone graft, yrs, median (IQR)	9.9 (9.3, 10.9)
Time from bone graft to CBCT, mo, median (IQR)	8.8 (6.4, 72.3)
Graft success, no. (%)	851 (94.6)

IQR, interquartile range.

In this series, the radiographic success rate of alveolar bone grafting was 94.6% (n = 851). The canine was erupted on the postoperative CBCT in 64% of the cohort and 94% of these clefts had good to excellent periodontal bone levels. There were 6 alveolar clefts with perioperative complications. Four had a postoperative infection treated with antibiotics: 3 had a successful graft and 1 had a failed graft. Two alveolar clefts had wound breakdown resulting in failure. The remaining 43 failed grafts had uneventful soft-tissue healing, but there was inadequate bone on CBCT at 6 months or later. There were 307 grafts (34% of cohort) with a CBCT taken more than 4 years after alveolar cleft repair with 14 failed grafts, giving a success rate at 4 years or later of 94.5%, with 5.5% requiring a second graft.

Location of the primary cleft repair included our institution (n = 510 alveolar clefts), elsewhere in the United States (n = 142), or a site outside the United States (n = 248). History of a previous failed bone graft existed in 57 clefts. There were 126 grafts completed in patients 12 years of age or older, and in 174 clefts, the canine was erupted at the time of the graft. Extraction of erupted cleft-adjacent teeth occurred before grafting in 518 sites and at the time of graft in 102 sites. Our cohort included 53 patients with 106 alveolar clefts who underwent premaxillary osteotomy at the time of bone grafting. An oronasal fistula was visible in 294 clefts. There were 225 alveolar grafts performed in the first quartile of grafts completed during the study period.

There was a pregraft CBCT scan available for 603 cleft sites and the median mesial–distal width was 7.5 mm. The bony cleft was considered large if the mesial–distal distance was 7.5 mm or greater and small if less than 7.5 mm. The bony defect was large (7.5 mm or greater) in 261 clefts with a pregraft CBCT. A bony palatal bridge in the axial plane was present in 265 of the 603 alveolar clefts with pregraft scans.

Univariate Analyses

Variables associated with an increased odds of graft failure included age 12 years or older (P = 0.003), primary cleft repaired at an outside institution (P = 0.011), visible oronasal fistula (P = 0.004), first quartile of grafts performed in the study period (P = 0.025), premaxillary osteotomy at the time of alveolar graft (P < 0.001), large alveolar cleft defect (P = 0.001), history of failed bone graft (P < 0.001), and presence of an erupted canine (P < 0.001). Extraction of erupted cleft-adjacent teeth 8 weeks before grafting (P < 0.001) and presence of a bony palatal bridge (P = 0.026) were associated with better outcomes and significantly decreased the odds of graft failure (Table 2).

Table 2.

