Plain radiography versus computed tomography scan in evaluation of adults with chronic rhinosinusitis in Zaria, Nigeria

Andrew Enigie Brai; Philip Oluleke Ibinaiye; Dotiro Chitumu; Iliyasu Y Shuaibu

doi:10.4103/JRMT.JRMT_2_20

ORIGINAL ARTICLE

Year : 2020 | Volume : 1 | Issue : 1 | Page : 8-12

Plain radiography versus computed tomography scan in evaluation of adults with chronic rhinosinusitis in Zaria, Nigeria

Andrew Enigie Brai¹, Philip Oluleke Ibinaiye², Dotiro Chitumu³, Iliyasu Y Shuaibu³
¹ Department of Radiology, National Ear Care Center, Kaduna, Nigeria
² Department of Radiology, Ahmadu Bello University Teaching Hospital, Zaria, Nigeria
³ Department of Otorhinolaryngology, Ahmadu Bello University Teaching Hospital, Zaria, Nigeria

Date of Submission	17-May-2020
Date of Decision	01-Jun-2020
Date of Acceptance	04-Jun-2020
Date of Web Publication	25-Sep-2020

Correspondence Address:
Dr. Andrew Enigie Brai
National Ear Care Center, No. 3 Golf Course Road, PMB 2438, Kaduna
Nigeria

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/JRMT.JRMT_2_20

Abstract

Background: Chronic rhinosinusitis (CRS) can be a debilitating illness with major socioeconomic costs and decreased workplace productivity. Till date, paranasal sinus (PNS) radiography continues to be the primary imaging modality for diagnosing CRS in most developing countries. Computed tomography (CT) scan – the gold standard imaging modality is either unavailable or unaffordable in these regions. Aim: The aim of this study is to compare plain radiographic and CT scan findings of the PNSs in adults with CRS. Subjects and Methods: After initial clinical evaluation and rhinoscopy, consenting patients were assessed using both plain radiography (PR) and direct coronal CT scan of the PNSs. Results: One hundred and eighteen (68 males and 50 females) adults with a mean age of 39.4 years were involved in this study. The percentage agreement between PR and CT for sinus abnormalities was <80%. Using CT scan as the gold standard, PR had the highest diagnostic value in the maxillary sinus with sensitivity and specificity of 76.4% and 89.7%, respectively. Sensitivities and specificities for the ethmoid (50.0% and 76.2%), frontal (35.3% and 89.3%), and sphenoid sinuses (17.1% and 96.1%), respectively, were also recorded. Conclusion: In the diagnosis of chronic maxillary rhinosinusitis, correlation between PR and CT scan was good with high agreement. However, for chronic ethmoidal, frontal, and sphenoidal sinusitis, PR demonstrated poor agreement and weak correlations with CT scan. It is recommended that noncontrast coronal CT scan should replace paranasal PR as the initial imaging modality for evaluating adults with CRS where facilities are available.

Keywords: Chronic rhinosinusitis, computed tomography, paranasal sinuses, plain radiography

How to cite this article:
Brai AE, Ibinaiye PO, Chitumu D, Shuaibu IY. Plain radiography versus computed tomography scan in evaluation of adults with chronic rhinosinusitis in Zaria, Nigeria. J Radiat Med Trop 2020;1:8-12

How to cite this URL:
Brai AE, Ibinaiye PO, Chitumu D, Shuaibu IY. Plain radiography versus computed tomography scan in evaluation of adults with chronic rhinosinusitis in Zaria, Nigeria. J Radiat Med Trop [serial online] 2020 [cited 2023 Mar 26];1:8-12. Available from: http://www.jrmt.org/text.asp?2020/1/1/8/296112

Introduction

Chronic rhinosinusitis (CRS) is defined by inflammation of the nasal and paranasal sinus (PNS) mucosa that persists beyond 12 week.^[1] In Nigeria, CRS constitutes 75% to 84% of patients with rhinosinusitis^[2],[3],[4] and has shown a prevalence of 7.3% in the North West.^[2]

It is recognized that symptoms have high sensitivity, but low specificity, hence, the need for objective assessment by imaging.^[5] Although the indication for plain PNS X-rays may have declined over the last decade, it is still the initial examination of choice in our setting. It is widely available, affordable, and now better with processed images following the advent of computed and digital radiography. On the other hand, computed tomography (CT) scan provides an anatomical roadmap and is being reserved for patients considered for surgery.^[6] In our setting, where both modalities are available and where most patients 'pay-out-of-pocket' for their investigations, using both investigations may cause more financial strain for patients. Thus, the need to re-establish the relevance of PNS radiography in the management of CRS in adults.

