Article Text
PDF
Abstract
Background Despite many studies suggesting an association between serum IgG4 and IgG4-related diseases (IgG4-RD), the evidence of the utility of serum IgG4 titres in differentiating between IgG4-RD and non-IgG4-RD remains uncertain.
Methods The primary analysis was based on published studies. Data were pooled by means of a random-effect model, and sensitivity, specificity, positive likelihood ratios (LR+), negative likelihood ratios (LR–), diagnostic ORs (DOR) and summary receiver operating characteristic curve (SROC) were calculated. Subgroup analyses were performed based on the racial/ethnic distribution of these studies.
Results A total of 27 studies with 1691 (8.6%) IgG4-RD cases and 17 944 non-IgG4-RD subjects were included. Moreover, 1462 (86.5%) of the 1691 IgG4-RD patients had elevated serum IgG4 levels whereas 10.5% (1,882 of 17,944) of the non-IgG4-RD subjects had elevated serum IgG4 levels. The pooled sensitivity of serum IgG4 was 86% (85%–88%), specificity was 90% (89%–90%), LR+ was 9.19 (7.16–11.78), LR– was 0.17 (0.12–0.24), and the DOR was 60.8 (40.9–90.4), respectively. The area under the SROC curve for the differential diagnosis between IgG4-RD and non-IgG4-RD was 0.95 (0.94–0.97). Ethnic subgroup analyses revealed different findings with respect to DOR for Asian (103.8; 95% CI 63.3 to 170.2), and Caucasian (25.7; 95% CI 17.6 to 37.5) populations.
Conclusions Overall, elevated serum IgG4 levels were associated with IgG4-RD. The results revealed a moderate-to-high sensitivity (86%, 85%–88%) and high specificity (90%, 89%–90%). Subgroup analyses in serum IgG4 diagnostic performance revealed differences among Asian and Caucasian populations.
- Autoimmune Diseases
- Autoimmunity
- Immune Complex Diseases
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Many studies suggest an association between serum IgG4 titres and IgG4-related diseases (IgG4-RD).
The evidence of utility of serum IgG4 titres in differentiating between IgG4-RD and non-IgG4-RD remains uncertain.
WHAT THIS STUDY ADDS
Elevated serum IgG4 levels were associated with IgG4-RD. The results revealed a moderate-to-high sensitivity (86%, 85%–88%), and high specificity (90%, 89%–90%).
Subgroup analyses in serum IgG4 diagnostic performance revealed differences among Asian and Caucasian populations.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Our study showed that an elevated serum IgG4 level is still one of the most useful serological markers for the diagnosis of IgG4-RD.
The diagnostic ORs suggested that the diagnostic accuracy of serum IgG4 was higher in the Asian population than in the Caucasian population.
Introduction
IgG4-related disease (IgG4-RD) is a systemic disorder that involves multiple tissues and organs, including the salivary glands, pancreas, bile ducts, pituitary gland, eyes, thyroid, lungs, aorta, kidney, retroperitoneum, lymph nodes and skin.1–6 The diagnosis of IgG4-RD depends primarily on clinical manifestations, imaging and serum IgG4 levels. A definite diagnosis of IgG4-RD relies on the histopathological characteristics of biopsy specimens, including a lymphoplasmacytic infiltrate enriched with IgG4+ plasma cells, obliterative phlebitis and storiform fibrosis.1–11 Although IgG4-RD is generally considered a rare condition, the number of patients diagnosed with IgG4-RD is increasing.3
IgG4-RD is characterised by elevated serum immunoglobulin G4 (IgG4) levels, multiorgan involvement and a dramatic response to corticosteroid therapy. Serological markers such as serum IgG4 play an essential role in the diagnosis of IgG4-RD, and serum IgG4 levels have been included in various diagnostic criteria for IgG4-RD.4 5 Serum IgG4 levels have been reported to be a serological marker for IgG4-RD1–6 11; however, wide variability exists in the reported frequency of elevated serum IgG4 levels in IgG4-RD,7 8 11–37 and the reliability of the serum IgG4 concentration as a biomarker for diagnosis has been questioned.7 8 36 Moreover, initial studies have often been small and focused on single-organ systems (particularly the pancreas) in Asian or Caucasian patients.
