Clinical Characteristics of 76 Patients with IgG4-Related Hypophysitis: A Systematic Literature Review

Background IgG4-related hypophysitis (IgG4-RH) is a rare disease, and its prevalence remains unclear. In recent years, an increasing number of cases have been reported because of the increasing recognition of this disease. We aimed to summarize case reports of IgG4-RH and outline the clinical features and outcomes. Methods We performed PubMed search of articles using the search terms “hypophysitis [AND] IgG4.” Consequently, only 54 English articles (76 cases) met Leporati's diagnostic criteria. Results Of the 76 cases, the ratio of men to women was 1.5 : 1, and the age at diagnosis was 54.1 ± 17.8 years. The median IgG4 concentration was 405.0 mg/dl. Anterior hypopituitarism, isolated central diabetes insipidus, and panhypopituitarism were observed in 14 (18.4%), 12 (15.8%), and 44 (57.9%) cases, respectively. The sequence of anterior hormone deficiency was as follows: gonadotropin (68.4%), ACTH (63.2%), TSH (59.2%), GH (48.7%), and prolactin (42.1%). The median number of involved organs was 1.5, and the lung (18.4%), retroperitoneum (17.1%), kidney (15.8%), submandibular glands (14.5%), and pancreas (13.2%) were the common involved organs. Elevated IgG4 concentration and normal IgG4 level were in 42 (76.4%) and 13 (23.6%) cases, respectively. Patients with elevated serum IgG4 concentration were older (60.9 ± 14.3 vs 45.6 ± 17.4, p=0.001) and male-prone (78.6% vs 40.4%, p=0.003) and had a susceptibility of multiple organ involvement (78.6% vs 35.0%, p=0.001) compared to those with normal serum IgG4 levels. Males were older at disease onset (61.5 ± 12.6 vs 42.9 ± 18.8, p < 0.001) and had a higher IgG4 concentration (425.0 vs 152.5, p=0.029) and a greater number of involved organs (2.0 vs 0.0, p=0.001), while isolated hypophysitis was more prominent in female (63.3% vs 26.1%, p=0.001). Conclusion In this review, we found that there were different characteristics between different genders. Patients with elevated serum IgG4 level in terms of some clinical features were also different from those with normal serum IgG4 level. However, the data in this review were limited by bias and confounding. Further clinical studies with larger sample sizes are warranted.


Introduction
IgG4-related disease (IgG4-RD) is a rare, newly recognized, multiorgan involved disease, which was characterized by infiltration of IgG4-positive plasma cells into the organs and elevated serum IgG4 level. e incidence of IgG4-RD was estimated to be 0.28-1.08/100,000 patients in Japan [1]. It was first created by Hamano et al. in 2001 when describing one case of sclerosing pancreatitis [2]. To our knowledge, pancreas, retroperitoneum, and salivary glands were the most commonly involved organs in IgG4-RD. Pituitary, which was a rare involved organ, was initially reported in 2004 [3]. It accounted for merely 1.5% of systemic cases of IgG4-RD [4]. Wong et al. reported the first case with both clinical manifestation and histopathologic evidence in 2007 [5]. In order to avoid the complication of transsphenoidal surgery, new diagnostic criteria were proposed by Leporati et al. in 2011 [6]. Since then, cases continued to be reported. In this review, we summarized case reports of IgG4-RH reported in the English articles and provided a detailed analysis of the clinical features of the 76 cases, aiming to get a better understanding of this recently recognized and rare entity.

Literature Search and Data Extraction.
We performed a PubMed database of the US National Library of Medicine search of articles using the search terms "hypophysitis [AND] IgG4" published up to March, 2019. A manual search of the literature was also performed. Articles were excluded if (1) cases were reported in languages other than English, (2) cases without individual data, and (3) cases not conformed with Leporati's diagnostic criteria (Table 1). Two investigators independently searched articles according to the inclusion and exclusion criteria. Any disagreements regarding the suitability of individual articles were resolved by discussion. Consequently, 54 articles of IgG4-RH (76 cases) were included. e PRISMA flow diagram is demonstrated in Figure 1. For each case, two investigators independently extracted the following parameters from eligible articles: sex, age, symptom, pituitary function, MRI, IgG4 serum concentration, involved organs, therapy, and the response to therapy. A list of all reviewed articles [3,5,6, is given in Table 2.

