Assessment of sagittal skeletal discrepancy: A comparative cephalometric analysis of linear and angular parameters Authors: Ajit Vikram Parihar, A Kavin Prasanth, Akansha Pandey, Dishant Mahajan, Mohana Nethivalavan, Sarita Parihar Published: 2025 ABSTRACT INTRODUCTION:Cephalometrics is the most frequently utilized clinical tool for assessing sagittal jaw relationships. There have been numerous linear and angular measures employed to diagnose these anteroposterior discrepancies. Each of the methods demonstrates its intrinsic characteristics based on the variation of variables besides the jaw relationship itself. The objective of this study is to make a comparative evaluation of 23 parameters, which includes linear and angular measurements of sagittal discrepancy in detail.MATERIALS AND METHODS:About 150 lateral cephalograms were gathered from subjects obtained in natural head position. This comparative study evaluated the mean and standard deviation, the reliability, and the correlation between various parameters and their sensitivity and specificity were calculated.RESULTS AND CONCLUSION:All the measurements are statistically highly significant in determining the sagittal skeletal discrepancy. Recent angles like Beta angle, Yen angle, and Pi analysis have lots of accuracy and consistency associated with them. However, the human population exhibits significant variability, making it impossible for a single cephalometric examination to reliably determine true skeletal relationships in every circumstance. Therefore, the spectrum of cephalometric analysis must be understood by clinicians so that it can be used appropriately according to the patient’s needs. Introduction Cephalometrics is the most frequently utilized clinical tool for assessing sagittal jaw relationships. In the diagnosis and planning of orthodontic treatment, the sagittal relationship is of utmost concern in determining the specific malocclusion of the patient. Previously, the patient’s overall profile view and palpation of the anterior surfaces of the basal part of the jaw with occluded teeth were used clinically to examine the skeletal pattern.[1] However, after the introduction of cephalometrics by Broadbent and Hofrath in 1931, several linear and angular parameters have been proposed to diagnose the skeletal pattern of the patient. These analyses have both merits and demerits.[23] The benefit of employing linear measurements is their enhanced accuracy and reduced measurement errors, which could be attributed to the presence of only two reference points, as opposed to angular measurements, which include three reference points. Although angular measurements offer a wider range of coverage, variation in facial height and jaw proclination can alter them. Linear measurements are affected by the inclination of reference lines.[45] The initially introduced parameters utilized reference planes, such as the SN (Sella-Nasion) plane and Frankfort horizontal plane, and reference points, including point A, point B, and nasion. The AP (Anterio-posterior) jaw relationship within the dentofacial complex cannot be estimated reliably using cranial reference planes, such as the FH plane and S-N (Sella-Nasion) line related to the cranium.[6] Points A and B are affected by growth and tooth movement. The stability of the nasion has been questioned as it does not remain stable with age[7]; hence, the concept of using the occlusal[78] and palatal planes[9] as reference planes came into effect. Nonetheless, dental development, tooth eruption, and orthodontic tooth movement affect the occlusal plane.[1011] Similarly, the palatal plane inclination can vary significantly, influencing the determination of the true skeletal relationship.[1011] Nanda and Merrill[91213] introduced the points G and M to represent the mandible and maxilla, respectively. Unlike points A and B, these are the least affected by local remodeling resulting from dental movements. These landmarks paved the way for developing newer cephalometric analyses, such as the Yen angle,[12] W angle,[13] and Pi analysis,[14] which are independent of cranial reference planes and dental occlusion. The reference plane used in the measurement of Pi analysis is the true horizontal, which is a line perpendicular to the true vertical obtained in the natural head position.[14] This plane is not affected by the rotation of the jaws and their inclination. However, despite the introduction of newer angles, the orthodontist is still faced with a dilemma in determining the optimal approach for evaluating the sagittal relationship in different patients. Therefore, the objectives of this study were to obtain the following: The mean and standard deviation of 23 parameters [as discussed in Table 1] of sagittal skeletal discrepancy in class I, class II, and class III individuals Table 1Mean and standard deviation, ANOVA value, and Tukey’s analysis of all parameters for Class I, II, and IIIParameterClass IClass IIClass IIIANOVA value P Tukey’s analysisANB angle2.7±1.06.6±1.4−3.7±1.7675.77<0.0001 (HS)2>1>3Wylies method−4.9±3.5−9.3±5.45.5±3.7156.12<0.0001 (HS)3>1>2AB plane angle−5.0±2.5−10.4±5.36.1±2.3263.69<0.0001 (HS)3>1>2Jenkins plane−0.4±1.9−6.1±3.97.5±3.2232.69<0.0001 (HS)3>1>2AB’ distance8.0±2.613.2±2.90.6±3.2231.35<0.0001 (HS)2>1>3AXD angle8.9±2.113.5±3.22.5±3.0187.95<0.0001 (HS)2>1>3AD’ distance12.9±3.118.7±4.85.2±3.6148.76<0.0001 (HS)2>1>3Wits appraisal0±1.34.2±2.7−6.8±2.6288.43<0.0001 (HS)2>1>3APDI81.3±4.572.0±6.997.7±3.9303.16<0.0001 (HS)3>1>2AXB angle4.4±2.210.4±3.9−1.0±2.6179.33<0.0001 (HS)2>1>3JYD angle5.3±1.910.6±3.8−0.3±2.3190.84<0.0001 (HS)2>1>3AF BF distance4.1±2.310.1±4.9−2.4±3.3143.17<0.0001 (HS)2>1>3APP