Breast MRI features of idiopathic granulomatous mastitis and invasive breast carcinoma: A single-center comparative study.

A 7-y-old American Quarter Horse mare was presented to a veterinary hospital for ongoing reproductive monitoring and management of endometritis. During physical evaluation, several nodular lesions of unknown origin and duration were discovered in the mammae. Histologically, biopsies from the lesions were confirmed as severe granulomatous mastitis with intralesional rhabditid nematodes. A putative diagnosis of Halicephalobus gingivalis or Pelodera strongyloides infection was assigned, and ivermectin (1.2 mg/kg PO) treatment was given every 2 wk. When lesions persisted after the initial dose of ivermectin, PCR testing of 18S and 28S gene targets confirmed Cephalobus cubaensis infection. Given the lack of response after 4 doses of ivermectin and the zoonotic concern, euthanasia was elected. Analysis of sediments of the skin tissues recovered eggs, larvae, and female nematodes with a blunt tail terminating in a mucro and the rhabditid esophagus of Cephalobus spp. Spread to the left inguinal lymph node was confirmed. Cephalobus sp. is a saprophytic nematode that has very rarely been reported as a cause of verminous mastitis in horses and in a single fatal case of meningoencephalitis in a human. Differentiation between C. cubaensis, H. gingivalis, and P. strongyloides by histopathology alone is unreliable because of morphologic similarities; molecular identification should be pursued.

Granulomatous lobular mastitis (GLM) is a chronic, refractory inflammatory condition of the breast with an unclear etiology. Emerging evidence suggests a potential role of pyroptosis in the pathogenesis of GLM. This study investigated the therapeutic effects of Tuolitounong Decoction (TLTND) on pyroptosis in both in vitro and in vivo models of GLM. An in vitro GLM model was established using MCF10A mammary epithelial cells exposed to tissue homogenates derived from GLM-affected human breast tissue. Cells were treated with TLTND at varying concentrations, with or without the pyroptosis inhibitor necrosulfonamide. Pyroptosis-associated proteins, including caspase-1, gasdermin D (GSDMD), lipopolysaccharide-binding protein (LBP), and NLR family pyrin domain containing 3 (NLRP3), were assessed using immunofluorescence and western blotting. Cell viability was evaluated using the Cell Counting Kit-8 (CCK-8) assay. For in vivo analysis, a rat GLM model was induced by injecting a combination of human GLM tissue homogenate supernatant and Freund's complete adjuvant into the third and fourth pairs of mammary glands in female Sprague Dawley rats. TLTND was administered via daily oral gavage for 21 days. Post-treatment evaluations included histopathological assessment, expression of estrogen receptor (ER) and progesterone receptor (PR), and pyroptosis-related biomarkers in mammary gland tissue. Exposure to GLM homogenate successfully induced pyroptosis-related pathological features in MCF10A cells, whereas serotonin (5-HT) inhibition showed no significant effect. TLTND treatment demonstrated dose- and time-dependent effects, with enhanced therapeutic efficacy at 10 mg/mL after 48 h (P = 0.009). Of note, TLTND significantly reduced the protein expression levels of key pyroptosis markers (cleaved caspase-1, n-GSDMD, cleaved IL-18 and cleaved IL-1β) by 52.6%, 91.0%, 78.1% and 87.1%, respectively, compared to the model group in rat mammary tissue. Co-treatment with TLTND and necrosulfonamide significantly downregulated the expression of caspase-1, GSDMD, LBP, and NLRP3, while increasing cell viability. In vivo, both prednisone acetate and TLTND ameliorated histopathological features of GLM. Notably, TLTND promoted localized abscess maturation and increased ER and PR expression, while reducing levels of interleukin-1β and GSDMD-N in mammary tissue. TLTND demonstrated therapeutic potential in reducing pyroptosis in mammary epithelial cells and modulating inflammation in granulomatous lobular mastitis, partly by downregulating the caspase-1/GSDMD-mediated pyroptosis signaling pathway. These findings support further investigation of TLTND as a complementary treatment approach for GLM and provide insight into its underlying mechanisms of action.

