Now Reading
Controversies in orbital reconstruction—I. Defect-driven orbital reconstruction: A systematic review

Controversies in orbital reconstruction—I. Defect-driven orbital reconstruction: A systematic review



International Journal of Oral & Maxillofacial Surgery, 2015-03-01, Volume 44, Issue 3, Pages 308-315, Copyright © 2014 International Association of Oral and Maxillofacial Surgeons


Abstract

In the 1980s, computed tomography was introduced as an imaging modality for diagnosing orbital fractures. Since then, new light has been shed on the field of orbital fracture management. Currently, most surgeons are likely to repair orbital fractures based on clinical findings and particularly on data obtained from computed tomography scans. However, an important but unresolved issue is the fracture size, which dictates the extent and type of reconstruction. In other fields of trauma surgery, an increasing body of evidence is stressing the importance of complexity-based treatment models. The aim of this study was to systematically review all articles on orbital reconstruction, with a focus on the indication for surgery and the defect size and location, in order to identify the reconstruction methods that show the best results for the different types of orbital fractures.

Introduction

Orbital defects are one of the most commonly encountered facial fractures because of the exposed position and thin bony walls of the midface area. Orbital fractures may occur alone or in combination with other midfacial fractures, including zygomatic complex fractures, Le Fort II and III fractures, naso-orbito-ethmoidal fractures, and frontal bone/orbital roof fractures. The classic blowout fractures are believed to result from buckling of the orbital rim and retropulsion of the orbital content. Approximately half of all orbital fractures consist of isolated wall fractures, which primarily comprise orbital floor defects and medial wall fractures.

The management of orbital fracture treatment remains controversial, and a particular subject of debate is the indication for surgery. Most surgeons are apt to repair orbital fractures based on clinical findings and particularly on data obtained from computed tomography (CT) scans, and the key question is, what fracture size needs reconstruction? In the early 1970s, Putterman et al. advocated guidelines with a conservative approach. In particular, the primary recommendation was to wait watchfully and follow the course of the patient's recovery to detect the possible development of post-traumatic diplopia, enophthalmos, and hypoglobus. Later insights led to well-defined indications for immediate surgery ( Table 1 ).

Table 1
Burnstine criteria for orbital fracture surgery timing.
Immediate Early Observation
Time frame Within 24 h 1–14 days >14 days
Indications • Diplopia with CT evidence of an entrapped muscle or peri-orbital tissue associated with a non-resolving oculocardiac reflex: bradycardia, heart block, nausea, vomiting, or syncope
• ‘White-eyed blowout fracture’, young patient (<18 years), history of peri-ocular trauma, little ecchymosis or oedema (white eye), marked extraocular motility vertical restriction, and CT examination revealing an orbital floor fracture with entrapped muscle or peri-muscular soft tissue
• Early enophthalmos/hypoglobus causing facial asymmetry
• Symptomatic diplopia with positive forced duction, evidence of an entrapped muscle or peri-muscular soft tissue on CT examination, and minimal clinical improvement over time
• Large floor fracture causing latent enophthalmos
• Significant hypo-ophthalmos
• Progressive infraorbital hypaesthesia
• Minimal diplopia (not in primary or downgaze), good ocular motility, and no significant enophthalmos or hypo-ophthalmos
CT, computed tomography.

Strong indications for immediate repair include (1) diplopia with radiological evidence of compressed orbital tissue resulting in early ischemic necrosis and oculocardiac reflex, (2) life-threatening white-eyed blowouts or trapdoor fractures in children with eye motility disturbances, and (3) radiological evidence of orbital tissue compression accompanied by oculocardiac reflex, early enophthalmos, or hypoglobus producing facial asymmetry that affects function and cosmesis. In addition, to prevent the fibrosis of injured orbital tissue, early repair within 2 weeks has been proposed for some indications, such as clinically unimproved diplopia with radiological evidence of orbital tissue compression. Further, several studies have shown that early reconstruction of large orbital defects is essential for good functional results. The most difficult management decisions occur with regard to patients with smaller orbital defects. For example, patients with orbital fractures who have good ocular motility and only slight displacement of the orbital content are often treated expectantly. Estimating the benefit of surgery in these cases is challenging, since the behaviour of the soft tissues over time is unpredictable. Thus, the indication for surgical intervention in these types of cases remains controversial.

