Key Points

QuestionÌý What are the clinical presentations, investigation findings, and treatment outcomes of immunoglobulin G4–related pachymeningitis (IgG4-RP)?

FindingsÌý In this systematic review, the clinical presentation of IgG4-RP was mostly headache and cranial nerve dysfunction; investigation showed predominant involvement of the cavernous sinus and middle fossa and limited sensitivity of serum IgG4 levels, cerebrospinal fluid analysis, and classical pathology features. With treatment with glucocorticoids and immunosuppressants, 37% of patients achieved complete recovery, with relapse occurring in 40% of cases.

MeaningÌý IgG4-related pachymeningitis often presents without typical pathological features or elevated IgG4 levels, challenging its diagnosis; more accurate biomarkers and tailored treatment approaches are required.

ImportanceÌý Immunoglobulin G4 (IgG4)–related disease is an increasingly recognized fibroinflammatory condition that can involve multiple organs, including the pachymeninges. The understanding of IgG4-related pachymeningitis (IgG4-RP) remains limited because of its rarity and the predominance of knowledge derived from case reports and case series.

ObjectiveÌý To systematically review and synthesize the clinical presentation, investigation findings, and prognosis of IgG4-RP to better understand its diagnosis and management.

Evidence ReviewÌý A comprehensive systematic review was conducted following guidelines from the Preferred Reporting Items for Systematic Reviews and Meta-analyses. PubMed/MEDLINE, Embase, and Scopus were searched from their inception until May 30, 2023, using terms related to IgG4-related disease and pachymeningitis without language or publication restrictions. Case reports and series that met the 2020 Revised Comprehensive Diagnostic Criteria or the 2019 American College of Rheumatology/European League Against Rheumatism classification criteria were included. Data on clinical presentations, investigation findings, and treatment outcomes were extracted and summarized.

FindingsÌý A total of 148 case reports contributed data from 208 patients. Their median (IQR) age was 52 (39-62) years; 132 patients were male (63.5%) and 76 female (36.5%). Headache and cranial nerve dysfunctions were the most common neurological manifestations. Systemic involvement was identified in nearly half of the patients. Diagnostic imaging often showed preferential involvement of cavernous sinus and middle fossa. Laboratory results highlighted elevated serum IgG4 levels in 97 of 147 patients (65%) of patients and cerebrospinal fluid pleocytosis in 43 of 82 patients (52%). Storiform fibrosis or obliterating phlebitis were uncommon pathological findings. Mortality was below 1% (1/134; 0.7%), but only a third of patients presented complete clinical improvement, and the recurrence rate was 60 patients (40%) in a median (IQR) follow-up time of 9 (1-20) months. Glucocorticoids were the most commonly prescribed treatment, in 143 of 169 patients (85%); rituximab was prescribed as maintenance therapy in 53 of 169 patients (31%).

Conclusions and RelevanceÌý IgG4-RP commonly presents with headaches and cranial nerve dysfunction, posing diagnostic challenges due to the significant absence of systemic manifestations, low IgG4 serum levels, and atypical pathological findings. Current treatment outcomes are limited by incomplete recovery and frequent relapses underscoring the necessity for new treatment strategies.

Immunoglobulin G4–related disease (IgG4-RD) is a multisystemic fibroinflammatory disorder characterized by lymphoplasmacytic proliferation and infiltration.¹ The 2019 American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) classification criteria use a combination of clinical, radiological, serologic, and histopathology features to identify patients with IgG4-RD and to exclude mimics.²

The recognition of IgG4-RD emerged in 2001 when elevated serum IgG4 concentrations were first observed in patients with autoimmune pancreatitis. Over the following decade, the characteristic pathology was identified in various conditions, including sialadenitis and dacryoadenitis (Mikulicz syndrome), sclerosing cholangitis, orbital pseudotumor, tubulointerstitial nephritis, interstitial lung disease, aortitis, retroperitoneal fibrosis, and chronic thyroiditis (Riedel thyroiditis).¹