This study aims to evaluate the diagnostic performance of PNS radiography in detecting sinus abnormality when compared to direct coronal PNS CT scan. The patterns of sinus involvement and radiological findings were also determined.

Subjects and Methods

This cross-sectional descriptive study was carried out in the department of radiology, Ahmadu Bello University Teaching Hospital (ABUTH), a tertiary health institution located in Zaria, Kaduna state, North-western Nigeria.

The study was conducted over a period of 7 months (March to September 2016) after approval had been granted by the Health Research Ethical Committee of ABUTH Zaria.

Consecutively recruited 118 patients attending the ear, nose, and throat clinic in ABUTH, Zaria who fulfilled the clinical criteria for CRS according to the European Position Paper on Rhinosinusitis and Nasal polyps 2012^[7] and the international consensus on CRS^[8] and aged 18 years and above, were included in the study. All recruited patients had PNS radiographs and coronal CT of the PNS.

Patients with previous history of nasal trauma, sinonasal tumor and nonconsenting patients, and those who could not have both investigations done, were excluded.

Informed consent was obtained while information on patient demographics, symptom type, duration of symptoms, presence/absence of allergy, and cigarette smoking (defined by admission of smoking one pack of cigarette per day) were recorded in a data sheet.

Selected patients subsequently underwent both conventional plain radiography (PR) and coronal CT scan of the PNSs.

Paranasal sinus radiography

All X-ray images were processed using computed radiography (Colenta, Austria).

Standard PNS radiographs were obtained with patient in erect position and include open mouth occipito-mental (Waters') and occipito-frontal (Caldwell's) views.

Waters' view: This was achieved by angling the patient's orbito-meatal line (OML) about 45° to the table or image receptor with his or her mouth open. The horizontal central line of the cassette holder was at the level of the lower orbital margins and the median sagittal plane was equidistant from the outer canthi of both eyes. The central ray (CR) was directed perpendicular to the patient's head at a subject to image distance of 100 cm with the centring point at the anterior nasal spine.

Caldwell's view: The patient's head was made to touch the table with his or her nose and forehead while the OML is made perpendicular to the table surface. The CR was angled 15^o downwards projecting through the nasion along the mid-sagittal plane.

To reduce intraobserver error and variability, the best single view that depict each PNS pair was used. As such, Waters' view was used to evaluate the maxillary and sphenoid sinuses, and Caldwell's view for the frontal and ethmoid sinuses.

Radiographic findings in the PNS were predefined in accordance to similar previous study^[9] as follows: Normal (all sinuses are clear in patient), gross mucosal thickening (PNS wall thickening of ≥4 mm), haziness (veiling/reduced translucency with preserved bony markings), air-fluid level (dependent opacity of fluid density with flat surface at the “air-fluid” interface), complete opacification (opacification of entire sinus by opacity more than or equal to bone density), isolated domed opacities were classified as retention cysts, and pedunculated mucosal thickenings were classified as sinus polyps. Other miscellaneous findings were documented.

Computed tomography scan of the paranasal sinuses

A Hispeed NX/i Dual Slice CT scanner (General Electric, Germany) was used for this study.

Due the peculiarity of our CT scanner (two-slice), direct coronal noncontrast enhanced imaging technique was adopted for this study. With patient prone and neck hyperextended, images were obtained at 3 mm through the anterior PNSs, and subsequently, 5 mm contiguous slices (following a scout image) at 120KV and 200 mA as suggested by Babbel et al.^[10] Images were then transferred to the computer workstation through an Ethernet connection. All scans were evaluated using bone window (window width-2300 Hounsfield units (HU); window level-500 HU) and soft-tissue window (window width-100 HU; window level-56 HU). All images were stored and analyzed using the K-PACS DICOM viewer.

For easy comparison, findings were classified similarly as those for PNS X-rays namely: Normal (all sinuses clear in patient), mucosal thickening (sinus wall thickening), retention cysts (isolated domed soft tissue density), polyps (pedunculated mucosal thickening), air-fluid levels (density with flat surface at the “air-fluid” interface), and soft-tissue density (densities that cannot be classified as any of the above and may/may not opacify the entire sinus).

The duration between PNS radiography and CT scan was 24 h for most patients. PNS radiographs and CT scan images were reported first by the author and then validated by two senior consultants from radiology and ENT departments. To minimize observer bias, PNS radiographs were reported on the same day of study while CT images were analyzed retrospectively. Separate data forms for each modality with same-patient identity number maintained for both reports for easy comparison. The cost of the PNS radiography was borne by authors while approval was sort to markedly reduce that of CT since the operational cost of IV contrast used was not involved. Concessions were still made for those who could not afford the reduced fee.