Background:Idiopathic granulomatous mastitis (IGM) is a benign, chronic inflammatory breast disease frequently associated with autoimmune disorders. This prospective cohort study explored the prevalence of Hashimoto thyroiditis (HT) in patients with IGM and the relationship between this co-occurrence and systemic inflammatory markers.Methods:Between March 2019 and September 2022, 179 women were enrolled: IGM without HT (n = 52), IGM + HT (n = 19), and age-matched healthy controls (n = 89). Hematological indices, including white blood cell count, C-reactive protein, neutrophil-to-lymphocyte ratio (NLR), and lymphocyte-to-monocyte ratio (LMR), were measured at diagnosis before any treatment.Results:The prevalence of HT among patients with IGM was 26.8 %. Compared with isolated IGM, the IGM + HT group had a lower smoking rate (16.3% vs 40.4%, P = .028) and displayed a significantly lower median NLR (1.91 vs 2.76, P < .001) but higher median LMR (6.47 vs 4.75, P = .001).Conclusion:Approximately 1 in 4 women with IGM also had HT. A low NLR-high LMR pattern may reflect a distinct inflammatory milieu in this overlap cohort and could help tailor follow-up strategies for these patients. Routine thyroid screening should be considered in patients with IGM.

Idiopathic Granulomatous Mastitis Presenting as Refractory Inflammatory Breast Disease in a Young Woman.

Granulomatous mastitis is a rare inflammatory disease of the breast that frequently mimics malignancy both clinically and radiologically. A 30-year-old female presented with a painful lump in the left breast for three months associated with skin changes and sinus formation. Imaging suggested mastitis with possible abscess formation, and subsequent biopsy confirmed granulomatous mastitis with histopathological features favoring tuberculous etiology. Despite negative microbiological tests including AFB smear, GeneXpert, and culture, clinicopathological correlation supported the presumptive diagnosis of tuberculous mastitis. The patient was treated with anti-tubercular therapy and showed clinical improvement. This case highlights the diagnostic difficulty of granulomatous mastitis and emphasizes the importance of histopathological evaluation in distinguishing it from malignancy and other inflammatory breast diseases.

To develop an interpretable machine learning model utilizing ultrasound radiomics for distinguishing between granulomatous lobular mastitis and breast cancer. This retrospective study encompassed 237 patients who underwent preoperative breast ultrasound examinations and received pathological diagnoses of either granulomatous lobular mastitis or breast cancer at Quzhou People's Hospital between April 2013 and April 2023. Radiomic features were extracted from the ultrasound images, and feature selection was conducted using intra-class correlation coefficients, Pearson correlation coefficients, and the least absolute shrinkage and selection operator regression. Machine learning models based on radiomics were constructed using Extremely Randomized Trees, Light Gradient Boosting Machine, and Random Forest. Additionally, a combined model was developed by integrating independent clinical predictors with the radiomics signature.The model's performance was assessed using the area under the receiver operating characteristic curve, accuracy, sensitivity, and specificity. To evaluate clinical utility, decision curve analysis was employed, while Shapley additive explanation was utilized to interpret model explainability. A total of 1,161 radiomic features were extracted from each ultrasound image. Following Pearson correlation filtering, 135 features were retained, and 15 features were selected using the least absolute shrinkage and selection operator regression for model construction. The combined model, which integrated clinical factors with the radiomics signature, exhibited superior performance, achieving an AUC of 0.935 (95% CI: 0.902-0.969) in the training cohort and 0.833 (95% CI: 0.710-0.950) in the validation cohort. DCA indicated favorable clinical applicability. The Shapley additive explanation analysis shows that the imaging biomarker features lbp_3D_k_glszm_SmallAreaLowGrayLevelEmphasis, gradient_glcm_Imc2, gradient_glszm_ZoneEntropy, original_shape_Elongation, squareroot_glrlm_RunEntropy, and wavelet_HLH_glszm_LowGrayLevelZoneEmphasis have a strong correlation with the prediction of granulomatous lobular mastitis. The combined model, which incorporates ultrasound radiomics and clinical factors, exhibited significant efficacy in preoperatively differentiating granulomatous lobular mastitis from breast cancer. This non-invasive and interpretable methodology shows potential for enhancing diagnostic precision and guiding clinical decision-making.