BPP distance4.3±2.59.5±4.2−2.3±3.0158.40<0.0001 (HS)2>1>3FABA83.3±4.074.1±8.599.4±5.7201.30<0.0001 (HS)3>1>2Beta angle31.8±3.123.3±3.643.3±3.5275.57<0.0001 (HS)3>1>2Mu angle18.1±4.111.6±5.931.8±4.6215.78<0.0001 (HS)3>1>2Overjet3.3±2.16.8±3.2−1.3±1.7140.84<0.0001 (HS)2>1>3Yen angle122.6±3.7111.9±7.3134.6±3.5240.06<0.0001 (HS)3>1>2Dentoskeletal overjet2.7±1.77.5±3.2−3.1±1.6259.41<0.0001 (HS)2>1>3W angle55.1±5.849.6±10.463.5±2.5418.13<0.0001 (HS)3>1>2Pi linear1.5±2.17.9±2.1−5.5±1.9521.16<0.0001 (HS)2>1>3Pi angle1.8±2.610.0±3.1−6.9±2.4477.35<0.0001 (HS)2>1>3Yen linear3.0±1.95.4±2.7−9.6±3.3427.53<0.0001 (HS)2>1>3*P<0.001 considered highly significant. HS=Highly significant; ANOVA: Analysis of variance The reliability of these parameters in determining the sagittal skeletal relationshipComparative assessment of these parameters to evaluate the sagittal maxillary–mandibular discrepancyThe correlation between these parameters in diagnosing sagittal skeletal discrepancyThe sensitivity and specificity of various parameters in predicting the sagittal skeletal relationships The mean and standard deviation of 23 parameters [as discussed in Table 1] of sagittal skeletal discrepancy in class I, class II, and class III individuals Table 1Mean and standard deviation, ANOVA value, and Tukey’s analysis of all parameters for Class I, II, and IIIParameterClass IClass IIClass IIIANOVA value P Tukey’s analysisANB angle2.7±1.06.6±1.4−3.7±1.7675.77<0.0001 (HS)2>1>3Wylies method−4.9±3.5−9.3±5.45.5±3.7156.12<0.0001 (HS)3>1>2AB plane angle−5.0±2.5−10.4±5.36.1±2.3263.69<0.0001 (HS)3>1>2Jenkins plane−0.4±1.9−6.1±3.97.5±3.2232.69<0.0001 (HS)3>1>2AB’ distance8.0±2.613.2±2.90.6±3.2231.35<0.0001 (HS)2>1>3AXD angle8.9±2.113.5±3.22.5±3.0187.95<0.0001 (HS)2>1>3AD’ distance12.9±3.118.7±4.85.2±3.6148.76<0.0001 (HS)2>1>3Wits appraisal0±1.34.2±2.7−6.8±2.6288.43<0.0001 (HS)2>1>3APDI81.3±4.572.0±6.997.7±3.9303.16<0.0001 (HS)3>1>2AXB angle4.4±2.210.4±3.9−1.0±2.6179.33<0.0001 (HS)2>1>3JYD angle5.3±1.910.6±3.8−0.3±2.3190.84<0.0001 (HS)2>1>3AF BF distance4.1±2.310.1±4.9−2.4±3.3143.17<0.0001 (HS)2>1>3APP BPP distance4.3±2.59.5±4.2−2.3±3.0158.40<0.0001 (HS)2>1>3FABA83.3±4.074.1±8.599.4±5.7201.30<0.0001 (HS)3>1>2Beta angle31.8±3.123.3±3.643.3±3.5275.57<0.0001 (HS)3>1>2Mu angle18.1±4.111.6±5.931.8±4.6215.78<0.0001 (HS)3>1>2Overjet3.3±2.16.8±3.2−1.3±1.7140.84<0.0001 (HS)2>1>3Yen angle122.6±3.7111.9±7.3134.6±3.5240.06<0.0001 (HS)3>1>2Dentoskeletal overjet2.7±1.77.5±3.2−3.1±1.6259.41<0.0001 (HS)2>1>3W angle55.1±5.849.6±10.463.5±2.5418.13<0.0001 (HS)3>1>2Pi linear1.5±2.17.9±2.1−5.5±1.9521.16<0.0001 (HS)2>1>3Pi angle1.8±2.610.0±3.1−6.9±2.4477.35<0.0001 (HS)2>1>3Yen linear3.0±1.95.4±2.7−9.6±3.3427.53<0.0001 (HS)2>1>3*P<0.001 considered highly significant. HS=Highly significant; ANOVA: Analysis of variance The mean and standard deviation of 23 parameters [as discussed in Table 1] of sagittal skeletal discrepancy in class I, class II, and class III individuals Table 1Mean and standard deviation, ANOVA value, and Tukey’s analysis of all parameters for Class I, II, and IIIParameterClass IClass IIClass IIIANOVA value P Tukey’s analysisANB angle2.7±1.06.6±1.4−3.7±1.7675.77<0.0001 (HS)2>1>3Wylies method−4.9±3.5−9.3±5.45.5±3.7156.12<0.0001 (HS)3>1>2AB plane angle−5.0±2.5−10.4±5.36.1±2.3263.69<0.0001 (HS)3>1>2Jenkins plane−0.4±1.9−6.1±3.97.5±3.2232.69<0.0001 (HS)3>1>2AB’ distance8.0±2.613.2±2.90.6±3.2231.35<0.0001 (HS)2>1>3AXD angle8.9±2.113.5±3.22.5±3.0187.95<0.0001 (HS)2>1>3AD’ distance12.9±3.118.7±4.85.2±3.6148.76<0.0001 (HS)2>1>3Wits appraisal0±1.34.2±2.7−6.8±2.6288.43<0.0001 (HS)2>1>3APDI81.3±4.572.0±6.997.7±3.9303.16<0.0001 (HS)3>1>2AXB angle4.4±2.210.4±3.9−1.0±2.6179.33<0.0001 (HS)2>1>3JYD angle5.3±1.910.6±3.8−0.3±2.3190.84<0.0001 (HS)2>1>3AF BF distance4.1±2.310.1±4.9−2.4±3.3143.17<0.0001 (HS)2>1>3APP BPP distance4.3±2.59.5±4.2−2.3±3.0158.40<0.0001 (HS)2>1>3FABA83.3±4.074.1±8.599.4±5.7201.30<0.0001 (HS)3>1>2Beta angle31.8±3.123.3±3.643.3±3.5275.57<0.0001 (HS)3>1>2Mu angle18.1±4.111.6±5.931.8±4.6215.78<0.0001 (HS)3>1>2Overjet3.3±2.16.8±3.2−1.3±1.7140.84<0.0001 (HS)2>1>3Yen angle122.6±3.7111.9±7.3134.6±3.5240.06<0.0001 (HS)3>1>2Dentoskeletal overjet2.7±1.77.5±3.2−3.1±1.6259.41<0.0001 (HS)2>1>3W angle55.1±5.849.6±10.463.5±2.5418.13<0.0001 (HS)3>1>2Pi linear1.5±2.17.9±2.1−5.5±1.9521.16<0.0001 (HS)2>1>3Pi angle1.8±2.610.0±3.1−6.9±2.4477.35<0.0001 (HS)2>1>3Yen linear3.0±1.95.4±2.7−9.6±3.3427.53<0.0001 (HS)2>1>3*P<0.001 considered highly significant. HS=Highly significant; ANOVA: Analysis of variance Table 1Mean and standard deviation, ANOVA value, and Tukey’s analysis of all parameters for Class I, II, and IIIParameterClass IClass IIClass IIIANOVA value P Tukey’s analysisANB angle2.7±1.06.6±1.4−3.7±1.7675.77<0.0001 (HS)2>1>3Wylies method−4.9±3.5−9.3±5.45.5±3.7156.12<0.0001 (HS)3>1>2AB plane angle−5.0±2.5−10.4±5.36.1±2.3263.69<0.0001 (HS)3>1>2Jenkins plane−0.4±1.9−6.1±3.97.5±3.2232.69<0.0001 (HS)3>1>2AB’ distance8.0±2.613.2±2.90.6±3.2231.35<0.0001 (HS)2>1>3AXD angle8.9±2.113.5±3.22.5±3.0187.95<0.0001 (HS)2>1>3AD’ distance12.9±3.118.7±4.85.2±3.6148.76<0.0001 (HS)2>1>3Wits appraisal0±1.34.2±2.7−6.8±2.6288.43<0.0001 (HS)2>1>3APDI81.3±4.572.0±6.997.7±3.9303.16<0.0001 (HS)3>1>2AXB angle4.4±2.210.4±3.9−1.0±2.6179.33<0.0001 (HS)2>1>3JYD angle5.3±1.910.6±3.8−0.3±2.3190.84<0.0001 (HS)2>1>3AF BF distance4.1±2.310.1±4.9−2.4±3.3143.17<0.0001 (HS)2>1>3APP BPP distance4.3±2.59.5±4.2−2.3±3.0158.40<0.0001 (HS)2>1>3FABA83.3±4.074.1±8.599.4±5.7201.30<0.0001 (HS)3>1>2Beta angle31.8±3.123.3±3.643.3±3.5275.57<0.0001 (HS)3>1>2Mu angle18.1±4.111.6±5.931.8±4.6215.78<0.0001 (HS)3>1>2Overjet3.3±2.16.8±3.2−1.3±1.7140.84<0.0001 (HS)2>1>3Yen angle122.6±3.7111.9±7.3134.6±3.5240.06<0.0001 (HS)3>1>2Dentoskeletal overjet2.7±1.77.5±3.2−3.1±1.6259.41<0.0001 (HS)2>1>3W angle55.1±5.849.6±10.463.5±2.5418.13<0.0001 (HS)3>1>2Pi linear1.5±2.17.9±2.1−5.5±1.9521.16<0.0001 (HS)2>1>3Pi angle1.8±2.610.0±3.1−6.9±2.4477.35<0.0001 (HS)2>1>3Yen linear3.0±1.95.4±2.7−9.6±3.3427.53<0.0001 (HS)2>1>3*P<0.001 considered highly significant. HS=Highly significant; ANOVA: Analysis of variance Mean and standard deviation, ANOVA value, and Tukey’s analysis of all parameters for Class I, II, and III *P<0.001 considered highly significant. HS=Highly significant; ANOVA: Analysis of variance The reliability of these parameters in determining the sagittal skeletal relationship The reliability of these parameters in determining the sagittal skeletal relationship Comparative assessment of these parameters to evaluate the sagittal maxillary–mandibular discrepancy Comparative assessment of these parameters to evaluate the sagittal maxillary–mandibular discrepancy The correlation between these parameters in diagnosing sagittal skeletal discrepancy The correlation between these parameters in diagnosing sagittal skeletal discrepancy The sensitivity and specificity of various parameters in predicting the sagittal skeletal relationships The sensitivity and specificity of various parameters in predicting the sagittal skeletal relationships Materials and Methods The study sample comprised 150 lateral cephalograms, which included 50 samples each of Class I, Class II, and Class III malocclusion. The participants were the xyz population of the abc region. A pilot study was conducted using the Wits appraisal on 10 cases of each group (Class I = 10, Class II = 10, and Class III = 10) to determine the sample size. The mean value of each group was calculated, and the effect size obtained was 0.643. Using the effect size and number of groups as input in the G*Power 3.1.9.2 software, the sample size was determined to be 42 samples per group. The sample size was then rounded up to 50 samples per group to avoid unexpected errors. The inclusion criteria were (1) Patients aged 15–30 years, (2) First permanent molar teeth fully erupted, (3) Clinically healthy samples, and (4) Symmetrical face with no transverse problems. The exclusion criteria were (1) Prior orthodontic treatment history, (2) Cranial or facial malformations or congenital defects, and (3) Poor quality of cephalograms. Pretreatment cephalograms were acquired using the X-ray machine (Epsilon Health Care Solutions, Jaipur, India). The patient was instructed to stand in an upright posture, with the head straight, and all lateral cephalometric radiographs were obtained in centric occlusion, with the head in its natural posture and the lips in repose. The overjet was measured from the study cast of the patient as the horizontal overlap of the most prominent incisors, using a probe horizontally. The reference points and planes utilized in the study are depicted in Figures 1 and 2. All linear and angular measurements used in the study are illustrated in Figures 3–23. A single operator performed all the tracings to prevent inter-operator variations. Reference points Reference planes Wylies method. 5 points were projected to Frankfort horizontal plane and 5 dimensions were measured from these points AB plane angle. The angle formed between the lines AB and N-Pog ANB angle. The angle formed between the lines NA and NB Jenkins A plane.’a’ plane drawn at right angles to occlusal plane on point A and distance from point B measured A-B’ distance. The distance of point A from B’ line AXD angle. Point X drawn perpendicular from point A to SN plane. The angle between AX and XD forms AXD angle A-D’ distance. The distance from point A to D’ line Wits appraisal Perpendiculars drawn from points A and B to occlusal plane and the distance between the two lines measured APDI. Combination of three angles.i.e. Facial anglepalatal plane angle ± AB plane AXB angle. Perpendicular line drawn from point A to FH plane which was marked as X. Angle between AX and XB measured JYD angle. Perpendicular from point J to SN plane marked as point Y. Angle between JY and YD measured AF-BF distance. Perpendiculars from point A and B were drawn to FH plane and measured APP-BPP distance. Perpendicular line drawn from point A and B to the palatal plane and the distance measured FABA. The angle between FH plane and AB plane Beta angle. Angle between line from point A perpendicular to C-B line and A-B line Mu angle. Angle formed between AB line and perpendicular line from point A to mandibular plane Yen angle. The angle formed between lines GM and SM W angle. The angle formed between the perpendicular line drawn from point M to SG line and MG line Pi analysis. Pi angle was obtained between G’G and G’M. Pi linear was the distance between points G’ and M’ Yen linear. The perpendiculars drawn from points M and G on the functional occlusal plane and the distance measured Dentoskeletal overjet = (NB-point M1) + Overjet – (NA-point M2) This research was a comparative study in which the samples were classified based on the ANB angle (Angle between points A, Nasion, and B), molar relationship, and facial profile into skeletal classes I, II, and III for both males and females. Vertical growth patterns of the samples were measured using parameters such as the palatal plane angle, FMA (Frankfort–Mandibular Plane Angle), and Jarabak ratio. The vertical and horizontal growth patterns masking the true sagittal discrepancy were addressed by applying the modified Wits appraisal[15] to the samples during their selection. Class I included an ANB value of 0–4 and a straight profile with a Class I molar relationship. Class II included an ANB value