|Year : 2010 | Volume
| Issue : 2 | Page : 173-178
Evaluation of extensions of sinonasal mass lesions by computerized tomography scan
V Annam1, AM Shenoy2, P Raghuram1, V Annam3, JM Kurien1
1 Department of Radiodiagnosis, Kidwai Memorial Institute of Oncology, Bangalore, India
2 Department of Head and Neck Oncology, Kidwai Memorial Institute of Oncology, Bangalore, India
3 Department of Pathology, Sree Siddhartha Academy of Higher Education, Tumkur, Karnataka, India
|Date of Web Publication||5-May-2010|
Department of Radiodiagnosis, Kidwai Memorial Institute of Oncology, Bangalore
Source of Support: None, Conflict of Interest: None
Introduction : Cross-sectional imaging has had a significant impact on the management of the sinonasal malignancy. Staging of these lesions has been closely monitored by dependence on computerized tomography (CT) scan and now in small proportion with MRI. The objective of the study was to evaluate the reliability of CT scan in assessing the extensions of the sinonasal mass lesions and their impact on tumor staging. Materials and Methods : All cases of sinonasal lesions were reviewed retrospectively from June 2001 to May 2006 (five years) at KMIO, Bangalore. Only those cases that had CT scan reports and Histopatholgy Examination (HPE) reports were included in the study. All the cases, which were subjected to radiotherapy prior to surgery, were excluded from the study. There were a total of 38 cases with slight male preponderance. Tumor staging was done according to the sixth edition of TNM classification of malignant tumors. Two observers evaluated all the cases (Blinded Study) and sensitivity, specificity, positive predictive value, negative predictive value and predictive accuracy were estimated. Results : The relation between the CT scan findings and HPE reports was high. Though false positivity was noted more in the ethmoid sinuses, significant impact on tumor staging was noted with false positivity of the nasopharynx. Also, false positivity was noted with orbital wall/content extensions, resulting in unnecessary exenterations of the orbit. However, CT scan reliably predicted infratemporal fossa extension in the absence of trismus. False negativity was noted more often in soft palate, indicating the need of using MRI for delineation of tumor extension. Thus, we conclude that judicious use of CT scan in all cases and MRI in selected cases will translate into better tumor-free resections and improves survival.
Keywords: Computerized tomography scan, sinonasal lesions, extensions, tumor staging
|How to cite this article:|
Annam V, Shenoy A M, Raghuram P, Annam V, Kurien J M. Evaluation of extensions of sinonasal mass lesions by computerized tomography scan. Indian J Cancer 2010;47:173-8
|How to cite this URL:|
Annam V, Shenoy A M, Raghuram P, Annam V, Kurien J M. Evaluation of extensions of sinonasal mass lesions by computerized tomography scan. Indian J Cancer [serial online] 2010 [cited 2020 Apr 10];47:173-8. Available from: http://www.indianjcancer.com/text.asp?2010/47/2/173/63016
| » Introduction|| |
Sinonasal mass lesions are challenging not only from diagnostic viewpoint, but also from therapeutic aspects. Owing to the rarity of the lesions, delayed clinical presentations and poor prognosis, no one center has represented a large experience. Evaluating the extensions of the lesion helps in staging the disease and this is done effectively and economically by the commonly available cross-sectional imaging modality, the computerized tomography (CT) scan. Staging of the lesion in this manner helps in deciding between extended invasive surgery, standard surgery and in management of the eye. However, the preoperative assessment of the orbital invasion, nasopharynx and soft palate with imaging has not been extensively studied as has been done with respect to intracranial extension.
The purpose of this retrospective study was to evaluate the reliability of CT scan in assessing the extensions of the sinonasal mass lesions and their impact on tumor staging. This was done by relating the appearances on CT scan with histopathology reports (HPR).
| » Materials and Methods|| |
A retrospective study was carried out between May 2001 and June 2006. Patients with operable sinonasal malignancies were considered. Of the 38 patients, 21 patients were males and 17 were females. The mean age range was 51 years in males and 40 years in females. Only patients with sinonasal lesions who have undergone both NECT and CECT scan PNS with Somatom Spiral CT scanner (Siemens) prior to surgery were included. Information regarding anatomical areas, from CT scan and histopathology were taken. Post radiotherapy cases and recurrent cases were not included in the study. Two observers evaluated all the cases (Blinded Study). Sensitivity, Specificity, Positive predictive value, Negative predictive value and Predictive Accuracy were estimated based on 2 Χ 2 table generated for estimating these parameters.
