Comparative study of nasopharyngeal soft-tissue characteristics in patients with Class III malocclusion Oscar Martin, a Leonor Muelas, b and Maria Jose Vi ~ nas c Madrid, Spain Introduction: The aims of our study were to assess nasopharyngeal soft-tissue characteristics in white patients with Class III malocclusion and to compare the results with patients with ideal occlusion. Methods: A sample of 71 patients with Class III malocclusion was selected from the Dental School at the Complutense University of Madrid. None of the subjects had a history of upper airway disease, adenoidectomy, or pathology in the pharynx. Lateral cephalograms were digitized, and linear and area measures were made to define the airway character- istics. Error analysis was performed to prevent systematic or random errors. Independent means t tests and Pearson correlation analysis were used to evaluate sex differences and the correlations among the variables. A discriminant analysis was also used to identify different groups with excellent occlusion and Class III maloc- clusion. Results: Nasopharyngeal soft-tissue characteristics in Class III subjects were different in male and female subjects. Adenoidal tissue and upper airway length were greater in the men. Three groups of variables tended to be related among themselves but not with others. A discriminant analysis showed that changes in cranial base length are directly related to nasopharyngeal morphology in Class III malocclusion. A discriminant equation was also obtained. Conclusions: This study suggests new lines of investigation about the relationship between skeletal and dental anomalies and airway obstruction, and possible specific respiratory characteristics for each type of malocclusion. (Am J Orthod Dentofacial Orthop 2011;139:242-51) T he influence of the soft tissues on craniofacial growth has been discussed in the orthodontic liter- ature for many years. Since 1873, authors have studied airway obstruction and its effects on malocclu- sion development. 1-5 In general, obstructive disorders are thought to be responsible for muscular imbalances such as greater extensions in head posture, changes in tongue position, or lip incompetence, and these have been cited as etiologic factors for malocclusions. 6-11 It has been also suggested that skeletal Class III with maxillary hypoplasia is a cause of upper pharynx constriction. 12 The effects of rapid maxillary expansion associated with maxillary protraction on pharyngeal dimensions demonstrated that maxillary protraction improves the nasopharyngeal airway dimensions. 13,14 According to these facts, it can be assumed that a skeletal Class III malocclusion is associated with an abnormal airway. Class III malocclusions are relatively frequent in Spain, affecting up to 16% of people, depending on the re- gion. 15-19 On the other hand, a person with mild skeletal Class III does not always show a dental Class III relationship. Some patients have an excellent occlusion. The muscular environment partly determines the correct dental relationship, but can the nasopharyngeal airway be 1 factor that makes an excellent occlusion possible in patients with slight skeletal disharmonies? 1-5 Normal values can be described under skeletal, occlusal, and esthetic points of view, but when can we consider an airway as “normal”? Is this normality valid for different malocclusions? Is a specific airway characteristic related to a specific malocclusion? The literature includes few articles relating airway with occlusion. A recent study analyzed the airway in Chinese subjects with skeletal Class I, but not directly with occlusion. 20 Skeletal classification is a useful tool to evaluate the airway when comparing groups with pro- nounced skeletal differences but, under our assumption, From the Universidad Complutense, Madrid, Spain. a Assistant professor, Department of Orthodontics. b Associate professor, Department of Orthodontics. c Assistant professor, Department of Orthodontics. The authors report no commercial, proprietary, or financial interest in the prod- ucts or companies described in this article. Reprint requests to: Oscar Martin, Arenalon 1, 2 dcha, 28231, Madrid, Spain; e-mail, [email protected]. Submitted, August 2008; revised and accepted, July 2009. 0889-5406/$36.00 Copyright Ó 2011 by the American Association of Orthodontists. doi:10.1016/j.ajodo.2009.07.016 242 ORIGINAL ARTICLE
Transcript
ORIGINAL ARTICLE
Comparative study of nasopharyngeal soft-tissuecharacteristics in patients with Class IIImalocclusion
Oscar Martin,a Leonor Muelas,b and Maria Jos�e Vi~nasc
