Abstract
Objectives: Magnetic resonance imaging (MRI) is able to demonstrate a wide range of abnormalities in the paranasal sinuses, which are often reported as incidental findings on scans performed for indications other than the evaluation of paranasal sinus pathology. However, the clinical significance of these findings remains undefined. We present a prospective study that determines the prevalence of abnormalities in the paranasal sinuses in a population undergoing MRI scans for suspected intracranial disease. These findings are correlated with clinical data pertaining to nasal and sinus symptoms.
Study Design: Prospective, cross-sectional study.
Methods: Patients undergoing MRI scans for suspected intracranial pathology were asked to complete a questionnaire pertaining to symptoms of nasal/sinus pathology. The T2-weighted scans of 86 patients (mean age = 51 years) were then reviewed for evidence of paranasal sinus pathology using a standardized method for evaluation and reporting of results. These results were then correlated with those obtained from the patient questionnaire.
Results: Radiologic abnormalities were found in the paranasal sinuses of 33 (38%) patients. Abnormalities were most commonly seen in the ethmoid sinuses (44.8%) followed by the maxillary (38%), sphenoid (14%), and frontal (3%) sinuses. Analysis of the clinical data revealed no significant relationship between the presence of clinical symptoms of nasal and sinus pathology and abnormalities on MRI scan.
Conclusion: The assessment of inflammatory sinus pathology remains controversial. Based on the results of this study, incidental abnormalities of the paranasal sinuses detected on MRI scan do not appear to be related to clinical symptoms.
Sommaire
Objectif: L'imagerie par resonance magnetique (IRM) est capable de demontrer une grande variete d'anomalies des sinus paranasaux qui sont en fait des decouvertes fortuites sur des examens faits pour des indications autres que l'evaluation d'anomalies des sinus. La signification clinique de ces decouvertes reste a preciser. Nous presentons une etude prospective qui determine la prevalence des anomalies des sinus paranasaux chez une population subissant une IRM pour suspicion de pathologies intracraniennes. Nous correlons ces anomalies avec les donnees cliniques nasales et sinusiennes pertinentes. Devis: Etude prospective de cohorte.
Methode: Nous avons demande a des patients devant subir une IRM cerebrale de completes un questionnaire sur leurs symptomes nasaux et sinusiens. Les images en ponderation T2 de 86 patients furent ensuite evaluees selon une methode standardisee. Les resultats cliniques et radiologiques furent ensuite compares.
Resultats: Nous avons detecte des anomalies radiologiques chez 33 patients (38%). Les sinus ethmoides sont les plus souvent affectes (44.8%) suivi par les maxillaires (38%), les sphenoides (14%), et finalement les frontaux (3%). L'analyse des donnees cliniques n'a pas permis de mettre en evidence de relation entre les symptomes cliniques et les anomalies A l'imagerie.
Conclusion: L'evaluation des pathologies inflammatoires des sinus demeure controversee. D'apres cette etude, les anomalies naso-sinusiennes fortuites notees a l'IRM, ne semblent pas relies avec la symptomatologie.
Key words: magnetic resonance imaging, sinusitis, symptoms
The appropriate evaluation of patients suspected of having inflammatory paranasal sinus pathology remains controversial. Magnetic resonance imaging (MRI) can provide detailed and unique images of the paranasal sinuses. However, the clinical significance of radiologic abnormalities on these scans has yet to be established. Furthermore, MRI scans of the head, performed for suspected intracranial or other nonsinus pathology often reveal incidental evidence of inflammatory sinus pathology. For example, Patel and his colleagues reported a retrospective study in which they found incidental abnormalities of the paranasal sinuses in 49% of patients undergoing head MRI scans in a British population.' The determination of the clinical significance of these findings has important implications in terms of patient management as the incorrect interpretation of these results may lead to the institution of unnecessary therapeutic interventions.
Routine MRI scans of the head performed because of suspected intracranial disease are usually conducted in the axial plane and include views of all four sinuses. The vast majority of acute and chronic inflammatory diseases of the sinuses have bright signal intensities on T2-weighted images secondary to increased water content. In contrast, the appearance of inflammatory processes on Ti-weighted images is more variable. In our experience, sinus abnormalities are frequently noted on MRI scans of patients referred for neuroradiologic assessment (Fig. 1). We report the prevalence, site, and type of abnormalities occurring in this population and relate these abnormalities to the presence of nasal and sinus symptoms so as to better determine the clinical significance of these findings.