Hypertrophic pachymeningitis is a classical presentation of IgG4-RD,³ being the only neurological manifestation included in the entry criteria of the ACR/EULAR 2019 classification criteria.² IgG4-RD accounts for a significant proportion of cases of hypertrophic pachymeningitis once considered idiopathic.⁴ IgG4-RD pachymeningitis can cause headache, cranial nerve dysfunction, seizures, focal deficits, and cognitive decline that can be treated with high-dose glucocorticoids and immunosuppression.⁴

Since the first description of IgG4-RD pachymeningitis in 2011,⁵ there has been an increasing number of publications about this manifestation of the disease. However, our current knowledge is mainly based on case reports and case series; hence, a systematic review to summarize the clinical, radiological, and laboratory presentation of the disease might deepen our understanding of the presentation of this rare condition. Specific questions addressed in this study are as follows: (1) What are the most common neurological and systemic signs and symptoms in IgG4-RD pachymeningitis? (2) What are the most common investigation findings (laboratory, radiological, and pathological) in IgG4-RD pachymeningitis? (3) What is the typical prognosis of IgG4-RD pachymeningitis?

This is a systematic review of the clinical presentation, investigation findings, and treatment outcomes of IgG4-related pachymeningitis. This study is compliant with the Preferred Reporting Items for Systematic Reviews and Meta-analyses ()⁶ reporting guideline and was registered on PROSPERO ().

PubMed/MEDLINE, Embase, and Scopus databases were searched from inception to May 30, 2023, for case reports and case series of IgG4-related pachymeningitis. No language or publication-date filters were used. Search terms were synonyms of IgG4-related disease and pachymeningitis (eMethods in the Supplement).

Inclusion criteria were (1) case report or case series and (2) diagnosis of IgG4-related pachymeningitis according to the 2020 Revised Comprehensive Diagnostic (RCD) criteria for IgG4-Related Disease⁷ or the 2019 American College of Rheumatology/European League Against Rheumatism Classification Criteria for IgG4-RD.² Pachymeningitis was defined as the presence of thickening and contrast enhancement of the cranial or spinal dura mater as identified by magnetic resonance imaging (MRI). We excluded cases that did not meet this definition, including those with isolated leptomeningitis, orbital pseudotumor, or isolated hypophysitis. However, cases featuring dural involvement with extension to adjacent structures (eg, parenchyma and orbit) were included.

Given that many articles were published before the establishment of diagnostic criteria, we carefully reviewed the clinical and investigational information presented in these reports to assess whether the cases could be diagnosed as IgG4-related disease according to the currently available criteria. Each case was then classified as possible, probable, or definite IgG4-related disease based on the 2020 RCD criteria using the clinical, radiological, and laboratory data provided in the reports. Although some cases reported IgG4-RD pathology in the presence of antineutrophil cytoplasmic antibody (ANCA) positivity, we chose to exclude ANCA-positive cases because ANCA positivity is an exclusion criterion for IgG4-RD according to the ACR/EULAR 2019 diagnostic criteria.² The exclusion criteria defined by ACR/EULAR consensus were also followed.

Articles reporting a mixed population of patients affected by IgG4-RD pachymeningitis and other causes of pachymeningitis were included only if individual patient data from IgG4-RD cases were clear. Reference lists of the selected articles were screened. Search and screening were performed by 2 independent authors; conflicts on inclusion of data were resolved by consensus with a third author. We opted to not perform any specific risk-of-bias assessment as all cases were case reports or series and the risk of bias was overall high.

Duplicate reporting of patients is a significant concern in systematic reviews of case reports, especially in the context of rare diseases, where a relatively small number of investigators publish on the topic. To address this issue, duplicate articles were initially screened and excluded using a semiautomated system via the Rayyan online software.⁸ The remaining articles were subsequently organized by the country of publication and manually reviewed by the authors to identify potential duplicate reporting of patients from the same research groups.

