Etiologic revelation and outcome of the surgical management of idiopathic granulomatous mastitis;An Egyptian centre experience

Abstract

INTRODUCTION:

Idiopathic granulomatous mastitis (IGM) is a chronic inflammatory condition that is confused with cancer. It usually affects women in child bearing age. The exact aetiology and pathogenesis are still unknown, and the optimal therapeutic modality has not yet been established. Treatment most frequently includes Antibiotics, corticosteroids and immunosuppressant, surgical excision, and even mastectomy.

MATERIAL AND METHODS:

We studied a thirty cases diagnosed with IGM in our locality to find out the leading risk factors and the outcome of our surgical approach which involves excision of the lesion in continuity with duct system. Patients demographic data, history related to lactation and outcome were recorded.

RESULTS:

All patients were parous women with history of previous breast feeding for all kids. Twenty-six patients (86.66%) had a history of early incomplete nursing care to the affected breast. After our surgical approach, Twenty eight (93.3%) patients showed fast recovery with no detectable recurrences in the median follow up period (18 months) with acceptable cosmoses.

CONCLUSION:

History of breast feeding together with early failure of complete nursing from a single breast is the most important risk factors for development of IGM in young aged women. Surgery plays an important role in treating IGM, however, it should be directed towards excision of the present mass (s) together with the pathological and colonized duct system.

Introduction

Mastitis may occur as a part of systemic illness or as localized breast phenomenon. The acute infective ones are common and usually easy to diagnose and treat, however, the rarer chronic conditions are more indolent and their presentation simulates breast cancer, a painless indurated mass [1]. Idiopathic granulomatous mastitis (IGM) is a rare chronic inflammatory condition which mimics breast cancer [2]. This condition is mostly seen in young parous women in the child bearing period (age ranging from seventeen to forty-eight years, the youngest patient reported was eleven years old), with history of breast feeding and it is usually presented within months to eight years of the most recent lactation [3]. Interestingly, this disease is reported during pregnancy and in men [4]. The lesion is usually unilateral, rarely bilateral, with no predilection to a side [5] and was found in every quadrant of the breast except the subareolar region [6]. Nipple discharge and axillary lymphadenopathy are occasionally found (15% of cases), however, the most common presentation was a painful breast mass [5].

The diagnosis of IGM does not depend on the clinical and radiological findings which are in many cases non-specific and unable to resolve the differential diagnosis of inflammatory process and malignancy, however, the definitive diagnosis can be established only by the demonstration of the characteristic histological finding of a non-caseating granulomatous inflammation with aggregates of neutrophils in small suppurative foci (microabscesses). In areas of microabscess formation/suppuration, there was a lipid vacuole in the centre of the inflammatory process [7], nevertheless, confirmation of the diagnosis require the exclusion of the other possible causes of granulomatous breast disease such as tuberculous, parasitic and fungal infections, histoplasmosis, sarcoidosis, wegener’s granulomatosis, giant cell arteritis, polyarteritis nodosum, vaccination granuloma and foreign-body reactions [3]. The most important histological feature is the fact that the inflammatory process has a predominantly lobular distribution [8].

This condition has a complex clinical behaviour, protracted course and sometimes aggressive disfiguring clinical presentations (with large or multiple masses, skin ulcerations, peau d’orange, sinuses) with high tendency for recurrence (16–30% and upto 50% of cases in some studies) [9–12]. However, the natural history of idiopathic granulomatous mastitis is often that of a self-limiting condition and 50% of the cases eventually recover regardless the treatment used [13]. Despite the presence of several contradictory studies, the most commonly accepted hypothesis is that IGM is an autoimmune disease owing to its positive response to steroid treatment [10,14]. In view of the obscure leading causes and the hidden pathogenesis of IGM, it is usually difficult to eradicate it and the suggested therapies include simple observation, antibiotics, steroid, drainage and excision [15].

In this study we tried to find out the leading risk factors and the outcome of our surgical approach which involves excision of the lesion in continuity with duct system.