Statistical analysis was performed using the Statistical Package for the Social Sciences software (SPSS) version 20.0 (IBM Corp., Armonk, NY, USA). Tables were used for the display of data. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy of PR for CRS were determined, using two-by-two contingency tables with CT scan as the gold standard.

Results

One hundred and eighteen participants with CRS and a total of 472 pairs (118 × 4) of PNSs were analyzed in this study. The age range of the patients was 18–75 years, with a mean age of 39.1 (+13.32) years. The age was normally distributed with the age group 30–39 years most affected (31.3%), whereas age group < 20 years were least affected (5.1%). The male-to-female sex ratio was 1.4:1 with males accounting for more than half (57.6%) of the patients [Table 1].

Table 1: Age-sex distribution of adults with chronic rhinosinusitis in Zaria (n=118)

Click here to view

Radiographic findings [Figure 1]a and [Figure 1]b were normal in 33 patients (28%). On plain radiographs, the maxillary sinus was the most common sinus affected as seen in 71 patients (60.2%). While the sphenoid sinus was least involved, as detected in 10 patients (8.5%) [Table 2]. Haziness was the most common abnormality detected, seen in 37 patients (31.4%) [Table 3].

Figure 1: Waters' (a) and Caldwell's (b) views of a 34-year old man showing complete opacification of the left maxillary sinus and left frontal sinus mucosal thickening. Representative direct coronal computed tomography scan images show complete and partial opacifications by soft tissue densities in the left maxillary and left ethmoidal sinuses, respectively, (c) with mucosal thickening of the left frontal sinus (d)

Click here to view

Table 2: Paranasal sinus abnormalities and comparative agreement on X-ray and computed tomography scan

Click here to view

Table 3: Radiological findings on X-ray and computed tomography scan in patients with chronic rhinosinusitis in Zaria

Click here to view

On CT scans [Figure 1]c and [Figure 1]d, the PNS most involved was the maxillary sinus, as seen in 89 patients (75.4%), while the frontal sinus was least involved in 34 patients (28.8%) [Table 2]. CT scan findings were normal in 18 patients (15.3%). The most common finding was mucosal thickening detected in 62 patients (52.5%) [Table 3].

Comparing both modalities, the level of agreement for abnormal findings between the two varied according to the sinus studied. Higher agreement was observed for maxillary sinus abnormalities (78.8%), whereas the least agreement (59.3%) was found in the sphenoid sinus [Table 2].

Using CT scan as the gold standard and with at least one abnormality in any of the PNSs considered as positive disease while no abnormality in all sinuses is negative disease, the diagnostic performance of PR for each of the PNSs was evaluated. PR was most sensitive for maxillary disease (76.4%) with specificity, PPV, NPV, and diagnostic accuracy of 89.7%, 95.8%, 55.3%, and 79.7%, respectively. The least performance was for sphenoid disease with sensitivity, specificity, PPV, NPV, and diagnostic accuracy of 17.1%, 96.1%, 70%, 68.5%, and 68.6%, respectively [Table 4].

Table 4: Diagnostic value of plain radiographs in assessing adults with chronic rhinosinusitis

Click here to view

Discussion

This cross-sectional descriptive study conducted in North-Western Nigeria evaluated the patterns of plain radiographic and CT scan findings of CRS in adults as well as determined the diagnostic value of PR (using CT scan as the gold standard) in 118 patients with clinically diagnosed CRS.

The mean age of adults with CRS was 39.1 (+13.32) years in this study which is in agreement with other studies conducted in Nigeria,^{[3],[11],[12],[13],[14],[15]} where the mean age ranged between 31.7 and 39.1 years. The age group most affected in this study was the “30–49” years similar to the observation by Maduforo et al.^[12] This age group constitutes a significant amount of the work force in Nigeria suggesting their willingness to seek and pay for orthodox health care.

The male preponderance seen has been reported by several studies.^[12],[14] Conversely, some previous studies recorded female preponderance,^{[3],[11],[15]} while others recorded no gender predilection.^[16]

On plain radiographs, the maxillary sinus was most affected accounting for 60.2% of patients we investigated. This is in agreement with other studies conducted in Nigeria^{[11],[12],[15]} and abroad^[17] which recorded 66.7% to 78% maxillary sinus involvement. In this study, the findings on the CT scans revealed 75.4% maxillary sinus involvement. Other studies^{[14],[17],[18]} have shown similar maxillary sinus (73.1–81.7%)

involvement on CT scans. It has been postulated that the maxillary sinus is more commonly affected by CRS because of its superiorly-directed (anti-gravity) drainage.^[19] This predisposes it to stagnation of secretions and infection more than the remaining sinuses, which are located above the nasal cavity and have gravity-assisted drainage.