We extracted patient-level data from included articles. Demographic data included sex, age, country of publication, and comorbidities. Neurological signs and symptoms were the presence of headache, visual loss, diplopia and/or strabismus, facial numbness, facial palsy, vertigo and/or hearing loss, bulbar syndrome, seizures, cognitive decline, ataxia, and focal deficits. We evaluated the following systemic involvement: pancreas, salivary glands, biliary tree, orbit, kidney, pulmonary, lymph nodes, aorta, retroperitoneal, thyroid, and other.

Investigation findings were grouped into radiological, laboratory, and pathological. Radiological findings included the topographic imaging involvement in brain (high-convexity, falx, sellar region, anterior, middle or posterior fossa, clivus, parietal, occipital, frontal, temporal or infratemporal region, cavernous sinus, orbital apex, Meckel cave, petrous apex, cerebellum, cerebellopontine angle, tentorium cerebelli, foramen magnum, transverse or sagittal sinus, or diffuse) or spinal (cervical, thoracic, or lumbar) MRI. Laboratory findings were serum IgG4 levels (mg/dL) and cerebrospinal fluid (CSF) analysis (white blood cell count, protein level, presence of oligoclonal bands, and IgG4 levels). Pathological findings included the site of biopsy, presence of storiform fibrosis, lymphoplasmacytic infiltrate, obliterative phlebitis, and IgG4/IgG ratio greater than 40%.

The prognosis of IgG4-RD pachymeningitis was evaluated by classifying patients into categories of complete response, partial response, no response, or worsening at the last follow-up reported in the case report. Both clinical and radiological outcomes were evaluated. The use of glucocorticoids and immunosuppression, which can influence prognosis, was also recorded. We defined refractory disease as the presence of relapse or no significant improvement after treatment.

We summarized quantitative data in median and interquartile range and qualitative data in percentages and total counts. Summary tables were presented. Statistical analysis was performed in Excel (Microsoft Office 365).

We conducted further analyses to identify predictors of outcomes in IgG4-RD pachymeningitis. First, we compared outcomes based on geography (Asia vs the United States/Europe). Second, we evaluated individuals based on the phenotype of dural involvement (spinal vs cranial). Finally, we assessed patients according to the type of the first immunosuppressant treatment they received (rituximab vs other immunosuppressants). Statistical significance was defined as P value less than .05. For those comparisons, we performed χ² or Mann-Whitney test as appropriate in R Studio version 2022.07.2 (Posit, PBC).

Database searching identified 1091 articles, of which 529 were duplicates. After title and abstract reading, we excluded 315 articles. Twenty-four articles were not retrieved. After full-text reading, we excluded 95 articles (18 because of incomplete data or insufficient information to diagnose IgG4-RD, 17 because of the absence of pachymeningitis, 18 presented ANCA-positive cases, 8 were not case reports or series, 4 were reports of repeated cases, 8 received other final diagnosis, and 22 were case series without individual information). At last, 128 articles were initially included according to database searching. We additionally reviewed 22 articles from references, of which 2 were excluded (1 was not pachymeningitis and 1 presented insufficient data), leaving 20 additional articles. The final review included 148 articles (eFigure and eAppendix in the Supplement).

We included 208 participants, of whom 132 (63.5%) were male and 76 (36.5%) female; the median (IQR) age was 52 (39-62) years. Most case reports described patients in the United States (44; 21.1%), Japan (36; 17.3%), India (24; 11.5%), and China (23; 11%).

Table 1 summarizes the neurological presentation of IgG4-related pachymeningitis. The most common presentations were headache and cranial nerve dysfunction. Neuro-ophthalmological syndromes (visual loss, diplopia, and/or strabismus) were the most common type of cranial nerve dysfunction. Seizures were reported in 21 cases (11%). The typical presentation was subacute, with a median (IQR) evolution of 6 (1-22) months until admission to health care services.

Systemic involvement of IgG4-related hypertrophic pachymeningitis is reported in eTable 1 in the Supplement. Other organs and systems were involved in 91 patients (49%), which for the most part was diagnosed in concomitance with pachymeningeal disease. A small portion of patients with systemic disease presented neurological signs earlier than other organs. The 3 most common systemic organ involvements were nasal or paranasal spaces (21 patients; 23%), lung (16 patients; 17%), and lymph nodes (14 patients; 15%). The clinical presentation did not differ significantly according to geographical location, except for a higher frequency of kidney involvement in Asian publications (7/47 patients [15%] vs 0 in Europe/North America) (eTables 2 and 3 in the Supplement).