Granulomatous mastitis (GM) is an inflammatory breast disease. Pentraxin 3 (PTX3) is frequently dysregulated in inflammatory diseases. C. parakroppenstedtii. infection was applied to establish a rat GM model. Histological analysis showed that the progression of GM was accompanied with the damage of DECs. The serum and mRNA levels and protein expression of PTX3 was increased in GM rat models. PTX3 overexpression drove macrophage to M2 macrophage polarization, inhibiting inflammation response. PTX3 overexpression promoted clearance of apoptotic mammary duct epithelial cells (DECs) by M2 macrophages. However, PTX3 depletion in macrophage inhibited M2 macrophage polarization and impaired the efferocytotic ability, mediating inflammatory response and accumulation of apoptotic DECs. In summary, PTX3 exerts protective function in GM. PTX3-mediated M2 macrophage polarization promotes the clearance of apoptotic DECs. Therefore, targeting the communication between macrophages and DECs may be a promising strategy for GM.

Granulomatous lobular mastitis (GLM) is a chronic inflammatory breast disease of unknown aetiology, characterised by granulomatous inflammation centred on the mammary lobules. Clinically, it presents with breast redness, swelling, and ulceration, with a protracted course prone to recurrence. Its development is closely associated with multiple factors including milk stasis, hyperprolactinemia, ductal injury, and microbial infection. The pathogenesis remains incompletely elucidated, with various hypotheses proposed such as autoimmune responses, delayed-type hypersensitivity, and autoinflammatory reactions. Key mechanisms involve abnormal activation of cellular and humoral immunity, release of multiple inflammatory cytokines, and activation of signalling pathways including NF-κB and PI3K/AKT/mTOR. This article systematically reviews relevant literature spanning nearly five decades, summarising the evolution of GLM definitions, predisposing factors, and pathogenesis. It aims to establish a theoretical foundation for future molecular subtyping, immunological target exploration, multicentre studies, and the advancement of precision prevention and treatment for GLM.

Abstract Background/Aims Idiopathic granulomatous mastitis (IGM) is a rare, benign but clinically aggressive inflammatory breast condition that often mimics malignancy on imaging and examination. Its exact cause remains uncertain, but growing evidence supports an autoimmune or autoinflammatory basis rather than a purely infectious one. In the UK, IGM is often under-recognised, leading to diagnostic delays and unnecessary surgery. The overlap between IGM and immune-mediated inflammatory disease places rheumatology in a key position to support accurate diagnosis and medical management. This study aimed to review all IGM cases seen in our regional breast referral centre, serving multiple NHS trusts, and to evaluate outcomes following the introduction of a rheumatology-led multidisciplinary approach. Methods A retrospective review was conducted of ten consecutive patients with biopsy-proven IGM diagnosed between April 2016 and August 2024. All underwent triple assessment in the breast clinic (clinical examination, imaging, and core biopsy). Microbiological cultures were extended using lipid-enriched media, and where indicated, 16S PCR was performed for Corynebacterium kroppenstedtii. Rheumatology review included detailed autoimmune and vasculitis screening (ANA, ENA, dsDNA, ANCA, complement, ACE, prolactin, and QuantiFERON-TB). Immunosuppressive treatment was based on disease severity, with prednisolone (20-40 mg daily) as induction therapy and methotrexate (15-25 mg weekly) as a steroid-sparing agent where required. Clinical outcomes included remission, relapse, treatment tolerance, and diagnostic delay before rheumatology involvement. Results All ten patients were female (mean age 37 years, range 26-48) with a mean BMI of 29.8 kg/m². Seven were Caucasian and three were from Syria, Afghanistan, and Ethiopia. All presented with unilateral painful breast masses; four had imaging suspicious for malignancy (BI-RADS 3-5). Corynebacterium kroppenstedtii was isolated in four patients. Two patients had positive autoimmune serology (ANA or anti-smooth muscle antibody). Five patients achieved remission with steroids alone, and six required methotrexate, including one switched to intramuscular methylprednisolone acetate due to steroid intolerance. Complete remission was achieved in nine of ten patients, sustained for up to nine years; one patient remains under treatment. No serious adverse effects were reported. The median delay before rheumatology referral was six months (range 2 weeks-18months), underscoring the importance of early multidisciplinary involvement. Conclusion Our experience supports IGM as an immune-mediated granulomatous disorder that responds well to immunosuppressive therapy within a multidisciplinary framework. Early rheumatology input allows prompt exclusion of systemic autoimmune disease, targeted immunosuppressive therapy, and avoidance of unnecessary surgical procedures. Collaboration between rheumatology, breast surgery, radiology, pathology, and microbiology is vital for diagnostic accuracy and durable remission. As a regional referral centre, our model demonstrates excellent long-term outcomes and aligns with the 2022 International GLM Consensus and EUBREAST GRAMAREG recommendations. Incorporating rheumatology leadership into IGM management across UK breast networks could standardise care, improve patient experience, and contribute prospectively to the GRAMAREG registry. Disclosure F.A. Khan: Honoraria; UCB. Other; Congress Sponsorship EULAR (UCB) and BSR 2025 (Novartis). T. Fasih: None. M. Khan: None.