of >4, a convex facial profile, and a Class II molar relationship. Class III comprised an ANB value of <0, a concave facial profile, and a Class III molar relationship. A total of 50 cases of Class I, Class II, and Class III malocclusions were taken. Statistical analysis Once the values were calculated from the lateral cephalogram, the data were transferred to a Microsoft Excel (Microsoft Corporation, Redmond, WA, USA) sheet. The mean and standard deviation of various linear and angular measurements of sagittal skeletal discrepancy were computed using the SPSS (SPSS Inc., Chicago, IL, USA) Statistics 17.0 software. Descriptive analysis was also performed. The differences among Class I, Class II, and Class III in various parameters were analyzed using the analysis of variance (ANOVA) test, followed by Tukey’s post hoc analysis. The reliability of different measurements was derived using the coefficient of variance (CV). The correlation between various parameters in Class I, Class II, and Class III was analyzed using the Pearson correlation test. The sensitivity and specificity were determined using the receiver operating characteristic (ROC) curve. Results The mean and standard deviation of all 23 parameters are presented in Table 1. The mean values are compared with the original mean values and discussed in the subsequent sections. Tukey’s post hoc analysis followed the ANOVA tests to determine whether any differences existed among the means of the three skeletal groups; the results are presented in Table 1. The reliability of the measurements was tested by calculating the CV, and the values are depicted in Figure 24. The most reliable and least reliable angles were the Yen angle and the Jenkins plane to point B, respectively, which are described individually in the discussion section. CV values of all parameters showing the most and least reliable parameters The correlation between the various parameters was measured with the Pearson correlation test. Only the significant correlations were considered, and their r and P values are listed in Tables 2 and 3. The sensitivity and specificity of various parameters were determined using the ROC curve to differentiate between Class II and Class III [Figure 25 and Table 4], Class II and Class I [Figure 25 and Table 4], and Class III and Class I [Figure 25 and Table 4]. The ROC curve plots the true positive rate (sensitivity) against the false positive rate (specificity) at various cutoff points. The four most sensitive and specific parameters are discussed in Table 4. Receiver operating characteristic for various parameters between three classes. (a) class II and III, (b) class II and class I,(c) class III and class I Table 2P value and r-values of various parameters for different classes (Pearson correlation test)ParameterClass IClass IIClass IIIANB versus Beta angle r −0.313 −0.347 −0.352 P 0.027 0.014 0.012 ANB versus Mu angle r−0.285−0.145−0.126 P0.0450.3160.382ANB versus Yen angle r−0.355−0.615−0.458 P0.0110.0000.001ANB versus W angle r−0.241−0.386−0.352 P0.0920.0060.012ANB versus Pi angle r0.2950.4940.509 P0.0380.0000.000ANB versus Pi linear r0.2910.6760.501 P0.0410.0000.000Beta versus Mu angle r0.6070.8020.775 P0.0000.0000.000Beta versus Yen angle r0.4560.6100.549 P0.0010.0000.000Beta versus W angle r0.2200.2680.427 P0.1250.0600.002Beta versus Pi linear r−0.470−0.332−0.394 P0.0010.0190.005Beta angle versus Pi angle r−0.438−0.283−0.333 P0.0010.0460.018Mu angle versus Yen angle r0.0700.5540.417 P0.6310.0000.003Mu angle versus W angle r0.1860.0650.316 P0.1960.6530.025Mu angle versus Pi angle r−0.3850.012−0.058 P0.0060.9310.687Mu angle versus Pi linear r−0.4180.006−0.107 P0.0030.9660.460Yen angle versus W angle r0.2690.7240.558 P0.0590.0000.000Yen angle versus Pi angle r−0.395−0.472−0.527 P0.0040.0010.000Yen angle versus Pi linear r−0.408−0.524−0.548 P0.0030.0000.000W angle versus Pi angle r−0.387−0.252−0.294 P0.0060.0770.038W angle versus Pi linear r−0.361−0.265−0.292 P0.0100.0630.039Pi angle versus Pi linear r0.9900.8270.969 P0.0000.0000.000*P<0.05 considered significant P value and r-values of various parameters for different classes (Pearson correlation test) *P<0.05 considered significant Table 3Interpretation of the Pearson correlation test of various parametersIncrease inClass IClass IIClass III IncreaseDecreaseIncreaseDecreaseIncreaseDecreaseBeta angleMu angleYen angleANB anglePi anglePi linearW angleMu angleYen angleANB anglePi anglePi linearW angleMu angleYen anglePi anglePi linearANB angleMu angleBeta angleANB anglePi anglePi linearBeta angleYen angleW angleBeta angleYen angleYen angleBeta angleW angleANB anglePi anglePi linearW angleMu angleBeta angleANB anglePi anglePi linearW angleMu angleBeta angleANB anglePi anglePi linearW angleYen angleANB anglePi anglePi linearBeta angleYen angleANB anglePi anglePi linearYen angleMu angleBeta angleANB anglePi anglePi linearPi angleANB anglePi linearBeta angleW angleYen angleMu angleANB anglePi linearW angleYen angleBeta angleANB anglePi linearW angleYen angleBeta anglePi linearANB anglePi angleBeta angleW angleYen angleMu angleANB anglePi angleW angleYen angleBeta angleANB anglePi angleW angleYen angleBeta angle Interpretation of the Pearson correlation test of various parameters Table 4Sensitivity and specificity scores between three classesTo differentiate Class II from Class IIIParameterCutoff scoreAUCSensitivity (%)Specificity (%)Beta angle30.50.98110099.98Mu angle15.50.97810099.84Yen angle1250.98110099.96W angle58.50.97110099.96 To differentiate Class II from Class I Parameter Cutoff score AUC Sensitivity (%) Specificity (%) Wits appraisal1.750.9519098.8Pi linear2.50.98210096.8Pi angle2.50.97710096.64Yen