Home Print this page Email this page Users Online: 40
Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 


 
 Table of Contents  
ORIGINAL ARTICLE
Year : 2020  |  Volume : 27  |  Issue : 1  |  Page : 7-11

Clinico-radiologic correlates of oral and maxillofacial radiographs: A 3-year study of 156 patients in a Nigerian teaching hospital


1 Department of Oral and Maxillofacial Surgery, University of Calabar Teaching Hospital, Calabar, Nigeria
2 Department of Oral and Maxillofacial Surgery, Aminu Kano Teaching Hospital, Kano, Nigeria
3 Department of Dental and Maxillofacial Surgery, Jos University Teaching Hospital, Jos, Nigeria

Date of Submission01-Jan-2018
Date of Acceptance25-Aug-2019
Date of Web Publication13-Mar-2020

Correspondence Address:
Dr. Otasowie D Osunde
Department of Oral and Maxillofacial Surgery, University of Calabar Teaching Hospital, Calabar
Nigeria
Login to access the Email id


DOI: 10.4103/wajr.wajr_1_18

Rights and Permissions
  Abstract 


Background: Any discrepancy between clinical and radiographic information may have dire consequences on the optimal care of patients.
Aim: The aim of this study was to determine the correlation between the clinical and radiographic findings in oral and maxillofacial surgery patients.
Materials and Methods: This was a retrospective study of all patients with plain radiographic views of the oral and maxillofacial region, seen at a Nigerian Teaching Hospital, over a 3-year study period. The radiographs and patients' case files were retrieved and demographic, clinical as well as radiographic information were obtained. Radiographic information obtained included source and types of the radiographs, patients' bio-data, side and site of the pathology, clinical indication, labeling on the radiograph as well as interpretation errors in form of missed diagnosis other than missed fractures. The data were analyzed using SPSS version 13. Cohen's kappa agreement test was done between clinical and radiographic information.P <0.05 was considered significant.
Results: Radiographs from 156 patients aged 8–80 years, mean (standard deviation), 37.9 (19.22) years, were reviewed. There were 102 males and 54 females. Trauma (n = 54; 34.6%) was the most common indication for radiographs. “Transposition” of side of the lesions accounted for 9 (5.8%) of the cases. Radiographic “transposition” was significantly associated with facial fractures, temporomandibular joint ankylosis, and impacted mandibular third molars (likelihood ratio: χ2 = 16.930; df = 10;P = 0.03). There was some disagreement between clinical and radiographic information with regard to side (kappa = 0.788;P = 0.001).
Conclusion: Discrepancies in the side of lesions, between clinical and radiographic information, were observed in this study. Adequate care should be taken by clinicians and radiologists to minimize errors in radiographs.

Keywords: Interpretation errors, missed diagnosis, plain maxillofacial radiograph


How to cite this article:
Osunde OD, Omeje KU, Efunkoya AA, Amole IO, Akhiwu BI. Clinico-radiologic correlates of oral and maxillofacial radiographs: A 3-year study of 156 patients in a Nigerian teaching hospital. West Afr J Radiol 2020;27:7-11

How to cite this URL:
Osunde OD, Omeje KU, Efunkoya AA, Amole IO, Akhiwu BI. Clinico-radiologic correlates of oral and maxillofacial radiographs: A 3-year study of 156 patients in a Nigerian teaching hospital. West Afr J Radiol [serial online] 2020 [cited 2020 Sep 25];27:7-11. Available from: http://www.wajradiology.org/text.asp?2020/27/1/7/280598




  Introduction Top


One of the basic investigative tools in oral and maxillofacial surgery is the use of plain radiographs. This is usually complementary to other diagnostic modalities such as hematologic, serology, blood chemistry, histopathologic investigations as well as advanced imaging techniques. To serve its purpose, a radiograph, apart from being of high quality, should also be able to accurately depict the indicative pathology in terms of side and site of the lesion. In addition, the true identity of the patient should be correctly displayed on the radiograph for obvious medicolegal reasons.

