Maxillary Sinus Pleomorphic Adenoma: A Systematic Review

Abstract

Background: Pleomorphic adenoma (PA), also known as a mixed tumor, is a neoplasm of the glandular tissue. The maxillary sinus (MS) is an atypical location of PA; however, as an empty cavity inside the bone, it predisposes tumors to develop unnoticed. The aim of the following systematic review was to identify and characterize the diagnostics and prognosis of maxillary sinus pleomorphic adenoma (MSPA). Methods: The cases of MSPA that underwent treatment and were observed for possible recurrence were qualified for the review. The medical databases were searched using the following engines: ACM, BASE, Google Scholar and PubMed. The risk of bias was assessed using the JBI Critical Appraisal Tool for Case Reports. The extracted data were tabulated and summarized in a descriptive manner. Results: To the best knowledge of the authors, only seven cases of MSPA were described in the last 20 years. Relapse occurred in three of these cases and malignant transformation occurred in two. The longest recorded time without recurrence was 4 years. The size of the detected MSPA in the largest dimension ranged in various cases from 40 to 60 mm, with an average of 48 mm. Discussion: All but one of the seven included reports showed flaws during the risk of bias assessment. Only in two of seven reported cases was there no reason to suspect that the tumor had penetrated the sinus from the oral or nasal cavity. The MS is disadvantageous as a location as tumors occupying the entire volume of the sinus are often diagnosed. Recurrences and malignant transformations seem to be frequent and often remain undetected for a prolonged period. Other information: This research received no external funding. OSF Registries number: 8KVGM.

1. Introduction

1.1. Rationale

Pleomorphic adenoma (PA), also known as a mixed tumor, is a neoplasm of the glandular tissue. PA typically develops from the parenchyma of the salivary glands. A mixed tumor of glandular tissue also occasionally occurs within the lacrimal glands and skin [1,2,3]. Within the skin, it is derived from the sweat glands and is called chondroid syringoma [1,3].

It is formally assumed that PA is a benign and encapsulated neoplasm, which theoretically should facilitate its diagnostics, surgical removal and limit recurrence [1,4]. In fact, the capsule surrounding the PA has defects, and the tumor itself has protrusions penetrating the surrounding tissues [1,4]. Thus, PA is often locally recurring, which is the result of non-radical resections [5,6]. Repeated surgeries of PA recurrences may force tumor tissue into the vessels and lead to the distant development of further foci, which is diagnosed as a metastatic form of PA [7].

The paired maxillary sinuses (MSs) are located above the oral cavity. They are naturally connected with the nasal cavity, but tumors from underneath, including those from the palate’s salivary glands, can also penetrate into the MSs and freely expand [8]. The MS is an atypical location of PA; however, as an empty cavity inside the bone it predisposes tumors to develop unnoticed. Therefore, the diagnosis of MS PA may occur once the neoplasm is already extensive or has even transformed into carcinoma ex PA [2,7,9].

1.2. Objectives

This systematic review was undertaken to identify and characterize maxillary sinus pleomorphic adenoma (MSPA) cases in terms of diagnostics and prognosis. The main research question is concerned with the recurrence rate of MSPA depending on the treatment. A secondary research question is the stage at which MSPA is typically diagnosed, and is determined according to the greatest dimension of the tumor at the time of detection.

2. Materials and Methods

2.1. Eligibility Criteria

The inclusion and exclusion criteria were established according to the PICOTS scheme (

Table 1. Criteria for inclusion and exclusion of studies.

	Inclusion Criteria	Exclusion Criteria
Problem	The presence of PA within the maxillary sinus	-
Intervention	Any form of treatment	-
Comparators	Not applicable
Outcomes	Time before tumor recurrence, or length of follow-up in the absence of recurrence	No follow-up after the intervention
Timeframe	Articles published in 2002–2022
Settings	Case reports as separate articles and as components of larger primary studies	Articles in languages other than English

) [10].

2.2. Information Sources

For the sake of transparency and repeatability, this systematic review was based on the following open access search engines: Association for Computing Machinery: Guide to Computing Literature (ACM), Bielefeld Academic Search Engine (BASE), Google Scholar and National Library of Medicine: PubMed (PubMed). Final searches using all the above-mentioned engines were performed on 24 April 2022.

2.3. Search Strategy

Search strategies for all engines used to collect review records are shown in

Table 2. Search strategies.