Introduction: The aims of our study were to assess nasopharyngeal soft-tissue characteristics in white patientswith Class III malocclusion and to compare the results with patients with ideal occlusion.Methods: A sample of71 patients with Class III malocclusion was selected from the Dental School at the Complutense University ofMadrid. None of the subjects had a history of upper airway disease, adenoidectomy, or pathology in the pharynx.Lateral cephalograms were digitized, and linear and area measures were made to define the airway character-istics. Error analysis was performed to prevent systematic or random errors. Independent means t tests andPearson correlation analysis were used to evaluate sex differences and the correlations among the variables.A discriminant analysis was also used to identify different groups with excellent occlusion and Class III maloc-clusion. Results: Nasopharyngeal soft-tissue characteristics in Class III subjects were different in male andfemale subjects. Adenoidal tissue and upper airway length were greater in the men. Three groups ofvariables tended to be related among themselves but not with others. A discriminant analysis showed thatchanges in cranial base length are directly related to nasopharyngeal morphology in Class III malocclusion. Adiscriminant equation was also obtained. Conclusions: This study suggests new lines of investigation aboutthe relationship between skeletal and dental anomalies and airway obstruction, and possible specific respiratorycharacteristics for each type of malocclusion. (Am J Orthod Dentofacial Orthop 2011;139:242-51)
The influence of the soft tissues on craniofacialgrowth has been discussed in the orthodontic liter-ature for many years. Since 1873, authors have
studied airway obstruction and its effects on malocclu-sion development.1-5 In general, obstructive disordersare thought to be responsible for muscular imbalancessuch as greater extensions in head posture, changes intongue position, or lip incompetence, and these havebeen cited as etiologic factors for malocclusions.6-11 Ithas been also suggested that skeletal Class III withmaxillary hypoplasia is a cause of upper pharynxconstriction.12 The effects of rapid maxillary expansionassociated with maxillary protraction on pharyngealdimensions demonstrated that maxillary protraction
the Universidad Complutense, Madrid, Spain.tant professor, Department of Orthodontics.ciate professor, Department of Orthodontics.tant professor, Department of Orthodontics.uthors report no commercial, proprietary, or financial interest in the prod-r companies described in this article.t requests to: Oscar Martin, Arenal�on 1, 2 dcha, 28231, Madrid, Spain;l, [email protected], August 2008; revised and accepted, July 2009.5406/$36.00ight � 2011 by the American Association of Orthodontists..1016/j.ajodo.2009.07.016
improves the nasopharyngeal airway dimensions.13,14
According to these facts, it can be assumed thata skeletal Class III malocclusion is associated with anabnormal airway.
Class III malocclusions are relatively frequent in Spain,affecting up to 16% of people, depending on the re-gion.15-19 On the other hand, a person with mild skeletalClass III does not always show a dental Class IIIrelationship. Some patients have an excellent occlusion.The muscular environment partly determines the correctdental relationship, but can the nasopharyngeal airwaybe 1 factor that makes an excellent occlusion possible inpatients with slight skeletal disharmonies?1-5 Normalvalues can be described under skeletal, occlusal, andesthetic points of view, but when can we consider anairway as “normal”? Is this normality valid for differentmalocclusions? Is a specific airway characteristic relatedto a specific malocclusion?
The literature includes few articles relating airwaywith occlusion. A recent study analyzed the airway inChinese subjects with skeletal Class I, but not directlywith occlusion.20 Skeletal classification is a useful toolto evaluate the airway when comparing groups with pro-nounced skeletal differences but, under our assumption,
Fig 2. Surface variables: 11, total area (12 1 13); 12,aerial area; 13, adenoidal area.
Martin, Muelas, and Vi~nas 243
makes the mistake of classifying a mild skeletal dishar-mony with an ideal airway and a muscular balance.
In a previous study, we showed the nasopharyngealnorms of ideal occlusion in a Spanish population andthe relationships between the variables.21 The aims ofthis study were to describe the nasopharyngeal charac-teristics of Class III malocclusion and to compare the re-sults with those obtained previously for excellentocclusion. The objectives of this study were to (1) iden-tify soft-tissue nasopharyngeal characteristics in adultswith excellent occlusion and Class III malocclusion,(2) evaluate sexual dimorphism in soft-tissue airwaystructure of subjects with excellent occlusion and ClassIII malocclusion, (3) establish a possible relationshipbetween soft-tissue measurements with cranial struc-tures in these patients, (4) establish nasopharyngeal dif-ferences between Class III subjects and a control group,and (5) determine a discriminant equation based oncranial structures.