Methods

This study included thirty patients who met the histologic criteria and diagnosed as IGM and were treated in our institution between January 2013 and May 2016. All patients presented by a unilateral breast mass or a residual persistent mass after medical treatment or incision and drainage of an abscess. Patients who had a history of breast surgery or cosmetic treatment by silicone prosthesis or injection of silicon, liquid paraffin, or beeswax into the breast parenchyma (olegranulomatous mastitis) were excluded from this study. A detailed medical history was taken, regarding the duration of the complaints, whether the complaints were of a repetitive nature, the treatments used, the number of pregnancies, the duration of breast nursing, any lactation problems, smoking habits, oral contraceptive use, history of breast trauma and the presence of chronic systemic autoimmune disease or intake of any galactorrhea-related drugs. In all patients, the mass forming lesions were confirmed by a breast examination (Fig. 1) and imaging evaluation using breast sonography, whereas mammography and magnetic resonance imaging (MRI) were used depending on the age and clinical condition of the patients. All patients were biopsied using tru-cut needle biopsy (14-gauge). Tuberculosis and possible fungal infections were ruled out in each patient from the patient’s medical history, chest X-ray and Ziehl-Neelsen staining (for tuberculosis), Periodic acid-Schiff (PAS) and GMS staining (for fungal infection) of the paraffin-embedded specimens. Bacterial culture, fungal culture, and mycobacterial culture were not routinely used in our study except for eleven patients in whom we underwent microbiological testing of the aspirate or fresh tissue samples. The definitive diagnosis was made by the presence of an exclusively granulomatous inflammatory reaction with neither caseous necrosis nor any specific organism in the tissue samples. All patients were treated by surgical excision of their masses with major duct excision in-continuity together with part of the overlying skin (Fig. 2). The resultant defect was closed primarily or with the use of parenchymal displacement techniques (Fig. 3). No volume replacement techniques (autologous or synthetic) were used in this study. None of the patients received steroids in the preoperative or the postoperative settings. The recovery period was defined as the time between the initial office visit and the disappearance of symptoms or the completion of wound healing. Recurrence was defined as the reappearance of the same symptoms (mass) after the surgical removal of the lesion.

Results

The characteristics of thirty patients are summarized in Table 1. The mean age was thirty-three years (range: 23–43 years). All the patients were in the reproductive period. All patients were non-smokers. All patients were parous and had given birth to two to five kids. At the time of diagnosis, none of the patients were pregnant or lactating. None of the patients had a history of breast trauma, galactorrhea, chronic systemic or autoimmune disease (e.g. erythema nodosum), or a history of specific granulomatous infection or history of chronic use of antipsychotic drugs or steroids. Only one patient reported the use of oral contraceptive pills for unknown period of time, the remaining twenty-nine patients (96.66%) used an intra-uterine device (IUD) after their last kid. Twelve patients (40%) reported the presence of gynaecological infections in the form of repeated vaginal discharge. Time of diagnosis ranged from eight months to seven years since the last delivery with Twenty-five (83.3%) patients diagnosed within four years of their most recent delivery. All patients did breast feeding for their kids. Exclusive breast feeding was documented in twenty-three patients (76.66%) and the other seven patients used supplementary feeding, two of these patients were having twins. None of the patients reported the occurrence of clinical lactation mastitis or any other lactation problems except for nipple excoriation, nevertheless twenty-five patients (83.3%) had a history of early incomplete (for 2–4 months only) nursing of their last kid(s) from the breast in which the IGM later developed. Another one patient did not succeed to feed her three kids at all from the offending breast due to nipple inversion despite repeated trials during the first month of lactation to re-evert it and giving it to the baby to suckle. Interestingly, two of the remaining four patients who fed normally from both breasts had twins in their last delivery and one member of the twins had to feed artificially.

All our patients presented by a painful indurated breast mass. Unilateral involvement was present in all patients; the right breast was involved in fourteen patients (46.66%) and the left breast in sixteen patients (53.33%). The most commonly affected breast quadrants in our study were the outer (43.33%) and lower (26.66%) quadrants, followed by the inner (16.66%) and upper (13.33%) quadrants. All masses were peripheral (not sub-areolar) except for six cases who presented by segmental disease affecting a whole breast segment from the nipple to the periphery.