Cystic neutrophilic granulomatous mastitis is a rare benign breast inflammatory disease often associated with Corynebacterium kroppenstedtii. Its clinical and radiologic features frequently mimic breast carcinoma, leading to diagnostic and therapeutic dilemmas. We present a North African 41-year-old woman with pituitary insufficiency who developed a painless, enlarging breast mass. Multimodality imaging (ultrasound, mammography, computed tomography, and magnetic resonance imaging) all suggested malignancy, with magnetic resonance imaging classified as Breast Imaging-Reporting and Data System 5. Core needle biopsy suggested cystic neutrophilic granulomatous mastitis. Due to discordance between suspicious imaging and benign histology, the patient underwent surgical excision. Final pathology confirmed cystic neutrophilic granulomatous mastitis. At 12-month follow-up, the patient remained recurrence free. This case highlights the diagnostic challenge of cystic neutrophilic granulomatous mastitis, particularly when imaging strongly suggests carcinoma. It emphasizes the indispensable role of histopathology and the need for multidisciplinary management to avoid overtreatment.

Abstract Background Idiopathic granulomatous mastitis (IGM) is a rare, benign breast disease characterised by granulomatous inflammation. Current literature mainly comprises of small-sized retrospective studies with no well-designed clinical trials and limited representation of UK population. With no consensus to guide clinical practice there is a significant risk of delayed diagnosis and variation in management strategies that may adversely affect patient outcomes Methods ENIGMA is a multi-stage programme of work aiming to develop guidelines for the diagnosis and management of IGM in the UK. A scoping review will assess the range and quality of existing evidence, focusing on its relevance to UK practice. Findings from this review will inform a national practice questionnaire (NPQ) to capture current approaches to IGM care across the UK. These data, combined with a modified Delphi process involving experts and stakeholders, will support guideline development and help define priority research questions to advance best practice. Progress The scoping review identified 225 papers, involving 13 062 patients, mostly from Turkey, the USA, and China. Only three studies, covering 25 patients, originated from the UK. Commonly reported ethnic groups were Hispanic and Turkish. Risk factors include smoking and hormonal influences such as oestrogen and prolactin. Reported treatment strategies range from conservative approaches to medical therapies (antibiotics, steroids, immunosuppressants) and surgical interventions (drainage or excision). Preliminary results of the NPQ highlight substantial heterogeneity in UK diagnostic and management practices. Once the NPQ analysis is complete, the working group will initiate the Delphi process to establish consensus and inform guideline development.

