View/Download PDF
Original Article
6 (
2
); 85-94
doi:
10.4103/2321-1407.177960

Assessment of panoral radiograph quality in a dental treatment center

Orthodontic Practice, Ministry of Interior, Manama, Bahrain, UAE
Professor of Orthodontics and Dean, European University College, Dubai Healthcare City, Dubai, UAE
Resident, Advanced Orthodontic Program, University of Alabama, Tuscaloosa, USA
Address for Correspondence: Prof. Donald J. Ferguson, Ibn Sina Building, Block D, 3rd Floor, Office 302, P. O. Box 53382, Dubai, United Arab Emirates. E-mail: fergusonloud@gmail.com
Licence
This is an open access article distributed under the terms of the Creative Commons Attribution-Non Commercial-Share Alike 3.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as the author is credited and the new creations are licensed under the identical terms.
How to cite this article: Abdul-Wahab H, Ferguson DJ, Abou-Kheir N. Assessment of panoral radiograph quality in a dental treatment center. APOS Trends Orthod 2016;6:85-94.

Abstract

Purpose

The quality of orthopantogram (OPG) images is primarily a function of patient positioning during image capture. The purpose of the investigation was to evaluate the quality of digital panoral radiographic images obtained by using the same imaging device in a large dental treatment center on the basis of the radiography technician operator securing the image.

Materials and Methods

Three hundred OPGs radiographs taken on a Kodak 8000C Digital Panoramic and Cephalometric System device in a large dental treatment center comprised the sample. The most recent OPGs beginning at May 2010 through 2007 were selected for three radiography technicians until subgroups of 100 OPGs per technician were obtained. Each panoral was evaluated by two investigators for 21 OPG image errors.

Results

Mean panoral total score was 14.71 and mean grade was 2.41 for the entire sample indicating a “good” quality. Significant differences were found among technicians for 3 of 21 panoral fault variables. The relative contribution to inferior OPG quality was greatest for the following positioning faults in rank order: Gazebo effect (11.3%), condyles pushed out (11.0%), unclear nasal structures (10.0%), airspace over U6s (9.5%), and condyles image top (9.0%).

Conclusions

There were no significant differences among technicians for overall quality indicators (total score and grade). However, statistically significant differences among the three technicians were found for image error wide anterior teeth, Gazebo effect, and unclear nasal structures.

Keywords

Digital
image quality
orthopantogram
positioning errors

INTRODUCTION

The use of panoramic radiology in general dentistry has had remarkable expansion in the past several decades. The panoramic radiograph provides a quick, low cost, general view of the teeth and jaws. The ease of operating the panoramic machine, the ability to image vital anatomical structures, and the low radiation dose are the reasons for panoramic radiology growing in popularity.[1] The panoral provides an image of the facial structures that includes both maxillary and mandibular arches and their supporting structures. The digital image is obtained by rotation of the X-ray source and the film at the same speed in the opposite direction around the head.

Since patients are being subjected to radiation, it is rational to make certain that panoramic radiology procedure is safe by using minimum radiation dose and optimized by achieving proper image quality. Although standards and instruction for obtaining optimal panoral quality have been well-described and are readily available, there is extensive evidence that many dental practices fail to reach optimal or acceptable standards.[2]

Craniofacial anatomical structures as depicted on a panoral radiograph may vary in clarity, recognition (presence or absence), and size due to varying quality of the digital radiograph. When the imaging machine is held as a constant and the device does not waiver from day-to-day as an operating mechanical device, faults in panoral image are most commonly a function of patient positioning errors.[3-5] It is patient positioning error that causes the majority of problems associated with the rendering of an excellent panoramic image.[6-10] When positioning the patient in the panoramic imaging device, three lasers should be used as follows: (1) The Frankfort laser is directed to emulate the patient’s own Frankfort plane in relationship to the X-ray beam, (2) the mid-sagittal laser is directed toward the patient’s facial midline, the canine laser directed to the distal upper canine position after the patient’s chin is rested on the machine chin rest, (3) the mid-sagittal laser is directed toward the patient’s facial midline, and (4) patient movement during image capture.

Panoramic X-ray machines are generally available with a mirror and laser light beam guide that shines on the patient’s face to facilitate positioning in different planes.[11] Patient positioning errors translate into distorted panoral image.[12]

Common distortions associated with incorrect Frankfort plane laser position include the following:

  • Condyles at the top of the image

  • Condyles at the edge of the image (could also be due to improper canine laser position)

  • Flat or reverse curve of smile line

  • Exaggerated smile line

  • Palate superimposition as thick radio-opaque

  • Palate overlapping maxillary roots

  • Ghosting of hyoid bone

  • Gazebo effect of the spine.

