Attention has recently been focused on the importance of multiorgan disorders characterized by the formation of inflammatory pseudotumors, with or without organ dysfunction, accompanied by increased immunoglobulin G4 (IgG4) levels.1

IgG4-related disease (IgG4-RD) consists of infiltration by plasma cells responsible for IgG4 synthesis and secretion. The diagnostic criteria are characterized by: clinical manifestations that vary depending on the affected organ and serological and histopathological findings. It is associated with a wide range of clinical signs, and the disease characteristically involves the exocrine glands (lacrimal and salivary glands and pancreas); however, it has been described as being practically universal as it can affect nearly any organ.2–6 The serological diagnosis is based on high IgG4 levels (≥135mg/dl); however, this criterion is not absolute, as we find patients in whom this immunoglobulin subclass is not elevated despite their confirmed diagnosis; moreover, increased IgG4 levels have been reported in other diseases.7 The diagnostic criterion having the greatest specificity is the histological examination. Thus, it is necessary to demonstrate that the histopathological criteria for the disease (IgG4/IgG ratio, storiform fibrosis, obliterative phlebitis) are met.

The recent recognition of the disease has enabled us to reclassify patients who we had previously diagnosed with “inflammatory pseudotumor”.6 The main objective of our study was to evaluate the prevalence of IgG4-RD in biopsies from patients sent to the pathology department of our hospital.

For this retrospective clinical and histological study, we analyzed the biopsy reports issued by the pathology service of Hospital Central “Dr. Ignacio Morones Prieto” in San Luis Potosí, Mexico, from 7 January 2011 to 28 June 2013. The study was divided into 3 phases; the first involved the search for those reports in the pathology database that included data indicative of IgG4-RD8 (Fig. 1). At the time of this phase, none of the specimens had been examined by immunohistochemical staining for IgG4.

Fig. 1.

Flowchart. Selection of biopsies for the study.

(0.14MB).

In the first phase, 2 of the researchers (DHV and TJP) carried out a manual search among the histology reports for all the specimens from January 2011 to June 2013, a total of 23,720 biopsies that exhibited the histological—not immunohistochemical—characteristics of IgG4-RD, such as: (a) nonspecific inflammatory response with lymphoplasmacytic infiltration; (b) inflammatory pseudotumors localized to any organ; (c) storiform fibrosis; and (d) obliterative phlebitis. Those specimens reported as neoplasms of any type or infections were excluded, as were all those for which the histological description was not available.

In the second phase, the biopsy findings were classified as probable or definite IgG4-RD according to the criteria proposed by Umehara et al.9 Probable disease was defined as the presence of 2 inclusion criteria, in which case, immunohistochemistry for IgG and IgG4 was performed, and definite disease as the presence of 3 or more inclusion criteria and IgG4 positivity, with an IgG4/IgG ratio ≥35%. The remaining specimens were excluded. We did a search for the clinical and demographic data of the patients whose biopsy results had indicated probable disease. Only 1 patient had undergone the determination of serum IgG4.

In the first phase, we reviewed 23,720 biopsies. In the second phase, of 71 biopsies classified as probable, 8 were finally recorded as definite cases of IgG4-RD; the remainder of the specimens (n=63) were categorized as probable. The median age of the 3 women and 5 men whose specimens had been classified as definite cases was 45 (12–68) years. The diagnoses in the specimens with a definitive diagnosis of IgG4-RD referred for pathological study were (Fig. 2): aortitis (n=1), inflammatory pseudotumor of the lung (n=1), pancreatic tumor (n=1), pericarditis (n=1), lymphadenopathy under study (n=2), dacryoadenitis (n=1) and parotitis (n=1). The 2 patients with lymph node involvement were admitted to the hospital for fever of unknown origin and the patient with pancreatic tumor developed sclerosing cholangitis and salivary gland involvement. All were positive for IgG4 according to immunohistochemical analysis.

Fig. 2.

Aortic reconstruction by magnetic resonance imaging indicating dilated portion of aortic arch. Patient with IgG4-related sialoadenitis with aortic involvement.

(0.2MB).

The probable group included 63 biopsies, which were reanalyzed; 29 were excluded because of insufficient microscopy data for diagnosis; the remaining biopsies were subjected to immunohistochemistry with IgG and IgG4. The pathological diagnoses prior to the immunohistochemical analysis of these biopsies are shown in Fig. 3. The IgG4+ specimens corresponded to: 5/17 cases of granulomatous mastitis, 1/1 case of chronic sclerosing dacryoadenitis, 2/9 cases of chronic sialoadenitis (1 associated with chronic pancreatitis), 1/2 cases of chronic thyroiditis, 1/1 case of chronic pancreatitis (Fig. 4). The percentage of IgG4+ plasma cells in the specimens ranged between 5% and 10% of the cells, in all the samples with an IgG4/IgG ratio >35%.