Univariate Analyses of Variables Associated with Bone Graft Failure

Predictor Variables	Graft Failure (n = 49), No. (%)	Graft Success (n = 851), No. (%)	Odds Ratio (95% CI)	P
Patient characteristics
Female sex	26 (53.1)	340 (40.0)	1.70 (0.95–3.03)	0.072
Older age at grafting (≥12 yrs)	14 (28.6)	112 (13.2)	2.64 (1.38–5.06)	0.003
Cleft type and history
Syndromic diagnosis	10 (20.4)	94 (11.0)	2.07 (0.99–4.27)	0.051
Cleft palate (compared with cleft alveolus)	43 (87.8)	753 (88.5)	0.92 (0.39–2.25)	0.877
Bilateral cleft	17 (34.7)	339 (39.8)	0.80 (0.44–1.47)	0.475
Cleft repaired in United States	36 (73.5)	616 (72.4)	0.95 (0.49–1.82)	0.869
Cleft repaired at any outside institution	30 (61.2)	360 (42.3)	2.15 (1.19–3.88)	0.011
Presurgical orthopedic appliance	15 (48.4)	345 (56.3)	0.72 (0.35–1.49)	0.378
History of failed bone graft	11 (22.4)	46 (5.4)	5.07 (2.43–10.55)	<0.001
Dental history
Presence of erupted lateral incisor	3 (6.1)	84 (9.9)	0.60 (0.18–1.96)	0.393
Presence of erupted canine	26 (53.1)	148 (17.4)	5.37 (2.98–9.67)	<0.001
Presence of impacted teeth in cleft site	1 (2.0)	90 (10.9)	0.17 (0.02–1.26)	0.084
Extraction of erupted cleft-adjacent teeth before graft	11 (22.4)	507 (59.6)	0.20 (0.10–0.39)	<0.001
Extraction of erupted cleft-adjacent teeth at the time of graft	8 (16.3)	94 (11.0)	1.57 (0.72–3.45)	0.261
Pregraft maxillary expansion	34 (69.4)	543 (63.8)	1.29 (0.69–2.40)	0.429
Pregraft orthodontic treatment	7 (14.3)	102 (12.0)	1.22 (0.54–2.80)	0.632
Premaxillary osteotomy at time of labiopalatal repair	3 (6.1)	35 (4.1)	1.51 (0.45–5.10)	0.504
Premaxillary osteotomy at time of graft	14 (28.6)	92 (10.9)	3.28 (1.70–6.33)	<0.001
Presence of a visible oronasal fistula	25 (52.1)	269 (31.7)	2.34 (1.31–4.20)	0.004
Graft-related factors
Large bony cleft defect (≥7.5 mm)	22 (73.3)	239 (41.7)	3.84 (1.68–8.78)	0.001
Presence of bony palatal bridge	7 (23.3)	258 (44.8)	0.38 (0.16–0.89)	0.026
Procedure performed earlier in grafting surgeon’s career (first quartile of cases)	19 (38.8)	206 (24.2)	1.98 (1.09–3.59)	0.025

Sex, syndromic diagnosis, cleft type, cleft repair in a home country outside the United States, use of presurgical orthopedic appliances before labial repair, presence of an erupted lateral incisor on the major segment or premaxilla, impacted teeth on the cleft site, extraction of erupted cleft-adjacent teeth at the time of alveolar grafting, pregraft maxillary expansion or orthodontic treatment, or premaxillary osteotomy at the time of labiopalatal repair did not significantly affect graft outcome (all P > 0.05).

Multivariable Analyses

When combined into a single model, presence of an erupted canine, large bony cleft defect, and premaxillary osteotomy at time of alveolar bone grafting remained significant predictors of graft failure, and presence of a bony palatal bridge remained protective against failure (all P < 0.05; Table 3). ROC analysis demonstrated accuracy of this model with an area under the curve of 0.83.

Table 3.

Predictor Odds Ratios from Multivariable Logistic Regression

Predictor Variables	Odds Ratio	Wald (df = 1)	P	95% CI
				Lower	Upper
Older age at grafting (≥12 yrs)	0.42	2.60	0.107	0.15	1.21
Presence of erupted canine	8.39	17.40	<0.001	3.09	22.79
Extraction of erupted cleft-adjacent teeth before graft	0.53	1.95	0.163	0.22	1.29
Premaxillary osteotomy at time of graft	2.69	4.31	0.038	1.06	6.86
Large bony cleft defect (≥7.5 mm)	3.64	8.65	0.003	1.54	8.62
Presence of bony palatal bridge	0.38	4.45	0.035	0.15	0.93

DISCUSSION

Although several studies in the literature document the success rate of alveolar bone grafting and identify factors associated with graft outcomes, there is considerable variability in the reported rates of success and inconsistencies in the variables shown to affect results. Most studies are limited by small sample size and subjective or inadequate methods of outcome assessment. The purpose of this study was to evaluate the results of alveolar bone grafting performed by 1 surgeon in a large single-center sample. Specific aims were to determine the success rate of alveolar bone grafts using a reliable CBCT assessment tool and identification of variables associated with graft outcomes. Our radiographic analysis of 900 alveolar bone grafts demonstrated graft success in 94.6% of alveolar cleft sites. This is comparable to the results reported by Anver et al.²⁷ in 105 clefts using an assessment tool similar to the one we developed. Our radiographic analysis, however, included several additional measures and a more rigorous analysis in a considerably larger cohort. In a recent study, Kimia et al.⁵ reported an 86.2% success rate in 195 clefts. The orthodontic cleft team used CBCT scans to classify graft take as excellent, adequate, none, or formation of a continuous labial bridge. Although this was a relatively large sample size, success was based on a subjective analysis without reliability measures.