Distortions associated with mid-sagittal laser position include the following:

  • Wider ramus on one side (could also be due to movement of the patient during image capture)

  • Wider teeth on one side (could also be due to movement of the patient during image capture)

  • Unclear nasal structures (could also be due to movement of the patient during image capture)

  • Unequal condylar heights (which is due to tipping head during image capture).

Distortions associated with canine laser position include the following:

  • Spine superimposition on rami

  • Anterior teeth small and blurry (if canine laser was too far backward)

  • Anterior teeth wide and blurry (if canine laser was too far forward)

  • Gazebo effect of the spine

  • Washington monument image of the spine (if patient was slumping forward rather than take a step forward).

Distortions associated with movement of the patient during image capture include:

  • Step defect in the inferior border of the mandible

  • Unclear nasal structures

  • Wider teeth on one side

  • Wider ramus on one side.

Other patient positioning faults leading to panoral image distortion include the following: (1) If the patient does not bite on the bite block, the image will result in overlapping of anterior teeth.[5] (2) Improper tongue position during image acquisition will lead to airspace shadow above the maxillary roots. If the tongue lays at rest during the exposure, radiation penetrates the empty space between the dorsal surface of the tongue and the palate, producing a radiolucent shadow superimposed over the maxillary roots and reducing diagnostic quality of the image. The tongue should be placed at the roof of the mouth at image capture; the patient should be instructed to swallow, and hold the position.[8] (3) If the lead apron is extending above the collar line, an opaque defect resembling a “Shark Fin” appears on the panoral. The lead apron should not extend above the collar line to avoid being superimposed onto the panoral.[12] (4) The patient should be routinely instructed to remove any metallic jewelry such as earrings or necklaces prior to taking the panoramic radiograph to avoid ghost images of these items appearing on the image.[12]

It is important to have the tongue in the correct position during image acquisition, and this is accomplished by asking the patient to place the tongue at the roof of the mouth, swallow, and hold the position.[8]

Several authors have reported on the problems of securing an excellent panoral image. Al-Faleh[1] found about 93.6% of panorals had errors in a sample of 500 radiographs evaluated. Kaviani et al. (2008)[13] evaluated a sample of 250 panoramic radiographs reporting 7.6% were free of error, 10.4% had to be retaken, and 92.4% were useable, but presented errors. Rushton et al.[2] evaluated the quality of 1813 panoramic radiographs in 41 general dental practices, and graded only 0.8% of films as “excellent,” 66.2% as “diagnostically acceptable,” and 33% were “unacceptable.” Kullman and Joseph[14] evaluated the quality of 199 digital panoramic radiographs exposed by a dentist with minimal experience in taking panoramic radiographs and reported that each radiograph represented from 1 to 9 errors, and that no radiograph was completely free from error. Glass and Seals[4] assessed 75 edentulous panorals and reported 8% were error-free and 89% demonstrating positioning errors. Akarslan et al.[15] evaluated 460 radiographs using 20 error categories; 37.6% were found to be error-free.

More recently, 297 panoramic radiographs were evaluated from 99 dental hospitals and clinics in Korea, and errors were found in 94.3% (280) of panorals with the majority of errors due to patient positioning.[13] Evaluation of 1782 panoramic radiographs resulted in 11% error-free radiographs and 89% presented positioning errors.[6] Granlund et al.[8] evaluated 1287 panoramic radiographs and reported 4% error-free.

Given the results found in previous studies, it is obvious that taking an acceptably clear panoramic radiograph is a challenge for the operator taking it. It is obvious that care should be taken while capturing the panoramic radiograph image, especially while positioning the patient in the panoral device to avoid image faults and perhaps exposing the patient to further radiation.

Each dentist has the responsibility of teaching the staff the proper way to take a panoral radiograph. Because of there may be great variability among operators securing panoramic radiographic images, it is reasonable to evaluate for differences in operator technique and influence on the quality of the panoramic radiograph in a high volume patient care environment wherein all digital images were obtained on the same radiographic device.

The purpose of the investigation was to evaluate the quality of digital panoral radiographic images obtained by using the same imaging device in a large dental treatment center on the basis of the radiography, technician operator securing the image. The null hypothesis tested was no difference in the quality of panoral radiographs among three technician operators.

MATERIALS AND METHODS

A total of 300 digital panoral radiographs taken on a Kodak 8000C Digital Panoramic and Cephalometric System device were used in the study. The panoral radiographs were made available at the European University College located at Dubai Healthcare City, United Arab Emirates.

Sample

The digital radiographs were randomly selected from patient files created between 2007 and 2010. The first image was selected beginning with May 2010 patient files and then proceeding sequentially to earlier dates. Three subgroups were established from the start based upon the operator who obtained the panoramic image; sample selection continued until 100 panoral radiographs were found per operator. Initial screening of all radiographs was performed by two observers (HAW and ND). Exclusion selection criteria were as follows: (1) patients <11 years old, (2) patients with severe craniofacial anomalies, and (3) patients with special needs.