Abs tractAim: Idiopathic granulomatous mastitis (IGM) is a benign, chronic inflammatory disease of the breast, and its imaging findings may overlap with those of malignant non-mass enhancement (NME).This study aimed to investigate the performance of deep learning and machine learning models in differentiating IGM from malignant NME based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Methods:In this retrospective study conducted between January 2019 and March 2023, DCE-MRI findings of 30 patients with histopathologically confirmed IGM and of 33 patients with breast cancer presenting as NME were analyzed.The second dynamic phase of DCE-MRI (Dataset 1, 475 images) and the corresponding subtracted images (Dataset 2, 402 images) were used in this study.Datasets were sequentially split into 80% training and 20% testing sets to ensure a patient-level split.Image features were extracted using SqueezeNet and classified with a narrow neural network. Results:The mean age was significantly lower in the IGM group than in the NME group (41.311.3 vs. 52.211.4years, p<0.001).For Dataset 1, the area under the curve was 0.997 in training and 0.870 in testing; for Dataset 2, the area under the curve was 0.998 in training and 0.807 in testing.Training accuracy was 0.984 (Dataset 1) and 0.978 (Dataset 2), whereas test accuracy was 0.811 (Dataset 1) and 0.704 (Dataset 2). Conclusion:The findings of this study suggest that deep learning shows significant promise for non-invasive differentiation of IGM from malignant NME on DCE-MRI, particularly in cases that are clinically indistinguishable.

This case series characterizes workup, specialties involved, and treatment for idiopathic granulomatous mastitis.

Idiopathic granulomatous mastitis (IGM) is a rare inflammatory breast condition lacking standardized treatment and with unpredictable outcomes. To address these issues, using clinical and ultrasound findings from an initial subset, we created the Pittsburgh Classification to stratify severity and developed a corresponding treatment algorithm for IGM, then evaluated its effectiveness in a larger cohort of IGM patients. This retrospective multicenter study reviewed clinical and sonographic findings and outcomes of women with biopsy-proven IGM treated at multiple breast centers between 2020 and 2025. The Pittsburgh clinical classification ranges from Type 1 (minimal skin irritation) to Type 5 (widespread involvement); ultrasound classification spans Type A (localized mass ≤2 cm) to Type D (diffuse disease). Treatments were assessed utilizing the Pittsburgh algorithm, with responses classified as full response (CR), near-complete response (nCR), or no response (NR). Chi-square tests assessed associations (p<0.05). Of 522 patients included (mean age 37.0±8.8 years), 86.4% (n = 451) received algorithm-concordant treatment, achieving CR in 68.7% (n = 310), nCR in 35.3% (n = 159) and NR in 11.8% (n = 53). Among these, 65.4% (295/451) of patients with CR were concordant with the Pittsburgh treatment algorithm, whereas 13.6% (n = 71) patients received discordant treatments, with a significantly lower CR rate of 21.1% (15/71) (p<0.001). Multifocal disease was significantly more prevalent in NR (83.0%, 44/53) and nCR (70.4%; 112/159) patients compared to CR (20.6%; 64/310) (p<0.001), although lesion-based response rates were similar (CR 56.8%, nCR 57.0%, NR 56.6%). Regarding concordance with treatment algorithm, clinical Type 4 IGM was more prevalent in NR (67.9%; 36/53) and nCR (72.9%, 116/159), whereas in clinical Type 1 IGM, NR, nCR, and CR were 1.8% (1/53), 4.4% (7/159), and 30.6% (95/310), respectively (p<0.001). Surgery at presentation was preferred in 16.9% (n = 88) of patients, with 6% (n = 30) requiring subsequent surgical treatments to treat residual disease. Concordance with the proposed IGM treatment algorithm based on clinical and ultrasound findings resulted in significantly higher CR rates. Multiple foci and stratified clinical types correlated with outcomes. Prospective global research is needed to validate these findings.