Hormone Deficiency.
Anterior hypopituitarism was defined as any anterior pituitary hormone deficiency. Secondary hypogonadism was diagnosed when FSH/LH concentration and estradiol or testosterone level was not elevated, respectively, for women or men. Secondary adrenal deficiency was defined as low morning serum cortisol and ACTH level. Secondary hypothyroidism was diagnosed when the serum FT4 level was below the normal range and serum TSH level was inappropriately low or normal. GH deficiency was defined as low age-adjusted IGF1 level. e diagnosis of diabetes insipidus was based on the clinical findings of polyuria and polydipsia, low levels of antidiuretic hormone (ADH), low urine osmolality in a water deprivation test, and an increase in urinary osmolality or a decrease in urine volumes in response to a desmopressin trial.

Statistical
Analysis. All parameters were described in the standard summary statistics, including mean, standard deviation (SD), median, minimum, maximum, and composition ratio. Statistical differences for continuous, normally distributed data were analyzed by Student's t-test; all continuous, non-normally distributed data were analyzed by nonparametric testing (Mann-Whitney U test). Categorical variables were assessed by chi-square or Fisher's exact test, as appropriate. All statistical tests were performed by SPSS version 20. A p value <0.05 was considered statistically significant.

Patient Demographics.
In a total of 76 patients, the mean ± SD age at diagnosis was 54.1 ± 17.8 years. 60.5% (n � 46/76) were men (the mean ± SD age at time of onset was 61.5 ± 12.6 years), 39.5% (n � 30/76) were women (the mean ± SD age at time of onset was 42.9 ± 18.8 years), and the ratio of men to women was 1.5 : 1. e minimum age was 14 years (ranged from 14 years to 87 years). Two patients were pregnant women. According to the data, the ages of onset were as follows: 7 (9.2%) patients were in their 40s, 19 (25.0%) in their 50s, 13 (17.1%) in their 60s, and 17 (22.4%) in their 70s ( Figure 2).

Diagnostic Method.
Forty-three (56.3%) cases were diagnosed by pituitary biopsy via a transcranial approach or transsphenoidal approach. Twenty-three (30.3%) cases were diagnosed based on biopsy-proven IgG4-related disease in other organs when enlarged pituitary gland and/or pituitary stalk were presented.

erapy.
ere was only one not receiving treatment; moreover, there was little change in the follow-up of 4 years. e use of glucocorticoid alone was the most common therapy for our patients (n � 55, 72.4%). Various kinds and doses of glucocorticoid were used to treat this disease. However, there were still some cases showing a relapse of the pituitary mass when the doses of glucocorticoid were tapered (Table 4). Surgery combined with glucocorticoid was taken in 13 (17.1%) patients. Four (5.3%) patients undergone glucocorticoid combined with immunosuppressive or anti-CD20 agents, including methotrexate, mycophenolate mofetil, azathioprine, cyclosporine A, and rituximab. In cases where clinical outcomes were reported, 93.8% (n � 60/64) of cases showed improvement following treatment, at least initially.
3.9. Clinical Differences according to Sex. Compared with women, men were older at the age of onset and had a higher IgG4 concentration and a greater number of involved organs (p < 0.001, p � 0.029, p � 0.001). Another important difference between men and women was that the isolated IgG4-RH was almost exclusively present in women (p � 0.001). However, they had no difference in efficiency of therapy (Table 5).

Clinical Differences according to Serum IgG4
Concentration. We analyzed the clinical features of patients according to elevated versus normal serum IgG4 concentration (Table 6). It showed patients with elevated serum IgG4 concentration were older, male-prone, and more inclined to multiple organ involvement compared to those with normal serum IgG4 levels (p < 0.01 for all comparisons).