In our study, there were several significant univariate predictors of graft failure. Patients 12 years or older at time of bone grafting, who had their primary cleft repaired at an outside institution, who had a visible oronasal fistula, or who had the procedure performed in the first quartile of procedures in the study period were twice as likely to have a failed graft. Undergoing a premaxillary osteotomy at the time of bone grafting or having a large alveolar cleft defect yielded a 3-fold higher odds of graft failure. History of a failed bone graft or presence of an erupted canine increased the odds of a graft failure by more than 400%. Conversely, extraction of erupted cleft-adjacent teeth before grafting and presence of a bony palatal bridge were associated with better outcomes and significantly decreased the odds of graft failure. In the multivariable model, presence of an erupted canine, large bony cleft defect, and premaxillary osteotomy at time of alveolar bone grafting remained significant independent predictors of failure, and presence of a bony palatal bridge significantly decreased the odds of graft failure. ROC analysis confirmed the strong performance of this model to predict graft failure.

There is considerable discrepancy in the existing literature regarding patient and surgeon predictors of graft outcomes. Age 12 years or older at time of grafting was shown in this study and in most other reports to be associated with failure.^2–5 Kimia et al.⁵ found international adoption as a predictor of graft failure in a univariate analysis. In our study, cleft repair in a home country outside the United States was not a predictor of graft outcome, although primary cleft repair at an outside institution (both domestic and abroad) increased the odds of graft failure in the univariate analysis. The presence of a visible oronasal fistula at the time of bone graft was associated with poorer outcomes in the univariate analysis in this study and the study by Kimia et al.⁵ Our study documented that bone grafting performed in the first quartile of procedures in the study period was associated with increased odds of graft failure in the univariate analysis. This suggests that surgeon experience plays a role in graft outcome; there are studies that support^22,23,28 and refute these findings.^20,29,30 Although our study documented history of a failed graft as a predictor variable in the univariate analysis, Chalien et al.⁸ found no correlation between graft outcomes and previous graft failure. Our study revealed no association between presurgical orthodontic treatment or maxillary expansion and bone graft outcomes, although others have documented improved success with orthodontic treatment before grafting.^20,21,31 Patients who had extraction of cleft-adjacent teeth before bone graft had better outcomes in our univariate analysis, which is consistent with published reports.³²

In the multivariable analysis, our study documented that the presence of an erupted canine at the time of alveolar bone grafting was an independent predictor of graft failure. Although there are a few small reports that found no significant relationship between canine eruption and bone graft outcomes,^20,29,33 the majority of studies in the literature demonstrate significantly better results if alveolar cleft grafting is performed before canine eruption.^7,9–13,34 When a canine erupts through a successfully grafted area, the alveolus develops vertically with tooth eruption, resulting in good interdental bone levels.¹⁰ If the canine emerges into the bony cleft before grafting, there can be a periodontal defect on the surface of the tooth adjacent to the cleft. The grafted bone will attach only to the portion of the canine root with bony coverage. Grating after the canine has erupted with inadequate periodontal bone leads to insufficient vertical and horizontal bone at the cervical level, resulting in a failed graft.

A large alveolar cleft defect was also found to be a significant independent predictor of graft failure. Studies by Feichtinger et al.,¹⁸ van der Meij et al.,¹⁷ and Leal et al.⁷ documented similar results using CBCT scans to study the correlation between defect size and bone graft success. Several other reports did not find a relationship between cleft size and graft outcomes, but these studies were either carried out in small cohorts or used 2D imaging, which is inadequate for measuring the size of the bony defect.^{15,16,29,35,36} There are several possible explanations for why larger defects are associated with higher odds for graft failure. Insufficient bone for grafting could have contributed to the inferior outcomes, but in this study, patients had cancellous marrow procured from the posterior iliac crest. The average total volume of cancellous bone from the posterior iliac crest is twice the volume available from the anterior approach, so there was always ample bone for grafting.^37,38 Sufficient angiogenesis is necessary for osteogenesis in cancellous grafts. Van der Meij et al.¹⁷ suggested that the slower rate of revascularization of the graft center in wider clefts may contribute to graft failure.