The clinical outcomes of treatment for the different types of orbital fracture are difficult to compare. The decision to choose a certain implant material must be based on the size and location of the defect and the remaining structural support in combination with clinical symptomatology. In the case of linear fractures with small defects and entrapment of the orbital content, the placement of a membrane may be suitable, whereas in larger defects affecting one wall or multiple walls, a stronger, supportive material may be necessary.

Jaquiéry et al. proposed a simplified two-dimensional model to describe these fractures semi-quantitatively in a trefoil-shaped diagram of the internal orbit. Five categories of the extent of the fracture were defined; fractures with a higher classification were associated with a lower accuracy of reconstruction due to repositioning of the globe ( Fig. 1 ). In our experience, the current process of surgical decision-making is rarely influenced by this classification.

Classification of orbital fractures (modification of the model by Jaquiéry et al. 16 ).
Fig. 1
Classification of orbital fractures (modification of the model by Jaquiéry et al. ).

The aim of this study was to systematically review all prospective and retrospective clinical trials on orbital reconstruction. Particular focus was placed on the indication for surgery in relation to defect size and location, in order to identify the reconstruction methods that show the best results for the different types of orbital fracture.

Methods

A systematic literature search in PubMed (updated until 4 October 2013; all indexed years) was performed using multiple search terms, combining the subjects ‘orbital fracture’, ‘reconstruction material’, ‘volume’, and ‘classification’. The search excluded case series with 10 or fewer subjects. The language was restricted to English and German. All human clinical studies (prospective and retrospective) on various surgical reconstruction methods used for orbital fracture treatments met our entry criteria. Preclinical animal and cadaveric studies, as well as clinical studies comparing different incisions or approaches rather than reconstruction methods, were excluded. Fig. 2 shows a flow diagram of the inclusion process. Two authors (SS and LD) appraised the relevance of the articles based on the abstracts (in a primary review process, according to the PRISMA criteria (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) ). In a secondary review, full articles were retrieved, and relevant articles were included. Disagreement was resolved through discussion with a third person (PG).

Flow diagram of the present systematic review, performed in accordance with the PRISMA criteria.
Fig. 2
Flow diagram of the present systematic review, performed in accordance with the PRISMA criteria.

The PubMed search terms were as follows: (((((“Orbital Fractures”[Mesh])) OR (orbital fracture*[tiab] OR orbit fracture*[tiab] OR orbital trauma*[tiab] OR orbit trauma*[tiab] OR orbital injur*[tiab] OR orbit injur*[tiab] OR orbital wall fracture*[tiab] OR orbital wall injur*[tiab] OR orbital wall trauma*[tiab] OR orbital floor fracture*[tiab] OR orbital floor injur*[tiab] OR orbital floor trauma*[tiab] OR blow-out fracture*[tiab] OR blowout fracture*[tiab] OR supraorbital fracture*[tiab] OR trapdoor fracture*[tiab] OR malar fracture*[tiab] OR tripod fracture*[tiab] OR orbitozygomatic fracture*[tiab] OR orbito-zygomatic fracture*[tiab] OR zygomatico-orbital fracture*[tiab] OR zygomaticoorbital fracture*[tiab] OR tripartite fracture*[tiab] OR (le fort[tiab] AND fracture*[tiab]) OR (lefort[tiab] AND fracture*[tiab])))) AND ((“Prostheses and Implants”[Mesh] OR prosthes*[tiab] OR implant*[tiab]) OR (“Internal Fixators”[Mesh] OR internal fixat*[tiab] OR plate*[tiab] OR reconstruct*[tiab] OR membrane*[tiab] OR sheet*[tiab] OR mesh*[tiab]) OR (“Bone Transplantation”[Mesh] OR bone transplant*[tiab] OR bone graft*[tiab] OR “Cartilage”[Mesh] OR cartilage[tiab] OR “Fascia Lata”[Mesh] OR fascia lata*[tiab] OR “Periosteum”[Mesh] OR periosteum*[tiab] OR “Dura Mater”[Mesh] OR dura[tiab] OR “Gelatin”[Mesh] OR gelatin[tiab] OR “Sclera”[Mesh] OR sclera*[tiab]) OR (“Biocompatible Materials”[Mesh] OR biomaterial*[tiab] OR bioceramic*[tiab] OR animal derived[tiab]) OR (“Durapatite”[Mesh] OR durapatite[tiab] OR hydroxyapatite[tiab] OR hydroxylapatite[tiab] OR bioactive glass[tiab] OR “Titanium”[Mesh] OR titanium[tiab] OR “Cobalt”[Mesh] OR cobalt[tiab] OR “Silicones”[Mesh] OR silicone*[tiab]) OR (“Polymers”[Mesh] OR polymer[tiab] OR polymers[tiab] OR polymeric[tiab] OR polyethylene*[tiab] OR nylon*[tiab] OR teflon[tiab] OR “poly(lactic acid)”[Supplementary Concept] OR “poly(lactic acid)”[tiab] OR polylactic acid[tiab] OR poly- d,l -lactic acid[tiab] OR poly- l -lactic acid[tiab] OR “poly(lactic- co -hydroxymethyl glycolic acid)”[Supplementary Concept] OR PLA/PGA[tiab] OR polydioxanone*[tiab] OR polyglactin 910[tiab]) OR (“Alloys”[Mesh] OR alloy*[tiab]))) NOT case reports[pt].