The least involved sinus on plain radiographs in this study was the sphenoid sinus accounting for 8.5%, as similarly reported by previous studies with affectation ranging from 0 to 7.1%.^{[11],[12],[15]} However, on CT scan, this study revealed that the frontal sinus was least affected (28.8%). This discordance was also observed by Stelmach et al.^[17] with the sphenoid sinus least affected on PR (3.2%) and the frontal sinus least affected on CT scans (10.9%). In contrary, previous CT-based studies in Nigeria^[14],[18] reported sphenoid sinus as the least involved. The reason for this difference may be due to the smaller sample sizes (52 and 60 patients) used in the latter studies. Our findings may reflect the actual trend in our environment in view of the larger population evaluated.

On plain radiographs, haziness was the most common abnormality detected in 31.4% of patients in this study. This observation is different from previous studies which reported gross mucosal thickening^[13] and sinus opacity^[15] as the most common findings. Differences in observer opinions on definitions of findings may account for this variation. On CT scans, however, it was observed that mucosal thickening was the most common finding which agrees with a previous CT-based study by Ahmad and Tahir.^[18] It has been postulated using “maxillary sinus disease model” that inflammation in the sinuses initially appear as haziness on X-ray but with progression and recurrence, findings change to pathologically significant mucosal thickening, air-fluid level exudates and then complete opacification of the sinus.^[20] On CT scans where haziness cannot be used as a descriptive term, mucosal thickening may represent the initial appearance of CRS.

In comparing the two imaging modalities for the assessment of the PNSs, the most adequate approach is the determination of the level of agreement for abnormal findings between these modalities.^[17] This study showed the levels of agreement <80% for all sinuses studied. Stelmach et al^.[17] similarly demonstrated <80% agreement in all but the frontal sinus while comparing PR with low-dose CT scan in asthmatics with CRS. Since it is known that CT allows for better evaluation of the PNS, PR was probably associated with false-positive and false-negative cases, suggesting that the use of PR instead of CT of the PNS may result in under-or over-diagnosis.

PR is currently employed as the first-line modality in screening for CRS in our environment. Its discriminative and predictive ability for an abnormality in at least one PNS (when compared to CT scan) was evaluated in this study. For maxillary sinuses, X-rays have been estimated to have high diagnostic values with above-average to excellent sensitivity, specificity, PPV and NPV (76.4%, 89.7%, 95.8%, and 55.3%, respectively). This agrees with previous studies^[13],[21] advocating its relevance in chronic maxillary sinusitis with sensitivities of 60%–88.6% reported. This pattern for the maxillary sinus is not unexpected, as unlike the other sinuses, it is larger and less prone to superimposition by other structures. However, for other sinuses (ethmoid, frontal, and sphenoid sinuses) closer to the brain and orbits where life-threatening complications are more common, this study reveals only average to incredibly low sensitivities. This also concurs with previous comparative study by Chiu et al^.[21] for the ethmoid and sphenoid sinuses (57% and 28.6%, respectively), reflecting difficulty in detecting disease due to superimposition of structures in these sinuses. In this study, however, specificity for the sphenoid sinus disease was the highest (96.1%). As such, the absence of disease in the sphenoid sinus on PNS X-rays may be a reliable finding. However, with sensitivity as low as 17.1%, there is a grave risk of under diagnosing potentially fatal CRS using PR alone. With the poor performance of PR and the knowledge that patients can get frustrated with the management of their CRS,^[22] it will be clinically wise to avoid unnecessary investigation burden by selecting CT as a one-stop modality in evaluating CRS.

Conclusion

This study found that maxillary sinus is the most affected by CRS on both PR and CT scan. The most common radiological findings were haziness and mucosal thickening on PR and CT scans, respectively. When compared to CT scans, the highest diagnostic performance for PR was for maxillary sinusitis, but it was poorer in the other sinusitis anatomically closer to the brain.

Limitation

Since this is a hospital-based study, findings may not reflect the actual situation in the community.

Recommendation

Non-contrast coronal CT scan (preferably multislice-multidetector CT), rather than plain radiograph, should be used as the first-line radiological investigation for adults with CRS in regions where the facility is available. CT scans evaluating CRS can be made safer and more affordable to patients using non-contrast technique which has lesser operational cost.