Idiopathic granulomatous mastitis (IGM) is a rare benign inflammatory breast disease with a high risk of relapse. The study objective was to evaluate relapse predictors and treatment outcomes in a large cohort of IGM patients. We retrospectively analyzed female patients diagnosed with IGM (2018-2024) at the Central Military Hospital, Baku. Diagnosis was confirmed by core needle biopsy. Patients were managed with systemic therapy (corticosteroids and/or immunosuppressants) when clinically indicated; local measures (e.g., aspiration/drainage, intralesional steroid) were used selectively in localized disease. Relapse was defined as reappearance of clinical or radiological findings after remission. Univariable and multivariable logistic regression models were applied to identify independent predictors. The cohort consisted of 92 patients. Relapse occurred in 22/85 methotrexate-treated patients (25.9%), with most relapses occurring between the third and fifth months. No relapse events were observed in the azathioprine subgroup (n = 7). However, this finding should be considered observational only due to the small numbers and zero-event data. In multivariable analysis, erythrocyte sedimentation rate (ESR) >20 mm/h (and angiotensin converting enzyme >52 U/L, where applicable) were associated with relapse, whereas apparent associations with tumor necrosis factor alpha inhibitors and cyclosporine likely reflect confounding by indication because these agents were used as rescue therapy in refractory/relapsing disease. Elevated ESR was also associated with prolonged treatment duration (p =0.006). A structured and individualized treatment approach may contribute to favorable clinical outcomes in patients with IGM. Observed relapse patterns support the importance of risk-adapted management rather than a uniform therapeutic strategy. Given the retrospective design and limited subgroup sizes, these findings should be interpreted cautiously and considered hypothesis-generating. Prospective, multicenter studies are required to validate relapse-associated factors and optimize treatment strategies.

Idiopathic granulomatous mastitis (IGM) is a rare inflammatory condition of the breast with unknown etiology, primarily affecting women of childbearing age. IGM commonly presents with culture-negative abscesses, often leading to significant scarring, hyperpigmentation, and deformation of the affected breast. Breast cancer and tuberculosis are important differential diagnoses. IGM is confirmed by the presence of non-caseating granulomas and microabscesses on histopathological examination. The literature mainly consists of case studies, and there is no evidence-based treatment. A regional multidisciplinary collaboration group was formed in Stockholm. The purpose of this collaboration was to develop a guideline to assist physicians in the investigation and management of patients with suspected IGM. A literature search was conducted in NCBI Pubmed and thirty relevant studies were included. Flowcharts were developed for the investigation and treatment of IGM. Investigation includes wound bacterial cultures, DNA sequencing (for fungi and bacteria), core needle biopsy, blood tests and skin punch biopsy. Ultrasound-guided drainage is recommended for abscesses, and surgery should be avoided. An initial treatment option when diagnosis has been established consists of daily application of topical group III corticosteroids for 2-3 weeks. Second-line therapy involves a combination of oral corticosteroids and methotrexate for 1-2 months, followed by a tapering of corticosteroids, with continuation of methotrexate monotherapy for an additional six months. Due to the complexity and suffering associated with IGM, patients should be managed by a team of specialists with a specific interest in the condition.The EUBREAST Network has an upcoming IGM registry study, which is expected to provide valuable insights into the understanding and management of this condition.