	Search Strategy
ACM	[[All: “pleomorphic adenoma”] OR [All: “mixed tumor”] OR [All: “adenoma pleomorphum”] OR [All: “tumor mixtus”]] AND [[All: “maxillary sinus”] OR [All: “maxillary antrum”] OR [All: “sinus maxillaris”] OR [All: “antrum maxillaris”] OR [All: “highmore sinus”] OR [All: “highmore antrum”] OR [All: “sinus highmori”] OR [All: “antrum highmori”]]
BASE	(“pleomorphic adenoma” “mixed tumor” “adenoma pleomorphum” “tumor mixtus”) AND (“maxillary sinus” “maxillary antrum” “sinus maxillaris” “antrum maxillaris” “highmore sinus” “highmore antrum” “sinus highmori” “antrum highmori”)
Google Scholar	(“pleomorphic adenoma” OR “mixed tumor” OR “adenoma pleomorphum” OR “tumor mixtus”) AND (“maxillary sinus” OR “maxillary antrum” OR “sinus maxillaris” OR “antrum maxillaris” OR “highmore sinus” OR “highmore antrum” OR “sinus highmori” OR “antrum highmori”) sorted by date
PubMed	(“pleomorphic adenoma” OR “mixed tumor” OR “adenoma pleomorphum” OR “tumor mixtus”) AND (“maxillary sinus” OR “maxillary antrum” OR “sinus maxillaris” OR “antrum maxillaris” OR “highmore sinus” OR “highmore antrum” OR “sinus highmori” OR “antrum highmori”)

.

2.4. Selection Process

The selection process for this study was based on the PRISMA guidelines [11] (Supplementary Materials). Searches using all engines were made on the same day, and the results in the form of records were entered into the Rayyan application (Qatar Computing Research Institute, Doha, Qatar and Rayyan Systems, Cambridge, MA, USA) [12]. Based on the comparison of the abstracts’ contents, possible duplicates were automatically indicated, and removed following additional manual verification. The same was performed for records outside the assumed timeframe. The remaining entries were screened and then analyzed in the full text. In the case of non-compliance with the assumed criteria, the records at the screening stage and reports at the full text stage were removed with the reasons given. In case of doubt in the screening assessment, the entire report was qualified for the analysis.

2.5. Data Collection Process and Data Items

The data were extracted without the use of automation tools.

Table 3. Data items.

Outcomes	Other Variables
(1) The time frame from treatment to first recurrence. (2) Total observation time in the absence of a recurrence. (3) The size of the tumor in the greatest dimension.	(1) The age and sex of the patient. (2) Symptoms that suggested the presence of MSPA. (3) Diagnostic procedures that confirmed the presence of a tumor within MS. (4) Initial histological diagnosis at the biopsy stage. (5) Treatment method. (6) Histological diagnosis of the whole tumor. (7) Number of MSPA recurrences. (8) The type of neoplasm that MSPA transformed into.

presents a list of parameters and information extracted from each of the studies. The data most consistent with the assumed objectives were selected. In case of recurrence, the time from the end of primary treatment to the first diagnosis of recurrence was taken into account. In case of no recurrence, the total observation time was noted. In order to comparatively assess the size of the tumor at the stage of its detection, the largest dimension was selected. The age of the patient at which the primary treatment was started was obtained. The dates of the primary occurrence of PA were extracted in each case, even if the article focused on the recurrence or malignant transformation.

2.6. Study Risk of Bias Assessment

The risk of bias for the included studies was assessed manually using the JBI Critical Appraisal Tool for Case Reports [13]. For the reports on the recurrence or transformation of PA into a malignant neoplasm, information on the primary occurrence of MSPA was assessed in terms of risk of bias.

2.7. Synthesis Methods

The synthesis was performed based on all of the studies qualified for the review. In case of the absence of numerical data, no attempt was made to estimate them for the purpose of synthesis. Time periods were standardized to years and months. For smaller units extracted from the reports, (days, weeks) conversion to months was applied, assuming that 1 month consists of 4 weeks, i.e., 28 days. Lengths in any other units were converted to millimeters. Outcomes and other variables for individual studies were tabulated. Additionally, the synthesized numerical data were graphically presented in the form of charts and diagrams. The cases were analyzed in subgroups that were heterogeneous with regard to the presence or absence of PA recurrence.

3. Results

3.1. Study Selection

The number of records found and the ranges of all engines valid on the day of the search (24 April 2022) are presented in

Table 4. Engine ranges and number of results on final searches.

	Engine Range (Records)	Number of Results
ACM	3,313,563	0
BASE	302,348,097	32
Google Scholar	about 160,000,000	3
PubMed	over 33,000,000	51

. The screening process is presented in the PRISMA diagram (Figure 1). The records removed at the eligibility stage are summarized in

Table 5. Records excluded at the eligibility stage.