Only 153 patients were included in the analysis of topographic dural involvement because the remaining articles did not explicitly mention topographic involvement and lacked available images for review (Table 2). Topographic cranial imaging involvement of pachymeningitis could be localized in skull base, cerebral convexities, and/or dural folds. Skull base involvement was more common in middle fossa, particularly cavernous sinus, orbital apex, and infratemporal fossa. Involvement of posterior fossa occurred predominantly in the clivus and retroclival space. Involvement of anterior fossa was uncommon. The frequency of cerebral convexities involvement presented an anteroposterior gradient. Involvement of dural folds occurred in only 24 cases (16%). Parenchymal involvement was observed in 19 cases (12%), occurring secondary to infiltration or mass effect (edema) adjacent to meningeal disease. More than 1 topographic (multifocal) involvement was present in 102 cases (66%). Figure 1 represents the spectrum of MRI findings in IgG4-related pachymeningitis.

Spinal involvement of pachymeningitis was predominantly cervicothoracic. Dural contrast enhancement extended longitudinally across more than 3 contiguous vertebral bodies in 18 spinal pachymeningitis cases (43%). Spinal pachymeningitis was more likely to be an isolated manifestation of IgG4-RD compared with cranial pachymeningitis (35/45 [78%] vs 82/146 [56%]; P = .01) (eTable 4 in the Supplement).

Table 3 summarizes laboratory and pathology findings of IgG4-related pachymeningitis. The majority of patients presented CSF abnormalities, including pleocytosis in 43 of 82 patients (52%) and elevated protein levels in 41 of 71 patients (58%). Serum levels of IgG4 were elevated in 97 of 147 patients (65%); for the majority of those (50/93; 54%), it was mildly elevated (1-2×). Elevated CSF IgG4 levels were observed in all cases where this information was available. In our research, we found 169 patients with documented pathological findings consistent with IgG4-related disease; 137 (81%) were meningeal biopsies. Specific pathological findings were explicitly reported in 157 articles, and of these, lymphoplasmacytic infiltration and elevated IgG4/IgG ratio were present in almost every biopsy. However, the storiform pattern of fibrosis and obliterating phlebitis were uncommon.

Based on these findings, we classified patients according to 2020 RCD diagnosis criteria: 46 were diagnosed as definite, 98 as probable, and 59 were possible IgG4-related hypertrophic pachymeningitis. Five patients were not classified between possible or probable disease because of incomplete information.

In a median (IQR) follow-up of 9 months (1-20), only 1 patient died (1/134, 0.7%). A total of 151 studies reported information about disease evolution, of which 91 (60%) were monophasic and 60 (40%) presented recurrences. There were reports of 21 recurrences of pachymeningeal disease (median [IQR] time until recurrence, 12 [7-24] months) and 6 recurrences of other organ disease (median [IQR] time until recurrence, 36 [22-48] months). Twenty-four patients had 1 recurrence, 9 patients had 2 recurrences, 4 patients had 3 recurrences, and 1 patient presented 4 recurrences. Nine patients had recurrence after immunosuppressant withdrawal, 8 during glucocorticoid use, 6 during immunosuppressant use, and 31 before the introduction of immunosuppressants or glucocorticoids. Complete clinical or radiological recovery occurred only in a third of cases (Figure 2).

Glucocorticoids were the most commonly prescribed treatment, in 143 of 169 patients (85%). Initial treatment with intravenous cyclophosphamide was infrequent, only in 7 patients. Rituximab was prescribed as maintenance therapy in 53 of 169 patients (31%). The most commonly prescribed oral immunosuppressants were azathioprine (21/169, 12%) and methotrexate (20/169, 12%), followed by mycophenolate mofetil (7/169, 4%).

