Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Pediatr Dent. Author manuscript; available in PMC 2017 December 11.
Published in final edited form as:
PMCID: PMC5724038

Assessment of Restorative Treatment of Patients With Amelogenesis Imperfecta

Dr. Chiung-Fen Chen, DDS, MS, pediatric dentist,1 Dr. Jan Ching Chun Hu, BDS, PhD, professor,2 Dr. Maria Regina Padilla Estrella, DMD, MS, clinical lecturer,3 Dr. Mathilde C. Peters, DMD, PhD, professor,4 and Dr. Eduardo Bresciani, DDS, MS, PhD, assistant professor5



The purpose of this study was to assess restorative treatment outcomes in the mixed dentition of amelogenesis imperfecta (AI) patients and determine the postrehabilitation oral health status and satisfaction of the patients.


Clinical and radiographic examinations were performed on eight AI patients, who had 74 restorations placed in permanent incisors and molars, to allow evaluation of the integrity of the restorations and periodontal status post-treatment. Subjects completed a survey regarding esthetics, function, and sensitivity.


Among the 74 restorations evaluated, seven were lost; of the remaining restorations, 31 were posterior, and 36 were anterior. Ten were rated clinically unacceptable. Teeth with stainless steel crowns had a moderate gingival index (mean=2.3) and plaque index (mean=2.0) scores. Widening of the periodontal ligament and pulp canal obliteration were common radiographic findings. Subject’s recall of satisfaction regarding esthetics (P=.002) and sensitivity (brushing—P=.03; eating—P=.01) showed a statically significant difference before and after treatment.


During mixed dentition, teeth with amelogenesis imperfecta may be restored with conventional treatment modalities. Direct restorations should be considered “interim” with multiple repairs anticipated. Post-treatment, gingival inflammation and plaque accumulation were observed. Subjects were satisfied with their appearance and reported a decrease of hypersensitivity.


Amelogenesis imperfecta (AI) is a hereditary defect of enamel affecting both the primary and permanent dentition.1 It is a clinically and genetically diverse group of conditions caused by mutations in genes critical for normal enamel formation, mineralization, and maturation. By definition, AI includes only those cases where enamel defects occur in the absence of other syndromes or metabolic disorders.1 The incidence of AI ranges from 1 in 718 to 1 in 14,000, depending on the population studied.1,2 According to Witkop, AI can be classified into four main types: (1) hypoplastic; (2) hypomaturation; (3) hypocalcified; and (4) hypomaturationhypoplastic with taurodontism.1 Each main type of AI may be further divided into several subtypes, depending on clinical features and inheritance patterns. Overlapping clinical features often complicate correct diagnosis.

AI affects enamel crystallite formation, resulting in abnormal crystallite morphology.3,4 Decreased mineral content associated with increased protein content in AI-affected enamel could affect the bond strength of adhesive restorative materials.57 Although AI primarily affects enamel formation, nonenamel-related manifestations may also occur, including accelerated dental development,8,9 high prevalence of dental impaction,9,10 congenitally missing teeth,10 crown and/or root resorption,8,10 pulp calcification,8,10 and associated abnormalities.812 The presence of such abnormalities may pose limitations on the clinical management of AI-affected patients.

Potential clinical implications of AI include low caries susceptibility, rapid attrition, excessive calculus deposition, and gingival hyperplasia.8,11 Anterior open bite is another common finding associated with AI.8,12 The severity of these clinical problems may vary with each type of AI. Low caries susceptibility has been reported in children with severe hypoplastic and hypomineralized AI.13 Bacteriological and salivary data were inadequate to explain the low caries susceptibility in AI-affected patients.13 This warrants additional investigations focused on determining a potential difference in oral microflora and biofilm between affected and unaffected individuals.