Errors in health care have been listed as the leading cause of death and injury.[1] A major reason for accidents in medicine is that the continuum of care includes a chain of events where faults can grow and evolve.[2] One report has shown that error in radiology is seven times more than in other departments resulting in poor continuity of patient care.[2]

The maxillofacial region is one area of the body where esthetics is of utmost concern. Tumors affecting this area are easily noticed and there may be very little or no clerical errors in the radiographic film report or in the recording of clinical details such as the site and side of the pathology. However, in conditions such as trauma or ankylosis of the temporomandibular joint (TMJ), error may occasionally occur as the side and site of the indicative pathology may not be readily obvious on clinical inspection alone. Improper labeling and misdiagnosis of facial fractures constitutes the primary cause of radiological malpractice lawsuits in the developed countries and is the most common source of diagnostic error in hospital accidents.[3]

In Nigeria, not much work appeared to have been carried out on radiographic errors, especially as it relates to radiographs of the maxillofacial region. In the present study, analysis of different maxillofacial radiographs from 156 patients was undertaken with a view to highlighting the correlation between the clinical and radiographic findings.


  Materials and Methods Top


A retrospective analysis of all plain extraoral radiographs in the Maxillofacial Unit of Aminu Kano Teaching Hospital, Kano, Nigeria, over a 3-year period, spanning from January 2010 to December 2012, was undertaken by a senior consultant oral and maxillofacial Surgeon. The inclusion criteria were availability of previous radiologic reports and clear and undamaged radiographs. All radiographs from each of the patients, irrespective of the number of views, were counted as one. This was to allow for easy comparison with the previously available radiologic reports. In addition, the case files of the patients were also retrieved, and the demographic information (such as patient's name, age, and gender) and clinical information (which includes presenting complaint, side of lesion, diagnosis, and indication for radiologic investigation) were obtained. The radiographic information obtained included source of the radiographs (from a private or public institution), nature of radiographs (whether digitalized or nondigitalized), patients' biodata, side and site of the pathology, clinical indication, labeling on the radiograph as well as interpretation errors in form of missed diagnosis, missed fractures, and nonexisting fractures. A fracture was said to be nonexisting where a clinically and radiographically absent fracture was reported as present. Radiographic errors or discrepancies were confirmed if there were variations between the original radiologic report and the surgeon's observation in terms of intraoperative findings, clinical information, and the surgeon's re-interpretation of the radiographs. The data were analyzed using Statistical Package for the Social Sciences (SPSS) version 13 (Chicago, Illinois, USA). Quantitative variables were presented as frequencies and percentages, while qualitative variables were presented using measures of central tendencies. Cohen's kappa agreement test between clinical and radiographic information was done for interobserver errors, between previous radiographic interpretations made by the general radiologist and that by the maxillofacial surgeon, as well as the correlation between clinical and radiographic information on each of the consecutive cases was determined. P < 0.05 was considered statistically significant.


  Results Top


The ages of the 156 patients ranged from 8 to 80 years, the mean (standard deviation) was 37.9 (19.22) years, and there were 102 males and 54 females, giving a male:female ratio of 2:1. Over half of the radiographs were taken in private radiological centers (n = 86; 55.1%) and the remaining 70 (44.9%) were taken at the radiology department of our institution. Eighty (51.3%) of the radiographs were digitalized, while 76 (48.7%) were nondigitalized.

The clinical indications for radiologic request are presented in [Table 1]. Maxillofacial trauma (n = 54; 34.6%) was the most common indication, followed by impacted mandibular third molar (n = 28; 17.9%), jaw tumors (n = 22; 14.1%), and TMJ ankylosis and fibro-osseous lesions each represented by 10 (6.4%) cases.
Table 1: Clinical indications for request for plain maxillofacial radiographs (n=156)

Click here to view


Combinations of posterior–anterior (PA) jaw and left and right oblique laterals of the mandible were the most common radiologic views requested and accounted for 70 (44.9%) of all maxillofacial radiographs examined. This was followed by combinations of left and right oblique laterals (n = 34; 21.8%), combinations of PA jaws and occipitomental views (n = 14; 9%), and isolated PA jaws (n = 12; 7.7%). The distribution of other types of radiologic views requested is displayed in [Table 2].
Table 2: Distribution of different radiographic views requested in oral and maxillofacial unit

Click here to view


[Table 3] shows the different types of errors, including missed diagnosis, missed fractures, and nonexisting fracture. Clerical error involving “transposition” of side of the lesions accounted for 9 (5.8%) of the 156 radiographs reviewed. There were 5 cases of missed diagnosis, 3 of which were TMJ ankylosis, and in the remaining 2 (1.3%), bony tumors were misdiagnosed as soft-tissue swellings. There were 9 (5.8%) cases of missed fractures with 6 involving the horizontal ramus of the mandible and 3 (1.9%) were fractures of the zygomatic arch. Three (1.9%) nonexisting temporal bone fractures were reported [Table 3].
Table 3: Radiographic findings