First Author and Publication Year	Exclusion Reason
Francesco 2021 [14]	Wrong settings: no PA case report
Giovacchini 2021 [15]	Wrong problem: another neoplasm
Chaturvedi 2018 [16]	Wrong problem: another location
Rastogi 2016 [17]	Wrong problem: another location
Amanpour 2015 [18]	Wrong problem: another neoplasm
Thompson 2014 [19]	Wrong settings: no PA case report
Habesoglu 2010 [20]	Wrong problem: no MSPA cases

.

3.2. Study Characteristics

The list of qualified studies together with the variables that are not the outcomes of the review are presented in

Table 6. Included studies. NP—Not performed; F—female; M—male; PA—Pleomorphic adenoma; CT—Computed tomography; ES—Endoscopic surgery; OS—Open surgery.

First Author and Publication Year	The Age and Sex of the Patient	Symptoms That Suggested the Presence of MSPA	Diagnostics That Confirmed the Presence of a Tumor within MS	Initial Histological Diagnosis at the Biopsy Stage	Treatment Method	Histological Diagnosis of the Whole Tumor
Lygeros 2021 [21]	66 M	Nasal obstruction; Proptosis	CT	NP	ES	PA
Li 2019 [22]	49 F	Nasal obstruction	Unknown	Unknown	OS	PA
Sınacı 2017 [23]	13 F	Palate swelling	CT	PA	OS	PA
Ray 2015 [24]	33 M	Cheek swelling; Pain	CT	Ossifying PA	OS	Ossifying PA
Gupta 2013 [25]	39 F	Cheek swelling	CT	NP	OS	PA
Sreedharan 2012 [26]	32 F	Cheek swelling; Nasal obstruction	Unknown	NP	OS	Unknown (recurred as CA ex PA)
Ishikawa 2008 [27]	55 M	Nasal obstruction; Pruritus; Sneezing	CT	NP	ES	PA

.

3.3. Risk of Bias in Studies

The results of the bias risk assessment are presented in

Table 7. Risk of bias assessment. NA—Not applicable.

First Author and Publication Year	Were the Patient’s Demographic Characteristics Clearly Described?	Was the Patient’s History Clearly Described and Presented as a Timeline?	Was the Current Clinical Condition of the Patient on Presentation Clearly Described?	Were Diagnostic Tests or Assessment Methods and the Results Clearly Described?	Was the Intervention(S) or Treatment Procedure(S) Clearly Described?	Was the Post-Intervention Clinical Condition Clearly Described?	Were Adverse Events (Harms) or Unanticipated Events Identified and Described?	Does the Case Report Provide Takeaway Lessons?
Lygeros 2021 [21]	Yes	Yes	Yes	Yes	Yes	No	NA	Yes
Li 2019 [22]	Yes	No	Yes	No	Unclear	No	NA	No
Sınacı 2017 [23]	Yes	Yes	Yes	Yes	Yes	Yes	NA	Yes
Ray 2015 [24]	Yes	Yes	Yes	Yes	Unclear	Unclear	NA	Yes
Gupta 2013 [25]	Yes	Yes	Yes	Yes	Unclear	Yes	NA	Yes
Sreedharan 2012 [26]	Yes	Unclear	Unclear	No	Unclear	Yes	NA	Yes
Ishikawa 2008 [27]	Yes	Yes	Yes	Yes	Unclear	Yes	NA	Yes

.

3.4. Results of Individual Studies and Synthesis

The results of individual studies are presented in

Table 8. Outcomes and other variables closely related to outcomes. NA—Not applicable; CA—Carcinoma; PA—pleomorphic adenoma.

First Author and Publication Year	The Size of the Tumor in the Greatest Dimension	Total Observation Time in the Absence of a Recurrence	Time from Treatment to First Recurrence	Number of MSPA Recurrences	The Type of Neoplasm MSPA Transformed Into
Lygeros 2021 [21]	Unknown	1 year	NA	None	NA
Li 2019 [22]	45 mm	3 years	NA	None	NA
Sınacı 2017 [23]	40 mm	NA	6 years	1	NA
Ray 2015 [24]	47 mm	8 months	NA	None	NA
Gupta 2013 [25]	60 mm	NA	2 years	2	Myoepithelial carcinoma
Sreedharan 2012 [26]	Unknown	NA	14 years	1	CA ex PA
Ishikawa 2008 [27]	Unknown	4 years	NA	None	NA

. The follow-up time after the surgical removal of MSPA followed by the first recurrence, was 2, 6, and 14 years according to three of the studies eligible for this synthesis [23,25,26]. Authors who did not report a recurrence after treatment decided to follow up at 8 months, 1 year, 3 years and 4 years before the publication of their cases, that is, on average, 2 years and 2 months [21,22,24,27]. The size of the detected MSPA in the largest dimension ranged in various cases from 40 to 60 mm, with an average of 48 mm [22,23,24,25]. The extremely small amount of data prevented further processing.