In each of the 38 patients, the following anatomical areas were assessed with CT scan and HPR findings: skin, subcutaneous tissue, orbital floor, medial orbital wall/lamina papyracea, anterior orbital contents, retromaxillary space, pterygoid plates, pterygoid muscles, maxillary antrum, ethmoid sinus, frontal sinus, sphenoid sinus, intracranial extension either through cribriform plate or fovea ethmoidalis, hard palate, soft palate and nasopharynx.
Only those areas where information was available from both CT scan and HPR were included.
| » Results|| |
A total of 608 (38 Χ 16) anatomical areas were assessed. The CT scan findings were confirmed by HPR findings (TP + TN) in 526 anatomical areas (86.5%). Discrepancy was noted between the CT scan findings and HPR findings (FP + FN) in the remaining 82 anatomical areas (13.48%) [Table 1].
Based on these findings, the Sensitivity, Specificity, Positive predictive value, Negative predictive value and Predictive accuracy of the study were calculated [Table 2].
Sensitivity was least with medial orbital wall extension. Accuracy in evaluating the sinonasal lesion extension was least with ethmoid sinus extension.
Using the above information, tumor staging (T) of the sinonasal malignancies were done using AJCC TNM classification of malignant tumors, sixth edition, 2002  [Table 3].
Among the 38 sinonasal cases, correlation between the CT scan findings and HPE findings were noted in 24 cases. They include two cases of T1 stage, three cases of T2 stage, eleven cases of T3 stage, six cases of T4a stage and two cases of T4b stage.
Among the remaining 14 cases, upstaging was noted in 12 cases and downstaging was noted in 2 cases, notably in advanced cases [Table 4].
Upstaging was noted in one case of T1, three cases of T2, seven cases of T3 and one case of T4a.
Most commonly T3 was upstaged as T4b (five cases) and all were due to the false positivity at nasopharynx. The remaining two cases of T3 were upstaged as T4a, both of which were due to false positivity of frontal and sphenoid sinuses.
Out of the three cases of T2, two were upstaged as T3 and one as T4b. Among the two cases, which were upstaged as T3, one was due to false positivity of Ethmoid sinus and posterior maxillary wall bone, and the other was due to false positivity of ethmoid sinus only. The case, which was upstaged as T4b, was because of false positivity at nasopharynx.
One case of T1 was upstaged as T4a because of false positivity at sphenoid sinus.
One case of T4a was upstaged as T4b because of false positivity at nasopharynx.
False positivity of nasopharynx was the most common cause of upstaging of sinonasal lesions.
Downstaging was noted in one case of T4a and one case of T4b. The T4a case was downstaged as T3 because of false negativity of frontal sinus. The T4b case was downstaged as T4a because of false negativity of intracranial extension.
Of the 38 cases, orbital exenteration was attempted in 10 cases. Out of the 10 cases, CT scan showed involvement of the orbital floor only in 4 cases and this was confirmed by HPR. The orbital contents were found to be free of the tumor extension on both CT and HPR in these 4 cases. In the remaining 6 cases, CT scan showed extension of the lesion into both the orbital floor and orbital contents. However, HPR revealed that in four cases, only the orbital floor was affected and in the remaining two cases, both orbital floor and orbital contents were affected. Thus, of the six cases, the orbital contents were not affected in four cases and the orbital floor was not affected in two cases. Of the two cases where the orbital contents were affected, the globe was affected in only1 case and in the remaining case; the inferior rectus muscle and adjacent periorbital fat were found to be involved (globe was normal). Thus, out of the 10 cases where orbital exenteration was attempted, only one case showed globe infiltration [Table 5].
The pathological diagnosis of sinonasal malignancies shows the incidence of squamous cell carcinoma to be high followed by adenoid cystic carcinoma [Table 6].
| » Discussion|| |
Carcinomas of the nasal cavity and paranasal sinuses are rare, comprising only 0.2 to 0.8% of all malignancies and just 3% of all tumors that arise in the head and neck region. Despite their statistically small incidence, sinonasal tumors are feared because of their poor prognosis  According to registry (2007) at Kidwai Memorial Institute of Oncology, sinonasal malignancies constitute 6% of all tumors arising in the head and neck region.
Sinonasal malignancies are frequently asymptomatic in the initial stages or produce symptoms, which are indistinguishable from the more common disease processes, such as chronic sinusitis. At diagnosis, they are often, therefore, at an advanced stage and have usually spread beyond the site of origin. ,
Demonstration of the location and true local extent of the sinonasal malignancies is essential in choosing the appropriate treatment modality  and ensuring an oncologically satisfactory resection with negative cut margins. CT scanning has established as an important component of any pre-operative assessment of malignant disease in head and neck as it provides important information regarding the soft tissue component of tumors and bony erosions, which gives the surgeon a more accurate picture of the extent of disease spread and the degree of resectability.  It is more sensitive in assessing the osseous margins of the sinonasal cavity, the osseous floor of the anterior cranial fossa and the walls of the orbit (i.e., roof, floor and medial wall).  The sites of greatest importance include bony orbital walls, pterygoid region, pterygopalatine fossa, nasopharynx, sphenoid sinus, posterior wall of the frontal sinus, posterior ethmoid air cells, cribriform plate, orbital apex and base of the skull. ,
Lund et al. studied 22 patients and correlated the findings on CT scans with the operative and histologic findings. Of the 339 anatomic areas analysed, there was complete correlation between the three methods of assessment in 263 areas (78%).  In our study, 608 anatomical areas were analysed and there was complete correlation between CT scan and histologic findings in 526 cases (86.5%).