AI-affected patients are often concerned about poor esthetics, increased tooth sensitivity, difficulty in maintaining oral hygiene, and decreased masticatory function and may have lower self-esteem, which can affect their overall quality of life.14 As a result, most AI patients require extensive dental treatment, which is time consuming and poses a significant economic burden on their families. AI treatment depends on the individual diagnosis and phenotype. Optimal patient management takes into consideration the phase of dental development.15

In the mixed dentition stage, the treatment objectives are to improve esthetics, eliminate tooth sensitivity, prevent further loss of tooth structure, maintain adequate vertical dimension and arch perimeter, and improve periodontal health.12,16 Since teeth have a different eruption sequence, rehabilitation in the mixed dentition is complex. Definitive treatment often cannot be rendered until eruption of the permanent dentition is complete. Interdisciplinary treatment is required for patients affected with AI and can consist of surgical, orthodontic, periodontal, prosthetic, and restorative management.17

AI not only affects the patient’s dentition, but it can also have a significant impact on his or her psychosocial development.14 AI patients have reported being teased about their teeth and unhappy with the color, shape, and size of their teeth. Therefore, restorative treatments may preserve tooth structure, improve function, and reduce tooth sensitivity as well as improve the psychosocial wellbeing of AI patients, especially in the critical preadolescent and adolescent stages.14 Retrospective outcome assessment of restorative treatment in AI-affected patients showed that all AI-affected patients are positively influenced by their restorative treatment, and nearly half of the patients preferred restorative treatments to be done at an earlier age.18

Treatment options advocated in the literature include glass ionomer cements, composite resins, stainless steel crowns (SSCs), lab-fabricated crowns, and multiple extractions necessitating an overdenture.12,1922 Unfortunately, long-term follow-up of restorative outcomes of AI patients is particularly scarce. The majority of evidence is provided by case reports that present treatment modalities and outcomes of a few AI patients, with some additional description of their family members. It is surprising to note that there is currently no standard of care established for managing the AI-affected patient, especially during the mixed dentition stage. By gaining insight and understanding of the outcome of the various restorative alternatives for each type of AI, clinicians may select the most favorable approach for their individual AI-affected patient.

The purposes of this study were to evaluate the outcome of various treatment modalities for amelogenesis imperfecta patients in the mixed dentition stage and determine their postrehabilitation oral health status and satisfaction regarding appearance, function, and sensitivity.


This study was approved by the Health Sciences Institutional Review Board at the University of Michigan, Ann Arbor, Michigan. At the Children’s Dental Clinic of the University of Michigan twenty-nine AI patients were considered to be potential candidates. Twelve of 29 potential candidates met the following inclusion criteria: eight to 18 years old; no other systemic disorders; and a history of restorations placed in the permanent incisors and/or permanent first molars at least six months prior to the study. The patients were excluded if they: had any other systemic disease; had a restoration placed less than six months prior to the study; were currently undergoing orthodontic treatment; or had periodontal surgery within the last six months. Written informed consent and assent were obtained prior to investigation for eight subjects. Dental history data (clinical diagnosis, treatment rendered, and treatment notes) were collected by initial chart review.

Clinical restorative and periodontal evaluations of the restored permanent teeth were performed independently by two calibrated examiners between October 2009 and August 2010. In case of disagreement, all photos, radiographs, and teeth were re-evaluated and discussed until consensus was reached. Clinical photographs (taken with a digital single lens reflex camera) were used to evaluate the color and texture of teeth and gingiva. An attempt was made to take 15 intraoral photographs. In one case, only five photographs were taken due to behavioral issues. Restorations were evaluated based on modified US Public Health Services criteria.23 A blunt explorer with a 400-µm tip was used to assess restorations on the permanent incisors and first molars.