MATERIAL AND METHODS
We selected 162 adults from the data base of theDental School at Complutense University, Madrid, Spain.This sample was divided into 2 groups: excellent occlu-sion (n5 91; 55 men, 36 women) and Class III malocclu-sion (n 5 71; 33 men, 38 women).
American Journal of Orthodontics and Dentofacial Orthoped
Our selection criteria were white adults with noprevious orthodontic treatment in both groups; andexcellent occlusion with correct overjet and overbite, bi-lateral molar and canine Class I relationships, balancedocclusion, and no history of sleep disorders, snoring, sleepapnea, upper airway disease, adenoidectomy, or pathol-ogy in the pharynx in the control group. Our exclusioncriteria included tooth rotations over 25�, crowdinggreater than 3 mm, posterior crossbite, and craniofacialanomalies.
The selection criteria for the Class III group includeda dental Class III relationship; no craniofacial syndromes,deformities, or asymmetries; sealed lips in a relaxed po-sition; optimal periodontal health; and no missing teeth.Overjet and overbite values were�26 1 and 16 1 mm,respectively.
Cephalograms of all subjects were taken in naturalhead position. Good definition, molars inmaximum inter-cuspation, and lips sealed in a relaxed position were alsorequired. All radiographswere digitized by the same inves-tigator (O.M.) using a computerizated tracing program.
The following cephalometric measurements wereselected (Figs 1 and 2).
1. PNS-AD1: lower aerial width, the distance betweenPNS and the nearest adenoid tissue measuredthrough the PNS-Ba line (AD1).
ics February 2011 � Vol 139 � Issue 2
Table I. Sexual dimorphism in subjects with excellentocclusion
2. AD1-Ba: lower adenoid width, defined as the soft-tissue thickness at the posterior nasopharynx wallthrough the PNS-Ba line.
3. PNS-Ba: lower airway width, the distance betweenPNS and Ba—the sum of variables 1 and 2.
4. PNS-AD2: upper aerial width, the distance betweenPNS and the nearest adenoid tissue measuredthroughaperpendicular line to S-Ba fromPNS (AD2).
5. AD2-H: upper adenoid width, defined as the soft-tissue thickness at the posterior nasopharynx wallthrough the PNS-H line.
� Hormion (H): the cephalometric point located nearthe adenoidal tissue at the cranial base, localizedwhere a perpendicular to S-Ba line crosses the sphe-noid bone. The variations of this point are minimalbecause it is located far from growing sites.
6. PNS-H: upper airway width, the distance betweenPNS and H—the sum of variables 1 and 2.
7. N-H: nasal fossa length, the distance between Nand H.
mum distance between the upper soft palate andthe nearest point on the posterior pharynx wall.22
10. McNamara’s lower pharynx dimension: the mini-mum distance between the point where the poste-rior tongue contour crosses the mandible and thenearest point on the posterior pharynx wall.22
11, 12,and 13. Total, adenoidal, and aerial areas, using the
method of Handelman and Osborne.23 Thismethod takes as references the Ba-N plane,the bispinal plane, and 2 perpendicular linesto the bispinal plane: one crosses the more an-terior point at the atlas vertebra, and othercrosses the PNS. The resulting trapezoid is di-vided into 2 spaces (aerial and adenoid). The to-tal area is the sum of the adenoidal and aerialareas.
To assess measurement error, duplicate tracings of 25films were made by the same investigator, and the ran-dom method was used as described by Dahlberg.24
The mean difference between the first and secondmeasurements, the standard error of a single measure-ment, and the percentage of total variance attributableto measurement errors were calculated for each variable.Method error was determined as suggested by Houston25
with the formulaME 5
ffiffiffiffiffiffiffiffiffiPd2
2n
qwhere d is the difference
between measured pairs, and n is the number of pairs.The error method is comparable with that used in
other cephalometric studies and explained less than10% of the biologic variance.
February 2011 � Vol 139 � Issue 2 American
The mean differences were less than 1.0 mm, exceptfor the adenoidal and aerial areas: the mean differenceswere less than 4 mm2 The errors tended to be greaterwhen the definition points were less precise: eg, thesoft-tissue contours.