The clinical size of the mass varied from 4–14 cm in its greatest dimension (mean diameter was nine cm), while the corresponding radiological sizes were always smaller. In most of our cases (in twenty two patients), ultrasonography reported an irregular or lobulated hypoechoic heterogenous branching mass with tubular finger-like extensions, with occasional coarse internal echos, dots of calcifications, minimal parenchymal distortion with ectatic ducts at the site of the mass. Sometimes the ultrasonography reported parenchymal distortion without a mass (in three patients), or the presence of multiple lobulated irregular masses and surrounding parenchymal edema. Fifteen patients (50%) presented with associated overlying skin affection in the form of peaud’orange, erythyma with or without excoriation, sinus discharging purulent fluid (in three patients) (Fig. 1), or non-healing scarred wound of previous incision and drainage. Twenty-two patients reported seeking previous medical advice and received intermittent repeated courses of antibiotic and anti-inflammatory drugs for a period ranging from (1–9 months), seven patients of them underwent incision and drainage of abscesses in the offending breast. Synchronous or metachronous involvement of the contralateral breast was not found in this study. In eleven patients for whom microbiologic testing of aspirate and/or tissue samples was done, mixed infection with aerobic and anaerobic micro-organisms was found in seven; and the other four patients showed no growths in the microbiological testing and had history of recent antibiotic use.

All the patients of the study were operated for excision of the mass in continuity with the breast duct system together with part of overlying skin. In the six cases presented by a segmental disease formal segmental resection (Melon slice) was done ensuring that the whole duct system is addressed in the resection. The resulting defect usually closed primarily or with the help of glandular or demoglandular displacement techniques. No volume replacement techniques (autologous or synthetic) were used in any of the cases of the study.

Postoperative histopathological examination of the excised lesions showed non-caseating granulomatous inflammatory responses centred on breast lobules and around peripheral ducts. In affected lobules, there was a loss of acinar structures; the entire epithelial lining of the lobules was destroyed. The granulomatous reaction composed of epitheloid histiocytes, and foreign-body giant cells accompanied by lymphocytes, and plasma cells, also most of the granulomatous reactions were rich in neutrophils, which are seen aggregated in small suppurative foci (microabscesses) in 3 cases.

Twenty-eight (93.3%) patients of the study showed fast recovery, smooth postoperative course, very nice wound healing, acceptable cosmetic results; and with no detectable recurrences in the median follow up period of 19 months (range from 8 to 44 months), only two patients showed complications, one of them was diabetic and showed postoperative wound complications which was successfully managed by antibiotics and frequent dressings and the other showed small mass recurrence, away from the index mass, which was not painful and required no further intervention except for observation only and complete spontaneous resolution occurred four months later.

Discussion

Idiopathic granulomatous mastitis (IGM) is an uncommon completely benign non-cancerous condition affecting one or both breasts, was first reported by Kessler and Wolloch in 1972 [16], then Fletcher et al. reported a further seven cases and described the clinical features and the histology of the disease [17]. Its cause and treatment have not yet been clearly determined despite the various studies worked on it.

In some works, parenchymal heterogeneity and areas of mixed echo pattern with parenchymal deformity have been described in addition to the irregular hypoechoic mass lesion [18]. In our study, the ultrasound in most of the cases confirmed the presence of hypoechoic heterogenous irregular mass with tubular extensions associated with parenchymal distortion & ectatic ducts at the site of the mass with occasionally thickened skin and obliteration of subcutaneous fat, and sometimes small abscess cavities in accordance.

The occurrence of IGM without a history of recent pregnancy is uncommon. To the best of our knowledge, IGM in nulli-gravidae has been rarely reported [10,19], and are frequently associated with secondary hyperprolactinaemia due to pituitary adenoma, drug-induced hyperprolactinemia [20].

Currently, the most commonly accepted hypothesis is that IGM is an autoimmune disease [15,16] owing to its positive response to steroid treatment [10,14]. According to our study, we found that the most significant risk factors for development of IGM are history of breast feeding (all patients) together with the history of early incomplete nursing for the affected breast, which was found in twenty six patients of the study (86.66%). This typical history of non-nursing of the breast in which IGM later developed despite normal nursing for the other breast was formerly observed by Altintoprak et al. 2013, however, he didn’t correlate this observation to the disease process [21].

The pathogenesis of IGM could be similar to the pathogenesis of duct ectasia/periductal mastitis (DE/PDM) complex in which damage to ductal epithelium produced by local trauma, local chemical irritant or infection [10,22].

According to our study, we suggest that the trigger for the epithelial damage is neglected subclinical mastitis which passed unnoticed and untreated at times of lactation.