Depending on the restoration type, the following characteristics were evaluated: surface; color match; anatomic form; margin integrity/adaptation; fracture of restoration; caries; and wear. If a restoration received a score of unacceptable in any category, the restoration was rated as unacceptable. The periodontal status was determined by assessing the modified Quigley-Hein plaque index,24 modified gingiva index,25 pocket depth, and bleeding on probing near restored permanent first molars and permanent incisors. In addition, a total of 10 radiographs were taken using film-holding instruments (Rinn’s XCP film holding system, Dentsply Caulk, Milford, Del., USA) and evaluated for any pathology, margin discrepancy, and crown-tooth size discrepancy by two examiners. The definition of crown-tooth size discrepancy is that the selected preformed crown was smaller than the original size of the tooth. If recent radiographs were available, additional study radiographs were omitted. Subjects were also asked to answer a survey with 10 questions regarding smile, tooth esthetics, function, and sensitivity using the Wong-Baker FACES Pain Rating Scale.26

Reliability of clinical restorative, periodontal, and radiographic evaluations was determined by independent evaluation of patients in the study by two examiners. Both examiners were calibrated by a senior pediatric dentist prior to examining all study subjects. The initial inter-rater agreement was determined using Cohen’s kappa coefficient (kappa=0.66). The level of agreement, however, was determined to be poor for clinical “margin integrity” evaluation (kappa=−0.12). When the disagreement occurred, both raters re-evaluated photos, radiographs, and teeth to reach a consensus.

Data were analyzed using the SPSS 17.0 software (SPSS Inc, Chicago, Ill., USA). Descriptive statistics summarized the findings. Mixed model analysis was used to compare the periodontal status in the direct restoration and indirect restoration groups. A paired t test was used for comparing survey responses. Results with a P-value of ≤.05 were considered statistically significant.


Data were collected from the following types of AI-affected subjects: four hypoplastic; two hypocalcified; and two hypoplastichypomaturation (n=eight; six females and two males; age range= 9.4–15.9 years old). The AI subtypes were distinguished based on clinical characteristics and radiographic features. Among the eight subjects, six completed all study procedures. One of the remaining subjects completed all research procedures but the survey, while the other refused to participate in the clinical evaluation of restorations and periodontal examination. This particular subject had four SSCs and four direct restorations.

The initial chart review identified 96 affected teeth in eight AI patients (Figure 1). Among the 96 teeth, 22 were excluded [20 had no need for restorations, one was missing (extracted), and one was recently restored]. The 74 remaining teeth were classified into four different groups, based on the type of restorative material used (Table 1). In this study, amalgam for posterior restoration was scored separately. The direct restoration group included composite resin strip crowns and fewer than four surface composite resin restorations on the permanent incisors only. The indirect restoration group included lab-fabricated resin veneers and lab-fabricated composite resin or acrylic crowns. Of the 74 remaining teeth, 67 restorations were present: 27 SSCs; four amalgams; 21 direct restorations; and 15 indirect restorations. Seven restorations (six direct and one indirect) were lost and, therefore, not evaluated. The mean age of the present restorations was 38.53 months, with a range of six to 100 months (Table 2).

Figure 1
The Distribution of Teeth.
Table 1
Table 2

Of the 67 total teeth with restorations, 31 are permanent molars and 36 are permanent incisors. Only 59 restorations were evaluated clinically due to lack of participation of one subject (Table 3). Seven teeth with lost restorations were not scored clinically and radiographically. Clinical assessment resulted in 10 restorations being unacceptable (constituting failure), with the highest number being in the direct restoration group. One SSC showed extreme wear (crown perforation), and one amalgam restoration showed an unacceptable, pitted surface. Two indirect restorations showed unacceptable margins, and one indirect restoration was lost. Including six lost direct restorations, the failure rate of the direct restoration group (12/23) was approximately 52 percent (Table 3). In addition, frequent retreatment was needed in the direct restoration group. Seven direct restorations needed repair, and three SSCs were replaced. One indirect restoration was recemented.