The coefficient of reliability was calculated by com-paring the method error with the variance of the totalsample and expressed as a percentage. Reliability variedfrom 97.86% to 99.71%.
Statistical analysis
Statistical analysis was performed in the StatisticalDepartment at the Complutense University of Madrid.The subjects of each group were distributed in 2 sub-groups (men and women) to compare sex differences.The Levine test was used to assess the equality of thegroups’ variances. All measurements showed similar var-iances (P .0.05), except for PNS-AD2 in the Class IIIgroup and N-H in the excellent occlusion group.
An independent-means t test, assuming equality ofvariances, was used for all variables except N-H andPNS-AD2. In these cases, an independent-means t testnot assuming equality of variances was applied. Pearsoncorrelations between cranial base and the variables werealso calculated, as well as the amount of varianceexplained in the variables. Statistical significance was es-tablished by using a P value less than 0.05. A factorialanalysis was applied to prove a group-dependantrelationship among the variables, and a discriminantanalysis was designed to obtain a predictive classifica-tion function.
Journal of Orthodontics and Dentofacial Orthopedics
High standard deviations were noted in the parame-ters for adenoid, aerial, and total areas. Since the errormethod showed reliability near 99%, this means greatinterindividual variability. Most subjects in the controlgroup had a skeletal Class I relationship, and all subjectsin the Class III group had a skeletal Class III relationship.
Tables I and II show sexual dimorphism in the groups.Statistically significant differences (P\0.05) were foundin the Class III group for these variables: AD1-Ba, PNS-H,upper pharynx, and adenoidal area. The results suggestthat men with Class III malocclusion have larger adenoidtissue areas than do women. Men also had greater sagit-tal thickness of the upper airway. Nasal fossa and cranialbase were also greater in men but not statistically differ-ent from women.
In the excellent occlusion group, all variables showedsignificant differences except for the upper pharynxdimension. Significant sexual dimorphism was alsofound for AD2-H in this group. Likewise, for the ClassIII malocclusion group, the men had a higher mean valuethan did the women.
Table III shows the Pearson correlations among cra-nial base and the variables in the groups. This tableshows the correlation coefficient (r), the percentage ofvariance explained (r2), and the P value for these corre-lations. In the Class III group, cranial base explained 59%of the changes in lower airway thickness and 61% of thechanges in nasal fossa length. Because we hypothesizedthe existence of different groups in the excellent occlu-sion and the Class III malocclusion groups, a discriminantanalysis was used to identify them. The results of the dis-criminant analysis are given in Tables IV and V.
Figures 3 and 4 show the distribution of the 13variables represented by the discriminant analysis. Weremoved the 3 surface variable representations from
American Journal of Orthodontics and Dentofacial Orthoped
these figures because of their high dispersion out ofthe figures.
Five groups in the excellent occlusion group wereidentified, whereas only 3 groups in the Class IIImalocclusion group were found. These groups werecharacterized by containing several variables with strongstatistical relationships (r values near 1 or �1).
The distribution of variables by groups is as follows.
1. Excellent occlusion: group 1, PNS-AD2, aerialspace, upper pharyngeal space, PNSAD1, andAD2-H; group 2, N-H and PNS-H; group 3, adenoi-dal area, total area, and AD1-Ba; group 4, S-N andPNS-Ba; and group 5, lower pharyngeal dimension.
2. Class III malocclusion: group 1, AD1-Ba, AD2-H,PNS-H, PNS-Ba, S-N, and adenoidal area; group2, aerial space, PNS-AD1, total area, and lower pha-ryngeal dimension; group 3, upper pharyngealspace, PNS-AD2, and N-H.
Table VI shows statistically significant differences(P\0.05) between the ideal and Class III malocclusionsamples for the following variables: PSN-AD1,AD1-BA, PNS-AD2, AD2-H, PNS-H, N-H, S-N, and lowerpharyngeal dimension. The interpretation of thesedifferences is in the discussion.
Table VII shows the canonical functions and coeffi-cients of the discriminant analysis. This last analysiswas performed to obtain a predictive equation to classifypatients with Class III malocclusion and ideal occlusionbased on nasopharyngeal measurements.
The following discriminant function was obtainedfrom this analysis: 0.252 3 N-H � 0.189 3 S-N �6.501 5 0.