Procedure

The 300 digital panoral images were displayed directly on a 17-inch monitor screen with 8-bit resolution in the same dark room with the same computer monitor viewing conditions held constant, i.e., window (contrast) and level (brightness). The radiographs were examined by the two investigators from one meter distance. To avoid observer fatigue, an interval of at least 12 h separated each viewing session.

A total of 29 panoral radiograph faults (study variables) based upon patient position errors were evaluated for each radiograph and scored on a 4-point scale (the greater the error, the higher the score) depending on how the error affected the visibility or appearance of the anatomical structures on the radiograph: 0 = no error, 1 = minor error, 2 = moderate error, and 3 = severe error.

Included in the 29 evaluation criteria were 8 bilateral criteria that were evaluated as either “Right” or “Left;” these 8 variables are as follows:

  1. Presence of radio-opaque items

  2. White “Shark Fin” image from lead apron

  3. Spine superimposed on the rami

  4. Condyle pushed out

  5. Condyle at the top of the image

  6. Ghosting of the hyoid bone

  7. Gazebo effect of the spine

  8. Step defect on the inferior border of mandible.

These 8 bilateral study variables were evaluated independently (right and left), then recombined to form a single variable (right + left/2) to give equal weighting; the final number of target variables for the study was 21.

On the basis of comparing to a high quality orthopantogram (OPG) [Figure 1], the panoral radiograph evaluation criteria used were as follows:

  1. Opaque items: Presence of radio-opaque items [Figure 2]

  2. Airspace over U6s: Shadow of air space over the roots of the maxillary molars and the airway space over the rami (improper tongue position) [Figure 3]

  3. Shark Fin: White “Shark Fin” in the image (lead apron above the collar line) [Figure 4]

  4. Spine rami: Spine superimposed on the rami (biting too far forward) [Figure 5]

  5. Condyles pushed out: Patient biting too far backward on the bite block or chin is too high shows condyles pushed out [Figure 6]

  6. Overlapped anteriors: Overlapping of the anterior teeth (not biting on the bite block) [Figure 7]

  7. Wide anterior teeth: Wide anterior teeth or blurry anterior due to bite block being too far backward or canine laser too far forward to the canine [Figure 6]

  8. Narrow anteriors: Narrow anterior or blurry anterior due to bite block being too far forward or canine laser to far backward to the canine

  9. Reverse smile: Chin too high error shows flat or reverse smile line [Figure 2]

  10. Palate thick opaque line: Chin too high error shows image of the palate as a thick radio-opaque line overlapping maxillary roots [Figure 2]

  11. Palate overlaps upper molars: Chin too high error shows image of the palate overlaps the maxillary molar [Figure 3]

  12. Exaggerated smile: Chin too low error shows exaggerated smile line [Figure 5]

  13. Condyle at image top: Chin too low error shows condyle at the top of the image [Figure 4]

  14. Hyoid ghosting: Ghosting of the hyoid bone over the mandible (chin too low) [Figure 8]

  15. Gazebo effect: Spine forms arch or Gazebo effect due to chin being too low [Figure 7]

  16. Wider teeth: Wider teeth more on one side that the other (twisting head) [Figure 9]

  17. Wider ramus: Wider ramus more on one side that the other (twisting head) [Figure 9]

  18. Unclear nasal structures: Movement of the patient (twisting head) error shows unclear nasal structures [Figure 9]

  19. Unequal condyle height: Condyles not equal in height (tipping of the head) [Figure 8]

  20. Step defect: Step defect in the inferior border of the mandible (movement of the patient during exposure)

  21. Washington monument: Patient slumping forward shows ghosting of spinal column as radio-opaque line resembling Washington monument [Figure 6].

Figure 1: Representation of a digital orthopantogram with high quality
Figure 2: Orthopantogram image faults opaque items (arrow), reverse smile (solid line), and palate thick opaque line (dotted)
Figure 3: Orthopantogram image faults air space over U6s (solid line) and palate overlaps U molars (dashed lines)
Figure 4: Orthopantogram image faults Shark Fin (arrow) and condyles at image top (lines)
Figure 5: Orthopantogram image faults Spine Rami (arrow) and exaggerated smile (line)
Figure 6: Orthopantogram image faults wide anterior teeth, Washington monument (dotted lines), and condyles pushed out
Figure 7: Orthopantogram image faults overlapped anterior (arrows) and Gazebo effect (lines)
Figure 8: Orthopantogram image faults hyoid ghosting (arrow) and unequal condyles height (lines)
Figure 9: Orthopantogram image faults wider ramus (arrowed lines), wider teeth (molar brackets), and unclear nasal structures

Total score per panoral radiograph was computed as the arithmetic total of the 21 positioning fault variables; possible total score per panoral image could range from 0 to 63.