linear4.50.7768699.76 To differentiate Class III from Class I Parameter Cutoff score AUC Sensitivity (%) Specificity (%) Beta angle36.50.9849899.4Mu angle22.80.9809898.84Yen angle127.50.9889899.4W angle58.50.97710099AUC=Area under the receiver operating characteristic curve Sensitivity and specificity scores between three classes AUC=Area under the receiver operating characteristic curve Based on the P value, all measurements, both angular and linear, were highly statistically significant in determining the sagittal skeletal discrepancy. Wits appraisal,[8] Yen angle,[12] Pi angle,[14] Pi linear,[14] and Beta angle[16] were good predictors of Class I, Class II, and Class III skeletal patterns, as their mean and standard deviation values did not coincide with each other. In contrast, overjet[17] was not a good predictor of Class I, Class II, and Class III as its mean and standard deviation values coincided with each other [Table 1]. Wylie’s method, AB plane angle, Jenkins “a” plane to point B, A–B’ distance, AXD angle- the interior angle formed by the intersection of the lines extending from points A and D at point X (X is point of intersection of perpendicular from point A to SN plane), A–D’ distance, APDI (Anteroposterior Dysplasia Indicator, AXB angle-(A perpendicular is constructed from point A to Frankfort Horizontal, establishing point X. A line from points X to B forms angle A-X-B), JYD angle-formed by the intersection of the lines extending from points J and D to point Y, AF–BF distance, APP–BPP distance (perpendicular projection of points A and B to palatal plane), FABA (FH to AB Plane Angle), Mu angle, W angle, dentoskeletal overjet, and Yen linear were good predictors of the Class III pattern but weak in differentiating Class II from Class I as the mean values coincided with each other [Table 1]. Discussion Wylie was the first to propose a quantitative method in 1947 to assess AP dysplasia, which indicates whether the skeletal base is orthognathic or prognathic.[18] The scores can be measured in five different dimensions to identify or locate the dysplasia. According to the present study, it is a good predictor of Class III but weak in differentiating between Class II and Class I. It has a high CV (72.9). The study by Kapoor et al. also showed that it is not a reliable method of determining AP dysplasia [Figure 24].[19] The first angular measurement for anteroposterior dysplasia, AB plane angle, was given by Downs in 1948.[20] The mean values of the present study are similar to the results of Downs and Doshi et al.[2021] This angle is also a good predictor of Class III, but weak in differentiating between Class II and Class I. Four years later, Riedel introduced the ANB angle, which became the most commonly used sagittal relationship parameter.[22] However, both angles are subject to error as the position of the nasion is unstable during growth and prone to displacement. The ANB values obtained in this study are similar to those of earlier studies by Doshi et al., Riedel, Ahmad and Jahjah, and Richardson.[21222324] In contrast, studies by Gu et al. reported slightly higher values owing to demographic variations.[25] In 1954, Jenkins introduced the “a” plane based on a functional viewpoint, using the occlusal plane as the most desirable reference point. According to the present study, this plane is a good predictor of class III but weak in differentiating between Class II and Class I, and it is the least reliable parameter as the CV is high (445.4) [Figure 24]. It is highly variable because of the difficulties or inaccuracies in identifying the occlusal plane and/or variations in it. Taylor, in 1969, stressed the importance of linear rather than angular measurement for relating the apical base and introduced the AB distance.[7] Based on the findings of the present study, it is a good predictor of Class III but weak in differentiating between Class II and Class I. Moreover, the CV is 33.2, which shows less reliability compared with other parameters [Figure 24]. This parameter is used to quantify changes in the apical base before and after treatment. In 1975, Beatty provided both linear and angular measurements, including the AXD angle and AD distance.[26] The mean values of the present study are similar to the original Beatty values.[26] The AXD angle is a good predictor of class III, but weak in differentiating between Class II and Class I. In addition, it has good reliability (AXD angle: 24.1, AD’ distance: 24.2), although less pronounced when compared with other parameters [Figure 24]. The second most extensively used measurement, Wits appraisal, was given by Jacobson in 1975, which evaluated points A and B using the functional occlusal plane as a reference to overcome the problems associated with the ANB angle.[8] The mean values obtained in the present study are aligned with the normal values reported by Jacobson, Sharma et al., Alam et al., and Qamaruddin et al.[8272829] This measurement is a good predictor of the skeletal pattern, whether it is Class I, II, or III, as the mean values of the three groups do not coincide with each other. However, it exhibits a high CV (67.5), indicating its low reliability [Figure 24]. This finding agrees with the studies of Sharma et al., Gul-e-Erum and Fida, and Kim and Vietas.[273031] The disadvantage of this method is that the occlusal plane is influenced by several factors that may affect the result. In 1978, Kim and Vietas proposed the APDI, which combined three angular measurements to achieve a high correlation against molar displacement.[31] The mean values of the present study are similar to those of Kim and Vietas but slightly lower for the Class II pattern when compared with the Doshi et al. study.