4. Discussion

4.1. General Interpretation of the Results

4.1.1. Origin

An extremely small number of reports were identified, as only seven described cases of PA originally localized in MS [21,22,23,24,25,26,27]. It is known that PA may originate from the oral mucosa glands and approach the maxillary sinus from the palate side [16,23]. Likewise, nasal PA also develops adjacently to the maxillary sinus [17,28,29]. Therefore, it cannot be guaranteed for any of the included case reviews that PA actually originated from the mucosa of the maxillary sinus [21,22,23,24,25,26,27]. In the report by Ishikawa et al., the tumor entered the maxillary sinus presumably from the nasal cavity [27]. In the case report published by Sreedharan et al., the obliteration of the nasal passage and the presence of a tumor mass within the oral cavity made it impossible to determine where it originated from [26]. In the description by Gupta et al. of an unspecified swelling of the palate that was surgically treated 2 years before the first examination in the authors’ clinic, an oral origin was suggested [25]. Analyzing the study of Sınacı et al. it may be questionable whether the original orally derived tumor crossed the bottom of the maxillary sinus or just deformed it [23]. The later presence of PA in the maxillary sinus could be due to the destruction of the bone barrier during the first surgery [23]. Incomplete data obtained from listing a 49-year-old woman as one of the 17 cases of the series by Li et al. prevented an analysis of the tumor origin [22]. The cases presented by Lygeros et al. and Ray et al. seem to be the only ones in which tumor emergence from the mouth or nose was not indicated [21,24]. PA in the report by Ray et al. differs additionally in the histological picture due to the presence of calcifications [24]. In addition to the MSPA described in the following work, a unique case of PA in the frontal sinus was reported [30]. Another tumor of salivary tissue origin found in MS is malignant myoepithelioma [31].

4.1.2. Diagnostics and Treatment

The dominant symptoms in the form of swelling of the cheek, palate or lateral wall of the nose can be easily explained by the expanding growth of PA [1,4]. Computed tomography was the only test that confirmed the presence of PA in the sinus and provided information on the extent and nature of the tumor [21,23,24,25,27]. Therefore, it should be considered a standard diagnostic tool when MSPA is suspected. The initial largest dimension of MSPA is close to or even exceeds the natural dimensions of the maxillary sinus, according to studies by various authors [22,23,24,25,32]. This can be explained by the expanding tumor growth, which models sinus walls, and by the fact that MSPA may originate from a different anatomical space and occupy MS secondarily. In most cases, no initial incision biopsy was performed. This may be explained by the difficulty of access as well as the solid and well-defined tumor RTG image, which does not indicate the invasion of the surrounding tissues [1,21,25,26,27]. Apart from an open surgery, which consisted primarily of an anterior approach with an incision through the upper lip and surrounding the nose wing, an endoscopic intervention was performed in two cases [21,22,23,24,25,26,27]. The short observation period of 1 and 4 years was not sufficient for the assessment of the effectiveness of the latter method [21,27]. Establishing the scope and method of resection requires a balance between limiting recurrence due to radicality and ensuring quality of life by minimizing the invasiveness of the procedure [33,34,35]. However, none of the case reports provided details of the surgical procedure [21,22,23,24,25,26,27].

4.1.3. Recurrence and Malignancy

For PA originating from other locations, the time to first recurrence may be lengthy, with a median of approximately 7–9 years [33,34,35]. This may explain the lack of recurrence in MSPA case reports in which postoperative follow-up was up to 4 years [21,22,24,27]. The three known cases of MSPA recurrence cannot be compared due to the different period before the first recurrence (2, 6 and 14 years) and different histological diagnoses [23,25,26]. Malignant transformation was noted in two of them, which may indicate a higher tendency of MSPA to develop into cancer than PA of other locations [4,5,26,33,34,35].

4.2. Limitations

Only one of the seven included reports did not have any flaws during the risk of bias assessment. Limitations to the review process are the omission of source articles published prior to the last 20 years and the exclusion of cases reported in languages other than English.

5. Conclusions

To the best knowledge of the authors, there are only seven cases of maxillary sinus pleomorphic adenoma described in the last 20 years. Only in two of reported cases was there no reason to suspect that the tumor had penetrated the sinus from the oral or nasal cavity. The described location is so unfavorable that tumors occupying the entire volume of the sinus can be diagnosed. The longest recorded time without recurrence is 4 years. Malignant transformation occurred in two of three recurrent cases.