Grade or “Quality” of each of the 300 radiographs was computed from the total scores using a scale from 1 to 4 (the lower the score, the higher the quality) as follows:

  1. Grade 1: Excellent (total score of 10 points or less)

  2. Grade 2: Good (total score of 10.1–15 points)

  3. Grade 3: Fair (total score of 15.1–19.9 points)

  4. Grade 4: Poor (total score of 20 points or more)

Statistical analysis

Descriptive statistics (means and standard deviations) was computed per each of the 21 study variables as well as for total score and grade per subject. Nonparametric Kruskal–Wallis H testing of the ordinal data was used to test for differences among operators for each of the 21 faults variables, total score, and grade. Data tabulation and analysis was performed using the SPSS (Statistical Package for Social Services (SPSS) software, version 15.0.1, IBM, Armonk, NY) PC + statistical package. The measurements were reliable as disagreements in opinion between the two observers were dealt with by exchanging different points of view until a consensus was reached.

RESULTS

Three hundred OPGs taken by three separate radiology technicians on the same digital Kodak machine were evaluated. Each study variable was scored on the basis of the severity of error observed by the two study investigators with 0 representing no error to three representing severe error.

Descriptive statistics including means and standard deviations were computed for each of the 21 study variables and scores per variable ranged from 0.0 for spine rami to 1.59 for Gazebo effect. Total score representing the sum of fault variables, averaged 14.71 ± 3.97; grade (0 = excellent and 3 = poor) averaged 2.42 ± 0.85 [Table 1].

Table 1: Rank and relative percent contribution of each positioning fault to total score
Variable Mean SD Percentage Rank
Opaque items 0.03 0.27 0.2 19
Airspace over U6s 1.38 1.04 9.5 4
Shark fin 0 0 0.0 21
Spine rami 0.20 0.48 1.5 14
Condyles pushed out 1.58 0.70 11.0 2
Overlapped anteriors 0.01 0.17 0.1 20
Wide anterior teeth 0.19 0.66 1.0 16
Narrow anteriors 0.07 0.40 0.5 17
Reverse smile 0.68 0.90 4.3 11
palate opaque line 0.94 1.02 6.1 9
Palate overlaps upper molars 0.66 0.83 4.3 12
Exaggerated smile 0.19 0.57 1.5 15
Condyle at image top 1.22 0.63 9.0 5
Hyoid ghosting 0.72 1.03 5.3 10
Gazebo effect 1.59 0.87 11.3 1
Wider teeth 0.55 0.92 3.3 13
Wider ramus 1.00 1.03 6.6 7
Unclear nasal structures 1.52 1.08 10.0 3
Unequal condyle height 1.13 0.98 7.6 6
Step defect 0.07 0.27 0.4 18
Washington monument 1.00 1.03 6.4 8
Total score 14.71 3.97 100.0
Grade 2.42 0.85

Means scores and SD for all study variables as well as total score and grade of errors detected from evaluation of 300 digital panoral radiographs (OPGs) taken by 3 radiology technicians. The relative contribution of each positioning error to panoral quality is represented by percent of overall error and rank of each variable from 1 to 21 with rank “1” representing the most contribution to inferior panoral quality and “21” the least contribution. SD – Standard deviation; OPG – Orthopantogram

Relative percent of positioning fault contribution to total score (mean fault variable score divided by total score) ranged from 11.3% to 0.0%. The relative contribution of each positioning fault was greatest for Gazebo effect (11.3%) and ranked first or 1 in contribution to inferior panoral quality. Condyles pushed out ranked 2 with 11.0%; unclear nasal structures ranked 3 with 10.0%; airspace over U6s ranked 4 with 9.5%, and condyles image top ranked 5 with 9.0%. Positioning fault variable Shark Fin ranked 21 or last and contributed nothing (0.0%) to inferior panoral quality [Table 1].

The mean total score of 14.71 and the mean grade of 2.42 for the entire sample qualified for a “Good” rating. No differences (P > 0.05) among the three radiology technicians (14.5, 14.8, and 14.9) were found for the total score or the grade study variables.

Kruskal–Wallis testing demonstrated statistical differences among the three operators taking the radiographs. Study variables wide anterior teeth, Gazebo, and unclear nasal structures were found different as a function of the operator: Wide anterior teeth, significantly greater error was found for operator 3 (0.34) than for operator 2 (0.9,P = 0.02); Gazebo effect error was significantly greater for operator 1 (1.70) compared to operator 3 (1.39,P = 0.02); and unclear nasal structure error was greater for operator 3 (1.72) than for operator 1 (1.32,P = 0.02) [Table 2].