Click here to view


Clerical error, involving “transposition” (such that there is a mismatch of radiographic and clinical sides), was significantly associated with fractures of the facial skeleton, TMJ ankylosis, and impacted mandibular third molars (likelihood ratio: χ2 = 16.930; df = 10; P = 0.037) [Table 4].
Table 4: Distribution of clerical radiographic errors according to clinical indications

Click here to view


Cohen's kappa agreement test between clinical and radiographic information with regard to side of the lesion indicated kappa value = 0.788 and P = 0.001 [Table 5]. Cohen's kappa agreement test between previous radiologic reports as obtained from the radiology department and that obtained by the maxillofacial surgeon indicated kappa = 0.302 and P = 0.021, which showed very high discrepancy [Table 6].
Table 5: Kappa statistical test between clinical and radiographic information

Click here to view
Table 6: Kappa statistical test between general radiologist's and surgeon' radiographic diagnosis of maxillofacial fractures

Click here to view



  Discussion Top


Maxillofacial trauma, jaw tumor, and impacted mandibular third molar were the most common indications for request of plain radiographs of the oral and maxillofacial region. Although orthopantomograph remains the radiologic gold standard investigation for impacted mandibular third molars,[4] it is not available in our center. As an alternative, oblique laterals of the mandible are used, especially when mouth opening is limited. In addition, oblique lateral views give a very good view of the inferior neurovascular canal and its relationship to the apices of the impacted tooth compared to periapical radiographs, particularly if the intraoral film is not properly positioned.

Radiology cannot always produce infallible interpretations or reports.[5] The errors noted in this study include clerical errors (5.8%) such as mislabeling, missed diagnosis (3.2%), missed fractures (5.8%), and overdiagnosis (i.e., presence of clinically and radiographically nonexistent fractures [1.9%]). All these may lead to misinterpretation error which may be potentially harmful, depending on the case.

Interpretation error in radiology is not new. A review of the literature suggested that the level of error for clinically significant or major error in radiology is in the range 2%–20% and varies depending on the radiological investigation.[6],[7] Studies with chest radiographs have shown that physicians and anesthesiologists do make inaccurate interpretations[8],[9],[10] and those faulty interpretations changed management decisions in up to 11% of cases.[7] With higher levels of training and experience, however, it was demonstrated that interpretation improved with greater confidence.[11],[12] Literature on interpretation errors in oral and maxillofacial radiographs is however sparse. In Nigeria, Akadiri et al. observed that condylar fracture was easily missed by the general radiologists.[12] This is in agreement with the result of the present study which showed relatively lower diagnostic skills of maxillofacial fractures by the general radiologist compared to the maxillofacial surgeon (kappa = 0.302; P = 0.021). This also confirmed the results of Lee et al.[13] who had previously observed that clinicians with little experience in the field of maxillofacial fractures showed a relatively low diagnostic acumen in the detection of condylar fractures. It is thus logical to assume that an oral and maxillofacial surgeon or better still, a maxillofacial radiologist, because of his superior understanding of the anatomy of the region, may report radiographs of the maxillofacial area more accurately than the general radiologist.

Clerical errors (or incorrect labeling of the radiograph such that a lesion that presents clinically on the right of the patient for instance is reported in the radiologic report to occur on the left) accounted for 9 (5.8%) in the present study. The results of this study shows that certain maxillofacial conditions like facial skeletal fractures, TMJ ankylosis, and impacted mandibular third molars were more likely to be misrepresented than other lesions because these pathologies are not as prominent on clinical inspection when looking at the patient extraorally, compared to tumors in the maxillofacial region. Even though the Cohen's kappa value for agreement between clinical and radiographic location of the lesion was 0.79, signifying good agreement, this is not acceptable as perfect agreement, with kappa value of 1 being the most desirable for optimal patient care.

While this type of discrepancy may be of no significant negative effect on the outcome of patient management in certain conditions, it may present with grave consequences in others. In patients with TMJ ankylosis, for instance, surgeons rely, to a very large extent, more on radiologic than clinical findings to decipher both the side and extent of ankylosed mass. Due to wrong labeling, the operator may mistakenly open up the left TMJ for condylectomy, only to find out that the pathology was on the right side. This will obviously have a negative effect on the psyche of the surgeon as well as increased operation time. For the patient, this will mean more surgical morbidity, increased period of anesthesia, avoidable surgical scar as well as increased likelihood of re-ankylosis.