Patients with bilateral complete cleft lip and palate often have a malposed premaxilla or large palatal fistula that are best managed with premaxillary osteotomy at the time of alveolar bone grafting. The results of this study and those by Geraedts et al.,³⁹ Abyholm et al.,⁴⁰ and Long et al.¹⁶ found that bone grafting in conjunction with premaxillary osteotomy is associated with higher odds of graft failure. Insufficient bone for grafting could have contributed to the inferior outcomes, but again, in this study, patients had cancellous marrow obtained from the posterior iliac crest, so there was always sufficient bone. It is more likely that the poorer outcomes in patients with premaxillary osteotomy were related to inadequate stabilization of the repositioned premaxilla.⁴¹ Stability of the segments is a key factor in bony healing. Although internal fixation with plates and screws provides the greatest rigidity, in young patients, the developing tooth buds are at risk for damage with this technique. A Kirschner wire placed through the premaxilla into the vomer has been described, but this can be challenging when the vomer is not of adequate size.⁴¹ Use of an orthodontic wire and an occlusal splint, either fabricated on model surgery or, more recently, 3D printed after virtual surgical planning, is the most common method of stabilization of the premaxilla after osteotomy and was used in this study. Surgical splints and orthodontic wires may not provide the rigidity required to stabilize the premaxilla adequately.

We documented a significant independent relationship between the presence of a bony palatal bridge and the odds of having a successful graft. Jia et al.,⁶ Leal et al.,⁷ and Sakamoto et al.²⁸ reported that patients with cleft lip and alveolus had significantly higher graft success rates than patients with cleft lip and palate. Chalien et al.⁸ found that absence of a bony bridge in patients with cleft lip and palate was associated with decreased alveolar bone levels. Patients with cleft lip and alveolus and some patients with cleft lip and palate have an intact hard palate (bony palatal bridge) that makes it easier to adequately dissect, close, and elevate the nasal lining to the extent of the fistula and beyond. Not surprisingly, these patients have better outcomes than do those without a bony palatal bridge, where the nasal mucosa is attached to palatal mucosa along the entire palate. In patients without a bony bridge, it is more challenging to separate the nasal lining from the palatal mucosa along the length of the palate without creating an opening into the nasal cavity. Inadequate closure of the nasal mucosa results in exposure of the graft to nasal secretions, leading to graft failure.

This study has several limitations. Given the retrospective study design, many patients did not meet inclusion criteria, as they were not routinely followed up in our cleft program and had 2D imaging for evaluation of graft success or were lost to follow-up. Although our standard protocol is to obtain CBCT imaging 6 months after bone grafting, many patients did not have CBCT imaging 6 months later, but rather years after the graft, which resulted in a broad range of timing of postgraft CBCT scans. Analyses were powered at more than 80% for all significant predictors in the multivariable logistic regression model. However, for the variables extraction of erupted cleft-adjacent teeth before graft and age 12 years or older, the power to find a statistically significant result was 56% and 66%, respectively. In addition, this was a single-surgeon experience, which limits the ability to generalize whether the findings are applicable to other surgical techniques and cleft centers.

CONCLUSIONS

This study demonstrated a success rate of 94.6% in 900 alveolar clefts performed by 1 surgeon using radiographic scoring criteria in 3 dimensions. The success rate of this study is on the higher end of results obtained in other smaller cohort studies using 2D and 3D imaging.^1,5 We identified 3 independent predictors of graft failure: presence of an erupted canine, large bony cleft defect, and premaxillary osteotomy at time of alveolar bone grafting. The presence of a bony palatal bridge significantly decreased the odds of graft failure. Other predictors associated with graft failure in the univariate analysis included age 12 years or older at the time of alveolar bone grafting, primary cleft repaired at an outside institution, visible oronasal fistula, surgeon experience, and history of a failed bone graft. None of these factors was an independent predictor, but surgeons should be aware that when they are seen in combination, increased odds of graft failure exist.

DISCLOSURE

The authors have no financial interest in any of the products, devices, or drugs mentioned in this article, and have no relevant financial relationship(s) with a commercial interest to report. No funding was received for this project.