The literature search identified a total of 7,410 records (6,964 retrieved from databases and 446 from other sources). After removal of duplicates, 3,625 references were screened in Phase 1 (title and abstract screening). Records not meeting the inclusion criteria were subsequently excluded.

A total of 133 full-text reports were sought for retrieval, of which 14 could not be obtained. In Phase 2 (full-text assessment), 119 articles were evaluated, and 40 were excluded for the reasons detailed in the PRISMA flow diagram (Fig. 1).

Fig. 1The alternative text for this image may have been generated using AI.

PRISMA 2020 flow diagram showing study identification, screening, eligibility assessment, and final inclusion, with itemized reasons for exclusion at the full-text stage

Ultimately, 79 studies met the inclusion criteria and were incorporated into the qualitative synthesis of this systematic review. Unless otherwise specified, all frequencies presented in the Results refer to the number of studies reporting a given feature and should not be interpreted as patient-level prevalence estimates.

General Characteristics of the Included Studies

The predominant language of publication was English (52; 65.82%), followed by Japanese (22; 27.85%). The remaining languages were represented by one article each (German, Italian, French, Chinese, and Russian).

The publication date range spanned from 1879 to 2025, with a greater concentration of publications in more recent years: before 2000 (n = 21; 26.58%), 2000–2020 (n = 28; 35.44%), and after 2020 (n = 30; 37.97%).

The included evidence was predominantly descriptive in nature, consisting of 64 full-text articles [1,2,3,4,5,6,7, 9,10,11,12, 14,15,16, 24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72] and 15 conference abstracts/posters [13, 18, 73,74,75,76,77,78,79,80,81,82,83,84,85] (Table 2). This distinction should be considered when interpreting aggregated findings.

Table 2 Distribution by study design and publication format among the studies included in the review

Across all included studies, 387 patients were reported. Despite this cumulative sample size, the evidence base remained overwhelmingly descriptive, with most publications being single-patient reports (n = 71; 89.87%). The largest sample sizes were reported by Zhao et al. (96 patients and 217 glands) [3] and Carey et al.. (37 patients, without specification of the number of affected glands) [2].

Methodological Quality

By study type, case reports predominantly demonstrated high methodological quality (41/51; 80.39%). Among case series, moderate quality was most frequent (13/23; 56.52%), followed by low quality (9/23; 39.13%). Cohort studies were classified as low quality in 100% of cases, while cross-sectional studies showed an equal distribution between low and high quality (Fig. 2). This low-quality rating in cohort studies was mainly driven by recurrent limitations in domains related to group comparability, identification and management of confounding, and completeness or clarity of follow-up.

Fig. 2The alternative text for this image may have been generated using AI.

Distribution of methodological quality according to JBI by study design. Stacked bar chart showing the proportion of studies classified as low, moderate, and high quality within each study design

All detailed quality assessment data are provided in Supplementary Material 2.

Nomenclature Used

If a single study could employ more than one term to refer to the disease, the most frequently used terms were fibrinous sialodochitis (38/79; 48.10%) and Kussmaul disease (34/79; 43.04%). Eosinophilic sialodochitis was used in 27/79 studies (34.18%), whereas eosinophilic sialadenitis/parotitis appeared in 4/79 (5.06%).

The grouped terminology allergic sialadenitis/parotitis was identified in 14/79 studies (17.72%), and allergy-related sialodochitis (ARS), a term coined by Zhao et al. [3], was used in 2/79 (2.53%). Over time, an evolutionary trend can be observed, with eosinophilic sialodochitis emerging as the most used term in more recent studies.

All detailed information is provided in Supplementary Material 3.

Demographics, Clinical Presentation, Atopy, and Gland Involvement

The weighted mean age was 43.17 years (n = 206; 53.23%), and the weighted median age was 48 years (n = 272; 70.28%). Pediatric involvement (< 18 years) was uncommon, without a specific age predominance, with some exclusively pediatric series [39, 45, 50, 59, 77] and others including mixed populations [56, 58, 69, 71].

The overall age range was broad, from 0.75 to 80.0 years. A clear female predominance was observed (243 women vs. 110 men; female-to-male ratio 2.21:1).

The most common symptoms, duration of symptoms, and associated allergic conditions are described in Table 3.

Table 3 Frequency of symptoms, potential triggers, duration, and atopic comorbidities reported in the included studies

Across the included studies, a cumulative total of 252 parotid glands and 154 submandibular glands was reported among 387 patients. In 12 studies, the specific number or type of affected glands was not reported [6, 7, 13, 55, 69,70,71,72,73, 77, 78, 81].

Among studies reporting these variables, bilateral involvement (of either the parotid or submandibular glands) and mixed involvement (both parotid and submandibular glands) were summarized descriptively from the reported patient counts (Fig. 3). Mixed gland involvement was reported in 21 studies, with a weighted mean of 30.4%. In rare instances, involvement of the sublingual gland was also described [1, 57]. Cumulative parotid andsubmandibular involvement were summarized at the gland level, whereas laterality and mixed-gland involvement were derived from reported patient counts aggregated across studies.

Fig. 3The alternative text for this image may have been generated using AI.

Cumulative affected glands and laterality. Counts of parotid and submandibular involvement are shown at the gland level, whereas laterality and mixed-gland involvement are summarized from patient counts aggregated across studies. These descriptive data should not be interpreted as patient-level or population-level prevalence estimates

All detailed information is provided in Supplementary Material 4.

Diagnostic Work-Up

The diagnostic evaluations most frequently reported included laboratory testing, cytology or histology of salivary mucus, histology of the salivary gland, and imaging studies (Tables 4 and 5).

Table 4 Laboratory, mucus cytology/histology, and tissue biopsy findingsTable 5 Imaging, sialendoscopy, and structural ductal findings

Among studies specifying laboratory values, peripheral eosinophilia was elevated in 37/51 (72.54%), and total IgE was elevated in 36/43 (83.72%).

Cytological or histological examination of salivary mucus revealed the presence of eosinophils in 91.38% of cases (53/58), while only four studies reported the presence of Charcot–Leyden crystals [5, 24, 50, 86]. In contrast, tissue biopsy demonstrated eosinophilic infiltration in 50% of studies (12/24). Notably, five studies specifically described biopsy sampling from the terminal salivary duct [4, 29, 53, 71, 72].

Sialendoscopy was reported as a diagnostic or confirmatory procedure in 5/79 studies, all of which were published from 2022 onward (6.32%) [2, 14,