Results

From the systematic search, a total of 231 studies including 15,032 patients with orbital injuries were identified ( Tables 2 and 3 ).

Table 2
Overview of retrospective studies on surgical orbital fracture repair, 1964–2013.
Fracture type Reconstruction materials Number of studies a Number of patients
Orbital floor/pure blowout Resorbable alloplastic
Porous polyethylene 12 824
PGA 910/PDA mesh 2 41
Ethisorb Dura 1 44
PDA plate/foil 2 26
Resorbable PDS sheet 4 429
PGA membrane 2 24
PLA membrane 2 79
PMMA 2 N/A
Fibrin glue and scaffold 1 10
Collagen membrane 1 23
Autograft
Temporalis fascia 1 32
Lyoph. tensor fascia lata 1 12
Lyoph. dura mater 4 185
Maxillary bone 4 150
Mandibular cortex 4 124
Iliac cancellous bone 2 106
Calvarial bone 4 102
Antral bone 1 11
Autogenous cartilage 7 91
Heterologous bone 1 20
Bovine processed bone 1 N/A
Non-resorbable alloplastic material
Titanium mesh 5 214
Kirschner wire fixation 2 56
Bioactive glass plates 3 85
Hydroxylapatite 2 103
Silicone implants 11 590
Implants and balloon 5 426
Stent 1 N/A
PTFE sheets 4 440
Not specified/various 23 2546
Orbital floor and medial wall Porous polyethylene 1 39
Titanium mesh 3 68
Bone graft 1 41
Mandibular cortex 1 46
CAD/CAM anatomical plates 1 15
Nylon foil ‘wraparound’ 1 98
Not specified/various 4 176
Medial wall Porous polyethylene 3 185
Hydroxylapatite 1 48
Customized titanium mesh 1 22
PGA 910/PDA mesh 1 31
Not specified/various 4 97
Lateral wall Bone graft 1 85
Orbital roof PGA 910/PDA mesh 1 85
Not specified/various 4 251
‘Large’, ‘extensive’, ‘complex’ or comminuted orbital fractures Porous polyethylene 3 198
CAD/CAM titanium sheets 2 29
Titanium and LactoSorb 1 20
Titanium implants 2 65
PLA/PGA plates and screws 1 11
Bone grafts 1 49
Not specified/various 2 89
Zygomatico-orbital fractures Porous polyethylene 2 27
Autogenous conchal cartilage 1 52
Maxillary wall graft 1 7
Hydroxylapatite 1 5
Titanium mesh 2 93
Plates and screws 3 112
Not specified/various 7 625
Heterogeneous/mixed Macropore 1 106
Titanium implant 5 284
CAD/CAM titanium sheets 1 1411
Resorbable sheets 12 176
Vitallium mesh 1 46
Autogenous graft 8 274
Biodegradable plates and screws 1 295
Bioactive glass plates 2 71
Polyethylene + hydroxyapatite 2 450
X-ray film implant 1 56
Not specified/various 16 2019
Total 217 14,650
Lyoph., lyophilized; N/A, not available; PDA, polydioxanone; PDS, poly- p -dioxanone; PGA, polyglycolic acid; PLA, polylactic acid; PTFE, polytetrafluoroethylene; PMMA, poly(methyl methacrylate).

a Some studies present more than one type of reconstruction material.

