Diagnostic accuracy of the tongue blade test combined with clinical signs to detect maxillary and mandibular fractures in the emergency department

Abstract

BACKGROUND:

To evaluate the diagnostic accuracy of clinical signs combined with the tongue blade test (TBT) to detect maxillary and mandibular fractures.

METHODS:

A cross-sectional study enrolled patients with maxillary and mandibular injuries in the emergency department. Physical examination and the TBT were performed, followed by radiological imaging (facial X-ray or computed tomography [CT]). The diagnostic accuracy was calculated for individuals and a combination of clinical findings at predicting maxillary and mandibular fractures.

RESULTS:

A total of 98 patients were identified, of whom 31.6% had maxillary fractures and 9.2% had mandibular fractures. The combination of malocclusion, tenderness on palpation and swelling with positive TBT had 100% specificity to detect maxillary and mandibular fractures. In the absence of malocclusion, the combination of tenderness on palpation and swelling with positive TBT produced a specificity of 97.8% for maxillary fracture and a specificity of 96.2% for mandibular fracture. A clinical decision tool consisting of malocclusion, tenderness on palpation, swelling and TBT revealed a specificity of 100% and a positive predictive value of 100%.

CONCLUSION:

The clinical decision tool is potentially useful to rule out mandibular fractures, thus preventing unnecessary radiation exposure.

INTRODUCTION

Mandibular and maxillary fractures are common facial injuries in the emergency department, with 22.5% and 39.4% of all facial traumas having mandibular and maxillary fractures, respectively.[1,2] Road traffic accidents, assault, sports injuries and falls are the main causes of facial injuries among the population.[3]

Apart from clinical examination and radiological investigations, the tongue blade test (TBT) is a sensitive screening tool for mandibular fractures.[4] TBT is an inexpensive tool that is ubiquitous in all emergency departments worldwide. Several studies have reported on the diagnostic accuracy of TBT for mandibular fractures.[4-8] TBT had a sensitivity of 97.5%, a specificity of 63.5%, a positive predictive value (PPV) of 66.2%, and a negative predictive value (NPV) of 95.2% in detecting mandibular fractures.[5] However, the utility and diagnostic accuracy of TBT for maxillary fracture have not been fully explored. In the anatomy literature, the muscles of mastication consist of the masseter, temporalis, medial and lateral pterygoid muscles, which originate from the maxilla region and are inserted into the mandible. It is possible that patients with maxillary fractures would have a positive TBT. Computed tomography (CT) is the gold standard for confirmation of mandibular or maxillary fractures.[5]

The objective of this study is to evaluate the diagnostic accuracy of clinical signs combined with TBT to detect maxillary and mandibular fractures.

METHODS

Study setting and patient selection

This was a cross-sectional study in the emergency department between 1^st April 2021 and 31^st March 2022.

All facial injury patients with suspected maxillary or mandibular fractures were eligible for inclusion. Patients were excluded if they had airway obstruction, required advanced airway, were unable to complete or perform TBT, were edentulous or had full dentures and open fractures. Convenience sampling was used for patient recruitment. Informed and written consent were obtained from each patient who agreed to participate in the study.

Data collection and definition

All trauma patients were initially managed by emergency doctors according to the Advanced Trauma Life Support (ATLS) approach, where a primary survey was performed to identify life-threatening injuries followed by a secondary survey to assess other injuries. Patients with suspected maxillary or mandibular fractures were referred to the research team, where a standardized protocol for history taking and extra- and intra-oral examinations were performed for eligible patients. The following clinical signs of maxillary and mandibular fractures were recorded: swelling, malocclusion, facial asymmetry, laceration, tenderness on palpation, ecchymosis and bruises, abrasion and TBT. Malocclusion was defined as an objectively identified traumatic misalignment of the maxillary and mandibular dental arches.[9] Tenderness on palpation and swelling was defined as the presence of pain during palpation and swelling of the maxillary region (maxillary sinus, infra-orbital rim, zygomaticomaxillary complex, Le Fort I, II, III complex and maxillary dentoalveolar complex) and mandibular region (symphyseal or parasymphyseal region, body, angle, ramus, coronoid process, condylar process and mandibular dentoalveolar complex).[9] For TBT, the patient was asked to bite on a tongue depressor until it was broken against twisting motion by an examiner, resulting in a negative test. Otherwise, the test was positive if the examiner was unable to break the tongue depressor as limited by pain.[5]

These physical findings were not shared with the treating emergency doctor and did not influence the management of patients. Subsequently, these patients underwent radiological imaging (facial X-ray or CT) to confirm the diagnosis. Maxillary fractures were defined as any fracture of the infra-orbital rim, maxillary sinus, zygomaticomaxillary complex, Le Fort I, II, III complex and maxillary dentoalveolar complex.[9] Mandibular fractures were defined as any fracture of the symphyseal or parasymphyseal area, body, angle, ramus, coronoid process, condylar process and mandibular dentoalveolar complex.[9] All radiological images were reviewed and reported by board-certified radiologists who were blinded in this study. Patients who reported maxillary or mandibular fractures on facial X-ray or CT were referred to the oral and maxillofacial surgery (OMFS) team for further management. The decision of treatment was determined by an oral and maxillofacial surgeon or consultant. The fracture treatment was assigned to either conservative or active interventions. Conservative treatments included adequate analgesics, avoidance of nose blowing or holding the nose when sneezing, a soft non-chewing diet and watchful observation. Active treatments were divided into closed or open treatments. Closed treatments included intermaxillary fixation, rigid and flexible splinting or appliances for dental injury. Open treatments included open reduction and internal fixation in an operating room.