Table 3

A total of 59 teeth with restorations in seven subjects were examined for periodontal status. Teeth that received an SSC had a moderate score for gingival index (mean=2.3±0.69 SD) and plaque index (mean=2.0±0.92). Using the mixed model analysis to compare the direct and indirect restoration groups, statistically significant differences were found in pocket depth in the indirect restoration group (P=.00) and in plaque index in the direct restoration group (P=.03). Bleeding on probing was observed in every restoration group (SSCs=20/23, 87 percent; amalgam=1/4, 25 percent; direct restoration=12/17, ~71 percent; indirect restoration =14/15, ~93 percent).

A total of 67 teeth with restorations in eight subjects were included in the radiographic evaluation. There were no furcation or apical radiolucencies observed in any group. Widening of the periodontal ligament (PDL; n=8/67, ~12 percent) and pulp canal obliteration (n=5/67, ~8 percent) were common radiographic findings requiring periodic evaluation but not clinical intervention. Margin discrepancy (9/27, ~33 percent) was only observed in the SSC group. In addition, crown-tooth size discrepancy (6/27, ~22 percent), with the performed crowns being smaller than the original tooth szie, was observed on the radiographs in this group.

In a 10-question survey, the subjects rated satisfaction with esthetics, function, and sensitivity of teeth (Table 4). Although a trend of improvement was noted in smile (P=.14), no statistical difference was found before and after treatment. Statistically significant differences were found in esthetics (P=.02) and tooth sensitivity while brushing (P=.03) and eating (P=.01).

Table 4


Most enrolled subjects in this study had a hypoplastic type of AI. This is consistent with other studies reporting hypoplastic AI, accounting for 60 to 73 percent of cases evaluated.16 SSCs were used to restore the majority of AI-affected permanent molars (27/31, ~87 percent). Many case reports have suggested that SSCs are the treatment modality of choice for the cost-effective restoration of young permanent molars.12,15 In this study, most SSCs were judged as clinically acceptable (22/23, ~96 percent). One of 23 SSCs was perforated and had at least 52 months of clinical service. Our data supports the outcome of a retrospective study of SSCs on AI-affected permanent molars reporting a 5-year survival rate of 55 percent.27

Margin discrepancy appeared to be a concern with the SSC restorations in this study. A space of approximately three mm should be preserved from the restoration margin to the alveolar bone, allowing two mm of periodontal attachment and one mm of sulcus depth.28 It is often inevitable, however, to place a SSC margin subgingivally in partially erupted permanent molars. When a restoration margin is placed subgingivally, this may not only risk invasion of the periodontal attachment but create a plaque-retentive area. Increased plaque retention stimulates a more pronounced, plaque-induced inflammatory response, which corroborates our study showing moderate gingival index scores in the SSC group. Gingivitis may be found in the presence of an SSC, especially when there is inadequate crown length, contour, and position, or when excessive cement remnants remain in the gingival sulcus.29

In this study, a commonly observed problem among SSC restorations was crown-tooth size discrepancy. Apparently, clinicians tended to choose a smaller preformed SSC than was indicated. This is understandable, since crown morphology in AI-affected patients is different and usually smaller than that in patients with normal enamel. The cementoenamel junction of the clinical crown is often located subgingivally and difficult to discern. This crown-tooth size discrepancy may potentially lead to open proximal contacts and exposure of tooth structure following progressive eruption of the tooth. The latter makes it difficult for patients to keep the restoration margin and surrounding gingiva clean—possibly resulting in a failed restoration.16 A periapical radiograph is recommended after fitting the SSC and prior to its cementation.

Amalgam was placed in four permanent first molars in this study. Teeth restored with amalgam are often not severely affected by AI and, therefore, are able to retain an intracoronal restoration. The small sample size, however, prevented a meaningful conclusion regarding this restorative therapy.