In our study, none of our patients gave a definite history of overt clinical mastitis at times of lactation, however most of them (86.66%) gave history of early incomplete nursing from the offending breast, the history we suggest to be related to occurrence of subclinical (histological) mastitis. This highlights the strong relationship between the occurrence of IGM and subclinical not clinical mastitis. Review of the literature revealed that there is still no definitive or ideal treatment regimen for IGM because of its rarity and uncertainty about its etiology and pathogenesis [15].

In 1980, DeHertogh et al. followed by several other authors described the efficacy of steroids (high dose of 60 mg per day for 3 weeks or longer courses according to the response followed by gradual tapering of the dose). Gurleyik et al. studied nineteen patients with IGM, they preferred the protocol of initial steroid therapy followed by conservative surgical excision of the remaining lesions. They emphasized that steroid treatment alone did not achieve complete response in any of the patients of the study [23]. Others reported poorer response to steroid therapy with maximum success rate about 30%, and with eventual surgery required in most patients [24], this is together with the known potential side-effects of long term use of steroids.

Immunosuppressives (especially methotrexate or azathioprine) also have been reported in treating IGM patients especially those who relapsed after tapering steroids [14,25]. Owing to the belief of the role of hyperprolactinaemia in the pathogenesis and the recurrence of IGM, anti-prolactin medications (e.g. cabergoline) were reported to be beneficial [26].

Other authors stated that the natural history of IGM can be self-limiting and 50% of cases recover without complications, regardless of the treatment used (no treatment, antibiotics alone, surgery alone, antibiotics and steroids, antibiotics and surgery, steroids and surgery, antibiotics and steroids and surgery). The time required for this remission was from 6 to 12 months [27].

The role of antibiotic therapy in granulomatous mastitis is highly questioned and there is no sufficient evidence yet to date on the therapeutic value of antibiotic in IGM [24].

Tse et al., stated that surgery should be mandatory in IGM treatment and without surgical treatment, IGM patients may undergo a chronic progressive clinical course [28].

It was claimed that IGM should not be treated surgically to avoid complications such as fistula formation and poor wound healing. In addition, a higher recurrence rate (ranging from 21% to about 50%) [3,18] was reported after surgical excision compared to the recurrence rate after steroid treatment [25]. Some studies referred the failure and recurrence after surgical excision of granulomatous mastitis to the incompleteness of the excision and recommended wide local excision [2,24]. Yau et al. recommended a wide surgical excision or even mastectomy for IGM with persistent symptoms because of a lower rate of recurrence [29]. Recently, surgical treatment was reconsidered for treating granulomatous mastitis and was proved by several studies to be a critical treatment method for granulomatous mastitis being effective, with a fast recovery, lower recurrence rates and provide more accurate histopathological diagnosis [24]. Sung Mo Hur, 2013 reported that with surgical excision the success rate was 91.7% success rate and the recurrence rate was 8.7% [24]. However in Sung’s study, the lesions were relatively small (The average size of the lesions was 3.8 ± 2.3  cm ) and they used to undergo wide local excision of the lesion with surrounding normal tissues.

Erozgen et al., in his study of thirty-three patients with granulomatous mastitis concluded that surgical treatment should be the first treatment option in patients with complicated granulomatous mastitis as abscess formation, fistulae, and persistent wound infection, nevertheless corticosteroid therapy could be reserved for non-complicated cases [30].

Based on our suggested risk factor and pathogenesis, we recommended that the best approach to eradicate this disease is surgery aimed at excision of the mass together with duct system which represents a source rather than a path for micro-organisms and the potential bacterial infection.

In our study we did not perform wide local excision but rather we were more keen to remove the core (nidus) of this chronic granulomatous inflammation, which is the most indurated part of the mass that almost containing the previously extravasated milk products which initiated the process, and we frequently did not excise the surrounding innocent inflammatory tissue and we left it to resolve spontaneously during the recovery period. In all cases, we used to thoroughly palpate the breast parenchyma during surgery, and this frequently (in twelve cases) revealed nearby or distant small indurated dormant lesions that had not been preoperatively detected clinically, for these dormant lesions only simple excision was required just by avulsing their core. We observed that this surgical approach is very effective in eradication of the disease with avoidance of the need for steroid therapy.

Conclusion

History of breast feeding together with early failure of complete nursing from a single breast is the most important independent risk factors for development of IGM in young aged women. Surgery plays an important role in treating IGM cases, however, it should be directed to excising the present mass (s) together with the pathological and colonized duct system.