Two methods were used to validate the formula: em-pirical and cross validation. The validation methods sug-gested reliability of 99.4%.
ics February 2011 � Vol 139 � Issue 2
Table III. Correlations among cranial base and airwayvariables
Variable
Excellent occlusion Class III malocclusion
S-N S-NPNS-AD1P value 0.017 \0.01% explanation 5 22.94r value 0.24 0.48
AD1-BaP value 0.0001 \0.01% explanation 15 38.07r value 0.39 0.62
PNS-BaP value \0.0001 \0.01% explanation 33 59.75r value 0.58 0.77
PNS-AD2P value NS \0.01% explanation 13.69r value 0.37
AD2-HP value NS NS% explanationr value
PNS-HP value 0.0008 \0.01% explanation 11 51.12r value 0.34 0.72
N-HP value \0.0001 \0.01% explanation 47 61.15r value 0.69 0.78
S-NP value% explanation 100 100.00r value 1 1
Lower pharynxP value NS \0.05% explanation 7.67r value 0.28
Upper pharynxP value NS \0.01% explanation 20.43r value 0.45
Total areaP value 0.0002 \0.01% explanation 14 25.20r value 0.38 0.50
Aerial areaP value 0.0071 \0.01% explanation 7 10.82r value 0.28 0.33
Adenoidal areaP value NS \0.01% explanation 17.56r value 0.42
Figure 5 shows the distribution of the 162 subjects inthe study and the predictive reliability of the discrimi-nant function.
Depending on sex, 2 more formulas were obtained,but the reliability values of cross validation decreasedto 98.9% for men and 98.6 for women. These formulasare the following: men, 0.2843 N-H � 0.2023 S-N �8.7915 0; and women, 0.2483 N-H� 0.1513 S-N�8.232 5 0.
DISCUSSION
In this study, we tried to establish the characteristicsof the nasopharyngeal airway in patients with Class IIImalocclusion and to compare them with a control grouphaving excellent occlusion. Although we located similarstudies, none included occlusion as a selection crite-rion.13,14,20,26
As has been verified, excellent occlusion prevails insubjects with skeletal Class I, although it can appear inthose with skeletal Class II or Class III.21 In contrast to
Journal of Orthodontics and Dentofacial Orthopedics
Fig 3. Excellent occlusion factorial data analysis.
Martin, Muelas, and Vi~nas 247
other studies, we did not exclude subjects with mildClass II and Class III if they had ideal occlusions. Previousscientific studies tried to identify the relationshipbetween airway obstruction and maxillary dysplasia.Because of this, most studies are based on skeletal rela-tionships and not on occlusion.27 Therefore, we believethat a person with an ideal nasopharynx can developsome dentoalveolar compensatory mechanisms to ob-tain an ideal occlusion with mild maxillary dysplasia.An occlusal classification, instead of another based onskeletal relationship, eliminates the risk of classifyinga subject with mild skeletal Class III with an ideal naso-pharynx and excellent occlusion in the Class III group.The description of the morphologic characteristics ofideal or Class III nasopharynx is then more reliable. Clin-ical studies using similar measurements did not include
American Journal of Orthodontics and Dentofacial Orthoped
excellent occlusion as a selection criterion. Furthermore,the types of malocclusion or skeletal classification of thesample were not defined.27,28 These studies tried tofind relationships between different measurements butnot to define specific airway characteristics. Otherstudies with similar measurements tried to analyzemorphologic differences in monozygotic twins.7,29
The aim was to prove that environmental factorsinfluence craniofacial skeletal development, but theairway characteristics were not defined. The conclusionwas that an exogenous factor might explain whythese genetically identical patients had differentphenotypes.
Concerning the Class III nasopharyngeal morphol-ogy, we found significant sex differences in 4 variablesin contrast to excellent occlusion, where differences
ics February 2011 � Vol 139 � Issue 2
Fig 4. Class III malocclusion factorial data analysis.
248 Martin, Muelas, and Vi~nas
were found in 7 variables. Similar to ideal occlusion, themean values were greater in men. These results suggestthat subjects with Class III malocclusion show less sexualdimorphism than do those with excellent occlusion. Tak-ing into account the differences between those with ex-cellent occlusion and Class III malocclusion, we observedthat characteristics of a Class III nasopharyngeal airwayare not only a more constricted upper airway, but alsoless sexual dimorphism.