Evaluation of the orbit : Graamans and Slootweg studied 15 patients for orbital invasion and concluded that though the strength of CT is its ability to evaluate both bone and fat, it is difficult to distinguish tumor that compresses versus the tumor that invades the periorbita.  In our study, invasion of the orbital floor, medial orbital wall/Lamina Papyracea and anterior orbital contents were assessed accurately in 30 cases (78.94%), 31 cases (81.57%) and 33 cases (86.84%), respectively. [Figure 1]. In contradistinction, CT scan revealed orbital floor invasion in 8 patients (21.05%), medial orbital wall/Lamina Papyracea invasion in 5 patients (13.15%) and invasion of the anterior orbital contents in 5 patients (13.15%), but none were confirmed on histologic examination (false positivity). Also, CT scan showed no evidence of medial wall/Lamina Papyracea invasion in 2 patients (5.26%); however, invasion was noted in both these cases on histology examination (false negativity)
In a study by Carrau et al, 58 patients with bone and/or soft tissue invasion of the orbit were included. Surgical resection, sparing the orbital contents was done in patients with only soft tissue invasion and en bloc resection, including orbital exenteration was done in patients with invasion of both orbital bones and soft tissues. At 3-years follow-up, 52% of the patients whose orbit was exenterated were alive and without evidence of the disease, compared with 59% of the patients whose orbit was spared. They concluded that sparing of the soft tissues of the orbit when the periorbita has not been transgressed by squamous cell carcinoma does not downgrade the rate of cure or local control.  Imola and Schramm Jr concluded that selective orbital preservation is oncologically safe and is a worthwhile undertaking in attempting to maintain a functionally useful eye with surgical management of sinonasal malignancy encroaching the orbit.  In our study, orbital exenteration was attempted in 10 of the 38 cases, where periorbital fat was thought to be involved on CT scan. However, histology showed involvement of the globe in only one case. Hence, the final decision with respect to exenteration or preservation of the globe that is functionally viable should be taken at surgery after frozen section confirmation of involvement of periorbital fascia or fat. This will lead to a few number of needlessly performed surgical exenteration of the orbit, thereby reducing the emotional burden for patients and their families.
Evaluation of Intracranial extension (ICE) : ICE can be shown most readily on CT scan.  In our study, presence or absence of ICE were noted accurately in 36 cases. In one case, suspicious infiltration into the cribriform plate was noted on CT scan. However, at surgery the lesion was seen extending up to the cribriform plate. In one case, at surgery the lesion was found extending into the middle cranial fossa. This might have been due to the aggressiveness of the tumor (Squamous cell carcinoma of left Maxilla) and prolonged period between diagnosis and surgery.
Evaluation of Infratemporal fossa (ITF) : Invasion of the posterior wall of the maxillary antrum and of the ITF is particularly well shown on high resolution CT. It is possible to show simple expansion of the posterior antral wall; expansion with early erosion, or correct invasion of the soft tissues and musculature. The low attenuation of the deep temporal fat pad serves as contrast to the invading tumor.  Such extension is usually not recognized clinically until trismus is present.  In our study, invasion into the ITF/posterior maxillary wall was accurately assessed in 36 cases. In the remaining 2 cases, the posterior wall was found to be intact with no involvement histologically, though suspicious on CT scan. Trismus, which is considered as an important clinical sign of ITF involvement, was not evident clinically in any of the cases [Figure 2].
Evaluation of the Paranasal sinuses : Opacification of the contiguous sinuses may indicate that tumor has spread from one to the other or that the ostium is blocked by the tumor mass, causing retention of the secretions within the other sinus. This is particularly common when tumor extends into the apex of the nasal airway, obstructing the ethmoid air cells.  In our study, retained or inspissated secretions and/or thickened mucosa within the paranasal sinuses were more commonly misinterpreted as extensions of the malignancy (false positive). This was noted more commonly in the ethmoid sinuses, followed by sphenoid sinuses and frontal sinuses. With regard to misinterpretation at the paranasal sinuses, it appears that the non-dependent location of these sinuses with their tenuous drainage ducts may lead to primary occlusion of the draining pathways by the mass lesions with secondary changes in these sinuses, probably of an inflammatory nature. The false positivity of all the paranasal sinuses altered the tumor staging (T3 and T4a) to a certain extent and therefore requires complementary assessment by MRI, which will reveal the differences between true disease infiltration and obstruction secondary to infiltration of the draining ostia.