Acknowledgment

The authors would like to thank the CRS patient participants, the clerical staff of the ENT, and radiological departments of ABUTH, Zaria.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Tomassen P, Van Zele T, Zhang N, Perez-Novo C, Van Bruaene N, Gevaert P, et al. Pathophysiology of chronic rhinosinusitis. Proc Am Thorac Soc 2011;8:115-20.
2.	Iseh KR, Makusidi M. Rhinosinusitis: A retrospective analysis of clinical pattern and outcome in North Western Nigeria. Ann Afr Med 2010;9:20-6. [PUBMED] [Full text]
3.	Fasunla JA, Nwaorgu OG. Adult chronic rhinosinusitis: Spectrum of clinical features and management in a tertiary health institution and literature review. East Cent Afr J Surg 2011;16:12-8.
4.	Sogebi O. Oyewole E. Rhinosinusitis: Clinical features seen in Sagamu, Nigeria. Niger Med Pract 2006;54:28-31.
5.	Fokkens WJ, Lund VJ, Hopkins C, Hellings PW, Kern R, Reitsma S, et al. European position paper on rhinosinusitis and Nasal Polyps 2020. Rhinology 2020;58:1-464.
6.	Bachert C, Zhang N. Medical algorithm: Diagnosis and treatment of chronic rhinosinusitis. Allergy 2020;75:240-2.
7.	Fokkens WJ, Lund VJ, Mullol J, Bachert C, Alobid I, Baroody F, et al. EPOS 2012: European position paper on rhinosinusitis and nasal polyps 2012. A summary for otorhinolaryngologists. Rhinology 2012;50:1-2.
8.	Bachert C, Pawankar R, Zhang L, Bunnag C, Fokkens WJ, Hamilos DL, et al. ICON: Chronic rhinosinusitis. World Allergy Organ J 2014;7:25.
9.	Gupta SC, Singh M, Jain A, Walia DK. A comparative study of radiological and antroscopic findings in the lesions of maxillary sinus. Indian J Otolaryngol Head Neck Surg 2004;56:9-13.
10.	Babbel R, Harnsberger HR, Nelson B, Sonkens J, Hunt S. Optimization of techniques in screening CT of the sinuses. AJNR Am J Neuroradiol 1991;12:849-54.
11.	Sogebi O, Oyewole E, Bajomo A. Radiologic features of chronic rhinosinusitis in Sagamu. Niger Med Pract 2009;54:28-31.
12.	Maduforo CO, Ibinaiye P, Onotai LO. Plain radiographic pattern of chronic sinusitis in Port Harcourt: Our recent experience. Int J Med Med Sci 2013;3:317-20.
13.	Kolo ES. The role of plain radiographs in the diagnosis of chronic maxillary rhinosinusitis in adults. Afr Health Sci 2012;12:459-63.
14.	Amodu EJ, Fasunla AJ, Akano AO, Daud Olusesi A. Chronic rhinosinusitis: Correlation of symptoms with computed tomography scan findings. Pan Afr Med J 2014;18:40.
15.	Da Lilly-Tariah OB, Aniemeka J. Plain radiological profile of paranasal sinuses in chronic nasal diseases in university of Port Harcourt Teaching Hospital. Niger J Med 2006;15:305-8.
16.	Greisner WA 3^rd, Settipane GA. Hereditary factor for nasal polyps. Allergy Asthma Proc 1996;17:283-6.
17.	Stelmach R, Junior SA, Figueiredo CM, Uezumi K, Genu AM, Carvalho-Pinto RM, et al. Chronic rhinosinusitis in allergic asthmatic patients: Radiography versus low-dose computed tomography evaluation. J Asthma 2010;47:599-603.
18.	Ahmad B, Tahir A. Rhinosinusitis in North-Eastern Nigeria: Computerized tomographic scan findings. Niger J Surg Res 2004;5:110-3.
19.	Phelps P. Radiology of the nose and paranasal sinuses. In: Kerr A, Mackay I, Bull T, editors. Scott Brown's Otolaryngology. 6^th ed. Oxford: Butterworth-Heineman; 1997. p. 1-22.
20.	Al-Azzawi A, Al-Umeri K. Comparison of clinical symptoms, plain radiographs, coronal CT and antral lavage in patients with chronic maxillary sinusitis. Med J Babylon 2011;8:73-8.
21.	Chiu PY, Chen JD, Wei JW, Chang CY. The diagnostic value of sinus radiography in the evaluation of sinusitis. Chin J Radiol 2010;35:143-8.
22.	Vennik J, Eyles C, Thomas M, Hopkins C, Little P, Blackshaw H, et al. Chronic rhinosinusitis: A qualitative study of patient views and experiences of current management in primary and secondary care. BMJ Open 2019;9:e022644.