February 2011 � Vol 139 � Issue 2 American
In general, the following nasopharyngeal differencesbetween subjects with Class I occlusion and Class III mal-occlusion were found: (1) aerial space is decreased inClass III subjects, (2) the anterior cranial base dimensionand the nasal fossa length are decreased in Class IIIsubjects; (3) McNamara’s lower pharynx dimension22 isincreased in Class III subjects; and (4) on the upperpharynx, the narrowest area is located at AD2 in bothgroups.
Journal of Orthodontics and Dentofacial Orthopedics
Table VI. Sexual differences
Excellent occlusion Class III malocclusion
Mean SD SE Mean SD SE t test P valuePNS-AD1 28.15 3.50 0.37 25.85 4.13 0.49 S \0.001AD1-Ba 20.69 3.08 0.32 22.29 4.08 0.48 S 0.007PNS-Ba 48.80 3.50 0.37 48.14 5.81 0.69 NS 0.398PNS-AD2 25.63 3.94 0.41 22.23 5.41 0.64 S \0.001AD2-H 11.05 3.77 0.40 8.21 3.41 0.41 S \0.001PNS-H 36.68 3.05 0.32 29.86 4.42 0.52 S \0.001N-H 89.33 5.16 0.54 68.61 6.71 0.80 S \0.001S-N 73.85 3.97 0.42 70.86 6.40 0.76 S 0.001Lower pharynx 11.00 3.67 0.38 12.60 3.35 0.40 S 0.005Upper pharynx 17.35 3.59 0.38 17.92 4.71 0.56 NS 0.400Total area 748.10 128.01 13.42 732.90 159.89 18.97 NS 0.514Aerial area 470.57 107.69 11.29 441.06 111.80 13.27 NS 0.091Adenoidal area 277.53 108.58 11.38 291.84 103.96 12.34 NS 0.398
Canonic discriminant function standarized coefficients
Function1
N-H 1.483S-N �.978
Structure matrix
Function1
N-H 0.756S-N 0.124
Discriminant canonic function coefficients
Function1
N-H 0.252S-N �0.189Constant �6.501
Df, degrees of freedom.
Martin, Muelas, and Vi~nas 249
Analyzing differences between excellent occlusionsand Class III malocclusions, we agreed with Trotmanet al9 and Athanasiou et al30 that the lower pharyngealdimension is increased in Class III subjects, perhaps be-cause of the more advanced position of the hyoid
American Journal of Orthodontics and Dentofacial Orthoped
bone. We also observed that the upper airway dimensionis decreased in Class III subjects. To explain this, we mustremember that an increased cranial deflection is a com-mon characteristic of Class III patients. A more angulatedcranial base determines a more advanced glenoid fossaposition (expressed as a more advanced mandible) anda more retruded maxilla.31,32
Similar to the findings of Steinberg and Fraser31 andRobertson,33 we found a significant relationship be-tween the cranial base and some nasopharyngeal vari-ables such as nasal fossa length and lower airwaythickness. This relationship is more evident in thosewith Class III malocclusion. Our study also agrees withthese authors about the lower dimension of the cranialbase length in Class III subjects. These authors concludedthat patients with obstructive diseases tend to havea smaller cranial base dimensions and a more acute cra-nial base angle. They also concluded that patients witha Class III profile tend to have a more acute cranialbase flexure angle than those with Class I or Class II.Although the authors did not describe the nasopharyn-geal airway characteristics, this could suggest that sub-jects with Class III malocclusion tend to have a moreconstricted upper airway.
The correlation between variables depending onthe occlusion is more intuitive if we compare the dis-criminant analysis results for the excellent occlusionand Class III malocclusion subjects. The discrimiantanalysis identified 5 groups of variables for excellentocclusion and only 3 groups for Class III malocclusion.We decided to study this behavior by pooling the datainstead of using a sex grouping because of the min-imal sexual dimorphism observed in the subjectswith Class III malocclusion. Comparing the groupcompositions between excellent occlusion and Class
ics February 2011 � Vol 139 � Issue 2
Fig 5. Discriminant equation.