[2131] The APDI is a good predictor of Class III but weak in differentiating between Class II and Class I. The APDI is highly reliable [Figure 24] when compared with other parameters, as the CV value is 5.5, indicating the most homogeneous and statistically significant results, providing only an anteroposterior dental relationship. Freeman, in 1950, published a paper on the AXB angle, modifying the ANB angle to reflect the effect of maxillary position.[32] The mean values obtained are similar to those of the Freeman study, which is a good predictor of Class III but weak in differentiating between Class II and Class I.[32] It is less reliable as its CV value is 49.1. As the position of the nasion might change horizontally and vertically during growth, it is eliminated, but point B, whose position changes during development, is not eliminated here. Jarvinen, in 1982, introduced the JYD angle, which utilizes the skeletal points J and D instead of points A and B.[33] The mean values obtained are similar to those of the Jarvinen study, which is a good predictor of Class III but weak in differentiating between Class II and Class I.[33] Although this angle has good reliability, with a CV value of 37.4, it is less when compared with other angles [Figure 24]. Jaw rotation and anterior face height affect the JYD angle; however, facial prognathism cannot alter it because the nasion is not used in this calculation. In 1987, Chang proposed the AF–BF distance based on the linear distance between points A and B projected onto the Frankfort horizontal plane.[34] The mean values of the study are similar to those of the Chang study.[34] Although this distance is a good predictor of the Class III pattern, it is weak in differentiating between Class I and Class II cases, as the mean values coincide with each other. The CV value is high (56.8), indicating that it is less reliable when compared with other parameters [Figure 24]. Nanda and Merill introduced the APP–BPP distance using the palatal plane as the reference plane owing to its stability throughout the growth period.[9] The mean values of the present study are similar to those of the Doshi et al. study but slightly lower than the original values of the Nanda and Merill study because of variations in magnifications.[921] This distance is a good predictor of the Class III pattern but weak in differentiating between Class I and Class II cases, as the mean values coincide with each other [Figure 24]. In 1994, Yang and Suhr proposed FABA by calculating the angle between the FH and AB planes.[35] The mean values of the present study are higher than the original Yang and Suhr values, indicating that Indians have more severe skeletal patterns than the White population.[35] FABA is a good predictor of the Class III pattern, but less predictive between Class I and Class II cases. According to the studies of Gul-e-Erum and Kannan[36], FABA is the second most reliable (CV = 4.83) parameter [Figure 24]. Baik and Ververidou, in 2004, introduced a method for evaluating sagittal discrepancies, known as the beta angle.[16] The mean values are per the original Baik and Ververidou values, Bohra et al., and Sara studies.[101637] According to the present study, the beta angle is a good predictor of the skeletal pattern, regardless of whether it is Class I, II, or III, as the mean values of the three groups do not coincide with each other. Moreover, it is highly reliable (CV = 10) and comparable to the Yen and FABA angles [Figure 24]. Based on the ROC curve, it has 98% sensitivity and 98.4% specificity in differentiating between Class III and Class I [Table 4 and Figure 25] and 100% sensitivity and 99.8% specificity in differentiating between Class II and Class III [Table 4 and Figure 25]. Fattahi proposed the mu angle, and the mean values are close to those of the original Fattahi study.[38] According to the present study, it is a good predictor of the Class III pattern but less predictive between Class I and Class II cases, as the mean values coincide with each other. Based on the ROC curve, this angle exhibits 100% sensitivity and 99.8% specificity in differentiating between Class III and Class I [Table 4 and Figure 25] and 98% sensitivity and 98.8% specificity in differentiating between Class II and Class III [Table 4 and Figure 25]. Although it has good reliability, with a CV value of 22.9, it is lower when compared with other angles [Figure 24]. Zupancic et al., in 2008, used overjet as a predictor of the sagittal skeletal relationship; the mean values obtained are consistent with the original Zupancic et al. studies.[17] However, the major drawback is that the measurements were based on the dental inclination of the most anterior teeth, which does not predict the underlying skeletal dysplasia. According to the present study, it is not a good predictor of Class III, Class II, or Class I patterns, as the mean values coincide with each other. The reliability is low because of a high CV value of 62.8 [Figure 24]. In 2009, Neela et al. developed the Yen angle.[12] The mean values of the present study are higher than those of the Neela et al. study.[12] This angle is a good predictor of Class I, Class II, and Class III cases, as the mean values do not coincide with each other. According to Jain et al., it is most reliable, with CV = 7.7, which is supported by the studies of Doshi et al. and Mittal et al.