Evaluation of the adjacent soft tissues : In our study, extension of the lesions into adjacent soft tissues like nasopharynx, soft palate and skin were misdiagnosed to a certain extent. However, the sensitivity and specificity of CT scan in identifying the sinonasal extensions into the subcutaneous tissue was very high.
False positivity of the nasopharynx had a major impact in the alteration of the tumor staging of the sinonasal lesions. The false positivity might have been due to misdiagnosis of associated inflammatory processes as sinonasal infiltrations. Also, the extension of the sinonasal malignancy up to the nasopharynx indicates the protrusion of tumor towards area of least resistance viz from nasal cavity towards the choanae. However, at the level of the nasopharynx, CT scan may not be able to differentiate between tumor extension and tumor infiltration. To increase the sensitivity it is recommended that the MRI should be used to complement and evaluate CT scan findings wherever there is a radiological enhancement at nasopharynx.
Infiltration of the sinonasal lesions into the soft palate was more often missed. This again reflected the inadequacy of CT scan evaluation when disease extension occurs at the soft tissue level.
Infiltration into the overlying skin was over diagnosed in three cases and missed in two cases, with correct diagnosis being made in the remaining 33 cases. In all the five cases, subcutaneous tissue infiltration was noted.
| » Conclusion|| |
The accuracy of the CT scanner in the preoperative assessment of the extension of the lesions into the orbit and its contents is debatable. This has resulted sometimes in unnecessary orbital exenteration, thereby causing patient inconvenience.
The retained secretions within the paranasal sinuses, especially the ethmoid sinus were more often mistaken as extensions of the sinonasal lesion, thereby altering the tumor staging to a certain extent.
Extensions into the nasopharynx were more often over diagnosed, as they were misinterpreted as infiltrations, probably due to associated inflammatory processes. This had an impact on tumor staging to a significant extent.
Infiltration into the soft palate was often missed. This indicates the need for investigation by higher modality like MRI for delineation of the tumor extensions, so that the tumor staging and thence the treatment modality can be planned accurately.
Infiltrations into the subcutaneous tissues were interpreted accurately, whereas infiltrations into the overlying skin were missed in a few cases.
Computerized Tomography accurately demonstrated the extension of the tumor into the infratemporal fossa before trismus was evident clinically. Also, the osseous changes were clearly demonstrated.
Thus, judicious use of CT scans in all cases and MRI in selected cases will translate into better tumor-free resections and improves survival.
| » References|| |
|1.||TNM Classification of Malignant tumors. Wiley-Liss. Sixth edition; 2002. p. 43-7. |
|2.||Som PM, Bergeron RT. Head and Neck Imaging. 2 nd edition 1991. p. 169-224. |
|3.||Forbes WS, Fawcitt RA, Isherwood I, Webb R, Farrington T. Computed Tomography in the diagnosis of diseases of the paranasal sinuses. Clin Radiol 1978;29:501-11. [PUBMED] |
|4.||David Sutton. Textbook of Radiology and Imaging. 7 th edition. 2003. p. 1526-8. |
|5.||Parsons C, Hodson N. Computed tomography of paranasal sinus tumors. Radiology 1979;132:641-5. [PUBMED] |
|6.||Lund VJ, Howard DJ, Lloyd GA. CT evaluation of paranasal sinus tumors for cranio-facial resection. Br J Radiol 1983;56:439-46. [PUBMED] |
|7.||Loevner LA, Sonners AI. Imaging of neoplasms of the paranasal sinuses. Neuroimaging Clin N Am 2004:14;625-46. |
|8.||Mafee MF. Imaging of the paranasal sinuses and oromaxillofacial region. Radiol Clin North Am 1993;31:61-90. [PUBMED] |
|9.||Graamans K, Slootweg PJ. Orbital exenteration in surgery of malignant neoplasms of the paranasal sinuses. Arch Otolaryngol Head Neck Surg 1989;115:977-80. [PUBMED] [FULLTEXT] |
|10.||Carrau RL, Segas J, Nuss DW, Snyderman CH, Janecka IP, Myers EN, et al. Squamous cell carcinoma of the sinonasal tract invading the orbit. Laryngoscope 1999;109:230-5. [PUBMED] |
|11.||Imola MJ, Schramm VL Jr. Orbital preservation in surgical management of sinonasal malignancy. Laryngoscope 2002;112:1357-65. [PUBMED] [FULLTEXT] |
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]