Data collected were age, gender, occupation, mechanism of injury, clinical signs, TBT results, facial X-ray or CT findings, and whether OMFS referral was needed.

Sample size calculation

The formula outlined by Hajian-Tilaki [10] was used to calculate the sample size for the TBT to obtain a sensitivity of 80%. The formula used was:

where: n = sample size; Z = power of 80% (Z value of 1.28); S_e = sensitivity of 0.8; d = confidence interval of 0.10; Prev = prevalence of mandibular and maxillary fracture = 32%.[2] The calculated sample size was 81.92. Accounting for a 20% drop-out rate, the final sample size was 98.30.

Statistical analysis

Data were analysed with SPSS Version 26.0 (IBM Corp., USA). Demographic data were analysed using descriptive statistics. Parametric data are presented as means and standard deviations while non-parametric data are presented as median and interquartile range. Pearson’s Chi-square or Fisher’s exact test was used to analyze categorical data appropriately. In addition, the odds ratio (OR) was used to determine which clinical signs had the strongest association with confirmed fracture.

The sensitivity, specificity, PPV and NPV, positive likelihood ratio (+LR) and negative likelihood ratio (-LR) with corresponding 95% confidence intervals (CI) were calculated to determine the diagnostic accuracy of clinical signs and TBT at predicting maxillary and mandibular fracture. A clinical decision tool was constructed consisting of clinical signs and TBT with the focus of ruling out patients with maxillary and mandibular fractures.

RESULTS

Out of 230 eligible patients with facial injuries during the study period, 98 were enrolled. Four patients left the ED before radiological imaging. The demographic data are summarized in Table 1.

Table 1.

Demographic data of enrolled patients with facial injuries

The majority of the patients were male (75.5%). The patients’ ages ranged from 18 to 89 years old, with a median of 32 (IQR 23.00–50.25) years. Road traffic accidents were the main mechanism of injury in 64.3%. Maxillary fractures were diagnosed in 31 (31.6%) patients, and mandibular fractures were diagnosed in 9 (9.2%) patients. All patients with fractures were referred to the OMFS team for further evaluation and management; 39.7% of soft tissue injuries were referred to the OMFS team based on the clinical judgment of the treating emergency doctor. Among those diagnosed with maxillary or mandibular fractures, all patients were treated conservatively and given out-patient follow-up under OMFS clinic for reassessment.

Clinical findings related to maxillary and mandibular fractures

The association between clinical findings and diagnosis (fracture, maxillary and mandibular fracture versus soft tissue injury) is summarized in Table 2.

Table 2.

Clinical findings associated with fracture, maxillary and mandibular fracture versus soft tissue injury

Malocclusion (OR=18.13, P=0.006), tenderness on palpation (OR=6.63, P=0.007) and positive TBT (OR=9.98, P<0.001) were the physical findings most often associated with fracture. The combination of all these clinical signs had significant predictive value for detecting fractures (OR= 18.13, P =0.006).

TBT (OR=5.73, P =0.009) was the most significant predictor to detect maxillary fracture among other clinical findings. The combination of tenderness on palpation, swelling and positive TBT had a high predictive value for maxillary fracture (OR=10.56, P =0.011).

There were significantly more patients with malocclusion (OR=28.54, P=0.002), tenderness on palpation (OR=12.09, P=0.028) and positive TBT (OR=96.00, P<0.001) in mandibular fractures than in mandibular soft tissue injuries. In absence of malocclusion, the combination of tenderness on palpation, swelling and positive TBT was strongly associated with mandibular fracture (OR=87.50, P <0.001).

Diagnostic accuracy

The diagnostic accuracy with corresponding 95% CI of individual and combination of clinical signs with TBT for fracture, maxillary and mandibular fracture is presented in Table 3. Malocclusion had 100% specificity and 100% PPV to detect maxillary and mandibular fractures. Eliciting both malocclusion and facial asymmetry produced 100% specificity and 100% PPV for mandibular fracture. Tenderness on palpation had high sensitivity and NPV to detect maxillary fracture (sensitivity 93.5%, NPV 81.8%) and mandibular fracture (sensitivity 100%, NPV 100%) compared to other clinical signs. Swelling revealed high sensitivity to detect fracture (83.9% for maxillary fracture and 88.9% for mandibular fracture). TBT had high specificity and +LR for maxillary fracture (specificity 93.3%, +LR 4.4) and mandibular fracture (specificity 92.3%, +LR 11.6). The presence of clinical signs with positive TBT increased the specificity to 96.2%–100%. When malocclusion, tenderness on palpation and positive TBT were combined, the diagnostic accuracy for maxillary and mandibular fractures increased (specificity 100%, PPV 100%). In the absence of malocclusion, the combination of tenderness on palpation and swelling with positive TBT produced a specificity of 97.8% for maxillary fracture and a specificity of 96.2% for mandibular fracture.