Fig. 3.

Histological features in patients with and without positivity for IgG4. Eosin: eosinophils; IPT: inflammatory pseudotumor; OP: obliterative phlebitis; PC: plasma cells; SF: storiform fibrosis.

(0.13MB).

Fig. 4.

(A) Lacrimal gland stained with hematoxylin and eosin (40×): periglandular plasma cell infiltration with glandular distortion. (B) Immunohistochemistry of IgG4+ lacrimal gland, IgG4/IgG ratio >40%.

(0.29MB).

In the evaluation of the definite and probable biopsies, we found a prevalence of IgG4-RD of 25.3%. It was interesting to note that the highest prevalence of positivity was observed among the specimens—classified as probable—in which the reference diagnosis was granulomatous mastitis.

Given that many of the biopsies were performed in an ambulatory setting and the fact that the pathology department is the state referral center for San Luis Potosí, it was not possible to obtain the complete clinical data for the majority of the patients.

The prevalence of IgG4-RD in Caucasian populations is described as being low, although there are no available registries of Hispanic or Latino populations. This circumstance highlights the importance of the present study, in which we found a prevalence of 24% in biopsies with histological changes that were indicative of probable, or were compatible with, IgG4-RD. Such a high prevalence is due to the fact that our hospital is a referral center and, as has been pointed out by different authors, the greatest prevalence of the disease is observed in university hospitals, where the incidence is 0.28–1.08 per 100 000 population, with 336–1300 new cases each year. These facts underline the importance of maintaining a high level of suspicion, not only clinical, but histological as well.10,11

Our study shows that, due to the lack of knowledge about the pathophysiology of IgG4-RD, these conditions are most probably underdiagnosed, and that, given the heterogeneity of the clinical signs, the final diagnosis should be based on the characteristic histopathological findings.

The reports in the literature dealing with the clinical picture of the disease stress the multiorgan association; the most common examples include salivary gland involvement and pancreatic, bile duct, lymph node and even renal involvement.12–15 Our findings contrast with these observations, given the higher prevalence of mammary gland involvement. This disease site has only been reported in a study carried out by Cheuk et al.,16 in which the authors, on evaluating the mammary gland biopsies with presence of pseudotumors, found a great abundance of IgG4-positive plasma cells. The scope of the present study was broader as we took into account all the histopathological diagnoses, not limiting ourselves to any one organ. Despite the fact that we do not provide serum IgG4 levels in our study, the histological findings, especially IgG4/IgG plasma cell ratio greater than 50% in the biopsy, the presence of storiform fibrosis or the presence of obliterative phlebitis, are representative, and may alone exhibit greater specificity.17,18

The retrospective nature of the study makes it difficult for us to categorically establish the diagnosis of IgG4-RD because of the lack of clinical and serological data of our patients, although, given their high specificity, the most important criteria for the diagnosis of diseases caused by IgG4 are the histological findings. This is not the case of the determination of serum IgG4 since, as reported by Ebbo et al.,7,15,19 elevated serum IgG4/IgG ratios were detected in infections and neoplasms, a fact that underscores the absence of specificity of this antibody in a number of clinical contexts. This circumstance also leads us to question the central role of IgG4, and ask whether, in some cases, the infiltrate and this immunoglobulin might be an epiphenomenon.

The diagnostic value of the histological changes in IgG4-RD is variable, taking into account the high sensitivity of the tests. Masaki et al.18 reported a low specificity (30%) and high sensitivity (99%) for different variables (e.g., an IgG4/IgG ratio of less than 10%). Thus, the diagnosis is based on the serum IgG4 level, the presence of storiform fibrosis, the plasma cell infiltrate and eosinophilia.20

The biological behavior of IgG4 is heterogeneous, since it has both anti-inflammatory properties, as has been observed in certain allergic or allergy-like diseases such as asthma,21 and pathological properties, through which it exerts a direct effect on different structures, as in pemphigus foliaceus, in which the antibody is directed against desmoglein 3.22 Although the behavior is clear in the preceding examples, the same cannot be said of other IgG4-RD in which, despite there being a number of reviews, it still is not clear whether the immunoglobulin is an epiphenomenon or plays a major role in the disease.1,2 The antigen that could trigger the disease is also unknown, despite a proteomic analysis in which a 13-kDa antigen was detected.23

Our report clearly stresses the high prevalence of IgG4-RD, pointing out that none of the biopsies initially subjected to histological analysis revealed evidence of the disease. Although we found the organ involvement characteristic of IgG4-RD, mammary gland involvement, defined as granulomatous mastitis, was highly prevalent.

The authors declare they have no conflicts of interest.