Table 2: Panoral quality differences among three radiology technicians
Group (n=100 each group) Subset for alpha = 0.05 Kruskal-Wallis Probability
1 2 mean rank
Wide Anteriors
Operator-1 0.13 0.13 148.29
Operator-2 0.09 145.32 P=.020 Op-2 v Op-3
Operator-3 0.34 157.90
Significance 0.965 0.057
Gazebo
Operator-1 1.70 160.77
Operator-2 1.68 1.68 158.57
Operator-3 1.39 132.17 P=.020 Op-3 v Op-1
Significance 1.000 0.988
Unclear Nasal Structures
Operator-1 1.32 135.02 P=.024 Op-1 v Op-3
Operator-2 1.52 1.52 150.68
Operator-3 1.72 165.80
Significance 0.387 0.387

Out of the 300 panoral radiographs evaluated, 36 (12.0%) were scored as “Excellent” quality, 139 (46.3%) were scored “Good,” 89 (29.7%) were scored “Fair,” and 36 (12.0%) were judged as “Poor” quality. As stated previously, no significant differences were found among the three operators for panoral total score or for grade.

Utilizing Kruskal–Wallis testing, each positioning fault variable was compared by grade category to determine how proportional grading was among each of the four grading categories with results as follows [Table 3]:

  1. When wide anterior teeth fault occurred, grade category “Poor” was significantly disproportional (P < 0.02) from categories “Fair,” “Good,” and “Excellent”

  2. When airspace over U6s error occurred, category “Poor” was significantly disproportional (P < 0.04) from categories “Good” and “Excellent”

  3. Grade categories “Poor” and “Fair” were significantly disproportional from grade categories “Good” and “Excellent” for the following nine variables: Condyles pushed out (P = 0.000); reverse smile (P = 0.000); palate opaque line (P < 0.001); palate overlaps molars (P < 0.001); wider teeth (P < 0.002); wider ramus (P < 0.001); unclear nasal structures (P = 0.000); unequal condyle height (P < 0.01); and Washington monument (P < 0.01)

  4. When the Gazebo effect occurred, categories “Poor,” “Fair,” and “Good” were significantly disproportional (P < 0.04) from category “Excellent.”

Table 3: Comparison of proportional grading by grade category for each panoral positioning fault
Study variable Mean grade (n=300) Mean rank P
Excellent (n=36) Good (n=139) Fair (n=89) Poor (n=36)
Opaque items 0.03 148.0 149.1 153.0 152.2 NS
Airspace over U6s 1.38 108.8 144.3 163.4 184.4 P > E and G; P<0.04 F > E; P=0.005
Shark fin 0 150.5 150.5 150.5 150.5 NS
Spine rami 0.20 170.2 149.6 151.5 131.7 NS
Condyles pushed out 1.58 94.5 135.4 174.0 206.7 P and F > E and G; P=0.001 G > E; P=0.03
Overlapped anteriors 0.01 150 150 151.7 150 NS
Wide anteriors 0.19 141.9 140.1 156.7 183.7 P > E, G and F; P<0.02
Narrow anteriors 0.07 149.8 150.9 148.9 153.8 NS
Reverse smile 0.68 108.5 131.3 179.4 195.1 P and F > E and G; P=0.000
Palate opaque line 0.94 100.7 136.2 175.8 193 P and F > E and G; P<0.001
Palate overlaps upper molars 0.66 108.2 140.8 159.0 209.3 P and F > E and G; P<0.001
Exaggerated smile 0.19 156.4 155.5 144.2 140.8 NS
Condyle at image top 1.22 158.3 157.4 137.6 148.1 NS
Hyoid ghosting 0.72 153.8 151.3 148.3 149.5 NS
Gazebo effect 1.59 106.0 149.0 160.3 176.8 P, F and G > E; P<0.04
Wider teeth 0.55 121.3 129.3 170.3 212.6 P and F > E and G; P<0.002
Wider ramus 1 119.8 127.3 176.6 206.0 P and F > E and G; P<0.001
Unclear nasal structures 1.52 65.8 134.0 188.4 205.2 P and F > E and G; P=0.000 G > E; P=0.000
Unequal condyle height 1.13 116.5 137.0 168.0 193.5 P and F > E and G; P<0.01
Step defect 0.07 138.5 149.2 150.5 167.4 NS
Washington monument 1 108.5 131.6 165.3 228.8 P and F > E and G; P<0.01

