Abstract
Granulomatous mastitis is a chronic inflammatory breast disease with an enigmatic clinical presentation that can delay diagnosis and perpetuate ineffective treatments. We report our experience with the diagnosis and management of eighteen cases of granulomatous mastitis in the setting of an urban public hospital. The patients were identified after a retrospective review of pathology and surgery databases. Demographic, clinical, radiographic, histopathological data and treatment responses were reviewed. Out of a total of 18 patients, 8 patients were noted to be from Mexico. The mean age at presentation was 35 years and 11 of the 18 patients reported a pregnancy in the last four years. Most cases were unsuccessfully managed as bacterial breast infections initially; definitive treatment involved surgical excision and steroids. A complete symptomatic and radiographic resolution was documented in 8 of the 18 patients. Despite posing a challenge for the unsuspecting physician, an early diagnosis of granulomatous mastitis improves patient outcomes.
Keywords
Introduction
Granulomatous mastitis remains an elusive disease entity despite recent reports highlighting various diagnostic and management protocols [1–4]. First described in 1972 as a distinct pathology among benign breast diseases [5], the diagnosis of granulomatous mastitis (GM) is often established after excluding inflammatory breast cancer, autoimmune breast disease and other etiologies such as tuberculosis, fungal or parasitic breast infections. A definitive diagnosis requires a histopathological confirmation of multinucleated giant cell granulomas with microabscesses [6]. A combination of varied clinical manifestations, the absence of specific radiological findings and the lack of consensus on treatment paradigms often contributes to prolonged delay in diagnosis, unnecessary treatments and prolonged anxiety for a patient with GM. We present here our experience with 18 cases of GM in terms of clinical presentation, work up and management outcomes in a public hospital setting.
Materials and methods
This report includes 18 patients with a histopathological diagnosis of GM at Harlem Hospital Center between 2010 and 2016. Harlem Hospital is a public hospital that caters to a diverse urban population. The patients were identified after a retrospective review of pathology and surgery databases. Demographic, clinical, radiographic and histopathological data was reviewed. The clinical records were reviewed to identify possible risk factors for GM previously reported in literature. The response to the treatments given was also determined from the patients’ follow up records. Data was collected and was then analyzed and compared with a statistical software (STATA 13.1).
Results
18 cases of GM were identified: the patients’ demographic and clinical characteristics are presented in (Table 1). The mean patient age at presentation was 35 years. 16 out of 18 patients were Hispanic, and 8 of these patients were noted to be from Mexico. None of the patients were pregnant at the time of presentation; only one patient was noted to be breast-feeding. 11 of the 18 patients were noted to have been pregnant at least once in the last 4 years prior to presentation. A history of oral contraceptive use was documented in 4 patients and 2 patients reported smoking. None of the patients had diagnosed autoimmune disorders or any other chronic comorbidity; 2 patients reported a history of pulmonary tuberculosis. Prolactin levels at presentation were measured in 8 patients with a mean value of 14.3.
The most common presenting symptom was a palpable breast lump in 13 of the 18 patients, followed by breast pain in 8 patients and overlying skin changes including erythema and ulceration in 6 patients (Table 2). All patients underwent a radiological work up including an ultrasound and/or mammography at some point during their follow-up. The most consistent finding on ultrasound was a hypoechoic lesion or a lesion with mixed echogenicity while an irregular asymmetric density was reported on most mammograms. All patients were diagnosed with GM based on histopathological specimens obtained via core needle biopsy (8 patients) and/or an excisional biopsy (10 patients); the biopsy specimens were routinely assessed for fungal/mycobacterial growth with special histochemical stains.
Patients with GM were managed according to clinical presentations. 9 of the 18 patients were noted to have abscesses on initial presentation and underwent an incision and drainage as the first treatment followed by a course of antibiotics. Microbiology culture results showed no growth for all except two patients: one patient had a positive culture for gram-positive cocci while the other patient’s cultures grew Klebsiella. The two patients who reported a prior history of tuberculosis underwent blood QuantiFERON testing and chest x-rays to exclude active tuberculosis infection. All 9 patients were noted to have persistent symptoms of breast pain and overlying inflammation despite prolonged courses of antibiotics and multiple drainage procedures in some instances. 6 of the 9 patients underwent an excision (lumpectomy) for a persistent breast mass and were noted to have resolution of symptoms at follow –up. In addition, 3 of the 9 patients were noted to have developed draining fistulas despite drainage and antibiotic therapy and were given a trial of steroid therapy, to which a clinically significant response was documented in all three patients.