Despite many studies suggesting an association between serum IgG4 and IgG4-RD, the evidence of the utility of serum IgG4 titres in differentiating between IgG4-RD and non-IgG4-RD remains uncertain. This investigation aimed to evaluate the clinical utility of serum IgG4 levels in IgG4-RD patients by performing a meta-analysis of published pieces of literature. Particularly, we aimed to assess the utility of serum IgG4 levels for diagnosing IgG4-RD and for differentiating IgG4-RD patients from non-IgG4-RD participants. Subgroup analysis by race/ethnicity was also performed to determine the robustness of the findings.
Materials and methods
A comprehensive search of clinical studies indexed in the electronic databases of PubMed, Ovid Medline and Embase was performed, and the reference lists of the retrieved articles were also reviewed to identify additional studies for inclusion. The timeframe for the electronic search was from January 1990 to December 2021 with no language restriction. Search terms included: “immunoglobulin G4,” “IgG4,” “IgG4-related disease,” “AIP,” “autoimmune pancreatitis,” “sensitivity,” “specificity,” “predictive value,” and “likelihood ratio.” Two review authors (CC and TC) conducted the electronic search independently and in duplicate. Any disagreement was resolved by referring to a third author (KY). Authors of studies were contacted when pertinent information was not available in the published version. When multiple articles for a single study had been published by the same authors or teams, we used the most relevant publication. The inclusion criteria for our meta-analysis study were (1) studies with sufficient information to generate the statistical elements as true positive, true negative (TN), false positive (FP), and false negative; and (2) the control sources were from non-IgG4-RD individuals. The exclusion criteria for our meta-analysis study were (1) extracted data were insufficient to establish the 2×2 table; and (2) case reports, letters, review articles, conference abstracts, studies of duplicate populations and molecular biochemistry basic studies.
The quality of each included study was evaluated according to the Quality Assessment of Studies of Diagnostic Accuracy-2 (QUADAS-2) (online supplemental table S1).38 The primary analyses were based on published studies. The primary outcomes were sIgG4 levels. Data were pooled by means of a random-effect model, and sensitivity, specificity, positive likelihood ratios (LR+), negative likelihood ratios (LR–), diagnostic ORs (DOR), and areas under summary receiver operating characteristic (SROC) curves were calculated. Heterogeneity across studies was assessed using the I2 statistic. An I2 statistic greater than 50% was considered substantial heterogeneity. If substantial heterogeneity existed, subgroup analyses would be conducted to explore the heterogeneity. All statistical analyses were performed in Meta-Disc V.1.4 (Universidad Complutense, Barcelona, Spain)39 and Review Manager V.5 (Oxford, UK: The Cochrane Collaboration).
Supplemental material
Results
Characteristics of the study populations
In the present study, we identified 103 articles from the online database, of which 64 irrelevant articles were excluded. The remaining 39 articles were further assessed for eligibility, and 27 articles were finally included in the meta-analysis consisting of a total of 19 635 patients.11–37 The quality of the included studies as assessed by the QUADAS-2 tool (online supplemental table S1). was generally high, with all studies meeting eight or more of the criteria. Figure 1 shows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram for studies retrieved through the electronic search and the selection processes for study inclusion. Specifically, the studies included 1691 IgG4-RD cases and 17 944 non-IgG4-RD subjects. The detailed characteristics of each study are presented in table 1. A total of 27 studies with 1691 (8.6%) IgG4-RD cases and 17 944 non-IgG4-RD subjects were included. Among the included studies, 17 were conducted in Asian populations.11–27 An additional 10 studies examined individuals from Caucasian populations (table 1).28–37 The pretest probability of IgG4-RD across all studies ranged from 1% to 76% (mean 24.9%, median 19.0%) and the sample sizes of IgG4-RD cases in each study ranged from 12 to 8165 (mean 727, median 187). Moreover, 1462 (86.5%) of the 1691 IgG4-RD patients who were examined in these studies had elevated serum IgG4 levels whereas 10.5% (1882 of 17 944) of the non-IgG4-RD subjects had elevated serum IgG4 levels (table 1).
Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram for studies retrieved through the electronic search, and the selection processes.
Characteristics of the studies included in the meta-analysis
Statistical pooling of outcomes and meta-analysis
The diagnostic indicators and frequencies of the elevated serum IgG4 in all included studies are shown in table 2. In the systematic literature review (total n=19 635), the overall mean sensitivity of serum IgG4 was 81% (ranging from 33%–99%), specificity was 90% (59%–98%), positive predictive value (PPV) was 66% (7%–94%), negative predictive value was 92% (38%–100%), LR+ was 13.65 (2.22–55.10), LR– was 0.20 (0.02–0.68), and the DOR was 172.4 (median 59.5, range 10.2–1083.0) (table 2).
Diagnostic indicators of the included studies for the diagnosis of IgG4-RD through serum IgG4 levels
Furthermore, all diagnostic indicators of this study were pooled from the included studies. Summaries of the pooled diagnostic indicators for the included studies are illustrated in figure 2. In summary, the pooled sensitivity of sIgG4 was 86% (85%–88%), specificity was 90% (89%–90%), LR+ was 9.19 (7.16–11.78), LR– was 0.17 (0.12–0.24) and the DOR was 60.8 (40.9–90.4), respectively. The area under the SROC for the differential diagnosis between IgG4-RD and non-IgG4-RD was 0.95 (0.94–0.97) (figure 2F). Overall, serum IgG4 was associated with IgG4-RD, the result revealed a moderate-to-high sensitivity (0.86, 0.85–0.88) and high specificity (0.90, 0.89–0.90) in diagnosing IgG4-RD.
Graphical representations of the pooled diagnostic indicators for the diagnosis of IgG4-related disease (IgG4-RD) through serum IgG4 levels.
Subgroup analysis and metaregression
In the meta-analysis, it was revealed that there was considerable heterogeneity (I2>50%) for all statistical measures (figure 2). IgG4-RD cases were found to be associated with increased serum IgG4 levels (DOR 60.8, 95% CI 40.9 to 90.4, p<0.001). Since a common effect size was shared among present studies and significant heterogeneity existed, a random-effect model was used, and subgroup analyses were performed. In the Asian population studies (17 studies, total n=16 327), the pooled sensitivity, specificity, LR+, LR–, DOR and area under the SROC curve were 0.89 (95% CI 0.88 to 0.91), 0.90 (95% CI 0.90 to 0.91), 11.0 (95% CI 9.1 to 13.5), 0.12 (95% CI 0.08 to 0.18), 103.8 (95% CI 63.3 to 170.2) and 0.97 (95% CI 0.96 to 0.98), respectively. In the Caucasian population studies (10 studies, total n=3308), the corresponding pooled sensitivity, specificity, LR+, LR–, DOR and area under the SROC curve were 0.78 (0.74 to 0.82), 0.85 (0.84 to 0.86), 6.35 (4.04 to 9.96), 0.28 (0.17 to 0.45), 25.7 (17.6 to 37.5), and 0.90 (95% CI 0.88 to 0.92), respectively. Ethnic subgroup analyses revealed different findings with respect to DOR for Asian (103.8; 95% CI 63.3 to 170.2), and Caucasian (25.7; 95% CI 17.6 to 37.5) populations. The year of publication subgroup analyses revealed similar trends with respect to DOR studies conducted before 2013 (71.7; 95% CI 39.7 to 129.5), and after 2013 (52.8; 95% CI 29.7 to 93.6). Moreover, metaregression was performed to explore the potential sources of heterogeneity. In metaregression analysis, the year of publication was no longer statistically significant, but race/ethnicity differences remained significant (table 3).