Kruskal–Wallis nonparametric testing by grade quality category E, G, F, and P showing mean for all sample (n=300), mean rank and probability levels. Results revealed grading was significantly disproportional for 13 of 21 variables. Note that when wide anteriors positioning error occurred, Grade “poor” was significantly disproportional from all other categories. Conversely, when the Gazebo fault occurred, Grade “excellent” was significantly disproportional from all other categories. P – Probability level; NS – Nonsignificant statistically, i.e., P>0.05; E – Excellent; G – Good; F – Fair; P – Poor

DISCUSSION

The “quality” of 300 digital panoramic radiographs taken by three operators was evaluated by two study investigators for panoral errors based upon 21 patient positioning faults. Each radiograph was graded: 0 = no error; 1 = minor error; 2 = moderate error; and 3 = severe error. Scores for all 300 radiographs were computed and graded on a scale from 1 to 4 as follows: (1) “Excellent” (total 10 points or less). (2) “Good” (total from 10.1 to 15 points). (3) “Fair” (total from 15.1 to 19.9 points). (4) “Poor” (total 20 points or more). The sample overall averaged a total score of 14.71 and a grade of 2.42 for a “Good” rating. No differences were found in either total score or grade variables as a function of the operator who secured the digital radiograph.

Overall, 36 of 300 panorals (12%) in the present study were judged “excellent” which was slightly higher than most reported literature on the subjects. Other investigators judging OPG quality used a variety of evaluation methods which resulted in “excellent” or “error-free” ratings hovering from 5.7%,[3] 6.4%,[1] 7.6%,[13] 8%,[12] to 11%.[12] In contrast, two investigations reported error-free panorals as low as 0.8%[2] and as high as 37.6%[15] of radiographs evaluated.

The relative contribution of each panoral positioning fault to total score showed that the following five positioning faults contributed the most to inferior panoral quality in rank order: Gazebo effect >condyles pushed out >unclear nasal structures >airspace over U6s >condyles at image top.

When the study fault variables were evaluated by the four grading criteria, it was demonstrated that grading was disproportional for study fault variables as follows: Wide anterior teeth showed the greatest disproportional grading suggesting that when this positioning error occurred, a “Poor” grade was more readily assigned and there was little discrimination on the grading scale from Fair to Excellent. When airspace over U6s error occurred, a “Poor” or “Fair” grade would likely be assigned, and that there was little discrimination on the grading scale from Good to Excellent. In contrast, when Gazebo effect fault occurred, there was high discrimination on the grading scale from Good to Poor.

When taking into consideration the percent contribution to total score in combination with the ability to discriminate while grading, the positioning fault airspace over U6s emerges as important; this positioning fault ranked fourth (9.5%) in contribution to inferior panoral quality and demonstrated weak discrimination during panoral grading. While positioning error wide anterior teeth also demonstrated little discrimination during panoral grading, wide anterior teeth was ranked 16 out of 21 in contribution to inferior panoral quality, and is therefore deemed much less important than fault airspace over U6s. Gazebo effect ranked first in contribution to inferior panoral quality, but there was high discrimination in the grading of this positioning fault.

In the present study, statistical differences as a function of operator were found in study variables: Wide anterior teeth, Gazebo, and unclear nasal structures. The three variables scored statistically significant differences (P < 0.05) among the three operators with operator three demonstrating significantly higher score (poorer quality) than either of the two other operators for wide anterior teeth and unclear nasal structure. The panoral fault wide anterior teeth are due to improper use of the canine laser while positioning the patient in the imaging machine, i.e. positioning the patient too far back. Gazebo effect imaging error was significantly greater for operator 1 due to improper use of the Frankfort horizontal laser, i.e. tipping the patients head improperly.

Al-Faleh[1] studied 500 panoramic radiographs randomly selected from inactive files of adult dentate patients seen at the dental school taken by a trained technician and found that 468 radiographs (93.6%) showed one or more errors. The most common positioning error reported was palatoglossal air space above the root apices of maxillary teeth (81.8%); the second most frequent error reported was slumped position (17.2%). The percentages of other errors were between 10% and 11.6%. However, in our study, the highest percentage of error was found in Gazebo effect (11.3%) where more attention should have been paid during positioning the patient through the canine laser. In general, the Gazebo effect of the spine is obtained when the canine laser is too far backward or the patient being too far forward. The second highest error was condyles pushed out (11.0%) which could be due to either improper canine laser positioning or the head was tilted upward. The canine laser should just be at the back of the upper canines, and faults in maintaining this position will have crucial effects on the radiograph. When condyles are pushed out (at the border of the image), it is caused by the canine laser being too far forward and the patient is positioned too far backward. However, if the reason for the condyles being pushed out is the chin being tilted upward, this means that improper Frankfurt positioning occurred, and more attention and time should be spent to position the patient in the right planes before image acquisition. The third being unclear nasal structures (10.0%), which is usually due to the patient moving during the image capture. The patient should have been told about the time it takes to capture the image to hold still in the same position. It greatly helps to have a mirror or a sticker of some kind that keeps the patient busy looking at to prevent temptations to move during the image capture. The high percentage of errors that was obtained in the Al Faleh study was due to the lack of verbal communication and inadequate positional instructions by the technician.