Table 3
Overview of prospective studies on surgical orbital fracture repair, 2001–2013.
Study [Ref.] Design Indication for surgery Defect size Defect location Reconstruction materials Number of patients Follow-up (postop.) Diplopia (any gaze) Enophthalmos/proptosis/dystopia Eye motility disorder Infra-orbital hypaesthesia
Preop. Postop. Preop. Postop. Preop. Postop. Preop. Postop.
Kruschewsky et al. RCT Decreased ocular mobility; diplopia; enophthalmos; ocular functional compromise N/A Floor ± medial wall ± other facial fractures Auricular cartilage graft (8) vs. blade absorbable polyacid copolymer (12) 20 6 Months 25% vs. 42% ( P = N/A) 0% vs. 0% 88% vs. 83% ( P = N/A) 0% vs. 0% 13% vs. 8% ( P = N/A) 0% vs. 0% 38% vs. 42% ( P = N/A) 25% vs. 17% ( P = N/A)
Becker et al. RCT Incarcerated or prolapsed orbital tissue (CT) <1 cm Isolated floor (13) associated zygomatic fracture (11) ( n = 11) Collagen membrane (12) vs. PDS foil 0.15 mm (12) 24 (20; 2 lost to follow-up in both groups) 6 Months 21% of total 0% vs. 0% 0% vs. 0% 0% vs. 0% 29% of total 0% vs. 0% 60% of total 0% of total
Bayat et al. RCT Enophthalmos <2 mm N/A Floor ± medial wall ± other facial fractures Nasal septal cartilage (11) vs. conchal cartilage (11) 22 3–6 Months ( P = NS) 9% vs. 9% ( P = NS) Less in nasal septal group ( P = 0.008) ( P = NS) ( P = NS) ( P = NS) ( P = NS)
Dietz et al. RCT (multi-centre) Increased orbital pressure; enophthalmos; persistent diplopia; visual impairment; hypoparesthesia; ION; severe dislocation (fracture gap >3 cm) N/A N/A Perforated PDS foil 0.15 mm (14) vs. titanium dynamic mesh (14) 28 >6 Months 75% vs. 88% ( P = N/A) 50% vs. 50% 79% vs. 86% ( P = N/A) 86% vs. 86% 75% vs. 88% ( P = N/A) 50% vs. 50% 10% vs. 9% N/A
Al-Sukhun and Lindqvist CCT >2 cm >2 cm Floor ± medial wall Autogenous bone graft (24) vs. PLA 70/30 plate (15) 39 36 Weeks 33% vs. 13% ( P = NS) 0% vs. 7% ( P = NS) 46% vs. 20% ( P = NS) 13% vs. 13% ( P = NS) 21% vs. 20% ( P = NS) 0% vs. 0% 8% vs. 7% ( P = NS) 0% vs. 0%
Lieger et al. Pilot without controls >2 cm or >1 wall; expecting functional or aesthetic deficit Jaquiéry category III–IV a Floor ± medial wall Low-profile titanium mesh 27 12 Weeks 25–52% 26% 26% 11% ( n = 11) 36% 14% N/A N/A
Noda et al. Pilot without controls Persistent diplopia (<30 upward gaze, <40 downward gaze) Jaquiéry category I–III a Linear (2), middle (7), posterior (6) Periosteal suturing 15 5–36 Months 100% 40% N/A 0% 100% 40% N/A N/A
Wajih et al. Cohort study N/A N/A Floor Autogenous graft (14) vs. porous polyethylene (12) 26 6 Months 61% 11.5% vs. 26.9% ( P = 1.24) 50% 11.4% vs. 15.3% ( P = 0.47) 39% 2.8% vs. 7.7% ( P = 0.574) 3.8% preop.
Folkestad and Granström Cohort study N/A N/A Floor (51) with associated facial fracture (45) Various 51 12 Months 33% 9.5% 11% 16% 0% 82% 60% postop. ( P = N/A)
Kontio et al. Cohort study N/A N/A Isolated floor (11) and floor with associated facial fracture (13) Iliac cortex 24 5–13 Months 85% 0% 19% 38% N/A 0% N/A N/A
Kontio et al. Cohort study Deficiency in eye movements; diplopia; hypophthalmos; enophthalmos N/A ( n = 11) PDS implant 16 13–46 Weeks 56% 25% 13% 38% N/A N/A 69% 6% ( P = N/A)
Scolozzi et al. Preliminary CCT >2 cm defect; evidence of soft tissue entrapment >2 cm ( n = 11) Non-preformed (1) vs. 3D-preformed titanium mesh plates ( n = 11) 20 6–12 Months N/A N/A N/A N/A N/A N/A N/A N/A
Cai et al. Matched control trial N/A N/A ( n = 11) Kolibri intraoperative navigation device (29) vs. controls (29) 58 12 Months N/A 2% vs. 10% ( P = 0.039) N/A 3% vs. 10% ( P = NS) N/A 3% vs. 3% N/A 0% vs. 4% ( P = NS)
Fernandes et al. Case series >1.5 cm Restriction of gaze suggesting entrapment; diplopia >1.5 cm ( n = 11) Polyethylene implant 10 1–26 Weeks 90% 11% N/A 0% N/A 7% 0% 0%
CCT, controlled clinical trial; CT, computed tomography; ION, infra-orbital nerve; N/A, not available; NS, not statistically significant; PDS, poly- p -dioxanone; PLA, poly- l/dl -lactide; RCT, randomized clinical trial.