Materials and Methods
A prospective study was performed on 86 consecutive patients who underwent head MRI for suspected intracranial pathology from April 1998 to May 1998 at the Department of Radiology, Methodist Hospital, Memphis, Tennessee. In each patient, the T2-weighted axial scans were examined. When sagittal and coronal scans were available, these were also examined. A standardized scale was used to evaluate abnormalities and was classified as follows:
1. Normal: fully aerated with no soft tissue or fluid visible within its cavity
2. Minimal mucosal thickening: the mucosal lining was visible but occupied less than 50% of the cross-section of the sinus
3. Marked mucosal thickening: the mucosa occupied more than 50% of the cross-sectional area, but the sinus still contained some air
4. Polyp or retention cyst: a lesion with rounded edges surrounded by air
5. Fluid: a straight line, air-fluid interface
6. Other: anatomic abnormalities presumed to be attributable to previous nasal and sinus surgery
The findings were recorded according to the site and type of abnormality. A sinus was considered normal if it was fully aerated and no soft-tissue density was apparent within the cavity. All images were independently evaluated by at least two of the authors, and the results of these assessments were compared and recorded. When the two evaluations differed, the two evaluators reviewed the scans together, and a final report was generated. Immediately prior to undergoing their MRI scans, each patient completed a questionnaire pertaining to the incidence and severity of symptoms indicative of nasal and sinus pathology (Table 1). The results of the clinical questionnaire were then correlated with data obtained from the MRI scans.
Results
A total of 86 patients were evaluated (35 males, 51 females), with an age range of 19 to 84 years (mean = 51 years). Thirty-three (38%) of the patients studied showed an abnormality of one or more sinus groups. Radiologic abnormalities were most common in the ethmoid sinuses (45%), followed by the maxillary (38%), sphenoid (14%), and frontal (3%) sinuses (Fig. 2). A single sinus (or sinus group) was reported as abnormal in 22 patients (26%), and 17 patients had more than one abnormal sinus (20%) (Fig. 3).
Mucosal thickening was the most common category of abnormality seen in the ethmoid, sphenoid, and frontal sinuses. Mucosal polyps, retention cysts, air-fluid levels, and a hypoplastic sinus were also seen in the maxillary sinuses (Fig. 4).
The results obtained from the questionnaire indicated that 18 patients (22%) reported regular smoking, 8 (9%) were currently taking sinus medications, 12 (14%) had been diagnosed with chronic sinusitis by their physician, 16 (19%) were told that they suffered from allergies, 31 (36%) had taken at least one round of antibiotics for a sinus infection, 14 (16%) suffered from chronic facial pain, 10 (12%) reported a chronic runny nose, and 17 (20%) suffered from chronic nasal obstruction.
Data obtained from the clinical questionnaire were compared with those obtained from MRI scans (Fig. 5). Statistical analyses (chi-square) failed to reveal any significant relationship between the presence of symptoms indicative of sinonasal pathology and abnormalities on MRI scan.
Discussion
The results of this study indicate that there is a high incidence (38%) of incidental abnormalities in the paranasal sinuses identified on MRI scans obtained for the evaluation of intracranial pathology. However, no relationship between clinical symptoms of sinonasal pathology and abnormalities on MRI scan was identified. These results are consistent with previous studies reported in the literature using different imaging techniques to establish the incidence of radiologic abnormalities seen in the paranasal sinuses of asymptomatic patients (Table 2).1-6 The incidence of abnormalities reported in this study is similar to that reported in Cooke and Hadley's study4 (37.5%) and slightly lower than Patel et al.'s^sup 1^ (49%). Both studies were performed on a British population who underwent an MRI for symptoms and diseases not related to the paranasal sinuses. In both studies, findings were analyzed according to the anatomic location and the presence of mucosal thickening, sinus opacifications, fluid levels, and retention cysts/polyps. Patel et al.'s studyl was retrospective, so clinical correlation with any paranasal complaints was not available. In Cooke and Haley's4 study, both a retrospective and a prospective group were evaluated. The only symptom that correlated with an abnormality on MRI was the presence of an upper respiratory infection.
The appropriate work-up for patients suspected of having paranasal sinus disease has been well documented.7 The time-honoured history and physical examination remain the basic elements for evaluating sinonasal disease. In addition, careful examination of the nose with optical telescopes helps confirm the history and can document evidence of sinus disease and status of the middle meatus. If the patient does not respond to 10 to 14 days of appropriate medical therapy, computed tomography studies provide the standard of care in imaging. Computed tomography has the ability to optimally display bone, soft tissue, and air, thus facilitating accurate depiction of anatomy and the extent of disease in and around the paranasal sinuses. Computed tomography is currently the modality of choice in the evaluation of the paranasal sinuses and adjacent structures and is mandatory prior to surgical treatment.8
The role of MRI in the evaluation of inflammatory paranasal sinus disease remains unclear. The potential utility of MRI in evaluating the paranasal sinuses is owing to the perpendicular plane used for imaging, the high signal intensity on T2-weighted images of almost all inflammatory sinus abnormalities, and the absence of signal from structures surrounding the sinuses (bone, air).5 It therefore becomes important to distinguish incidental findings or normal changes such as an asymptomatic retention cyst or mucosal thickening from true pathology.