250 Martin, Muelas, and Vi~nas
III malocclusion, we observed that the group with S-Nand PNS-Ba in excellent occlusion does not exist inClass III. Instead, these 2 variables are included ina group with the variables mainly representing the na-sopharyngeal soft tissues. This means that the depen-dence of nasopharyngeal soft tissues with the cranialbase dimension (S-N) and the position of the maxillarelated to the cranial base (PNS-Ba) is higher in theClass III group than in the excellent occlusion group.For the moment, no cause-and-effect relationship canbe demonstrated—only a strong relationship of depen-dence between skeletal structures and soft tissues inthose with Class III malocclusion.
Finally, discriminant analysis offers a new equation toidentify an ideal occlusion or a Class III malocclusion withreliability of 99.4% based on easy-to-define structuressuch as anterior cranial base and nasal fossa length. Thisequation is as follows: 0.252 3 N-H � 0.189 3 S-N �6.5015 0.
Two equations were also found according to sex, buttheir reliability was lower than that above. Taking intoaccount the slight sexual differences in Class III maloc-clusion (differences between N-H or S-N were not statis-tically significant), we decided to use this one as themain formula when classifying patients.
These findings might help orthodontists to screenpatients for possible respiratory disorders when aneasy-to-identify structure, such as cranial base or nasalfossa, is altered.
February 2011 � Vol 139 � Issue 2 American
Although a 2-dimensional study is not the best wayto make an airway screening, cephalogram recordingsare systematic, are not taken in a hospital setting, arerelatively cheap, and thus might help orthodontists toevaluate the airway as a first approximation. In thisstudy, we opened a new line of investigation that couldpoint to a relationship between skeletal and dentalanomalies with airway obstruction and some possiblespecific respiratory characteristics for each type ofmalocclusion.
CONCLUSIONS
1. Cranial base length is statistically related to severalairway variables, mainly nasal fossa length andlower airway thickness. This relationship is strongerin Class III patients.
2. Airway variables have a more dependent relation-ship with skeletal characteristics in those with ClassIII malocclusion, mainly cranial base morphologyand maxilla position.
3. Upper airway thickness is greater in thosewith ideal oc-clusions than in Class III patients, in contrast to lowerpharynxdimension,which is greater inClass III patients.
4. Although upper airway thickness is greater in thosewith ideal occlusion, adenoidal tissue is a larger areathan in Class III patients.
5. Patients with Class III malocclusion show less sexualdimorphism than do those with excellent occlusion.
Journal of Orthodontics and Dentofacial Orthopedics
Martin, Muelas, and Vi~nas 251
6. No significant differences in lower airway thicknesswere found. However, the Class III group showeda statistically significant constricted airway at thisregion because of reduced aerial thickness andgreater adenoidal tissues.
7. A predictive equation to classify patients based ontheir airway has been identified: 0.252 3 N-H �0.189 3 S-N � 6.501 5 0.
REFERENCES
1. Tomes CS. On the developmental origin of the V-shaped con-tracted maxilla. Monthly Rev Dent Surg 1872;1:2-5.
2. Harvold EP, Chierici G, Vargervik K. Experiments on thedevelopment of dental malocclusions. Am J Orthod 1972;61:38-44.
3. Linder-Aronson S. Effects of adenoidectomy on mode of breath-ing, size of adenoids and nasal airflow. ORL J OtorhinolaryngolRelat Spec 1973;35:283-302.
5. Hellsing E, Hagberg C. Changes in maximum bite force related toextension of the head. Eur J Orthod 1990;12:148-53.
6. Behlfelt K, Linder-Aronson S, Neander P. Posture of the head, thehyoid bone, and the tongue in children with and without enlargedtonsils. Eur J Orthod 1990;12:458-67.
7. Font A. Obstrucci�on nasofar�ıngea y crecimiento facial. Rev EspOrtod 1998;28:137-52.
8. Yamada T, Tanne K, Miyamoto K, Yamauchi K. Influences of nasalrespiratory obstruction on craniofacial growth in young Macacafuscatamonkeys. Am J Orthod Dentofacial Orthop 1997;111:38-43.
9. Trotman CA,McNamara JA Jr, Dibbets JM, van der Weele LT. Asso-ciation of lip posture and the dimensions of the tonsils and sagittalairway with facial morphology. Angle Orthod 1997;67:425-32.