Abstract Background Granulomatous mastitis (GM) is a rare, chronically relapsing inflammatory breast disease for which no universally accepted treatment standard exists. Although corticosteroids, surgery, methotrexate and other modalities are widely used, the regimen that simultaneously optimises efficacy, relapse control and safety remains unclear. Methods Six databases (PubMed, Embase, Cochrane Library, Scopus, ClinicalTrials.gov, and Google Scholar) were systematically searched from inception to 29 September 2025. Eligible studies included retrospective or prospective cohort studies evaluating at least one treatment for granulomatous mastitis (GM) and reporting relevant outcomes. The methodological quality of all included studies was assessed using the Newcastle–Ottawa Scale (NOS). Statistical analyses were performed using Stata 17.0 and R 4.4.2; a Bayesian network meta-analysis was conducted with the gemtc package. The primary outcome was relapse rate (RR); secondary outcomes were complete response (CR) and objective response rate (ORR). Two prespecified sensitivity analyses were performed to assess robustness: (1) restricting the analysis to high-quality studies (NOS score ≥ 7), and (2) limiting the relapse rate analysis to studies with a median follow-up duration of ≥ 12 months. Results We included 19 cohort studies enrolling a total of 1,559 patients with granulomatous mastitis. Compared with systemic corticosteroids, the combination of surgery and systemic corticosteroids significantly improved complete response (OR = 9.46, 95% CrI 1.22–74.47). Triple therapy—comprising systemic corticosteroids, local corticosteroids, and surgery—was associated with a significantly lower relapse risk (OR = 0.120, 95% CrI 0.020–0.709). No other treatment regimens demonstrated statistically significant benefits for the primary outcomes. In terms of safety, the adverse event rate was lower with surgery + systemic corticosteroids (18.6%) than with systemic corticosteroids alone (23.73%), whereas surgery alone was associated with a higher rate (33.3%). Sensitivity analyses support the robustness of our main findings. Conclusions This study suggests that multimodal combination therapies are generally superior to monotherapy regimens. Among them, surgery combined with systemic corticosteroids demonstrated notable advantages in both complete response rate and safety profile and ranked among the top strategies for relapse prevention. However, these findings are exploratory in nature and warrant validation through additional high-quality clinical studies.

Granulomatous lobular mastitis (GLM) is often misdiagnosed clinically as breast cancer (BC). Therefore, it is crucial to differentiate between GLM and BC. This study used 259 samples from 129 patients with GLM and 130 with BC. A total of 874 radiomics and 11 clinical features were obtained. The least absolute shrinkage and selection operator algorithm was used to select radiomics features. Univariate and multivariate analyses were performed to screen clinical features. Three machine learning algorithms were applied to assess the efficiency of the radiomics, clinical, and combined models, which were compared to select the optimal model. Finally, a nomogram based on the optimal model was developed. Decision curve analysis (DCA) and calibration curves were used to assess the clinical utility of the nomogram. Twelve radiomics features were identified as the most relevant for distinguishing GLM from BC, including the original gray level co-occurrence matrix autocorrelation feature. Important clinical features included age, nipple inversion, and C-reactive protein levels. The combined model demonstrated superior performance in terms of accuracy, specificity, and sensitivity compared with the clinical and radiomics models. A nomogram was constructed based on the combined model. The calibration curve and DCA further confirmed the superior clinical value of the nomogram. A combined model incorporating 12 radiomics and 3 clinical features is potentially valuable for distinguishing GLM from BC.

ABSTRACT Background Granulomatous mastitis (GM) is a rare, chronic inflammatory breast condition that clinically and radiologically mimics breast carcinoma. GM is most commonly reported as tubercular (TB) mastitis. This study aims to identify predictive factors differentiating idiopathic and tubercular GM based on cytomorphology. Methods Data were retrieved from the departmental archives between January 2019 and June 2024. Additional investigations and therapeutic responses were recorded during follow‐up. Based on correlation with gold standard diagnostic tools and therapeutic responses, cases were categorised into idiopathic and tubercular GM groups using cytomorphological features and analysed for risk stratification. Results A total of 25 GM cases were included in the study, with a mean age of 29 ± 5.7 years. Ziehl‐Neelsen (ZN) staining was positive in three cases. A histological correlation was available for 21 cases, PCR for 5 and therapeutic response for all 25 cases. Seven cases responded to antitubercular therapy (ATT), while 18 responded to steroids. Cytological diagnosis identified TB GM in 8 cases and nonspecific GM in 17 cases. The agreement between original and revised diagnoses was excellent after correlation with gold standard methods and therapeutic outcomes. Univariate and multivariate analyses revealed that a cytological diagnosis of GM with well‐formed granulomas, absence of neutrophils, presence of many lymphocytes and lack of plasma cell infiltration was significantly associated ( p &lt; 0.05) with TB GM. In univariate analysis, age under 30 years was significantly associated with TB GM ( p = 0.025); however, this was not significant in multivariate analysis. Conclusion Cytomorphology is a valuable and rapid tool for differentiating between TB and idiopathic GM, helping to prevent unnecessary surgical interventions while awaiting results from ancillary diagnostic techniques.