Kaviani et al. (2008)[13] evaluated 250 panoramic radiographs (100 male and 150 female) with mean age of 24.3 years taken at the Department of Oral and Maxillofacial Radiology. The radiographs were categorized into five groups according to the type of errors: Patient positioning errors, darkroom errors, failure to remove metallic accessories, equipment setup errors, and patient movement during exposure. Only 19 radiographs (7.6%) evaluated as error-free, 26 (10.4%) were unacceptable and had to be retaken, and 231 (92.4%) panoramic radiographs presented errors. However, in the present study, given that the panoramic radiograph method used is digital, most of the faults obtained were due to positional errors. The results of our study revealed that 36 radiographs (12%) were considered excellent, 139 radiographs (46.3%) regarded as good, 89 (29.7%) considered fair, and 36 radiographs (12%) were unacceptable. Percentage of errors per error type in Kaviani study was as follows: Patient positioning (78%), film development (69.2%), equipment setup (3.2%), failure to remove metallic accessories (3.2%), and patient movement during exposure in (2.4%). In the Kaviani study, the most common errors were related to patient positioning (78%), which agrees with the findings of a pervious study, where the most common error found was “positioning of the patient” error (89.3%). However, the fact of having a high percentage of positional errors indicates how important it is to follow the instructions for taking panoramic images. This relates to our present study as the errors that were considered statistically significant were due to positional errors. The mid-sagittal, canine, and Frankfurt lasers were incorporated to help position the patients correctly to have superior panoramic radiographs. Unlike in the Kaviani study wherein movement during image capture ranked as fifth most common, in the present study, movement expressed as unclear nasal structures ranked third or 10% of errors observed.

Rushton et al.[16] evaluated the quality of 1813 panoramic radiographs in 41 general dental practices. The authors graded only 0.8% of films as “excellent,” 66.2% as “diagnostically acceptable,” and 33% were “unacceptable.” The most common faults directly contributing to failure of the radiographs were antero-posterior positioning errors, low density, and low contrast. Rushton found the greatest technical error to be in antero-posterior positioning 54.1% which coincides with the findings in our study, where Gazebo was the greatest error that led to obtaining poor panoramic radiograph 11.3%. This confirms how essential it is to follow instructions while taking the image to avoid having such errors in the panoramic radiographs obtained. In our study, among the 300 panoramic radiographs evaluated, 12% was found to be “excellent,” 46.3% as “Good,” 29.7% as “fair,” and 12% as “unacceptable.”

Kullman and Joseph[14] evaluated the quality of 199 digital panoramic radiographs taken in a newly established dental school in Kuwait. All radiographs were taken by a dentist with minimal experience in taking panoral radiographs. An experienced oral and maxillofacial radiologist found that the number of errors in each radiograph ranged from 1 to 9 and no radiograph was completely free from errors. The average number of errors per radiographs was 3.7. The most common error found in the Kullman study was that the teeth were not separated and constituted 24% of errors. This is generally due to not biting on the bite block when the image was taken. The second most common error (16%) cited by Kullman was maxillary and mandibular apices or crowns cut; the author explained this as a result of having these parts out of focus during exposure. Other errors cited were that the tongue was not in contact with the palate during exposure (0.6%) and the inferior border of the mandible was not smooth and continuous due to double images of the hyoid bones (15%). The greatest percentage of error in the present study was found in Gazebo effect of the spine (11.3%), followed by condyles pushed out (11.0%); unclear nasal structures ranked third (10.0%). Airspace over U6s ranked fourth (9.5%), which was similar with the Kullman study as this error is due to improper tongue position during exposure.

Akarslan et al.[15] studied the frequency of common errors seen on 460 panoramic radiographs taken in a dental school using 20 categories of common errors; the authors found 37.6% were error-free. The most common errors were found to be the palatoglossal airspace shadow of air above the tongue due to the patient not raising the tongue against the palate (46.3%) and the superimposition of hyoid bone with the mandible (26.3%); the least common error found was dirty or bent films (0.2%). Whereas in our present study, the most common error found was Gazebo effect of the spine due to patient being positioned too forward or the canine laser being too backward (11.3%) followed by condyles pushed out due to patient biting too far backward or chin tipped upward (11%). The least common error was presence of Shark Fin due to the lead apron being above the collar line (0%).