10. Ung N, Koenig J, Shapiro PA, Shapiro G, Trask G. A quantita-tive assessment of respiratory patterns and their effects ondentofacial development. Am J Orthod Dentofacial Orthop1990;98:523-32.
11. Ellingsen R, Vandevanter C, Shapiro P, Shapiro G. Temporal varia-tion in nasal and oral breathing in children. Am J Orthod Dentofa-cial Orthop 1995;107:411-7.
12. Handler SD. Upper airway obstruction in craniofacial anomalies:diagnosis and management. Birth Defects 1985;21:15-31.
13. Kilinc AS, Arslan SG, Kama JD, €Ozer T, Dari O. Effects on thesagittal pharyngeal dimensions of protraction and rapid palatalexpansion in Class III malocclusion subjects. Eur J Orthod 2008;30:61-6.
14. Sayinsu K, Isik F, Arun T. Sagittal airway dimension followingmaxillary protraction: a pilot study. Eur J Orthod 2006;28:184-9.
American Journal of Orthodontics and Dentofacial Orthoped
15. Canut JA, Vila F. Distribuci�on de las maloclusiones en pacientesortodoncicos. Rev Esp Ortod 1975;2:85-92.
16. Wucherpfenning B, Ferre F. Estudio morfol�ogico y evolutivo de laclase III entre los 5 y 10 a~nos. Ortod Esp 1990;31:144-66.
17. Orts J. [thesis] Valencia, Spain: University of Valencia; 1983.18. Asensi C. Estudio morfol�ogico sobre los prognatismos mandibu-
19. Carol Murillo J, Alvarez Sanchez MT. Epidemiologia de las maloc-lusiones en una poblaci�on escolar de Barcelona. Rev Esp Ortod1987;28:18-26.
20. Samman N, Mohammadi H, Xia J. Cephalometric norms for theupper airway in a healthy Hong Kong Chinese population. HongKong Med J 2003;9:25-30.
21. Martin O, Muelas L, Vi~nas MJ. Nasopharyngeal cephalometricstudy of ideal occlusions. Am J Orthod Dentofacial Orthop 2006;130:436.e1-9.
22. McNamara JA. A method of cephalometric evaluation. Am JOrthod 1984;86:449-69.
23. Handelman CS, Osborne G. Growth of the nasopharynx andadenoid development from one to eighteen years. Angle Orthod1976;46:243-59.
24. Dahlberg G. Statistical methods for medical and biological stu-dents. New York: Interscience Publications; 1940.
25. HoustonWJB. The analysis of errors in orthodontic measurements.Am J Orthod 1983;83:382-90.
26. Abu Alhaija ES, Al-Khateeb SN. Uvulo-glosso-pharyngeal dimen-sions in different anteroposterior skeletal patterns. Angle Orthod2005;75:1012-8.
27. Sanz MP, Sanz MA, Herrero MP. Relaci�on de la forma del �angulogon�ıaco con la base craneal y el espacio a�ereo. Ortod Esp 2005;45:26-35.
28. Rubio JA. Las adenoides: su repercusi�on cl�ınica. Ortod Esp 1990;31:84-96.
29. Baca A. Relaci�on entre la obstrucci�on respiratoria nasal ymorfolog�ıa facial. Un estudio radiogr�afico sobre 33 parejas de ge-melos id�enticos. Ortod Esp 1989;30:220-7.
30. Athanasiou AE, Papadopoulos MA, Mazaheri M, Lagoudakis M.Cephalometric evaluation of pharynx, soft palate, adenoid tissue,tongue, and hyoid bone following the use of a mandibular reposi-tioning appliance in obstructive sleep apnea patients. Int J AdultOrthod Orthognath Surg 1994;9:273-83.
31. Steinberg B, Fraser B. Cranial base in obstructive sleep apnea.J Oral Maxillofac Surg 1995;53:1150-4.
32. Pracharktam N, Hans MG, Strohl KP, Redline S. Upright and supinecephalometric evaluation of obstructive sleep apnea syndrome andsnoring subjects. Angle Orthod 1994;64:63-73.
33. Robertson C. Cranial base considerations between apnoeics andnon-apnoeic snorers, and associated effects of long-termmandib-ular advancement on condylar and natural head position. Eur JOrthod 2002;24:353-61.