Discussion
is review presented the largest number of IgG4-RH cases ever enumerated in English, with 76 cases meeting the inclusion criteria. We analyzed the characteristics of each Table 1: Diagnostic criteria for IgG4-related hypophysitis. 1. Histopathology mononuclear infiltration of the pituitary gland, rich in lymphocytes and plasma cells, with more than 10 IgG4-positive cells per high-power field 2. Sellar mass and/or thickened pituitary stalk on pituitary MRI 3. Biopsy-proven involvement in other organs (association with IgG4-positive lesions in other organs) 4. Elevated serum IgG4 levels (>140 mg/dl) 5. Rapidly reduction of the pituitary mass and symptom improvement with steroids When any of the following is fulfilled, criterion 1 only, criteria 2 + 3, or criteria 2 + 4 + 5. International Journal of Endocrinology    International Journal of Endocrinology parameter. In addition, we compared the differences of each parameter based on sex and serum IgG4 level aiming to understand the IgG4-RH precisely. Our analysis showed IgG4-RH presented at the 6th decade of life (mean age 54.1 ± 17.8 years) and was associated with a 1.5 : 1 male predominance, by contrast with the common lymphocytic hypophysitis, which was common in young females, particularly in association with late pregnancy or the postpartum period, and peaked in incidence in the 4th decade of life [59]. Iseda et al. reported that the age of onset was 66.3 ± 9.8 years, and the ratio of men to women was 9.3 : 1 in 2014 [23]. Shikuma et al. reported the age of onset was 64.2 ± 13.9 years, and the ratio of men to women was 2.4 : 1 in 2017 [60]. e possible reason was one previously been diagnosed as primary hypophysitis met the histologic criteria of isolated IgG4-RH. Moreover, we found patients with isolated pituitary lesion more tend to be female (p � 0.001). Importantly, 9 cases had been diagnosed IgG4-RD prior to IgG4-RH. We learned that it is important to follow-up these cases by considering them potential IgG4-RH cases.
With regard to features of MRI and pituitary function, the data characteristics were similar to Shikuma's review [60]. e sequence of anterior hormone deficiency to IgG4-RH was as follows: gonadotropin, ACTH, TSH, GH, and prolactin, which was different from lymphocytic hypophysitis that was characterized by ACTH > TSH > gonadotropin > prolactin > GH [61]. Elevated IgG4 concentration, as a common laboratory finding of IgG4-RH, was observed in 42 (76.4%) cases, and patients with elevated serum IgG4 concentration were older, male-prone, and more inclined to multiple organ involvement compared to those with the normal serum IgG4 level. is conclusion was consistent with the reports of Wallace and Carruthers [62,63]. However, the subjects in their articles were IgG4-RD patients, which was different from our review. Certainly, the serum IgG4 level could elevate in noninflammatory conditions, such as Wegener granulomatosis, multicentric Castleman's disease, and idiopathic plasmacytic lymphoadenopathy [64] and could be normal in up to 40% of patients, who were diagnosed by biopsyproven IgG4-related disease [65] and in postpartum IgG4-RH [48]. Moreover, low-dose steroid therapy may mask the evidence of systemic increases in the IgG4 level. Wallace et al. found there was no significant gender differences in IgG4-RD patients with regard to age at disease onset, disease severity, organ involvement, or serum IgG4 concentration [62]. Differently, we found males in IgG4-RH were older in age at disease onset, and male had a higher IgG4 concentration and a greater number of involved organs compared to female. e lung, retroperitoneum, kidney, submandibular glands, and pancreas were prevalent involved organs of IgG4-RH. We need to systematically explore these organs in IgG4-RH, especially patients with clinical symptoms of these common involved organs. Certainly, when patients had manifestations of other rare involved organs, we also need to explore them. IgG4-RH as a part of IgG4-RD, we also need to make a comprehensive exploration of the pituitary gland by assessing the level of pituitary hormone and doing a pituitary MRI examination when a patient was diagnosed with IgG4-RD.
Despite pituitary biopsy being an invasive examination and hard to operate, 43 (56.6%) cases were diagnosed by this method. We may reduce the necessity of pituitary biopsy by finding suspicious other organ damage with a careful examination. Currently, there was no clear standard for the treatment of IgG4-RH. Steroid therapy was the first-line treatment. e initial dose of prednisone was usually 0.6 mg/kg/day, and the dose could be regulated for rapid progression or higher body weight. It continued for 1-2 months. e dose was tapered to a maintenance dose (2.5-5 mg/day) over a period of 2-3 months, with a taper of   [68] detected the changes of the serum IgG4 level in 44 cases of IgG4-RD and proposed that the serum IgG4 level can be a predictor of the relapse. We need to confirm it with more data in future. While the exact etiology of IgG4-RH still remains unclear, its potential association with autoimmune conditions has been frequently reported. is review showed 5 cases (6.1%) were associated with autoimmune diseases, such as hashimoto's thyroiditis [25,54], systemic lupus erythematosus (SLE) [17], and Sjögren syndrome [14]. It was reported that IgG4-RH could be misdiagnosed as malignancy, granulomatous diseases, or tuberculosis [69]. Moreover, the risk of cancer increased threefold in patients with IgG4-RD in comparison to the general population [70], and we should attach importance to this disease.
is review had certain limitations. On the one hand, this review was grounded on case reports and small case series, but case reports not published in the English language were excluded. In addition, some case reports described the clinical data and outcomes very briefly. Consequently, some of the variables were missing. ese biases may have an influence on the conclusions of this review. On the other hand, reporting bias due to overreporting of severe cases may lead to an overestimation of the clinical disease. is bias could not be thoroughly ruled out. For rare diseases, such as IgG4-RH, we can establish an online disease registry system to facilitate systematic data collection furthermore.

Conclusions
We described clinical features of IgG4-RH. In addition, we revealed that there were different characteristics between different serum IgG4 levels and sexes. However, further clinical studies with larger sample sizes are warranted due to bias and confounding in this review.

Conflicts of Interest
e authors declare that they have no conflicts of interest in this paper.