We compared outcomes based on the geography of publication (eTables 2 and 3 in the Supplement). Case reports from Asia had significantly lower recurrence rates compared with those from Europe and the United States (23/71 [32%] vs 34/68 [50%]; P = .04). However, median (IQR) follow-up time was also significantly shorter for Asian reports (7 [3-18] months vs 16 [6-26] months) (eTable 3 in the Supplement).

We examined outcomes based on the phenotype of dural involvement (eTable 4 in the Supplement). Clinical and radiological outcomes were similar between patients with spinal and cranial involvement in IgG4-RD pachymeningitis. The frequency of each type of immunosuppressant used was also similar between the 2 groups.

We compared the frequency of refractory disease in patients treated with rituximab vs those treated with other immunosuppressants (eTable 5 in the Supplement). Rituximab was associated with a lower frequency of refractory disease (3/31; 9.7%) compared with other immunosuppressants (11/36; 31%, P = .03) in patients with IgG4-RD pachymeningitis who had neurological or systemic biopsy findings suggestive of IgG4-RD (eTable 5 in the Supplement).

This systematic review of 208 published cases synthesizes the available evidence on the clinical presentation, investigation findings, and treatment outcomes of IgG4-related disease (IgG4-RD) pachymeningitis. Our findings highlight the key clinical features of the disease and guide neurologists in the differential diagnosis of pachymeningitis.

The key neurological presentation of IgG4-RD pachymeningitis is a subacute headache with cranial nerve dysfunction, most often manifesting as neuro-ophthalmological syndromes. Neurologists should include IgG4-RD pachymeningitis in the differential diagnosis of painful ophthalmoplegia.⁹ Unlike the original description of Tolosa-Hunt syndrome,¹⁰ IgG4-RD pachymeningitis may present with neurological features beyond the cavernous sinus, such as seizures and evidence of systemic disease.

The first step in diagnosing a patient with subacute headache and cranial nerve dysfunction is neuroimaging. Brain MRI shows thickening and contrast enhancement of the dura mater, often with multifocal involvement (66%). While cavernous sinus involvement is common in IgG4-RD pachymeningitis, this condition differs from classical Tolosa-Hunt syndrome by its extension to other structures, such as the infratemporal fossa and clivus.

Neurosarcoidosis is a differential diagnosis of IgG4-RD, as it is another systemic disease that can present with pachymeningitis.¹¹ Key clinical and radiological features suggesting neurosarcoidosis over IgG4-RD pachymeningitis include a higher frequency of seizures (27% vs 11%), more frequent involvement of dural folds such as the falx cerebri and tentorium (45%-62% vs 6%-10%), and a distinct distribution of systemic disease.¹¹ In sarcoidosis, hilar lymph node, hepatic, splenic, skin, and cardiac involvement is more common, while in IgG4-RD, biliopancreatic, kidney, aortic, and retroperitoneal involvement is more frequent.¹¹ Meningeal biopsy in neurosarcoidosis typically shows compact, noncaseating epithelioid cell granulomas, an exclusion criterion for IgG4-RD.^2,11

ANCA-related vasculitis is another systemic disease that can present with pachymeningitis. Key clinical features that differentiate ANCA-related pachymeningitis from IgG4-RD pachymeningitis, based on our findings and previous studies, include a higher female predominance (57% vs 34%), a greater frequency of otological symptoms (22% vs 13%), and the presence of fever (46% vs 0%),¹² which is an exclusion criterion for IgG4-RD according to the 2019 ACR/EULAR criteria.² Meningeal biopsy in ANCA-related vasculitis shows focal necrosis and arteritis, with or without granulomas, which serves as an exclusion criterion for IgG4-RD pachymeningitis.^2,12

Serum IgG4 levels should be interpreted with caution in patients with pachymeningitis. Only half of the reported cases of IgG4-RD pachymeningitis had elevated serum IgG4 levels, and these were typically only mildly elevated. In contrast, all cases with reported CSF IgG4 levels showed elevated values, highlighting the need for further research on the diagnostic accuracy of CSF IgG4 levels. CSF analysis is otherwise nonspecific, with pleocytosis observed in only 52% of cases. However, CSF testing for neoplastic and infectious causes of pachymeningeal enhancement remains essential to exclude other differential diagnoses of subacute and chronic meningitis.¹³

Despite the high frequency of lymphoplasmacytic infiltration and an elevated IgG4/IgG ratio in meningeal biopsies, classic features like storiform fibrosis and obliterating phlebitis were uncommon. As the current ACR/EULAR 2019 classification criteria² rely on a scoring system that values typical pathological findings and IgG4 serum levels, this scoring system may have lower sensitivity in patients with IgG4-RD pachymeningitis, underscoring the crucial role of immunohistochemistry in establishing the diagnosis.