Although limited by cost, modern imaging techniques, especially three-dimensional computerized tomographic scan and cone-beam computerized tomographic scan would also help a great deal in avoiding errors in maxillofacial radiography and are supported by the results of a number of studies.[9],[11] Being able to visualize oral and maxillofacial pathologic entities in three dimensions assists in diagnosing and planning the appropriate treatment for the patient and minimizes misdiagnosis in the maxillofacial region.[9],[11]

One of the limitations of the study is that the working environment and conditions of the radiographers and radiologist were not taken into consideration while carrying out interpretation. It is possible that the interpreter or radiologist might have been fatigue, arising from having to cope with too many patients. Another limitation is that it is difficult to know if the clinical information in the patient case files were correctly documented.


  Conclusion Top


Discrepancies (between clinical and radiographic information) such as clerical error, “transposition,” missed fracture, and clinically nonexistent fracture were observed in this study. In this current litigious world, adequate care must be taken by both clinicians and radiologists/radiographers when attending to patients to minimize errors or discrepancies in radiographs. Where possible, joint clinico-radiologic meetings should be encouraged, where the need arises, between the two specialties of radiology and oral and maxillofacial surgery. It is also recommended that more residents should be encouraged to specialize in oral and maxillofacial radiology, which is still an evolving specialty in Nigeria, such that in future, all radiographs of the maxillofacial region would be handled by these groups of experts. In addition, regular refresher courses for general radiologists in the areas of maxillofacial radiology should be encouraged.



Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Graber ML. The incidence of diagnostic error in medicine. BMJ Qual Saf 2013;22 Suppl 2:ii21-7.  Back to cited text no. 1
    
2.
Graber ML, Wachter RM, Cassel CK. Bringing diagnosis into the quality and safety equations. JAMA 2012;308:1211-2.  Back to cited text no. 2
    
3.
Pinto A, Brunese L. Spectrum of diagnostic errors in radiology. World J Radiol 2010; 2:377-83.  Back to cited text no. 3
    
4.
Goyal S, Verma P, Raj SS. Radiographic evaluation of the status of third molars in Sriganganagar population – A digital panoramic study. Malays J Med Sci 2016;23:103-12.  Back to cited text no. 4
    
5.
FitzGerald R. Radiological error: Analysis, standard setting, targeted instruction and teamworking. Eur Radiol 2005;15:1760-7.  Back to cited text no. 5
    
6.
Goddard P, Leslie A, Jones A, Wakeley C, Kabala J. Error in radiology. Br J Radiol 2001;74:949-51.  Back to cited text no. 6
    
7.
Benger JR, Lyburn ID. What is the effect of reporting all emergency department radiographs? Emerg Med J 2003;20:40-3.  Back to cited text no. 7
    
8.
Gatt ME, Spectre G, Paltiel O, Hiller N, Stalnikowicz R. Chest radiographs in the emergency department: Is the radiologist really necessary? Postgrad Med J 2003;79:214-7.  Back to cited text no. 8
    
9.
Grosvenor LJ, Verma R, O'Brien R, Entwisle JJ, Finlay D. Does reporting of plain chest radiographs affect the immediate management of patients admitted to a medical assessment unit? Clin Radiol 2003;58:719-22.  Back to cited text no. 9
    
10.
Xavier-Souza G, Vilas-Boas AL, Fontoura MS, Araújo-Neto CA, Andrade SC, Cardoso MR, et al. The inter-observer variation of chest radiograph reading in acute lower respiratory tract infection among children. Pediatr Pulmonol 2013;48:464-9.  Back to cited text no. 10
    
11.
Eng J, Mysko WK, Weller GE, Renard R, Gitlin JN, Bluemke DA, et al. Interpretation of emergency department radiographs: A comparison of emergency medicine physicians with radiologists, residents with faculty, and film with digital display. AJR Am J Roentgenol 2000;175:1233-8.  Back to cited text no. 11
    
12.
Akadiri O, Olusanya A, Udeabor S, Agi C. Identification and interpretation of maxillofacial plain radiographs by junior dental trainees. J West Afr Coll Surg 2012;2:42-57.  Back to cited text no. 12
    
13.
Lee JH, Jung YH, Cho BH, Hwang DS. Diagnostic ability of panoramic radiography for mandibular fractures. Korean J Oral Maxillofac Radiol 2010;40:33-8.  Back to cited text no. 13
    



 
 
    Tables

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]



 

Top
 
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction
Materials and Me...
Results
Discussion
Conclusion
References
Article Tables

 Article Access Statistics
    Viewed1210    
    Printed95    
    Emailed0    
    PDF Downloaded0    
    Comments [Add]    

Recommend this journal


[TAG2]
[TAG3]
[TAG4]