Our study, as well as others, indicates that MRI scans may reveal a high incidence of abnormalities that are not clinically significant. Zinreich et al.9 studied the normal nasal cycle using MRI and found that in a normal adult, there are cyclical changes in the nasal mucosal volume. These changes also are observed in the mucosa of the nasal cavity and the ethmoid sinuses but do not affect the maxillary, frontal, or sphenoid sinuses. This may explain the increased incidence of abnormalities observed in the ethmoid sinuses in our study and in previous studies.1,3,6
Although MRI evaluation may not be accurate when appraising routine inflammatory sinus pathology, it does have certain defined indications in the evaluation of sinus pathology. With respect to sinus imaging, MRI has proven most helpful in the evaluation of regional and intracranial complications of inflammatory sinus disease and their surgical management and in improved display of anatomic relationships between intra- and extraorbital compartments.10 It has also been shown to be the best method of determining the extent of malignant disease6 and is also helpful in the evaluation of mucoceles and cephaloceles.11
Conclusion
The assessment of inflammatory sinus pathology remains controversial. Based on the results of this study, MRI will demonstrate incidental sinus abnormalities that usually have no clinical significance. Therefore, it would appear that MRI scan is not an accurate method to screen for paranasal sinus disease. It should be reserved for specific clinical indications in patients with known paranasal sinus disease, for example, to assess intracranial extension or distinguish a tumour from retained secretions or inflamed sinus mucosa. This study further highlights the importance of the clinical presentation when interpreting incidental abnormalities on the MRI of the paranasal sinuses.
References
1. Patel K, Chavda SV, Violaris N, et al. Incidental paranasal sinus inflammatory changes in a British population. J Laryngol Otol 1996; 110:649-651.
2. Fascenelli FW. Maxillary sinus abnormalities: radiographic evidence in an asymptomatic population. Arch Otolaryngol 1969; 90:98-101.
3. Havas TE, Mothey JA, Gullane Pj. Prevalence of incidental abnormalities on computed tomographic scans of the paranasal sinuses. Arch Otolaryngol 1988; 114:856-859.
4. Cooke LD, Hadley DM. MRI of the paranasal sinuses: incidental abnormalities and their relationship to symptoms. j Laryngol Otol 1991; 105:278-281.
5. Moser FG, Rubin JS, Honigsberg RM, et al. Incidental paranasal sinus abnormalities on MRI of the brain. Clin Radiol 1991; 43:252-254.
6. Lloyd GA. CT of the paranasal sinuses: a study of a controlled series in relation to endoscopic sinus surgery. J Laryngol Otol 1990; 104:477-481.
7. Lund VJ, Kennedy D. Quantification for staging sinusitis. The Staging and Therapy Group. Ann Otol Laryngol 1995; 67(Suppl):17-21.
8. Zinreich S. Rhinosinusitis: radiologic diagnosis. Otolaryngol Head Neck Surg 1997; 117:527-534.
9. Zinreich SJ, Kennedy DW, Malat J, et al. Fungal sinusitis: diagnosis with CT and MRI imaging. Radiology 1988; 169:439-444.
10. Som P, Curtin H. Chronic inflammatory sinonasal diseases including fungal infections: the role of imaging. Radiol Clin North Am 1993; 31:33-44.
11. Weber A. Inflammatory diseases of the paranasal sinuses and mucoceles. Otolaryngol Clin North Am 1988; 21:421-437.
Manish K. Wani, MD, Michael J. Ruckenstein, MD, MSc, FA CSC, and Salil Parikh, MD
Received 09/06/00. Received revised 14/11/00. Accepted for publication 08/12/00.
Manish K. Wani: Department of Otolaryngology, Head and Neck Surgery, University of Tennessee-Memphis; Michael J. Ruckenstein: Department of Otorhinolaryngology, Head and Neck Surgery, The University of Pennsylvania, Philadelphia, Pennsylvania; Said Parikh: Department of Radiology, University of Tennessee, Memphis, Tennessee.
Address reprint requests to: Dr. Manish K. Wani, Department of Otolaryngology, Head and Neck Surgery, University of Tennessee-Memphis, 956 Court Ave., Suite B226, Memphis, TN 38163.
Copyright Decker Periodicals, Inc. Sep/Oct 2001