Treatment outcomes for IgG4-RD pachymeningitis were characterized by partial response and frequent relapses, regardless of whether the phenotype was spinal or cranial. Future prospective studies should determine whether the lower recurrence rates observed in Asian reports compared with those from Europe and the United States are due to ethnic differences in IgG4-RD or simply a result of shorter follow-up times in those studies.

Glucocorticoids, known for inducing remission,^14,15 were the most commonly prescribed treatment, with 85% of patients receiving them. However, only a third achieved complete recovery, and 40% experienced relapses within a year, primarily during steroid-free periods. Cyclophosphamide, despite its potential efficacy,³ was not frequently prescribed, appearing in only 7 cases, highlighting the need for more consistent use of effective treatments. Azathioprine and methotrexate were the most frequently used immunosuppressants, contrary to the higher evidence for mycophenolate and leflunomide from randomized clinical trials.^16,17

Rituximab is recognized for its effectiveness in IgG4-RD.^18,19 We found that the use of rituximab as a first-line immunosuppressant was associated with a lower frequency of refractory disease compared with other immunosuppressants, while acknowledging the risk of selection bias in observational studies. However, rituximab was used in only 31% of cases, which may have biased the results of our study toward less effective outcomes.

This study has several limitations. The short median follow-up period of 9 months likely underestimates the true relapse frequency. Additionally, the geographic concentration of cases in Japan and the United States may not represent the global distribution of IgG4-related pachymeningitis. Moreover, publication bias may affect the percentage of positive results in investigation findings. For example, cases with a lower IgG4/IgG ratio in meningeal biopsy might face more difficulty for publication and could be underrepresented in this study. Also, it is not possible to completely eliminate the risk of duplicate reporting, although repeated cases were manually reviewed. Despite these limitations, this review offers a substantial patient cohort for a rare disease, contributing valuable insights into the clinical presentation, radiological presentation, and treatment outcomes of IgG4-related pachymeningitis.

IgG4-RD pachymeningitis most commonly presents as headaches with cranial nerve dysfunction, particularly affecting middle-aged males. The diagnosis of this condition is challenging because of the absence of systemic manifestations in half of the cases, normal IgG4 levels in 45% of patients, and a lower frequency of typical pathological features in the meninges. Although mortality is low, clinical recovery is often partial, and relapses are frequent. These findings highlight the need for improved diagnostic criteria and more effective initial and maintenance treatment strategies to enhance patient outcomes.

Accepted for Publication: September 13, 2024.

Published Online: November 18, 2024. doi:10.1001/jamaneurol.2024.3947

Corresponding Author: Guilherme Diogo Silva, MD, Universidade de Sao Paulo, Doutor Enéas de Carvalho Aguiar, 255, São Paulo 05403-000, Brazil (guilherme.diogo@hc.fm.usp.br).

Author Contributions: Dr Silva had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Terrim, Filho, Lucato, Adoni, Silva.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Terrim, Filho, Lucato, Giardini, Silva.

Critical review of the manuscript for important intellectual content: All authors.

Statistical analysis: Terrim, Filho, Silva.

Administrative, technical, or material support: Lucato, Giardini, Adoni.

Supervision: Filho, Lucato, Giardini, Adoni, Silva.

Conflict of Interest Disclosures: Dr Filho reported personal fees from Roche and Merck outside the submitted work. Dr Silva reported travel grants for a medical congress from EMS, Biogen, and Roche outside the submitted work. No other disclosures were reported.