Direct restorations are commonly placed in partially erupted incisors. Partial eruption, however, increases the difficulty in isolation and margin placement and may lead to failure. The high failure rate of these restorations may be due to bonding failure. The bond strength between enamel and adhesive restorative materials is highly dependent on the enamel surface modification. The bond strength of the permanent teeth with hypocalcified AI is lower than that of the normal teeth.30 On the other hand, literature reports suggest that a thin, nonprismatic enamel layer could be sufficient to adequately retain bonded materials.12,21,31 Failure could also be due to factors other than bonding failure, such as cohesive failure occurring in the enamel, at the dentoenamel junction, or in dentin.

Understanding the specific AI type and the degree of enamel hypomineralization is critical when choosing an adhesive restorative material to restore teeth affected with AI. Glass ionomer cement (GIC) was reported to be the ideal semipermanent restorative material, due to its chemical bonding to enamel and dentin, fluoride release, biocompatibility, and acceptable esthetics.32 None of the teeth in this study, however, were restored with GIC; thus, no information could be provided to support this statement. A case report by Moretti et al., however, demonstrated the successful use of resin-modified GIC to restore primary teeth and permanent incisors.33

Lindunger and Smedberg assessed the outcome of prosthodontic management in 15 AI-affected adult patients, resulting in 213 restorations rated “acceptable to excellent,” with one exception (unacceptable crown).18 During the follow-up period, four (two percent) restorations were recemented and 16 (eight percent) were replaced, including five (two percent) due to porcelain fractures and 11 (five percent) due to caries. The median age of the restorations was 60 months. Our restoration failure rate and frequency of repair is higher than the aforementioned study, which may be due to the different dentition stage in our patient population.

A statistically significant difference was found in pocket depth (P=.00) when comparing the indirect to the direct restoration group. One can speculate that, because most indirect restorations will require subgingival margin preparation to achieve acceptable esthetics, maintaining optimal oral hygiene can be a greater challenge. The results of mixed model analysis indicated that, when comparing the direct and the indirect restoration group, there was a statistically significantly higher plaque index score in the direct composite group (P=.02). This may suggest that the direct restorations in this study had a greater degree of plaque accumulation, making it more difficult to maintain optimal oral hygiene. Although bleeding on probing was observed in every restoration group (SSCs=20/23; amalgam=1/4; direct restoration=12/17; indirect restoration=14/15), this result needs to be read with caution. Bleeding on probing is often used to monitor changes over time, and a single assessment of bleeding on probing may not be meaningful. High gingival index values and high decayed, missing, and filled teeth, permanent and primary (DMFT/dmft), scores were found in a case series of AI patients.32

In contrast to this case series report, our results show the effect of restorative treatment rather than overall oral hygiene. In a study by Lindunger and Smedberg, hypomineralized AI patients had a higher plaque index, bleeding index, and pocket depth versus hypoplastic AI patients.18 In the current study, this statement could not be tested due to the small sample size.

PDL widening and pulp canal obliteration were common radiographic findings. Without baseline radiographs for comparison, however, we could not conclude that PDL widening and pulp canal obliteration were a result of the restoration itself.

Our study data were in agreement with published findings that restorations have a positive influence on AI patients, although, only a slight improvement in chewing function was achieved. We speculate that following restorative treatment that improves chewing function, AI-affected patients may still have to select proper food types to ensure adequate chewing and prevent any injury to their existing restorations. This result, though only observed in eight subjects, provided to be statistically significant evidence in support of restoring AI-affected teeth at a young age. The results of our study corroborate that dental treatments may be medically necessary for AI patients.14

Several limitations exist in this study. The small sample size did not allow for the investigation of associations between AI-phenotype and different restoration modalities. The restorations were evaluated once and in the absence of baseline data; therefore, changes over time could not be assessed. Also, the subjects concurrently answered survey questions related to before and after treatment (post-treatment only). They may have forgotten how they really felt prior to restorative treatment. Further research is needed to strengthen the results. Comparing the oral health status before and after restorative treatment of patients with different variants of AI will enable determination of the effectiveness of restoration types for various AI subtypes. Also, evaluation of restorations at fixed time points may provide information on the durability and sustainability of restorative treatments.