Results of metaregression analysis exploring source of heterogeneity
Discussion
IgG4-RD is a clinical disease entity characterised by elevated serum IgG4 levels and tumefaction of the involved organs, and tissue infiltration by IgG4-positive plasma cells. Awareness of IgG4-RD has increased worldwide since 2001.11 In view of its multifaceted presentation, IgG4-RD represents a great mimicker of many inflammatory, infectious and neoplastic conditions.1–3 Histopathological examination of the affected organs or tissues is still the gold standard for the diagnosis of IgG4-RD, whereas serum IgG4 level has been shown to be neither necessary nor sufficient for the diagnosis.1–11 Multiple non-IgG4-RD conditions are associated with elevated serum IgG4,9 leading to a low PPV for this test.7 8 24 26 28 34 36 37
Elevated IgG4 levels can be found in other diseases, especially asthma, allergies, malignancies, plasma cell dyscrasia, Castleman disease and other autoimmune diseases.1–3 9 Moreover, the inclusion of false-positive cases such as allergic rhinitis patients will compromise the PPV of the serum IgG4 test.7 8 24 26 28 34 36 37 Furthermore, predictive values are affected by the pretest probability of having a disease in the population tested. In a high pretest probability setting, it is more likely that persons who test positive truly have the disease than if the test is performed in a low prevalence population, as revealed in the present study.
Nonetheless, elevated serum IgG4 levels are pivotal to the diagnosis of IgG4-RD.11 In the present meta-analysis, the results showed that elevated serum IgG4 titres are still one of the most useful serological markers for the diagnosis of IgG4-RD, similarly to the previous meta-analysis.40–42 Notably, 10%–30% of patients with IgG4-RD may have normal serum IgG4 levels despite the histopathological criteria supporting an IgG4-RD diagnosis.1–10 30 43–48
In the present study, we found a strong association between serum IgG4 and IgG4-RD. In contrast to the few previous studies,7 8 here we found serum IgG4 to be the best single serum marker associated with IgG4-RD. The overall pooled sensitivity and specificity of serum IgG4 in this study in distinguishing IgG4-RD from non-IgG4-RD were 86% (95% CI 85% to 88%) and 90% (95% CI 89% to 90%), respectively. Thus, as a single serological marker for diagnosis of IgG4-RD, serum IgG4 has moderate-to-high sensitivity and high specificity. However, 10.5% (1882 of 17 944) of the non-IgG4-RD subjects had elevated serum IgG4 levels, in accordance with previously published studies.1–3 9 14 40–42 Elevated serum IgG4 levels can also be detected in a broad spectrum of the pancreatic, biliary tract, liver, allergic diseases, cancers, infections, as well as other autoimmune rheumatic diseases.9 Serum IgG4 levels vary among the different extents of organ involvements,1–3 so they can be an unreliable and unspecific single diagnostic marker. Therefore, serological markers such as serum IgG4 levels together with consistent radiographic findings might be helpful in the discrimination of IgG4-RD from patients without IgG4-RD.