6 of the 18 patients were primarily managed with antibiotics without drainage on initial presentation. None of the 6 patients responded to antibiotics and subsequently underwent a core needle biopsy after which a diagnosis of GM was established. All patients received varying regimens of steroids: all 6 patients responded to steroids initially, with 4 of the 6 patients noted to have a relapse after discontinuation of steroids (Tables 3 and 4). One of the patients required a lumpectomy for a persistent breast lesion while three patients were still noted to have recurrent complaints of breast pain and/or a palpable lump despite repeated steroid use.
The mean duration of follow up was 14 months. On the final follow up of the 18 patients included in this study, 8 patients were noted to have complete symptomatic and radiographic resolution, 7 patients reported no symptoms but follow-up imaging reported persistent lesions while 3 patients were noted to have persistent symptoms despite therapy.
Discussion
Granulomatous mastitis is a rare inflammatory breast disease of unknown etiology. Baslaim et al. reported an incidence of 1.8% of histopathologically confirmed cases of GM in 1106 women with benign breast diseases [7]. Despite case series of GM reported worldwide [7–9], there are very few reported cases from the United States: our series of 18 patients is the second largest series to date from this patient population [2]. GM is known to be more common in women of child bearing age and usually presents within a few years of parturition [1,2]; this is consistent with our study finding of a mean patient age of 35 years and a majority presentation within 4 years of pregnancy. Contrary to other published reports, we found no significant association between a diagnosis of GM and a history of contraceptive use, smoking or hyperprolactinemia [10,11]. Also, none of the patients in our series had any associated systemic autoimmune disorders. Most cases of GM in our patient population were in Hispanics from Mexico, which corresponds to the demographic distribution in other published studies from the United States [2,12].
Granulomatous usually presents with a painful breast lump in any quadrant of the breast, usually without associated axillary lymphadenopathy. There are no specific sonographic or mammographic features that can specify a diagnosis of GM. Most radiographic studies for out set of patients reported an ill-defined lesion with mixed echogenicity. We concur with prior reports in that a diagnosis of GM requires a histopathological diagnosis either via a core needle or an excisional biopsy [13]. However, other causes of granulomatous inflammation including sarcoidosis, Wegener’s and infectious etiologies such as tuberculosis and fungal infections should be excluded as part of the work up.
There is a general lack of consensus and clarity in the treatment of GM. Most of the patients with GM are invariably managed with antibiotics owing to the initial clinical presentation of breast pain with inflammatory skin changes, as is evident from the large number of patients that received antibiotics in our series. This often perpetuates a cycle of antibiotics and recurrent drainages. However, the main therapeutic approaches include surgery, steroids, immunosuppressant drugs or a combination of two or more modalities [2,3], although recurrences are reported to be as high as 50% [14]. Most authors advocate against surgery as the initial treatment of choice and recommend conservative management with steroids [2,3,14]. Although 6 of the 18 patients in our series were managed successfully with a lumpectomy alone, some of our patients in this review responded well to a steroid regimen with no recurrence. The results of our study concur with the general perception among breast physicians of limiting disfiguring procedures for apparently infectious abscesses on the breast: attempts should be made to work up these lesions, avoid unnecessary antibiotics and restrict aggressive interventions, unlike abscesses encountered elsewhere. Further, we believe that treatment with steroids results in regression of the inflammatory disease, allowing the possibility of more conservative surgery should it be necessary for persistent lesions. The appropriate duration and dosage of steroid therapy remains undecided, although it is generally recommended that steroid therapy be continued until a complete remission occurs.
To conclude, the diagnosis and treatment of GM can be a daunting task. In the absence of clinical trials, management of this disease entity will continue to be based on retrospective reviews. Further, the observation of apparently increased incidence of GM in women from Mexico is intriguing and warrants focused studies. Nevertheless, as our understanding of the disease process of grows, we can institute well-defined diagnostic and management protocols to reduce the undue use of antibiotics and prevent repeated disfiguring drainage procedures.