No one can deny the importance of a panoramic radiograph as a diagnostic tool as the survey plays an important role in the diagnosis of a disease or an abnormality. The value of any diagnostic procedure depends upon the amount of information gained by its utilization. In many instances, multiple errors occurred in one image; this could be due to that an inadequate time was spent for patient preparation and positioning. In panoramic radiography, there are numerous factors only pertinent to panoramic radiography, which can reduce the diagnostic quality of radiographs. Ahead of these factors is the improper patient positioning. The dentist should be aware enough to monitor the quality of panoramic radiographs making sure they are free of positioning errors.

The results from the present study suggest that panoramic radiography is a difficult radiographic technique to master and needs a well-trained operator to get high-quality radiographs. Both theoretical and practical training are recommended for radiology staff, as in Sweden, where dental staff should be properly trained to make exposures. We share the responsibility for an unnecessary retake of a panoramic radiograph, and exposing the patient to more radiation. We should not underestimate the hazards of radiation that we put our patient each time, we take low quality radiograph. An operator who is skilled enough should be hired to take radiographs. If a good panoramic radiograph could be acquired why do we accept a low-quality radiograph and expect a diagnosis to be achieved from it.

CONCLUSIONS

The quality of 300 digital panoramic radiographs taken by three operators was evaluated by two investigators for radiographic errors based upon 21 study variables. Each radiograph was graded by converting the total score variable to “Excellent,” “Good,” “Fair,” and “Poor.” The sample overall averaged a total score of 14.71 and a grade of 2.42 representing a “Good” rating. No differences were found in the total score or grade variables as a function of the radiology technician who secured the digital radiograph, but statistically significant differences were found for three study faulty variables as a function of technician operator as follows:

  1. Operator 3 had significantly greater wide anterior teeth or blurry anterior teeth due to the bite block being too far backward or canine laser too far forward to the canine

  2. Operator 1 had significantly greater Gazebo effect of the spine due to chin being too low or improper head tipping relative to the Frankfort horizontal laser

  3. Operator 3 had significantly greater movement of the patient during the imaging procedure resulting in unclear nasal structures.

The null hypothesis was accepted in relation of overall panoral total score and grade. However, the null hypothesis is rejected in relation to specific study fault variables as differences were found in image quality of OPG radiographs among operators for variables: Wide anterior teeth, Gazebo effect, and unclear nasal structures.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

  1. . Common positioning errors in panoramic radiography. Saudi Arabia: King Saud University.
  2. , , . The effectiveness of undergraduate teaching of the identification of radiographic film faults. Dentomaxillofac Radiol. 2005;34:337-42
    [Google Scholar]
  3. , , , , , , et al. Clinical image quality evaluation for panoramic radiography in Korean dental clinics. Imaging Sci Dent. 2012;42:183-90
    [Google Scholar]
  4. , . Williams EO. Common errors in panoramic radiography of edentulous patients. J Prosthodont. 1994;3:68-73
    [Google Scholar]
  5. , , . The value of a quality improvement programme for panoramic radiography: A cluster randomised controlled trial. J Dent. 2013;41:328-35
    [Google Scholar]
  6. , , , , , , et al. Positioning errors and quality assessment in panoramic radiography. Imaging Sci Dent. 2012;42:207-12
    [Google Scholar]
  7. , , . Common positioning errors in panoramic radiography: A review. Imaging Sci Dent. 2014;44:1-6
    [Google Scholar]
  8. , , , , , . Frequency of errors and pathology in panoramic images of young orthodontic patients. Eur J Orthod. 2012;34:452-7
    [Google Scholar]
  9. , , . Common errors in digital panoramic radiographs of patients with mixed dentition and patients with permanent dentition. Int J Dent. 2012;2012:584138
    [Google Scholar]
  10. , , , . Panoramic radiography: Effects of head alignment on the vertical dimension of the mandibular ramus and condyle region. Eur J Orthod. 2012;34:164-9
    [Google Scholar]
  11. . Acquiring panoramic images. In Kodak 8000 Digital Panoramic System User Guide. Ch. 5. Available from: http://www.carestreamdental.com/ (accessed )
  12. . Successful Panoramic Radiography. Rochester, New York: Kodak Dental Radiography Series, Eastman Kodak Company.
  13. , , . Evaluation of common errors of panoramic radiographs in tabriz faculty of dentistry. J Dent Res Dent Clin Dent Prospects. 2008;2:99-101
    [Google Scholar]
  14. , . Quality of digital panoramic radiography in a newly established dental school. Swed Dent J. 2006;30:165-70
    [Google Scholar]
  15. , , , . Common errors on panoramic radiographs taken in a dental school. J Contemp Dent Pract. 2003;4:24-34
    [Google Scholar]
  16. , , . The quality of panoramic radiographs in a sample of general dental practices. Br Dent J. 1999;186:630-3
    [Google Scholar]
Show Sections