a Orbital wall defect categorization by Jaquiéry et al. . In cases where the studied category was unclear, the defect types are listed descriptively in this table.

Retrospective studies

The majority of studies in the literature were retrospective in nature (94%; n = 217), providing either descriptive data on a single institution's experience with a heterogeneous series of orbital fracture reconstructions, or describing uncontrolled data on a single treatment modality. The clinical and radiological outcome measurements of these studies were heterogeneous and sometimes subjective. Moreover, the complications were not always reported, and the follow-up periods were variable. The defect locations and sizes were often poorly specified. An overview of the surgical treatment of orbital fractures reported for all retrospective studies from 1964 to 2013 is shown in Table 2 .

Prospective studies

As shown in Table 3 , 14 prospective studies were performed. Five of these were controlled clinical trials, of which four were randomized. In total, 380 orbital fractures were included in the prospective studies. The highest level of evidence was found in a randomized controlled trial by Bayat et al. . This study showed a statistical difference only in the advantage of nasal cartilage over conchal cartilage as the best reconstructive material in preventing enophthalmos after 3–6 months of follow-up ( P = 0.008). The remaining controlled clinical trials provided no statistically significant data. Descriptions of the actual defect size and location of the fractures were limited.

The indication for surgery was not described in four of the 14 prospective studies. For the remaining studies, the reasons for surgical intervention were diverse. Although the Burnstine criteria were applied in several studies ( Table 1 ), most studies used less objective criteria. The clinical parameters of motility disturbance and diplopia were reported as an indication for intervention in only six studies. Enophthalmos was an indication for surgery in four studies, but not all authors described the degree of enophthalmos. The defect size (as measured by CT scan) was used as an indication for surgery in five studies. Meanwhile, incarcerated or prolapsed orbital tissue in the maxillary sinus was the indication for intervention in two different reports.

The defect size and location were mentioned in six of 14 studies. Becker et al. focused on the reconstruction of small orbital defects (<1 cm). In the study of Lieger et al. , which used the Jaquiéry classification, only large orbital defects were reconstructed.

Discussion

In this systematic review, most of the studies showed substantial heterogeneity in the types and sizes of the orbital fractures, which might be due to limited case loads and the small number of patients available in these centres. In addition, the number of randomized controlled trials on orbital reconstruction was limited, and only one of these studies described both the defect size and localization in relation to the type of fracture. Hence, because of the small sample sizes, the heterogeneity of groups, and the poor description of the defect sizes and locations, no solid evidence-based conclusions or guidelines can be drawn on defect-driven reconstruction. Further reproducible trials using multi-centre settings are needed to develop guidelines for defect-driven orbital reconstruction.

In the general treatment of skeletal fractures, a common concept is that the surgeon bases the decision regarding which hardware to use on the complexity of the fracture. For most fractures, classifications and treatment models have been established based on the fracture complexity, bony buttresses, and support. However, in orbital fractures, the bony walls are generally comminuted and the parts are often useless for reconstruction. In contrast to other midfacial fractures, orbital defects need reconstruction rather than reduction and fixation of fragments, but the indication for this intervention is arbitrary. In some institutions, a surgical approach is advocated even in cases of small defects with no functional impairment, whereas in other centres, a non-surgical approach is the treatment of choice.