[21394041] According to the present study, the Yen angle seems to be the most reliable as it exhibits the lowest CV value of 3.8 [Figure 24]. Based on the ROC curve, it has 100% sensitivity and 99.6% specificity in differentiating between Class III and Class I [Figure 25] [Table 4] and 98% sensitivity and 99.4% specificity in differentiating between Class II and Class III [Figure 25 and Table 4]. AL-Hammadi, in 2011, introduced the dentoskeletal overjet.[42] The mean values are similar to the original AL-Hammadi values.[42] According to the present study, it is a good predictor of the Class III pattern but weak in differentiating between Class I and Class II cases, as the mean values coincide with each other. It is less reliable as its CV value is 76.2 [Figure 24]. Bhad et al., in 2011, proposed the W angle, whose mean values differed slightly from those of the original Bhad et al. study, which could be attributed to demographic variations and their complex interactions with genetic and environmental factors.[13] According to the present study, it is a good predictor of the Class III pattern but weak in differentiating between Class I and Class II cases, as the mean values coincide with each other. Based on the ROC curve, it has 100% sensitivity and 99% specificity in differentiating between Class III and Class I [Figure 25 and Table 4] and 100% sensitivity and 99.9% specificity in differentiating between Class II and Class III [Figure 25 and Table 4]. Mittal et al. also proved that it has high sensitivity and specificity. It exhibits good reliability as its CV value is 10.6 [Figure 24].[40] Kumar et al., in 2012, gave the Pi analysis, which involves both linear and angular measurements recorded in the natural head position.[14] The mean values are slightly different from the Kumar et al. values owing to demographic variations.[14] According to the present study, it is a good predictor of the skeletal pattern, whether it is Class I, II, or III, as the mean values of the three groups do not coincide with each other. Based on the ROC curve, the Pi angle has 100% sensitivity and 96.6% specificity in differentiating between Class II and Class I, and the Pi linear has 100% sensitivity and 96.8% specificity in differentiating between Class II and Class I [Figure 25 and Table 4]. According to Mittal et al.,[4041] it has 100% sensitivity and specificity. However, according to Jain et al., it is the least reliable because of the high CV value (142), which is also supported by the present study [Figure 24].[39] Shetty et al., in 2013, proposed the Yen linear, with the functional occlusal plane as a reference plane.[43] The mean values are slightly higher than the original values of Shetty et al., which could be ascribed to changes in technical settings and demographic variations.[43] According to the present study, it is a good predictor of the Class III pattern but less predictive between Class I and Class II cases, as the mean values coincide with each other. Based on the ROC curve, the Pi angle has 86% sensitivity and 99.7% specificity in differentiating between Class II and Class I [Figure 25 and Table 4]. Conclusion Based on the P value, all the angular and linear measurements are highly statistically significant, and this study provides a comprehensive assessment of various linear and angular cephalometric measurements to evaluate sagittal skeletal discrepancies, identifying reliable indicators for differentiating between Class I, II, and III skeletal patterns. Key findings Statistical significance of measurements: The study found that all tested cephalometric measurements (angular and linear) were statistically significant for identifying sagittal discrepancies among skeletal classes. Specifically, measurements like the Wits appraisal, Beta angle, Yen angle, Pi angle, and Pi linear are effective in determining skeletal classifications for Class I, II, or III, indicating their strong predictive capabilities.Reliability of individual measurements: The Yen angle showed the highest reliability, with a CV of 3.8, making it the most consistent for distinguishing skeletal patterns. This was followed by the FABA, APDI, Beta angle, and W angle, which also demonstrated reliable predictive qualities. In contrast, the Jenkins plane to point B was identified as the least reliable due to high CV, reflecting its susceptibility to measurement variability and limitations in differentiating between certain classes.Sensitivity and specificity analysis:Beta, Mu, Yen, and W angles: These measurements showed high sensitivity and specificity for distinguishing Class III from Class II skeletal patterns. Further, these angles also demonstrated high effectiveness in differentiating Class III from Class I patterns, enhancing diagnostic precision across all skeletal classes.Pi angle, Pi linear, Yen linear, and Wits appraisal: These metrics had higher sensitivity and specificity in differentiating Class II from Class I patterns. Statistical significance of measurements: The study found that all tested cephalometric measurements (angular and linear) were statistically significant for identifying sagittal discrepancies among skeletal classes. Specifically, measurements like the Wits appraisal, Beta angle, Yen angle, Pi angle, and Pi linear are effective in determining skeletal classifications for Class I, II, or III, indicating their strong predictive capabilities. Statistical significance of measurements: The study found that all tested cephalometric measurements (angular and linear) were statistically significant for identifying sagittal discrepancies among skeletal classes. Specifically, measurements like the Wits appraisal, Beta angle, Yen angle, Pi angle, and Pi linear are effective in determining skeletal classifications for Class I, II, or III, indicating their strong predictive capabilities. Reliability of individual measurements: The Yen angle showed