In the present analysis, we investigated the diagnostic performance and clinical utility of serum IgG4 testing for the diagnosis of IgG4-RD and explored the cause of heterogeneity of the included studies. In the present study, serum IgG4 was found to be associated with IgG4-RD (DOR 60.8, 95% CI 40.9 to 90.4, p<0.001) with heterogeneity (I2 statistic 73.2%) across different ethnic populations, similarly to a previous systematic review (DOR 45.2, 95% CI 23.4 to 87.1).41 Moreover, subgroup analyses of studies on Asian and Caucasian populations revealed different findings; for Asian (DOR 103.8; 95% CI 63.3 to 170.2), and Caucasian (25.7; 95% CI 17.6 to 37.5) populations. The DOR suggested that the diagnostic accuracy of serum IgG4 was higher in the Asian population compared with the Caucasian one. The reason for this discrepancy is probably due to autoimmune pancreatitis (AIP) being classified into two subtypes by an international consensus meeting for AIP coordinated by the Autoimmune Pancreatitis International Study Group.44–48 AIP is a rare form of chronic pancreatitis, with as yet undetermined incidence and prevalence in the general population. Based on the nomenclature published in 201249 and pathological classification,44–48 AIP is divided into two types: type 1 AIP, recognised as the pancreatic manifestation of IgG4-RD, and type 2 AIP, a pancreas-specific disorder not associated with elevation of serum IgG4.44–48 50–52 Type 1 AIP is characterised by a dense lymphoplasmacytic infiltrate with storiform fibrosis and obliterative phlebitis, whereas type 2 is characterised by the presence of granulocytic epithelial lesions.44–48 Type 1 AIP may present as an isolated disorder or as part of an IgG4-RD syndrome with other organ involvement. In contrast to Western countries, where type 2 AIP is supposed to be more frequent (up to 50%), Asian studies with large patient populations demonstrated that nearly all patients (>90%) suffered from type 1 AIP.44–48 50–52 Differences in geographical distribution might be responsible for the different results. This heterogeneity makes it difficult to compare among different studies.
There are limitations of the present study that warrant discussion. First, it is possible that some cases of IgG4-RD and non-IgG4-RD might have been misclassified, especially in patients who did not have a classic IgG4-RD presentation. Patients with IgG4-RD in the selected 27 studies were diagnosed based on varied diagnostic criteria, including the comprehensive diagnostic criteria, or they fulfilled the specific organ diagnostic criteria, such as Revised Japanese diagnostic criteria, Asian diagnostic criteria, Mayo HISORt criteria, International Consensus Diagnostic Criteria, Clinical diagnostic criteria of IgG4-related sclerosing cholangitis, or consensus statements on the pathology (Boston consensus histopathological criteria). These different criteria may cause a non-homogeneous selection of patients included in different studies. Moreover, whether serum IgG4 levels were tested before treatment was not clearly described in all studies. Furthermore, when interpreting low serum IgG4 levels, one must rule out the prozone phenomenon in the immunoassay, a condition in which excess antigen competes for binding sites on the antibody, thereby causing immune complexes to resolubilise and reduce the signal detected.53 The strengths of the present study are that it derives from a comprehensive literature search and that it provides further assessment of the causes of heterogeneity detected in previous studies.
In conclusion, the diagnosis of IgG4-RD is complex and usually requires a combination of clinical analyses, imaging, histological and serological examination. No finding alone is specific to IgG4-RD. Several groups have reported a lack of sensitivity and specificity of the IgG4 serum level to establish the diagnosis of IgG4-RD or to distinguish it from non-IgG4-RD or non-AIP diseases.7 8 Because IgG4-RD is uncommon, IgG4 elevations in patients with a low pretest probability of having IgG4-RD are likely to represent FPs. Elevated serum IgG4 levels are characteristic of IgG4-RD; however, elevations in serum IgG4 titres are seen in around 10% of subjects without IgG4-RD and thus sIgG4 levels cannot be used alone to distinguish IgG4-RD from non-IgG4-RD. Therefore, IgG4-RD diagnosis requires careful interpretation of test results together with the patient’s clinical presentation, as well as the exclusion of a broad variety of differential diagnoses.5 Further correlation among clinical, radiologic, serological and histopathological data is required for diagnosis.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Correction notice This article has been updated since it was first published online. The equal contributorship statement ‘C-TC and T-MC contributed equally’ has been removed.
Contributors All authors contributed to each of the following: (1) the conception and design of the study, or acquisition of data, or analysis and interpretation of data; (2) drafting the article or revising it critically for important intellectual content; and (3) final approval of the version to be submitted. KHY is the corresponding author acting as guarantor.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.