In summary, during the mixed dentition stage, permanent teeth affected with amelogenesis imperfecta can be successfully restored with conventional treatment modalities. Direct restorations, such as composite resin strip crowns and fewer than 4-surface composite resin restorations, showed a high percentage of failure. When provided, these should be considered “interim restorations,” and multiple repairs and replacements have to be expected. In all cases, gingival inflammation and plaque accumulation were observed following restorative treatment of AI patients. After restorative treatment, AI-affected patients were satisfied with their appearance and expressed a decrease in tooth sensitivity when eating and brushing.


The authors wish to thank staff and patients of the Children’s Dental Clinic at the University of Michigan School of Dentistry for their participation in this study, Ms. Angela Menighini Spencer, research assistant at University of Michigan, for her help in recruiting subjects, and Ms. Kathy Welch at the University of Michigan Center for Statistical Consultation and Research for her help and guidance in data analysis. This research was supported by a grant from the Delta Dental Foundation of Michigan and the Rackham Graduate Fund


1. Witkop CJ. Amelogenesis imperfecta, dentinogenesis imperfecta, and dentin dysplasia revisited: Problems in classification. J Oral Pathol. 1989;17:547–53. [PubMed]
2. Backman B, Holm AK. Amelogenesis imperfecta: Prevalence and incidence in a northern Swedish county. Community Dent Oral Epidemiol. 1986;14:43–7. [PubMed]
3. Gopinath VK, Al-Salihi KA, Yean CY, Ann MC, Ravichandran M. Amelogenesis imperfecta: Enamel ultrastructure and molecular studies. J Clin Pediatr Dent. 2004;28:319–22. [PubMed]
4. Seymen F, Kiziltan B. Amelogenesis imperfecta: A scanning electron microscopic and histopathologic study. J Clin Pediatr Dent. 2002;26:327–35. [PubMed]
5. Wright JT, Deaton TG, Hall KI, Yamauchi M. The mineral and protein content of enamel in amelogenesis imperfecta. Connect Tissue Res. 1995;32:247–52. [PubMed]
6. Venezie RD, Vadiakas G, Christensen JR, Wright JT. Enamel pretreatment with sodium hypochlorite to enhance bonding in hypocalcified amelogenesis imperfecta: Case report and SEM analysis. Pediatr Dent. 1994;16:433–6. [PubMed]
7. Seow WK, Amaratunge FA. The effect of acid etching on enamel from different clinical variants of amelogensis imperfecta: An SEM study. Pediatr Dent. 1998;20:37–42. [PubMed]
8. Poulsen S, Gjorup H, Haubek D, et al. Amelogenesis imperfecta: A systemic literature review of associated dental and orofacial abnormalities and their impact on patients. Acta Odontol Scand. 2008;66:193–9. [PubMed]
9. Seow WK. Dental development in amelogenesis imperfecta: A controlled study. Pediatr Dent. 1995;17:26–30. [PubMed]
10. Collins MA, Mauriello SM, Tyndall DA, Wright JT. Dental anomalies associated with amelogenesis imperfecta: A radiographic assessment. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 1999;88:358–64. [PubMed]
11. Sundell S. Hereditary amelogenesis imperfecta. I. Oral health in children. Swed Dent J. 1986;10:151–63. [PubMed]
12. Seow WK. Clinical diagnosis and management strategies of amelogenesis imperfecta variants. Pediatr Dent. 1993;15:384–93. [PubMed]
13. Sundell S. Hereditary amelogensis imperfecta: An epidemiological, genetic, and clinical study in a Swedish child population. Swed Dent J Suppl. 1986;31:1–38. [PubMed]