The strength of the reconstruction material in most cases is probably of limited relevance in the effective repair of fracture defects; instead, the choice of material is typically dependent on biocompatibility. For instance, previous work by van Leeuwen et al. has demonstrated that only minimal stress resistance is necessary in recontouring the orbital volume. Because of the low mechanical loads and minimal physical requirement imposed on the implant by the peri-orbital tissue, only relatively little orbital implant strength is required. However, the mechanical model introduced by van Leeuwen et al. does not incorporate cantilevered reconstructions as seen in larger defects, in which the posterior support can be limited because of a missing or fractured edge. Moreover, these defects conflict with the two-point circumferential model ( Fig. 3 ). As such, the properties of the reconstruction material can be relevant in larger defects.

Cantilevered reconstruction of orbital defects (Jaquiéry IV).
Fig. 3
Cantilevered reconstruction of orbital defects (Jaquiéry IV).

The debate on the ideal material for orbital reconstruction continues as well. The magnitude of the fracture (either the size of the defect or the orbital volume change) can be the most critical factor in choosing the appropriate orbital implant. Small defects are known to heal solely by the formation of scar tissue, whereas larger defects associated with enophthalmos and hypoglobus require a more rigid material (e.g. titanium or bone graft) to support the orbital contents and retain the contour of the orbit.

Three-dimensional measurements might be the best indicator of the risk of occurrence of clinically disabling enophthalmos. The best non-invasive instrument to quantify orbital fractures is CT with thin-cut axial and coronal recordings, although it often fails to measure the exact level of bony involvement. Wide exposure of the fractured area seems to be the gold standard for emphasizing the extent of the defect, but descriptions of the actual defect size and location are rare in most studies. As such, international research would greatly benefit from a clinical classification system for orbital fractures that considers the three-dimensional aspect of the orbit and thus the subsequent volume of the orbital soft tissue displacement.

The decision with regard to the appropriate implant could be made based on the complexity of the fracture. In the literature it is suggested that surgery is indicated in large fractures with involvement of more than 50% of the orbital wall, or where the surface is larger than 2 cm. The question remains as to whether these indications can be quantified accurately and whether they are adequately reproducible. Overall, patients with combined fractures of the orbital floor and medial wall, with loss of the medial strut, should be considered at higher risk of developing clinically significant enophthalmos. In these fractures, the three-dimensional changes of the orbit are more significant and the accurate form of an orbital implant becomes more important. Therefore in such cases pre-formed or patient-specific implants may be advantageous.

The indication for surgery in solitary medial wall fractures is even more controversial. No specific prospective studies have been published on this subject. Generally accepted indications are a positive forced duction test or persistent diplopia, with CT evidence of an entrapped muscle. Although enophthalmos is often mentioned as an indication for surgery as well, the enophthalmos rarely becomes significant (more than 2 mm) in the first 2 weeks after trauma.

An easily usable anatomical classification is essential for clinical communication, and the classification introduced by Jaquiéry gives a good anatomical impression of the defects ( Fig. 1 ). The first studies using this practical tool have been published. However, although this model describes the extent of the orbital fracture, it does not address the amount of volumetric change in the orbit. Nonetheless, the latter may very well be essential in the clinical decision-making process. Specifically developed software for the quantification of volumetric changes would be helpful.

In conclusion, currently available studies were analysed in this systematic review and it was found that the data are insufficient to provide a robust basis for guidelines recommending the best reconstruction method(s) for each type of orbital fracture. Furthermore, in the near future, it will hopefully be possible to identify those defects for which the use of certain biomaterial properties will increase the predictability of the orbital reconstruction. It is the authors’ opinion that uniform standardized tests (e.g., Hertel exophthalmometry and Goldman screens) and validated questionnaires, in combination with a three-dimensional volume-based defect classification, should be used by default in orbital fracture research in the coming decade. Further reproducible trials using a multi-centre setting are needed to address the controversies in orbital fracture management.

Funding

None.

You're Reading a Preview

Become a DentistryKey membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here