the highest reliability, with a CV of 3.8, making it the most consistent for distinguishing skeletal patterns. This was followed by the FABA, APDI, Beta angle, and W angle, which also demonstrated reliable predictive qualities. In contrast, the Jenkins plane to point B was identified as the least reliable due to high CV, reflecting its susceptibility to measurement variability and limitations in differentiating between certain classes. Reliability of individual measurements: The Yen angle showed the highest reliability, with a CV of 3.8, making it the most consistent for distinguishing skeletal patterns. This was followed by the FABA, APDI, Beta angle, and W angle, which also demonstrated reliable predictive qualities. In contrast, the Jenkins plane to point B was identified as the least reliable due to high CV, reflecting its susceptibility to measurement variability and limitations in differentiating between certain classes. Sensitivity and specificity analysis:Beta, Mu, Yen, and W angles: These measurements showed high sensitivity and specificity for distinguishing Class III from Class II skeletal patterns. Further, these angles also demonstrated high effectiveness in differentiating Class III from Class I patterns, enhancing diagnostic precision across all skeletal classes.Pi angle, Pi linear, Yen linear, and Wits appraisal: These metrics had higher sensitivity and specificity in differentiating Class II from Class I patterns. Sensitivity and specificity analysis: Beta, Mu, Yen, and W angles: These measurements showed high sensitivity and specificity for distinguishing Class III from Class II skeletal patterns. Further, these angles also demonstrated high effectiveness in differentiating Class III from Class I patterns, enhancing diagnostic precision across all skeletal classes. Beta, Mu, Yen, and W angles: These measurements showed high sensitivity and specificity for distinguishing Class III from Class II skeletal patterns. Further, these angles also demonstrated high effectiveness in differentiating Class III from Class I patterns, enhancing diagnostic precision across all skeletal classes. Pi angle, Pi linear, Yen linear, and Wits appraisal: These metrics had higher sensitivity and specificity in differentiating Class II from Class I patterns. Pi angle, Pi linear, Yen linear, and Wits appraisal: These metrics had higher sensitivity and specificity in differentiating Class II from Class I patterns. This is crucial in clinical settings where accuracy in classifying sagittal relationships can influence treatment planning. Limitations of overjet: Overjet measurements, commonly used to assess the horizontal overlap of teeth, were found to be unreliable for distinguishing skeletal classes. This is attributed to the influence of dental factors that may not accurately reflect the underlying skeletal structure.ROC curve analysis: The ROC curve was used to assess each parameter’s true positive rate (sensitivity) against the false positive rate (specificity) at various cutoff points. This analysis confirmed the high diagnostic value of Beta, Mu, Yen, and W angles for Class II versus III and Class III versus I distinctions, as well as Pi angle and Wits appraisal for Class II versus I. Limitations of overjet: Overjet measurements, commonly used to assess the horizontal overlap of teeth, were found to be unreliable for distinguishing skeletal classes. This is attributed to the influence of dental factors that may not accurately reflect the underlying skeletal structure. Limitations of overjet: Overjet measurements, commonly used to assess the horizontal overlap of teeth, were found to be unreliable for distinguishing skeletal classes. This is attributed to the influence of dental factors that may not accurately reflect the underlying skeletal structure. ROC curve analysis: The ROC curve was used to assess each parameter’s true positive rate (sensitivity) against the false positive rate (specificity) at various cutoff points. This analysis confirmed the high diagnostic value of Beta, Mu, Yen, and W angles for Class II versus III and Class III versus I distinctions, as well as Pi angle and Wits appraisal for Class II versus I. ROC curve analysis: The ROC curve was used to assess each parameter’s true positive rate (sensitivity) against the false positive rate (specificity) at various cutoff points. This analysis confirmed the high diagnostic value of Beta, Mu, Yen, and W angles for Class II versus III and Class III versus I distinctions, as well as Pi angle and Wits appraisal for Class II versus I. Clinical highlight The significance of accurate diagnosis in the anteroposterior dimension is of utmost importance in orthodontics.[44] The study concludes that recent cephalometric angles, such as the Beta angle, Yen angle, and Pi analysis, are highly accurate and consistent in identifying sagittal skeletal relationships. However, due to the inherent variability in human craniofacial anatomy, no single cephalometric measurement can universally determine skeletal relationships for all individuals.[44] The study emphasizes that clinicians should utilize a range of cephalometric analyses and tailor their choice based on each patient’s unique characteristics and diagnostic needs. This individualized approach allows for more accurate treatment planning and better clinical outcomes in orthodontics and craniofacial diagnosis. In essence, the study highlights that while several modern cephalometric angles offer reliable diagnostic support, a multi-parameter approach is essential for accurately identifying skeletal discrepancies across different patient profiles. Conflicts of interest There are no conflicts of interest.