14. Coffield KD, Phillips C, Brady M, Roberts MW, Strauss RP, Wright JT. The psychosocial impact of developmental dental defects in people with hereditary amelogenesis imperfecta. J Am Dent Assoc. 2005;136:620–30. [PubMed]
15. Bouvier D, Duprez JP, Bois D. Rehabilitation of young patients with amelogenesis imperfecta: A report of two cases. J Dent Child. 1996;63:443–7. [PubMed]
16. Kwok-Tung L, King NM. The restorative management of amelogenesis imperfecta in the mixed dentition. J Clin Pediatr Dent. 2006;31:130–5. [PubMed]
17. Ramos AL, Pascotto RC, Iwaki Filho L, Hayacibara RM, Boselli G. Interdisciplinary treatment for a patient with openbite malocclusion and amelogenesis imperfecta. Am J Orthod Dentofacial Orthop. 2011;139(suppl 4):S145–S153. [PubMed]
18. Lindunger A, Smedberg JI. A retrospective study of the prosthodontic management of patients with amelogenesis imperfecta. Int J Prosthodont. 2005;18:189–94. [PubMed]
19. Ranta H, Lukinmaa PL, Waltimo J. Heritable dentin defects: Nosology, pathology, and treatment. Am J Med Genet. 1993;45:193–200. [PubMed]
20. Rosenblum SH. Restorative and orthodontic treatment of an adolescent patient with amelogenesis imperfecta. Pediatr Dent. 1999;21:289–92. [PubMed]
21. Yip HK, Smales RJ. Oral rehabilitation of young adults with amelogenesis imperfecta. Int J Prosthodont. 2003;16:345–9. [PubMed]
22. Akin H, Tasveren S, Yeler DY. Interdisplinary approach to treating a patient with amelogenesis imperfecta: A clinical report. J Esthet Restor Dent. 2007;19:131–6. [PubMed]
23. Cvar JF, Ryge G. Reprint of criteria for the clinical evaluation of dental restorative materials. Clin Oral Investig. 2005;9:215–32. [PubMed]
24. Quigley GA, Hein JW. Comparative cleansing efficiency of manual and power brushing. J Am Dent Assoc. 1962;65:26–9. [PubMed]
25. Löe H. The gingival index, the plaque index, and the retention index system. J Periodontol. 1967;38:610–6. [PubMed]
26. Wong D, Whaley L. Clinical Handbook of Pediatric Nursing. 2. St. Louis, Mo; Mosby: 1986. p. 373.
27. Zagdwon AM, Fayle SA, Pollard MA. A prospective clinical trial comparing preformed metal crowns and cast restorations for defective first permanent molars. Eur J Paediatr Dent. 2003;4:138–42. [PubMed]
28. Padbury A, Jr, Eber R, Wang HL. Interactions between the gingiva and the margin of restorations. J Clin Periodontol. 2003;30:379–85. [PubMed]
29. Durr DP, Ashrafi MH, Duncan WK. A study of plaque accumulation and gingiva health surrounding stainless steel crowns. J Dent Child. 1982;49:343–6. [PubMed]
30. Faria-e-Silva AL, De Moraes RR, Menezes Mde S, Capanema RR, De Moura AS, Martelli H., Jr Hardness and microshear bond strength to enamel and dentin of permanent teeth with hypocalcified amelogenesis imperfecta. Int J Paediatr Dent. 2011;21:314–20. [PubMed]
31. Rada RE, Hasiakos PS. Current treatment modalities in the conservative restoration of amelogenesis imperfecta: A case report. Quintessence Int. 1990;21:937–42. [PubMed]
32. Markovic D, Petrovic B, Peric T. Case series: Clinical findings and oral rehabilitation of patients with amelogenesis imperfecta. Eur Arch Paediatr Dent. 2010;11:201–8. [PubMed]
33. Moretti AB, Sakai VT, Oliveira TM, et al. Oral management of a child with mixed dentition affected by amelogenesis imperfecta. J Dent Child. 2007;74:157–60. [PubMed]