Table 3.

Diagnostic accuracy of clinical examinations for fracture, maxillary fracture and mandibular fracture

An algorithm (Figure 1) was constructed using clinical signs in combination with TBT to detect patients with or without maxillary and mandibular fractures. It consisted of malocclusion, tenderness on palpation, swelling and TBT. It had a high specificity to detect maxillary and mandibular fractures; and a high negative predictive value (NPV=96.0%) for ruling out mandibular fractures when all of these clinical findings were absent.

Figure 1.

Algorithm for maxillary and mandibular fractures. SP: specificity.

DISCUSSION

Although OMFS doctors are trained to specifically assess maxillary and mandibular injury patients, primary and secondary surveys of trauma patients are mostly performed by emergency doctors. For the maxillofacial region, patients are assessed clinically to determine the risk of maxillary and mandibular fractures followed by radiological imaging. If maxillofacial fractures are diagnosed, the patient will be referred to the OMFS team to assess the need for further intervention. Early recognition of these fractures from the clinical signs is essential, as missing these fractures may result in major long-term morphological, functional and aesthetic consequences, leading to additional costs, increased burden and poor outcomes.

In this study, all patients recruited were treated conservatively by the OMFS team even if they were diagnosed with a fracture. We excluded patients who were unable to obey commands or give consent, required advanced airways, and had open fractures. These strict criteria may have excluded patients with serious facial injuries that required urgent surgical intervention.

Although individual clinical signs can be useful, it is of particular interest how a combination of these findings can be used as a clinical decision tool for the early identification of patients with maxillary and mandibular fractures. A recent study focused on how a combination of clinical findings can be used to predict maxillary and mandibular fractures.[9] However, the study of combining TBT with other clinical signs as a clinical decision tool is limited. A previous study concluded that negative TBT may be useful to rule out mandibular fracture, as the combined sensitivity was 95.4%,[6] but TBT should be used in adjunct with other clinical findings, as its sensitivity may be as low as 85.0%.[4] Hence, a clinical decision tool (Figure 2) consisting of malocclusion, tenderness on palpation, swelling and TBT was constructed to rule out mandibular fractures. The selection of clinical signs was based on high specificity, PPV and +LR, which aid in identifying patients with a high risk of fractures. The +LR for malocclusion was infinitive; thus, it is pathognomonic for patients with maxillary and mandibular fractures. Therefore, malocclusion was intentionally included in the clinical decision tool. In patients with mandibular or maxillary fractures, the muscles of mastication attached to the zygomatic, mandibular and maxillary bones may reduce the amplitude of mouth opening and bite forces due to pain, edema and muscle spasm, resulting in positive TBT.[5] In this study, TBT in combination with selected clinical signs had high specificity and a +LR to detect maxillary and mandibular fractures compared to individual clinical signs; thus, radiological imaging should be strongly considered. Radiological imaging may be avoided in mandibular injury patients without tenderness on palpation and negative TBT, as mandibular fractures can be ruled out with an NPV of 96.0%. The main advantage of this clinical decision tool is that it provides simple, fast and non-invasive bedside management during secondary surveys particularly for mandibular injury patients. The use of a clinical decision tool can potentially prevent unnecessary radiation exposure and improve the workflow for facial injuries in the emergency department.

Figure 2.

Clinical decision tool for mandibular fractures.

This study had several limitations. The main limitation was the inability to standardize the physical examination as patients with maxillofacial injury were assessed by emergency doctors or physicians with varying years of experience. In addition, the emergency doctors were less trained to assess maxillofacial injury patients than OMFS doctors or surgeons. Furthermore, patients treated conservatively were not followed up by the investigators after discharge from the emergency department to observe if they had a required delayed surgical intervention by the OMFS team. Second, the sample size was relatively smaller compared to other studies and the algorithm was derived from a single dataset; thus, the performance will be overestimated due to over-fitting. Therefore, this clinical decision tool should be validated by further research with a wider range of data collection, especially in the pediatric population where avoidance of unnecessary radiological imaging is of the highest priority.

CONCLUSION

A combination of clinical signs and TBT is potentially accurate compared to radiological imaging and may be used as a clinical decision tool to rule out patients with mandibular fractures. A validated clinical decision tool not only prevents unnecessary radiation exposure but also reduces fruitless consumption of time and resources in the emergency department.